THE DIGITAL

TRANSFORMATION
ROADMAP
 PRAISE FOR THE DIGITAL
TRANSFORMATION ROADMAP

“David L. Rogers’s work has both affected and validated our digital transformation efforts at
Acuity. He’s provocative and pragmatic, teaching us a different way to think about and stand
out from our competitors. His insight and advice continue to have lasting impacts on our
digital journey.”

—Melissa Winter, president, and Ben Salzmann, CEO, Acuity Insurance

“Essential reading for leaders seeking to thrive in a rapidly evolving digital landscape.
Embark on your digital transformation journey with Rogers as your guide and gain the tools
to ensure success amid constant technological shifts.”

—Sami Hassanyeh, chief digital officer, AARP

“Rogers provides a powerful blend of strategy and tactics to help large companies move
smarter and faster into the digital world. Featuring real-world lessons and a pile of hands-on
tools, this book is your essential roadmap to get any firm moving at startup speed.”

—Bob Dorf, coauthor of The Startup Owner’s Manual and lifelong entrepreneur

“Digital transformations all too often go badly wrong because those leading the change
have not taken the time to work through the organizational and human barriers that keep
things locked into old patterns. Armed with this terrific book, you can enter the
transformational space with your eyes wide open. It will be your guide to what the future
holds.”

—Rita McGrath, Columbia Business School, and author of Seeing Around Corners and
Discovery-Driven Growth

“After over a decade of digital transformations, if we’ve learned one thing, it is that the battle
to generate business value is won through breaking down organizational barriers and
aligning people, processes, and metrics. With The Digital Transformation Roadmap, Rogers
offers powerful insights to maximize the chances of transformation success.”
 —Didier Bonnet, professor of strategy and digital transformation at IMD Business School
and author of Leading Digital and Hacking Digital

“We’ve been undergoing digital transformation now for several decades, and it’s only
accelerating. The continuous change approach outlined in this book is the only way for
organizations to thrive in the long run. It’s ‘go digital’ or ‘go out of business.’ ”

—Thomas H. Davenport, President’s Distinguished Professor of Information Technology

and Management, Babson College
 Columbia University Press
Publishers Since 1893
New York Chichester, West Sussex
cup.columbia.edu

Copyright © 2023 David L. Rogers

All rights reserved
E-ISBN 978-0-231-55173-1

Library of Congress Cataloging-in-Publication Data

Names: Rogers, David L., 1970– author.
Title: The digital transformation roadmap : rebuild your organization for continuous change /
David L Rogers.
Description: New York : Columbia University Press, [2023] | Includes index.
Identifiers: LCCN 2023014410 | ISBN 9780231196581 (hardback) | ISBN 9780231551731
(ebook)
Subjects: LCSH: Technological innovations—Management. | Information Technology—
Management. | New products. | Strategic planning.
Classification: LCC HD45 .R6337 2023 | DDC 658.4/062—dc23/eng/20230418
LC record available at https://lccn.loc.gov/2023014410

A Columbia University Press E-book.

CUP would be pleased to hear about your reading experience with this e-book at cup-
ebook@columbia.edu.

Cover design: Noah Arlow

For Karen, my beloved partner in creative mucking around
 CONTENTS

Preface

1 The DX Roadmap

2 DX and the Challenge of Innovation

3 Step 1: Define a Shared Vision

4 Step 2: Pick the Problems That Matter Most

5 Step 3: Validate New Ventures

6 Step 4: Manage Growth at Scale

7 Step 5: Grow Tech, Talent, and Culture

Conclusion

More Tools for Your Business

Self-Assessment: Is Your Organization Ready for DX?

Cases and Examples by Industry

Visual Overview of the DX Playbook and DX Roadmap

Notes

Index

About the Author

 PREFACE

This is my fifth book and my first sequel. The prequel, The Digital
Transformation Playbook, which I published with Columbia University
Press in 2016, was the first book on digital transformation. It put the
topic on the map, becoming a global bestseller in more than a dozen
languages. It also defined the discipline by arguing that digital
transformation is not about technology; it is about strategy,
leadership, and new ways of thinking. That book has helped
thousands of business leaders rethink their business around new
strategies and embark on transformations for the digital era.
In many of these transformations, I had a front row seat as an
adviser in boardrooms and leader of workshops with executives at
many different companies, including Google, Microsoft, Citibank,
HSBC, Procter & Gamble, Merck, General Electric, and dozens more
equally ambitious enterprises. I have taught over 25,000 business
leaders through my classes at Columbia Business School’s
Executive Education programs in New York City, in Silicon Valley,
and online. Through this work and my one-to-one advising of CEOs
and chief digital officers (CDOs), I have delved into the inner
workings of digital transformation efforts in diverse industries and
around the world.
Today, every leading organization is pursing some kind of digital
transformation (DX). The question is no longer, “Should we pursue
DX?” but rather, “How do we ensure DX’s impact and business
value?”
Answering that question is not easy. Established businesses that
have embarked on DX may be ready to rethink their strategy, but
across all industries, these businesses are struggling with the
challenges of organizational change: How do we align digital efforts
across our silos? How do we balance future growth versus our
current core business? How do we overcome the inertia that plagues
large, complex organizations?
This book presents an approach based on a decade of evidence
of DX attempts, successes, and failures. It shows that, to lead an
established business today, you don’t just need to rethink your
business with strategy for the digital era. You need to rebuild your
organization for a world of constant digital change. As daunting as
that may sound, let me assure you it is possible—and it is the only
way to sustain and grow your business for the future.
In The Digital Transformation Roadmap, I provide a practical
roadmap for any leader attempting to transform their organization for
the digital era. Like all my books, this one includes practical tools
and frameworks to apply while driving change in your organization. I
road-tested these tools in my work with executives pursuing digital
change in a wide range of organizations. In the process, I have
refined each tool to answer the needs of firms of different sizes,
industries, geographies, and strategic challenges.
My goal is to help you, the reader, to maximize your DX success
in any organization you are leading or helping to grow, now and in
the future.

Acknowledgments

No book is possible without the help of many generous contributors.

I want to thank all the many business leaders whose work and
insights are reflected in the book. You were immensely generous in
sharing your experiences and lessons with me, whether in my
consulting engagements, classrooms, or interviews.
This book is indebted to the continued support of my longtime
agent, Jim Levine, and my publisher, Myles Thompson. My editor
Brian Smith helped guide the text through every stage, in and out of
pandemic disruptions. My dear and brilliant friends Bob Dorf and
Lucy Kueng each graciously read my entire first draft and offered
critical feedback. The final text owes much to the aid of my peerless
manuscript editor, Karen Vrotsos. Her incisive queries clarified my
thinking at every turn, trimming the excess and bringing focus,
buoyancy, and economy to the final text.
Throughout the book’s development, Columbia Business School
has sustained the development of my work and ideas and has been
an invaluable home base, where I’ve taught and conducted research
for over twenty years. I especially want to thank Mark Roberts, Pierre
Yared, Dil Sidhu, and Mike Malefakis for their leadership of our
Executive Education division and support of my teaching and
research. I also want to thank the thousands of executive students
I’ve had the pleasure to teach while writing this, in both my live and
online programs on digital transformation, strategy, and leadership.
The questions they’ve shared and the personal stories from their
careers contributed to the evidence I probed in developing this
book’s key insights. Thanks also to Ashwin Damera and Chaitanya
Kalipatnapu at Emeritus for an incredible journey in online education,
and to Clark Boyd for sharing the virtual stage in my online courses
for countless ask me anything’s (AMAs) with students.
Thanks also to Tom Neilssen and his team at BrightSight
Speakers for their advice and support, from the proposal stage to
ideas for the book’s cover. Heather Hinson contributed expert and
wide-ranging research with great cheer, and Mindy Bowman
designed the book’s illuminating graphics. Amy Kazor and Pat Curtis
were indispensable, as always, in keeping me focused on the book
amid countless competing projects.
Last, I thank my wife, Karen, and son, George. Their support,
companionship, and good humor kept me writing and smiling
throughout the long journey. Their love is the inspiration behind all
my work.

David Rogers
www.davidrogers.digital
 1
The DX Roadmap

In the early days of the digital revolution, as the World Wide Web
emerged as a mass platform for global communication, the New York
Times Company embarked on a bold new project to reimagine its
business for a new era. The project was what we would call today a
digital transformation (DX).
The digital transformation of the Times Company began with the
mandate of its chief executive: publisher Arthur Ochs Sulzberger Jr.
was the driving force and put his full authority behind it. He established
a separate division to champion the company’s digital efforts, a
subsidiary called the New York Times Electronic Media Company. He
hired Martin Nisenholtz, an expert in digital media and advertising, as
president of the division, which went on to hire new digital talent to
drive the project forward.
In the years that followed, the Times undertook a wide range of
projects aimed at showcasing new forms of digital journalism. Pieces
like the stunningly beautiful Pulitzer Prize–winning multimedia feature
“Snow Fall” promised exciting new possibilities for interactive media in
journalism.1 But such projects seemed to stand apart from the daily
output of the paper. The Times’s technology teams built digital versions
of the legacy product (the print newspaper), and they even digitized the
archive of articles going back to 1851. But editors lacked the data to
understand readers’ online behaviors in real time or to connect readers
to past articles on topics of interest. The Times showed a willingness to
try every new technology trend—from email and the Web to social
media, tablet editions, virtual reality, and chatbots. But it lacked clear
strategic priorities and the discipline to grow what worked and quickly
close pilots that failed to achieve results.
 Over time, serious problems began to emerge in the organization.
Creating a separate division for digital initiatives established a pattern:
a few people were tasked with driving all things digital, and the rest of
the organization stuck to their old ways of working. Meanwhile, the
legacy organizational structure—with its historic separation of the
business side from the journalism side—remained untouched. Despite
Sulzberger’s initiative, the Times’s leadership clearly prioritized the old
line of business (the print edition) over new digital ones. For example,
every new hire was invited to attend the daily Page One meeting to
observe the most senior editors choose which articles would go on the
front page of the physical paper.2 Meanwhile, editors refused to use
data to make decisions about matters of content—even when choosing
between two options for an article’s headline.
Equally serious problems appeared on the business side. From the
start, the vision of digital transformation at the Times was all about
“digitizing” the core business—literally, taking the same articles
produced every day for the print edition and using new technologies to
deliver them to readers. When he hired Nisenholtz, Sulzberger pledged
to deliver the New York Times by website, CD-ROM, or any other new
digital medium to come. “The internet? That’s fine with me. Hell, if
someone would be kind enough to invent the technology, I’ll be pleased
to beam it directly into your cortex.”3 This thinking—seeing technology
as just a new means of delivering the old product—persisted for years,
defining the Times’s future in terms of its past business. Meanwhile, the
Times failed to reckon with the fundamental shift in the economics of
advertising wrought by the internet. In the long term, no news media
business could survive if it clung to the old ad-based business model.
As the digital revolution continued, the Times’s transformation
looked more and more inadequate. Despite the Pulitzer and Peabody
awards for celebrated digital articles, some of the best digital talent the
Times had hired was moving to other publishers. New digital-native
publishers like Vox, BuzzFeed, and the Huffington Post—with
interactive formats, social media savvy, content optimized for search
engines, and feisty style—surged past the Times in attracting young
readers. For some stories, they even beat the Times by repackaging
content from its own archives. As one Huffington Post executive
confessed, “You guys got crushed . . . I’m not proud of this. But this is
your competition.”4
 Most seriously, the financial performance of the Times continued a
long slide (see figure 1.1). Revenue from print advertising dropped
precipitously, and digital advertising failed to deliver on its promise.
Total revenue dropped every year from 2006 to 2013, declining 52
percent in just seven years.5

Figure 1.1.
New York Times Company revenue, 2006 to 2012

In 2014, a bombshell dropped. Buzzfeed published a secret report,

commissioned by the Times’s top leadership and leaked from the inside
after a months-long investigation. The “Innovation Report” described
the New York Times Company as an organization at war with itself over
its digital future. Older employees clung fiercely to the status quo.
Departmental silos were organized around old traditions rather than the
changing needs of the business. Perfectionism prevented
experimentation with new digital models. Resources were poured into
one-off digital projects rather than digital tools and skills to change
everyday work. Senior leaders clung to an outdated focus on the old
print business, and digital talent decamped, seeing no future at the
paper. As the report explained, the constant focus on the company’s
legacy business became “a form of laziness because it is work that is
comfortable and familiar to us, that we know how to do. And it allows us
to avoid the truly hard work and bigger questions about our present and
our future: What shall we become. How must we change?”6
The story of the Times does not end there. But the questions and
crisis it faced are facing every legacy business attempting to transform
today. Like so many organizations that have struggled or lost ground in
the new era, the Times spent years of wasted effort because it lacked a
digital transformation roadmap for the future.

The Crisis of DX Failure

In the past several years, as the digital revolution has expanded from
media into every other industry, nearly every established business has
embarked on some kind of effort at digital transformation. But in sector
after sector, the large majority of these firms have failed to achieve the
results they hoped for. Multiple global studies, including surveys by
BCG and McKinsey, have found that 70 percent or more of digital
transformations fall short of their objectives or fail to achieve any
sustained benefit.7 The results of this institutional failure can be seen
everywhere. Famous companies and global brands (Kodak,
Blockbuster, etc.) have gone bankrupt in the face of digital challengers.
Market leaders rushed to exit businesses where they had once
dominated (Nokia phones). Entire categories of traditional companies
are in steep decline (movie theaters, department stores, news
publishers, etc.).
In other cases, established businesses have managed to hold onto
their existing business, but they watch helplessly as new digital
entrants swoop in to capture massive growth opportunities in their
sector. In the auto industry, legacy automakers watched as Tesla made
the first move into electric vehicles and quickly soared past them in
market value. In financial services, legacy banks focused mainly on
adapting their existing lines of business to the digital era, while fintechs
like Stripe, Block, Paypal, and Ant Financial defined entire new
categories of financial services and reaped phenomenal growth.
Faced with this glaring trend, business leaders have struggled in
their efforts at digital transformation. Famous examples include General
Electric (GE), which launched a hugely-ambitious DX effort in 2015
called GE Digital. It was championed by CEO Jeffrey Immelt, who
announced that GE would become a “top 10 software company” in five
years. Three years later, after disappointing results, GE Digital was
scaled back and spun out by the next CEO.8
I have spoken to executives in countless companies who tell me
that a chief digital officer (CDO) has been hired, but no one in the
business knows what the strategy for digital is supposed to be. Others
have pointed to complacency and unwillingness to risk capital on new
digital investments after years of success with their old business. One
large retailer I advised had hired senior-level talent from Google and
Amazon to turn their business into a data-driven organization but
confessed that their managers lacked the real-time data to experiment
and make decisions.
Failure to transform is fatal. Every year, digital technologies have a
deeper impact on business in every industry. The COVID-19 pandemic
accelerated many of these shifts, compressing several years of change
into weeks in areas like telemedicine, streaming media, online learning,
e-commerce, and remote work. These shifts are not slowing down.
They are continuing and accelerating. As shown by a global study of
senior executives I conducted with HCL, today’s leaders know that
digital transformation is no longer a question of “if” but of “how fast?”9
DX is on the agenda for every board in every industry. We cannot
afford to keep getting it wrong.

DX Defined

Before we go further, let’s clarify the definition of DX because the term,

now widely adopted, has become clouded in the confusion that
envelops all buzzwords. Having researched, practiced, and taught the
topic for many years, I define digital transformation as a very particular
activity:
Transforming
an established business
to thrive in a world of constant digital change.

Three ideas are worth noting. First, DX is about business, not

technology. Too often, DX efforts are defined in terms of the
technologies they intend to harness (artificial intelligence [AI],
blockchain, robotics, cloud computing, etc.). Of course, technology will
be part of the implementation of any digital strategy you develop. But
any DX effort should be framed around your business, your employees,
and your customers—not around a list of technologies to adopt.
Second, DX is about changing an existing organization, not creating
a start-up. A start-up’s mission is to search for a profitable business
model and then scale up a new organization to support its growth. But
an established business already has both a business model and an
organization—with employees, customers, products, distribution
channels, partners, and an established culture and way of working.
Thus, DX is fundamentally about changing an organization that is
already in motion. In physics, the tendency of a body in motion to keep
moving in the same direction and at the same speed is called inertia.
The more massive the body, the harder it is to change course. This is
true in business as well, where your biggest enemy of transformation is
inertia—the organization’s resistance to change.
Third, DX is a continuous process; it is not a project with start and
end dates. This is because the digital revolution is not a single change
that has happened (the birth of the internet, the shift to mobile, or cloud
computing) that you must adapt to. The digital revolution is an ongoing
acceleration of change being driven by successive waves of new
technologies. It will continue to reinvent customer behaviors, business
models, and economic systems far into the future. Buckle up for the
ride.

How DX Goes Wrong

Over the years, I have heard countless commentators on DX proclaim

that the single most important factor in success is a clear CEO
mandate for change. And yet, in GE, the New York Times Company,
and countless other cases, I have witnessed the opposite. In these
cases, the chief executive declares their full-throated commitment to a
DX effort. They even commit substantial financial and human capital.
And yet years go by, and the results fail to live up to what was expected
or to what is desperately needed for the future.
Why is DX failing? What have so many companies done so wrong?
Let’s start by recognizing that DX is hard. In many ways, it involves
a kind of balancing act. DX cannot simply be an effort to “digitize” the
legacy business—that is, to upgrade existing technology, cut costs, and
improve the customer experience of your current offerings. To stay
relevant, survive, and grow in the digital economy, every business must
be ready to digitize its core and grow beyond it, to maximize its current
cash flow and invest for the future, to pursue both incremental
innovations and more radical ones. But you cannot rebuild your current
business and build your next business with the same people,
processes, and organizational structures. You cannot use the same
approach to rebuild the past and to build the future.
The DX challenge is particularly hard for complex organizations.
Today, organizational complexity is driven by three primary factors: the
number of employees (measured in head count), lines of business (with
different offerings to different customers), and geographies of operation
(each with different regulations imposed). As any of these factors
increases, the complexity of managing the organization compounds
dramatically. Simply put, if you have an established business with just
500 employees and a single line of business operating in one
geography, transformation will be much easier to achieve. For such a
business, if you define a clear digital strategy based on market and
technology trends, you should be able to execute that strategy with
good leadership. But if your business has 10,000 (or 100,000)
employees, you operate multiples lines of business, and they straddle
several geographies with differing regulations, then driving DX will be
much, much harder.
The symptoms of this difficulty are seen everywhere. Talk to leaders
of complex organizations, and you will hear a litany of DX troubles: Our
employees are afraid of change. New ventures always lose out to the
core business. Legal and compliance reject many of our promising
ideas. Our digital efforts make for good press releases, but they aren’t
moving the needle. Risk aversion and slow decision making mean we
can’t keep up with digital competitors. Our legacy information
technology (IT) is inflexible. Our data is trapped in silos. Our workforce
lacks the skills they need.
Sound familiar? The list goes on, but these are all just symptoms of
more fundamental problems in how DX is practiced. In my own
research and work with a wide range of companies, I have found five
root causes of failure in digital transformation. Let’s review each of
these barriers to DX and the problems that they bring.

Top Barriers to DX Success

1. NO SHARED VISION
One of the biggest barriers to effective DX is the lack of a shared
vision. In countless large companies, I have seen “digital” declared to
be a priority, but when you talk to managers, it becomes clear that there
is no shared understanding of the digital future of their industry, where
their business aims to compete, or what gives them a right to win in that
future. Instead, there is only a generic rallying cry to “become digital.”
The symptoms of this lack of shared vision are many. Employees
move slowly and show fear of change, lacking a clear sense of where
the firm is going and how they might contribute. Investors balk at large
financial investments in digital, as do executives responsible for profit
and loss statements (P&Ls). The company’s digital initiatives are
generic, following the moves of peers and reacting belatedly to market
trends. And leaders rely on generic “digital maturity” metrics to guide
their efforts because they have no clear business metrics to judge their
DX progress.

2. NO GROWTH PRIORITIES
The next major barrier to DX is a lack of clear priorities for growth. This
lack may arise because the company is focused only on “digitizing” its
past business and not looking beyond it. Or leaders may lack the
discipline to define a few strategic priorities to focus on—whether
customer problems to solve or business opportunities to capture.
The symptoms of this lack of priorities are many. Without a clear set
of priorities, DX lacks strategic direction. Rather than focusing on
business problems, DX is defined by technologies (AI, cloud
computing, blockchain, etc.) and is easily hijacked by the latest shiny
new thing. Without a growth focus, DX focuses only on cutting costs
and optimizing the current business. Digital efforts are run by
technology specialists as the rest of the organization continues its work
unchanged. As a result, DX grows disconnected from business needs
and loses support over time.

3. NO FOCUS ON EXPERIMENTATION
The third major barrier to DX is an emphasis on planning over
experimentation. Decision makers spend years developing their plans
for DX. They demand laborious development of business cases before
work begins on any new digital product or service. When work does
begin, the focus is on meticulous planning and execution, following a
stage-gate approach to carry each project through to a predefined
solution. This approach stands in direct opposition to the model of rapid
experimentation that guides digital-native businesses. And though
legacy firms may adopt the trappings of experimentation—rolling out
agile software teams and enrolling in design thinking classes—they
squeeze these iterative methods into a planning-heavy management
model.
The symptoms of relying on planning over experimentation are
painful. Decision makers wait for benchmarks and best practices rather
than validating new ideas directly with customers. Teams are assigned
to build solutions rather than to solve problems. Projects have no
flexibility to change direction, leading to costly failures and a culture of
risk avoidance. Digital ventures move slowly, are beaten to the market,
and struggle to make an impact on the business.

4. NO FLEXIBILITY IN GOVERNANCE
The fourth barrier to DX is the use of business-as-usual (BAU)
processes and governance for all initiatives. Traditional silos, reporting
lines, and budgeting dominate, stifling efforts at growth. Companies
lack processes for iterative funding or for allocating resources beyond
the core. They are unable to stand up multifunctional teams to move
fast on new opportunities. In short, they have no repeatable process for
managing and scaling growth.
The symptoms of inflexible governance can be seen everywhere.
Executive sponsors must personally approve digital projects and grant
waivers to company rules. Functional silos impede team collaboration
and slow innovation. Resources are trapped in annual budgeting
cycles, hampering efforts to scale innovation. With few projects
approved, no one is willing to shut theirs down once it is started.
Uncertain ventures are deemed too risky, and innovations outside the
core are simply ignored.

5. NO CHANGE IN CAPABILITIES
The final barrier to DX is a reliance on preexisting capabilities, including
technology, talent, and culture. Legacy technology remains in place,
with only patches and cosmetic fixes to IT architecture, data assets,
and the rules that govern both. Legacy talent remains undeveloped,
with little investment in digital skills or in the workforce and leadership
who were hired and trained for the needs of the past. Legacy culture
remains unchanged, with mindsets and behaviors that are rooted in
top-down, command-and-control leadership.
The symptoms of stagnant capabilities are many. IT systems are
inflexible and reinforce the silos inside the organization. Managers lack
the shared real-time data that they need to make decisions. Every
digital project must go through a central IT division, causing
bottlenecks. IT needs are outsourced to vendors because your own
workforce lacks the skills for digital innovation. A top-down culture and
mindset lead to cynicism and a wait-and-see attitude from disengaged
employees.

How DX Goes Right

If the above scenarios sound familiar and bleak, don’t be disheartened.
True, surveys show that 70 percent of DX efforts are failing, but let’s
focus on the 30 percent that are succeeding! The DX Roadmap, which
is just ahead, draws from dozens of successful DX cases. Here are just
a few of the amazing stories of DX success in business-to-consumer
(B2C) and business-to-business (B2B) industries:

• The Walt Disney Co.—The legacy media business has stood up to

tech titans (Netflix, Apple, Amazon) by expanding its library of content
and launching its fast-growing streaming services Disney+, ESPN+,
and Hulu. Within three years of launching Disney+, the company’s total
number of streaming subscriptions had surpassed those of Netflix.10 At
the same time, the firm has reorganized its growth strategy, its metrics,
and its organizational structure around Disney’s new digital future as a
direct-to-consumer business.
• Mastercard Inc.—The credit card processing company is now one
of the world’s biggest fintechs. Leveraging its tremendous data and
global business network, Mastercard has built a growing business in
digital commerce. The company runs one of the top accelerators for
fintech start-ups, and its own innovation labs are building and scaling
new business models around cybersecurity, digital identity, and
analytics—while selling digital services to business customers around
the world.
• Domino’s Pizza, Inc.—The restaurant chain has achieved
phenomenal growth by reinventing its core delivery business for the
digital age. Domino’s invested in an AI-powered mobile app (one of the
first to take your order by voice); an “Anyware” omnichannel strategy
that allows customers to order by emoji, tweet, Apple Watch, or
Amazon Alexa; and innovations in delivery vehicles that include rolling
pizza delivery bots (for narrow streets in European cities) and even
flying pizza drones (first tested in mountainous New Zealand). With a
relentless focus on innovation in the delivery experience, Domino’s
digital investments helped power a stunning growth in share price of
3,200 percent in just seven years.11
• Deere & Company—The company made its name selling John
Deere farm equipment like tractors, seeders, and harvesters. Today,
Deere still makes these machines, but all of them are software- and
sensor-enabled and connected to the cloud. Deere is gathering data on
every square foot of a customer’s farmland and using that data to help
them optimize how seeds are planted, fertilized, and harvested, with
precision farming and data analytics.
• Air Liquide—The global leader in industrial gases is harnessing
data to unlock new sources of value in key industries like health care
and manufacturing. Today, Air Liquide is capturing data from its 400
industrial plants and its 20 million gas cylinders around the world and
leveraging that data to create value in areas like predictive
maintenance. Air Liquide is reshaping its customer experience for
business clients across an omnichannel mix of web, phone, apps, and
face-to-face interactions. And it is collaborating in new ways with large
partners and digital start-ups to address new markets like home health
care and urban air pollution.

The message from these and other companies from the winning 30
percent could not be clearer: DX is absolutely possible if your business
embraces transformative change.

DX = Digital Strategy + Organizational Change

I wrote the first book on the topic of DX, The Digital Transformation
Playbook. That book has become a global bestseller, published in over
a dozen languages and in both print and audio formats. Its reach has
led me to speak at venues around the world and advise the leadership
of scores of businesses. Meanwhile, my executive teaching for
Columbia Business School—both on campus and online—has brought
me into close contact with thousands more business leaders.
This book is the culmination of that practical experience and years
of research on DX. It draws on both the common, persistent problems I
have seen companies face and the key steps I have seen lead them to
success. Based on all this, I have developed a complete DX Roadmap.
At the outset, let’s be clear that digital transformation is not only
about adapting your strategy. True DX requires a combination of digital
strategy and organizational transformation. To sum it up as a formula,
DX = D strategy + organizational X
If you have read my previous book, you will recognize that this means
rethinking strategy for the digital era across five domains: your
customers, your competition, your data, your process of innovation, and
your value proposition. The DX Roadmap presented in this book
encompasses each of those five domains and shows how to make that
kind of change happen at every level of your organization—no matter
its industry, size, or complexity.
As we examine the DX Roadmap, it is important to remember that
DX is not a traditional change management project with start and end
dates, not a journey with a fixed destination. Instead, it is the way to
rebuild your organization to be ready for continuous transformation in
the digital future. As we’ll see, that’s precisely how digital-native
businesses like Amazon, Alphabet, Microsoft, and Netflix have learned
to thrive under continuous change. Your business can, too.

The DX Roadmap

Since my last book was published, I have received multiple inquiries

from CEOs, essentially asking, “Dear Professor Rogers—I have read
your book, l agree with it, and I am now ready to start my digital
transformation. Can you please send me an article, or better yet a
PowerPoint slide, that shows me the steps I should take?”
Let me be clear. What follows is not a set of IKEA instructions.
There is no simple list of investments, budget allocations, and strategic
steps that every company can follow for its DX. Every organization has
a unique context and history, unique strengths and challenges;
therefore, every organization will have its own starting point and its own
path forward. But, however unique each DX path may be, I have seen
that every organization faces the same barriers that must be overcome
for DX to have a real and lasting impact.
Therefore, I have developed a framework of five iterative steps to
navigate your own path to DX success for your business. The DX
Roadmap is based on my experience advising CEOs and CDOs at
more than twenty companies in varying industries and locations and
with different sizes and ownership structures, and on the experience of
scores of other companies that I have followed in my research.
Before we take a close look at each step, it is important to note that
they are not like steps on a ladder that you climb once and leave
behind. Although there is a sequence to starting them (begin Step 1,
then begin Step 2 . . .), the work of each step continues as you
advance to the others. DX is not a finite project. Its change will be
iterative and cumulative as you deepen and broaden the transformation
of your organization over time. Keeping this in mind, let’s take a look at
each step of the DX Roadmap (see figure 1.2).

Figure 1.2.
The Digital Transformation Roadmap

1. VISION: DEFINE A SHARED VISION

In the first step of the DX Roadmap, your goal is to define a shared
vision of the digital future for your organization. This starts with
describing the future landscape of your industry, shaped by digital
forces. It includes defining the unique advantages that give you a right
to win in this digital future. It means choosing a North Star goal for the
impact your work will have on customers and others. And it means
spelling out a business theory of how you will capture value and earn a
return on your digital investments.
 When done right, this step will enable your business to lead
proactively rather than react to external trends, to invest only in digital
initiatives where you have a competitive advantage, to clearly define
the business impact of digital investments, and to win the backing of
investors and finance chiefs as well as the support of the employees
who will drive your digital agenda forward.

2. PRIORITIES: PICK THE PROBLEMS THAT MATTER MOST

In the second step of the DX Roadmap, your goal is to define the
strategic priorities that will guide your digital growth agenda. This starts
with looking at strategy through the twin lenses of problems to solve
and opportunities to pursue. This step uses a variety of tools to identify
the most valuable problems and opportunities for your business. And it
uses problem/opportunity statements to define strategy and spark ideas
for digital innovation at every level of your enterprise.
Done right, this step will enable you to provide direction to teams
across your organization, focus digital on solving problems and not
deploying technologies, ensure digital delivers growth and not just
efficiency, and accelerate change with new ventures at every level and
in every department.

3. EXPERIMENTATION: VALIDATE NEW VENTURES

In the third step of the DX Roadmap, your goal is to rapidly test new
digital ventures in order to validate which ones will create value for your
customers and the firm. The step starts with thinking like a scientist:
defining your hypotheses and designing experiments to test your
business assumptions. It uses iterative metrics to gather data directly
from customers. It uses iterative prototypes and minimum viable
products (MVPs), each designed to answer a specific question. And it
uses a new model, the Four Stages of Validation, to sequence learning
and guide any venture on its path from new idea to business at scale.
When done right, this step will enable your business to test many
new ideas and learn which work best, make decisions based on data
from customers rather than benchmarks, keep your failures cheap and
your bias toward risk taking, and iterate and adapt quickly to build
innovations with value at scale.

4. GOVERNANCE: MANAGE GROWTH AT SCALE

In the fourth step of the DX Roadmap, your goal is to design
governance models to scale digital growth across the enterprise. This
means defining rules and decision rights for small, multifunctional
teams; creating structures (like labs, hackathons, and venture funds)
that provide flexible pools of resources; establishing boards that green-
light new ventures and oversee iterative funding; and managing three
paths to growth—inside the core, partnered with the core, and outside
the core—with the right rules and governance for each.
When done right, this step will enable you to empower teams to
drive growth, allocate resources flexibly, quickly shut down ventures
that are not working, and manage a steady pipeline of digital
innovations both in your core and beyond it.

5. CAPABILITIES: GROW TECH, TALENT, AND CULTURE

In the last step of the Roadmap, your goal is to invest in the technology,
talent, and culture that will be critical to your digital future. This includes
investing in technology with a microservices architecture, synchronized
data assets, and effective IT governance. It means growing your digital
skills by managing the talent lifecycle from hiring to training, exiting,
and beyond. It means defining the culture—shared mindsets and norms
of behavior—that will support your digital strategy; communicating that
culture with stories, symbols, and action; and enabling that culture
through everyday business processes.
When done right, this step will enable your business to integrate
technology across silos and with outside partners, use data to provide a
single source of truth to managers, give teams the skills to build their
own digital solutions, and empower employees at every level to drive
bottom-up change.

The Roadmap from the Bottom Up

The importance of empowering employees brings up an essential point.
As you apply the DX Roadmap, it is critical to understand that DX does
not flow only from the top of the organization down. Your DX effort may
begin in the CEO’s office or under the direction of a change agent like a
CDO. But it will never succeed if it is driven only from the top.
The defining fact of the digital era is its accelerating pace of change.
Every organization must learn to adapt not once but continuously. To
keep up, businesses are forced to shift away from top-down, command-
and-control management and to embrace a more agile vision of the
organization that prizes employee autonomy and initiative. Digital-
native businesses like Amazon, Netflix, and Alphabet have all
embraced this new model—achieving speed at scale by pushing
decision making down to the lowest possible level.
As we will see throughout this book, applying the DX Roadmap
means taking a bottom-up approach to your own organization as well.
“Bottom up” does not mean a flat organization with no hierarchy. It
means three things: decision making is pushed down, market insights
flow from the bottom up, and innovation starts at every level of the
organization.

Following the Five Steps: Success for the New York Times

The five steps of the DX Roadmap can ensure a strong start to any
new transformation effort, but they can also turn around a DX effort that
has long struggled. Among today’s most impressive DX success stories
is the New York Times Company. In the years since the painful self-
examination and exposure of its Innovation Report, the Times has
turned the corner to become a leading example of DX success. It did so
by addressing each of the five barriers to transformation and pursuing
all five steps of the DX Roadmap.
The turnaround began with a clear vision of the future. With the
internet continuing to disrupt the advertising market, the only hope for
the Times Company’s survival was to reinvent its business model.
Instead of a business run primarily on advertising revenue, it would
become a business based primarily on subscription revenue. This was
arguably the biggest strategic shift in the company’s history, and it
required changing nearly everything about the business. The year after
the damning Innovation Report, the Times’s leadership released a
strategic document titled “Our Path Forward.” It clearly spelled out the
ambition to transform the Times’s business model so that digital
revenue exceeded print revenue. It included a target to double digital
revenue in five years to reach $800 million—a figure that could sustain
the paper’s journalistic efforts around the world. Only then would the
mission of the Times be secure from the disruption of its print business.
The document also laid out a few strategic priorities to focus and
guide its digital efforts: transform the product experience to make a
New York Times subscription as indispensable to readers’ lives as a
Netflix or Amazon Prime account. Expand the Times’s global reach and
international readership. Grow digital advertising by creating new and
compelling ad formats. Organize employees’ work around digital
platforms and the reader experience. Shift energy from the print edition
while still maintaining quality for its audience.
With priorities in place, the Times Company accelerated its pace of
digital experimentation and new ventures. In its core news product, this
meant telling stories that were more visual and data driven, many of
them pioneered by The Upshot, a new journalism desk. It meant
experimenting with media like video, podcasts, virtual reality, and
interactive newsbots to see which would engage subscribers on mobile
devices. At the same time, the Times launched new stand-alone digital
subscriptions like NY Times Cooking and Games. And it explored new
business models based on licensing (a TV show on the FX network and
Hulu), affiliate sales (the Wirecutter product review site), and live
conference events. Several digital ventures came by acquisitions,
including Wirecutter, Serial (a podcasting studio), Audm (an audio app),
the Athletic (a sports site with global reach), and the hit online game
Wordle. Meanwhile, advertising and technology teams developed new
formats for advertising on mobile and for audio products.
Transformation at the Times Company extended to its governance
as well. Digital was no longer placed in a separate subsidiary but at the
core of the entire organization. For the first time, journalism, product,
and engineering perspectives were brought together in cross-functional
teams. Different governance models were established for different
growth opportunities. New subscriptions like Cooking and Crosswords
were placed in the Standalone Products and Ventures Group to focus
on their distinct metrics for growth. And The Athletic—an acquisition
with a distinct audience of sports fans—was made a separate operating
unit, led by its original founders.
At the same time, the company refocused its efforts to build the
capabilities that would be critical to its digital future. The Times’s
engineering teams built up its technology and data infrastructure,
allowing it to capture data on every digital action by its readers, link any
news articles to related stories from the New York Times’s century-plus
archive, and target advertising with proprietary first-party data that
maintained readers’ privacy. Meanwhile, new digital talent was hired
and promoted into leadership positions. Reporters of all stripes were
trained in the use of data-driven journalism and visualization. A. G.
Sulzberger, who became publisher in 2018, boasted, “We employ more
journalists who can write code than any other news organization.”12
The culture of the Times began to shift as well, from a risk-averse
mindset dictated by old traditions of the newsroom to a culture focused
on risk taking and learning from mistakes.
The results of this second wave of DX at the Times were
breathtaking. The company reached its ambitious goal of $800 million
in digital revenue a year ahead of target. An added goal to reach 10
million subscriptions by 2025 was met nearly four years early.13 Most
important, the Times achieved both its overarching goals—digital
revenue exceeded print revenue, and subscriptions eclipsed
advertising. Investors took notice, and in five years, from 2016 to 2021,
the stock price soared by 261 percent. Looking ahead, the Times
Company’s annual report laid out a new vision: “We aim to be the
essential subscription for every English-speaking person seeking to
understand and engage with the world.”14

Why the DX Roadmap Matters

The stakes could not be higher for businesses. The five steps of the DX
Roadmap will set you on the course to achieve truly impactful DX. They
will enable you to avoid the failures of organizational stagnation that
stifle so many DX efforts and to join the ranks of the DX winners. Table
1.1 summarizes the difference in outcomes between DX efforts that fail
or succeed in each of the five steps of the DX Roadmap.
Table 1.1.
Symptoms of Failure Versus Success in the Five Steps of the DX
Roadmap
Symptoms of Failure Symptoms of Success
Step 1: Vision
• Employees fear change and • Employees at every level
lack a clear sense of where understand the digital
the firm is going. agenda and push it forward.
• Backing for digital • Support for digital
investments is weak from investments is strong from
investors, CFOs, and P&L investors, CFOs, and P&L
heads. heads.
• Digital initiatives are • Only digital initiatives with
generic, following the a competitive advantage
examples of peers. receive investment.
• Generic digital maturity • The business impact of
metrics are used to guide digital is clearly defined, with
efforts. metrics to measure and
track results.
• The firm follows the market, • The firm leads the market,
reacts to others, and is alert to critical trends while
surprised by new entrants. there is time to choose a
course.
Step 2: Priorities
• Digital transformation is a • Clear priorities provide
series of scattered projects direction to digital
with no clear direction. transformation across the
organization.
• Digital efforts are defined by • Digital efforts are defined
the technologies they use. by the problems they solve
and opportunities they
pursue.
• Digital is focused solely • Digital is focused on future
on operations, cost growth as well as improving
cutting, and optimizing the current business.
the current business.
• A few people in the • Every department is
organization drive digital pursuing its own digital
while the rest stick to old ventures, with a backlog of
ways of working. ideas to try next.
• Transformation is • Transformation is linked to
disconnected from business the needs of the business
needs and loses support and gains support over time.
over time.
Step 3: Experimentation
• Innovation is focused on • Innovation is focused on
coming up with a few great testing many ideas to learn
ideas. which work best.
• Decisions are made based • Decisions are made based
on business cases, third- on experimentation and
party data, and expert learning from the customer.
opinion.
• Once they start a project, • Teams stay focused on the
teams are committed to problem but flexible on the
building the solution in full. solution.
• Failures are costly, so the • Failures are cheap, so
fear of risk is high. there is a bias toward risk
taking.
• Good ideas move slowly • Good ideas grow fast and
and don’t seem to move the deliver business value at
needle on the business. scale.
Step 4: Governance
• A top executive must • Established structures
personally approve any new provide resources and
innovation. governance for innovation.
• New ventures move slowly, • New ventures move fast,
led by traditional teams in led by highly independent,
functional silos. multifunctional teams.
• Allocating resources to new • Resource allocation
ventures is slowed by the happens quickly through
annual budgeting cycle. iterative funding.
• Innovation is limited to a • A steady pipeline of
few big projects, which are innovations is managed with
hard to shut down once they smart shutdowns to free up
are started. resources.
• The only ventures to gain • Governance supports
support are low-risk ventures with low and high
innovations in the core uncertainty, both in the core
business. and beyond.
Step 5: Capabilities
• Inflexible IT systems • Modular IT systems
reinforce silos and limit integrate across the
collaboration. organization and with
outside partners.
• Data is contradictory, • Data provides a single
incomplete, and inaccessible source of truth to managers
to managers in real time. across the company.
• Centralized IT governance • IT governance provides
causes bottlenecks for new oversight while keeping
projects. innovation in the hands of
the business.
• Employees lack digital • Employees can build and
skills, so digital projects must iterate digital solutions
be outsourced. themselves.
• A top-down culture and • An empowering culture and
bureaucracy stifle processes help employees
employees, breeding drive bottom-up change.
cynicism and inertia.

How to Use This Book

The book is meant to be extremely practical. It is based on my own

work advising firms of diverse sizes and in different industries and
locations—but also on the countless questions and years-long
exchanges that I have had with thousands of global executives through
my programs at Columbia Business School.
Because digital transformation must happen at every level of the
organization, the DX Roadmap is designed to zoom in or zoom out.
Each step is applicable whatever your role—whether you are CEO,
CDO, directing digital for a business unit, leading a functional team like
human resources (HR), or designing a new digital product.
The next chapter (chapter 2) examines the connection between DX
and innovation and why established firms are failing to innovate at the
pace demanded in the digital age. It explores the challenge of
innovation under great uncertainty and the challenge of innovation
beyond your core business. And it shows what we can learn from
digital-era businesses and from digital-era methods such as agile, lean
start-up, design thinking, and product management.
Chapters 3 through 7 lay out the five steps of the DX Roadmap in
detail. Each step is illustrated with real-world case studies of
businesses. You’ll find dozens of examples from a range of industries:
banking, insurance, retail, consumer goods, media, telecom,
technology, automotive, energy, health care, nonprofit, industrial
manufacturing, and even container shipping. Each case illustrates the
strategic concepts of the book with stories of individuals and their work
in the real world.
I have also included eight strategic planning tools to give you the
practical means of applying the DX Roadmap with your own team:

• Shared Vision Map

• Problem/Opportunity Statements
• Problem/Opportunity Matrix
• Four Stages of Validation
• Rogers Growth Navigator
• Corporate Innovation Stack
• Tech and Talent Map
• Culture-Process Map

These tools are based on my experience advising and consulting.

They are designed to pose the crucial questions that will guide you to
the right answers for your particular organization. Downloadable
versions of the tools, as well as additional tutorials on their application,
can be found in the Tools section of my website at
www.davidrogers.digital.
 The book’s conclusion returns to the theme of the bottom-up
organization and examines the three jobs of leaders in an era of
constant change. In the appendix, you will find visual summaries, a
self-assessment tool, and directions to more resources online.
The DX Roadmap is designed to be started quickly so that you can
learn by doing and see immediate results. Remember, real DX is
iterative in nature. Once started, the five steps will repeat, overlap, and
support one another. As you progress, your transformation will both
deepen and broaden in scope. The main point is to get started and to
learn by doing. Don’t wait for a five-year plan; start something in ninety
days! DX is not about spending months planning a multiyear process
and then faithfully carrying it out. It is about starting the first steps now
and learning as you go.
To thrive in the digital era, your organization must be built for
continuous transformation. By the time you adapt to today’s massive
digital shifts, you will already need to be ready for tomorrow’s. With a
nod to Darwin, I suggest that it is not the strongest of businesses that
will survive the digital era but those that are most adaptive to change.15
Onward to continuous transformation!
 2
DX and the Challenge of Innovation

Before we begin the five steps of the Digital Transformation (DX)

Roadmap, it is important to understand the role of innovation in DX.
In this chapter, we will learn how the failure to innovate is holding
legacy firms back, what the two key challenges to corporate
innovation are, and how to solve each of them. We will also discover
what incumbents must learn from digital-era organizations and
methods to successfully innovate for the digital age.

DX Is Innovation

In the twentieth century, the goal of any company was to find a

business model that worked, optimize that model, and exploit it for
as many years as possible. A successful business would see many
changes over the years in its product features, customer segments,
technology, operations, and more. But the business model—the
means by which the company created, delivered, and captured value
from the market—could prosper unchanged for generations. A few
long-lived firms would shift business models over many decades,
such as IBM evolving from selling office machines to services, or
General Electric moving from electric appliances to power plants,
financial services, and broadcast television. But these rare firms
were exceptions that proved the rule of business model constancy.
In the digital era, however, transformation is the rule, not the
exception. The life span of a business model—from when it emerges
as a viable means to run a business until its profitability begins to
fade—seems to grow shorter every year.1 Digital businesses thrive
not by sustaining and excelling at a single business model but by
constantly developing new models to adopt or to replace the old.

Netflix’s Repeated Reinvention

Think of Netflix, the preeminent global media company born to date

in the digital era. In its first twenty-four years, we can see a pattern of
transformation through at least four distinct business models. Marc
Randolph and Reed Hastings launched Netflix with a business
model aimed at disrupting the way people watched movies at home
—which, in 1998, meant renting individual VHS tapes from stores
that demanded quick returns and charged punishing late fees.
Netflix’s first business model entered the market by offering a
premium membership plan and delivering DVDs (then a cutting-edge
format) by mail with no late fees.
But even as Netflix grew rapidly, its founders were already
looking to their second business model. They knew the future would
be streaming videos over the internet, but the technology
infrastructure was not ready when they started the company. As
consumers’ home internet bandwidth increased, Netflix began a pilot
test of streaming video, offering a tiny library for free to their DVD
subscribers. After that test proved favorable, Netflix began to license
a large library of movies from rights holders like Starz and Epix.
When conditions were right, Netflix launched its second business
model to much fanfare: Hollywood movies streamed straight to your
home screen.
The company’s third business model arose because the finances
of its second model became unsustainable. Netflix’s soaring
membership numbers and revenue demonstrated the huge market
for streaming video, and rights holders demanded much larger fees
to renew their licenses. As streaming other people’s movies became
unprofitable, Netflix pivoted to its third business model—becoming its
own television and film network. With hit series like House of Cards,
Netflix began investing billions of dollars each year producing
original content to keep subscribers coming back. Demand for
growth led Netflix to expand around the world, creating global
content in diverse languages. With non-English-language series like
Squid Game, Netflix proved there was profit in repurposing content
from anywhere for a global audience.
Netflix’s fourth business model came after years of subscriber
growth finally began to slow down. From 2011 to 2021, subscribers
had grown from 21 million to 220 million. But in 2022, Netflix posted
a (slight) decline in subscribers for the first time in a decade.
Shocked investors cut the value of the stock in half. In response,
Hastings turned to a new business model, long discussed but never
implemented: advertising. This fourth business model (harkening
back to the older model of broadcast television), would generate a
new revenue stream from advertisers. It would also allow Netflix to
pursue new customers: those not willing to pay the full price of a
subscription, as well as the 100 million freeloaders, by Netflix’s
estimate, borrowing the accounts of others. Investors responded
favorably. As Netflix prepared to launch this new business model, the
company’s leaders stressed that it would evolve too: “Like most of
our new initiatives, our intention is to roll it out, listen and learn, and
iterate quickly to improve the offering. So, our advertising business in
a few years will likely look quite different than what it looks like on
day one.”2

The Innovation Divide

The same pattern of business model invention can be seen in every

digital-era business. Uber began by offering its customers rides, then
expanded to meal delivery. Tesla began selling electric sports cars,
then charging systems, then at-home batteries for a decentralized
power grid. Google started with a search engine, but it has grown to
include video (YouTube), a mobile operating system (Android), digital
home devices (Nest), and enterprise computing. Alibaba began with
an online marketplace and added search, digital payments, and a
broad set of financial services, including credit scoring and the
world’s largest mutual fund.
 For older legacy businesses, the digital era demands no less.
Digital transformation cannot mean just upgrading your data and
technology to optimize the business you have always run. For any
established business to thrive, DX must deliver innovation and new
engines of growth. But the reality is that most pre-digital-era
companies are terrible at this. In every industry, we see legacy
enterprises struggling to create new business models beyond their
core business. For every digital innovation that succeeds in the
market, there is an incumbent serving the same customers, in the
same industry, saying to themselves, “We should have done that!”
Why did Nokia, the king of mobile phones, fail to launch the
iPhone or a similar first-generation smartphone? Why did
established book publishers (e.g., Random House) and retailers
(e.g., Barnes & Noble) fail to create an e-book format before Amazon
launched the Kindle? Why did hospitality chains like Marriott fail to
capture the market that Airbnb saw for a digital travel experience
based on a multisided market? Why did Hollywood studios and
television networks wait years to follow Netflix’s path into streaming?
Why did global automakers wait a decade to follow Tesla into electric
vehicles? Why did traditional banks wait a generation after PayPal,
Alibaba, and others pioneered peer-to-peer payments, only to launch
a belated copycat like Zelle?
Again and again, we see traditional companies fail to capture
tomorrow’s opportunities for growth. They are failing because of two
fundamental challenges to innovation in established companies. I
call these the challenges of uncertainty and proximity. Fortunately,
both have management solutions—solutions that are at the heart of
the DX Roadmap.

Innovating Under Uncertainty

The first challenge for innovation in an established company is

managing innovation under great uncertainty. The topic of
uncertainty has long been central to the theory and practice of
innovation. Many practitioners describe business innovation as an
effort to address three types of uncertainty: desirability (does the
market want this?), feasibility (can we deliver this?), and profitability
(can this make a profit?). But other types are critical as well—
defensibility (can it fend off competitors?), scalability (can it grow?),
and legality (will it comply with laws and regulations?), for example.
For any specific venture, the sources of uncertainty become even
more specific, such as cost structure, competitive differentiation,
channel partners, user experience, and so on.
In the business environment of the digital era, uncertainty is only
increasing thanks to new technologies, competition across
industries, and constantly changing customer needs and
expectations. The greatest uncertainty is faced by companies
exploring new digital business models. Think of the questions that
Airbnb’s founders faced when starting their company. Which
customers would be willing to try this new business model? How
could Airbnb increase trust in the system? How would it collect
money? How would it navigate local taxes and regulations? And how
would the company scale while holding off competitors?
Established digital firms also face uncertainty as they pursue new
growth ventures. Google acquired YouTube eighteen months after
the video service launched, when the start-up was growing fast but
had no proven business model, had no revenue, and was in possible
violation of copyright law. Apple began work on the iPhone under
tremendous technical uncertainty about its touchscreen design and
with no notion that success would hinge on creating an app store for
outside developers. Facebook bet big on mobile, enduring years with
zero revenue as it first moved from the desktop to phones. Microsoft
bet on Azure when a focus on cloud computing seemed incredibly
risky and could cannibalize its existing enterprise Server and Tools
business line.

The Inadequacy of Traditional Planning

Managing under high uncertainty is extremely hard for established

businesses. As most organizations grow larger and more complex,
their every function—marketing, HR, operations, financial planning—
is designed for consistency and control, using methods like Six
Sigma or total quality management. An aversion to risk is matched
by a mindset that “failure is not an option.” The whole point of
management becomes to generate a predictable return on
investment (ROI).
Not surprisingly, the response of such businesses to uncertainty
is to manage it through planning. We can observe four steps in the
classic planning process: gather third-party data, write a detailed
business plan, make an expert decision, then focus on execution. In
short, study > plan > decide > build.
This approach to planning that was taught for years in business
schools works well with a well-known business model (which is
exactly what it was developed for). But it fails dramatically when
managing innovation under great uncertainty. (See the box “The
Perils of Planning.”) When pursuing innovation in our highly
uncertain digital world, traditional planning is a recipe for disaster.
Often, it leads to so-called analysis paralysis as laborious
benchmarking and writing of business plans stretch on for many
quarters or even years, and opportunities go to nimbler competitors.
In the worst cases, traditional planning leads to incredibly costly
failures in the digital realm.

The Perils of Planning

When dealing with well-known problems or when operating long-

standing business, that is, when managing under relative
certainty, the traditional planning approach (study, plan, decide,
build) can work quite well. But in digital-era conditions of
uncertainty, traditional planning fails at each stage.
Study (gather third-party data). Faced with the unknown,
managers are taught to gather preexisting data, often with the
help of large consulting firms. They look to competitors for
benchmarking and search for established case studies that might
provide them with best practices. While this is useful when facing
a known problem that others have already solved, it can be a fatal
 distraction when pursuing innovation in a rapidly changing
marketplace.
Plan (write a detailed business plan). Next, traditional
managers apply their best analysis to write a business case full of
detailed steps and projections on future outcomes. A major
decision can lead to multiple detailed business cases—each one
spelling out the implications of a different course of action. But
when you are pursuing innovation under uncertainty, a business
plan is a work of pure fiction! (My colleague Bob Dorf has said for
years that the practice should be moved from business schools to
creative writing departments.)
Decide (make an expert decision). Having analyzed the
options, a decision must now be made about what course of
action to take. Traditional practice is to rely on the opinions of
experts—those with the most seniority and experience. When the
future looks a lot like the past, the intuition of experienced leaders
may prove an excellent guide. But when dealing with great
uncertainty, expert opinion based on past experience can be
highly misleading. In Silicon Valley, this practice is called
management by “HIPPOs”—the highest-paid person’s opinion.
Build (focus on execution). After a course of action has been
chosen, the traditional enterprise becomes entirely focused on
execution. “You scope it, we build it!” is the mantra of traditional IT
and operations teams. Companies begin making large
investments that lock them into a multiyear plan for a particular
solution. Instead of starting small and proceeding with humility
about whether a planned innovation is the right one, companies
rush to start building at scale.

THE PLAN FOR CNN+

The failure of CNN+, launched in 2022 as a new digital business
model for news, is a perfect example. In it, we can see all four
stages of traditional planning at work:
 • Study: CNN began this initiative by surveying the booming
market for streaming video with the help of consulting firm
McKinsey. Netflix and Amazon Prime continued to dominate,
with hundreds of millions of customers, but Disney+ had
recently launched and gained 10 million subscribers on its first
day. Meanwhile, CNN’s sister network HBO was making a big
push for streaming as well.
• Plan: CNN laid out a plan to launch a paid subscription news
service with many of the same celebrity broadcasters as its TV
channel (although contracts with cable companies prohibited
streaming their live TV content). CNN and McKinsey projected
that the service would attract 2 million subscribers in its first
year, and at least 15 million within four years.3
• Decide: The decision to move forward with CNN+ came straight
from the top. CNN president Jeff Zucker saw in it the future of
news and a new era for the company. His boss Jason Kilar
proclaimed, “In my opinion, CNN+ is likely to be as important to
the mission of CNN as the linear channel service has been
these past 42 years. It would be hard to overstate how important
this moment is for CNN.”4 Not everyone agreed, but as one
CNN insider put it, “It was a vanity project . . . They wanted to
launch it.”5
• Build: With the most-senior executives’ decision to move
ahead, the company went all in on execution, with $300 million
spent developing CNN+ and hundreds of employees hired.
Contracts were signed for name-brand anchors such as NPR’s
Audie Cornish and Fox News’s Chris Wallace. Plans were made
to spend over $1 billion in the first four years to grow the
service.6

No doubt CNN+ was a bold and innovative new business model.

But the biggest uncertainty it faced was this: Did anyone want to pay
$5.99 a month to stream news? By one estimate, CNN+ joined more
than 300 other unique streaming services operating in the United
States, yet the average household paid for only four of them.7 At a
minimum, it would have been useful to know what percentage of
CNN’s current viewers were interested in paying for a streaming
version of the brand (10 percent? 5 percent? less?). But rather than
conduct a test to find out, the company plowed ahead with its plan.
As it turned out, uptake by CNN’s existing customers was small.
At the time, CNN’s TV channel was attracting an average of 773,000
viewers per day. In its first weeks, CNN+ was watched by fewer than
10,000 people per day—about 1 percent of the TV audience.8 Less
than a month after launching, it was announced that CNN+ would
shut down. Its corporate parent, Warner Bros. Discovery, concluded
that customers simply didn’t want to pay for another streaming
service, let alone one focused on news but lacking their favorite
prime-time CNN shows. Say goodbye to $300 million.

WHICH PHONE TO LAUNCH?

Even digital-native businesses can fall prey to overreliance on
traditional planning, especially when innovation decisions are driven
by a charismatic CEO. When Amazon decided it should enter the
smartphone market to compete with Apple’s iPhone and Google’s
Android business, two ideas were developed. One, code-named
Otus, would be a low-cost handset using the same software as
Amazon’s popular Fire tablets. The other, code-named Tyto, would
be a high-end smartphone with a 3D display that allowed users to
control their phone by gesture. Which idea would succeed in the
market? No one knew, but apparently a test was never conducted.
Otus was shelved and Tyto planned for launch based on the
founder’s decision. Reportedly enthralled with its 3D display, Jeff
Bezos maintained a personal commitment to Tyto even as the
screen posed huge technical challenges and the project dragged on
for four years—with 1,000 employees and more than $100 million
devoted to development. By the time Bezos himself unveiled the
product, dubbed the Fire Phone, the smartphone market had
evolved considerably. Amazon’s phone was out of touch with
customer demand and too expensive for what it offered (it lacked
compatibility with popular apps like Gmail and YouTube). In a year
when new iPhones sold 4 million units in their first twenty-four hours,
the Fire Phone sold just tens of thousands of units in its first six
weeks.9 The product was heavily discounted and cancelled soon
after.
How do digital start-ups like Airbnb and Netflix innovate
successfully in the face of such uncertainty? How do venture
capitalists invest in these kinds of ventures from the start without
going bankrupt? And how do large firms like Google and Apple
continue to innovate new business models? We will see the solution
to the challenge of uncertainty shortly. But first, let’s take a look at
another challenge.

Innovating Beyond the Core

The second challenge for innovation in an established company is

managing innovation outside its core business. (This is a challenge
not faced by start-ups.) I call this the challenge of proximity.
Managing growth beyond the core is one of the key difficulties that
executives tell me they struggle with amid digital change. Once a
company becomes successful, there is an inherent tension between
growing the existing business model and pursuing innovations that
stretch beyond it. In too many firms, this tension is always resolved
in favor of staying close to the core.
But for any organization to grow in the rapidly changing digital
era, it is essential that DX include innovation both in and beyond the
core. New ventures beyond the core may serve different customers
than the current business does. They may earn revenue in different
ways, and they may carry a different cost structure. They may
require new capabilities and partnerships. All these differences mean
that innovation beyond the core requires the organization to learn
and adapt to new management challenges—and this work is
essential. Every business needs to place bets on innovative ventures
outside its core if it hopes to achieve sustainable long-term growth.
As its existing business model matures and loses the potential for
growth, innovation of this kind is critical.
 The Lessons of Amazon Web Services (AWS)

Amazon poses a powerful example of the importance of innovation

beyond your core business. The company began with a retail
business model, taking orders online and shipping physical goods to
consumers. This core business started with books and expanded
into other product categories. Amazon extended its business model
by adding a marketplace to include third-party sellers on its website.
And it added a hugely successful retail membership model, Amazon
Prime, based on free shipping. But Amazon also pursued growth far
beyond its core business of retailing—including streaming media
(music, TV, and movies) and electronic devices (Kindle e-readers,
smart home devices, and its Alexa assistant). Most important, its
innovation beyond the core included Amazon Web Services (AWS).
The idea for AWS was floated in a 2003 paper by network
engineer Benjamin Black and his boss Chris Pinkham. The paper
proposed building a new computing architecture for the Amazon.com
website that would be more flexible, stable, and scalable. At the end
of the paper, Black and Pinkham added the suggestion that the
company could even sell the use of this architecture to other
companies as a B2B cloud-computing service.10
That idea, to enter the cloud-computing market, was far from
Amazon’s core business. At the time, Amazon was a purely
consumer-facing retail business. Cloud computing was radically
different. It would serve a completely different customer (enterprises
rather than shoppers), require a completely different sales process,
and interface with a completely different ecosystem of IT partners.
Bezos approved work on the new venture, and over time AWS
became not just a hugely successful venture but the most profitable
part of Amazon. By the time the company publicly broke out the
unit’s financials for the first time in 2015, AWS was already a $5
billion business, which stunned investors.11 Six years later, with AWS
representing 63 percent of all company profits, its head, Andy Jassy,
succeeded Bezos as Amazon CEO.12
Other digital-era businesses have thrived by innovating beyond
their core. Think of Apple, which started as a computer maker and
extended into other hardware businesses—MP3 players,
smartphones, tablets, and watches. But Apple grew beyond its core
of hardware. It enriched its ecosystem and its profits by venturing
into the sale of music, games, and apps. It launched its own
television and movie studio. It even created a data-driven fitness
service.
We will see the same pattern in successful examples of DX
throughout this book. The New York Times Company’s digital
innovation includes apps and emails for its news subscription but
also includes ventures in podcasting, live events, and digital
subscriptions beyond news. Walmart’s transformation has extended
its physical retail into the world of e-commerce, but it also includes
data-driven ventures in health care and financial services. And
Mastercard has not simply extended its core credit card business
into digital wallets and online payments; it has found new growth in
cybersecurity and digital identity services.

Struggling Beyond the Core

Despite these examples, most established businesses find it

extremely hard to pursue innovation beyond their core. This is
almost inevitable. Incumbents are designed and managed for the
primary goal of optimizing their current business model. As a result,
employees who come up with great ideas outside the current
business may find it impossible to bring those ideas to market.
Consider these famous examples:

• Cisco—Cisco was already a market leader in videoconferencing

technology with its Webex product when employee Eric Yuan
proposed innovating in a new direction. To complement its
success in the enterprise market, Yuan conceived of plans for a
more consumer-oriented videoconferencing tool. But within
Cisco, this idea was shot down as too far afield from the
enterprise focus of the core. Frustrated, Yuan quit to start his
own company and brought his innovation to the market as
Zoom.13 When the COVID-19 pandemic sparked global
 quarantines and a surge in demand for videoconferencing, it
could have been an enormous opportunity for Cisco. Instead,
the company watched as users flock to the previously little-
known Zoom platform. Daily users of Zoom rose to 200 million
from just 10 million the year before.14 Overnight, Webex
became an also-ran in its own category.
• IBM—In 1999, IBM’s strategy group reported to their CEO, Lou
Gerstner, that the firm had developed 29 recent breakthrough
technologies—in areas such as speech recognition, radio
frequency identification (RFID), business intelligence, and the
first internet router—each of which IBM had failed to
commercialize because executives were focused on serving
their current markets and were rewarded for delivering on short-
term, predictable results. Instead, second-mover companies like
Nuance, Akamai, and Cisco captured huge markets in areas
that could have been IBM’s to dominate.15
• Xerox—When Xerox founded its legendary lab Xerox PARC, it
assembled many of the world’s greatest computer scientists and
engineers. Their work produced innovations close to the
company’s core business (photocopying), such as the laser
printer, which was a commercial success for the company. But
the greatest innovation by far to come out of PARC was the
Xerox Alto, the world’s first computer with a graphical user
interface (GUI). If you’ve never heard of it, that’s because Xerox
left the Alto languishing in the lab. The company failed to realize
the Alto’s commercial potential because it was so far from
Xerox’s core business. It was Steve Jobs who saw the Alto on a
visit to PARC, recognized it as a breakthrough, and brought GUI
to the market with his Macintosh computer, followed soon by
Microsoft’s Windows operating system. Jobs later remarked, “If
Xerox had known what it had and had taken advantage of its
real opportunities [it could have become] the largest high-
technology company in the world.”16

Why do established companies fail to invest in their own

employees’ innovations and fail to bring them to market? There are
many reasons why established companies find it hard to innovate
beyond their core:

• Organizational structure—Growth opportunities outside the core

do not fit easily into existing business units, so there is no power
center in the company to support and sponsor them effectively.
• Metrics—Established business metrics are often poorly suited to
judge the success of a new business model. As a result, the
opportunity appears unattractive, risky, or too small to merit
attention.
• Resources—In a typical organization, executives who generate
the most revenue today control the investment of resources for
the future. Ideas that are irrelevant to their business units are
left with nothing but resource scraps.
• Customer focus—Companies are understandably focused on
their existing customers. Clayton Christensen identified this
paradox: most companies fail to invest in disruptive innovations
because they are too focused on their current customers to see
the opportunity to serve a different market.17
• Cannibalization—Promising innovations are actively resisted
and even shut down if they are perceived as threats that could
cannibalize the core business that generates today’s profits.
• Narrow vision—Over time, success causes a company’s vision
of the future to be defined by the products of its past.
Businesses develop a narrow, backward-looking view famously
described as strategic myopia by Ted Levitt. Hollywood studios
that thought they were in the movie business rather than the
entertainment business disappeared when newer media arrived.
Railroad companies that failed to focus on the broader
transportation market met a dead end.18 Levitt saw that
successful companies failed during times of change because
they focused on the products they built instead of the problems
they solved.

The net result is a perennial challenge for established

businesses: the less proximity an innovation has to the core
business, the harder it is for the company to manage it to growth.
We’ve now seen why so many established companies struggle to
innovate, faced with the twin challenges of uncertainty and proximity.
But we have also seen examples of firms that overcame both these
obstacles to achieve tremendous growth. Let’s look at the lessons
from the digital era to see how any business can solve the
challenges of uncertainty and proximity.

Solving for Uncertainty

Innovating under great uncertainty is possible. This much is obvious

from the growth of incredibly valuable digital businesses started
around the world with venture capital (VC) investment. Every one of
these businesses began under tremendous uncertainty because it
was inventing a new digital business model. Each succeeded by
using two levers available to every manager: experimentation and
iterative funding.

Experimentation: What Start-Ups Do

It is a Silicon Valley truism that great start-ups don’t start with a great
idea; they start with an idea and then test and pivot their way to what
works in the market. Instead of the traditional corporate reflex to
plan, plan, plan, and then build; start-ups succeed through a
constant process of experimentation. In The Digital Transformation
Playbook, I define experimentation as “an iterative process of
learning what does and does not work.”19 In both start-ups and the
scientific method, the aim of experimentation is the same: to validate
your key hypotheses or assumptions.
In a scientific experiment, you seek to validate whether condition
X will lead to outcome Y (e.g., if patients take a pill versus a placebo,
will their health improve?). To develop a single medicine, many
hypotheses must be validated. Does the pill speed recovery? By how
much? What dosage is required? How quickly does it take effect?
Are there side effects?
 In business experimentation, your goal is to validate the key
assumptions in your business model for any new venture.
Assumptions to test include: Who is the customer? What is their
problem or unmet need? Are they interested in our proposed
solution? How much would they pay for it? When, where, and how
will they use it? How would we deliver it? What profit margin could
we make? What’s the total possible size of the market?
Experimentation is the defining philosophy of digital-native
businesses. If you spend time with fast-growing start-ups or with
titans like Google and Amazon, you will quickly pick up a whole
different vocabulary around innovation: “MVP,” “fail fast,” “design
sprint,” “pivot,” “lean metrics,” “agile squads,” “product thinking,” and
so on. Any business leader who aims to master the art of innovation
today should be aware that these terms and practices derive from
four powerful approaches to innovation that have arisen in the digital
era. These four schools of thought are lean start-up, agile software
development, design thinking, and product management. I think of
these as the Four Religions of Iterative Innovation because each has
its own rituals and passionate adherents, yet they all share the same
core principles of managing innovation through iterative
experimentation. To learn about each, see the box “The Four
Religions of Iterative Innovation.”

The Four Religions of Iterative Innovation

Despite their differences, the Four Religions of Iterative Innovation

were each developed to address the same management
challenge: how to innovate and solve problems under tremendous
uncertainty. Each one tackles the challenge of uncertainty with an
approach that is designed around learning rather than planning.
And each envisions learning as a highly iterative process that
seeks to spend less upfront, build something tangible as early as
possible, and use that to validate assumptions and reduce
uncertainty in rapid fashion.
 Lean start-up, also known as customer development, arose in
the world of Silicon Valley start-ups. Its goal was to replace
traditional business planning when searching for new business
models. The core ideas were formulated in the early 2000s by
Steve Blank and his coauthor Bob Dorf and extended by Blank’s
student Eric Ries. Blank defines a start-up as “a temporary
organization designed to search for a repeatable business
model.”* This search is conducted by learning directly from
customers with interviews and MVPs designed to capture
customer feedback through rapid experimentation.
Agile software development, often shortened to agile,
originated in the world of IT developers. It was formulated in
response to traditional software development practices inside
large organizations that were too slow, costly, and inflexible. Agile
is comprised of several different methodologies, including Scrum,
Kanban, XP, and others, whose founders met in 2001 at a Utah
ski lodge and produced a shared manifesto on improving the
practice of software development.† Guiding principles include self-
directed teams, a focus on customer needs, rapid iteration cycles,
and the continuous delivery of software through incremental
deployments.
Design thinking originated with designers working for large
organizations; their goal was to bring principles of good design out
of the last phase of product development (What should our
product look like?) and into much earlier phases (What should the
product be? What problem is it trying to solve?). While rooted in
theories of human factors and creativity dating to the 1960s,
design thinking was popularized in the 2000s by design firms like
IDEO. Design thinking has no unifying text or framework; its core
principles are framing innovation around the needs of customers,
carefully studying the problem to be solved, and iterative
prototyping to bring in tangible feedback early and often. ‡ In
practice, it uses teams that combine expertise in different fields
such as industrial design, anthropology, and data science.
Product management is a term heard most often in large
digital-native companies (e.g., Alphabet, Meta) to describe how
they organize teams working on new ventures and ongoing
 products. It has no single methodology and typically applies many
tools of the other three religions (MVPs, design sprints, journey
maps, etc.). Individual companies develop their own tools as well:
Amazon product managers will write a “future press release” for
any proposed innovation to visualize the imagined customer
impact and work backward to begin development.§ Product
management is often contrasted with project management—a
traditional practice best suited when you know the deliverable you
need to build, and your goal is to meet a fixed timeline with fixed
resources.||

* Steven G. Blank and Bob Dorf, The Startup Owner’s Manual: The Step-By-Step Guide
for Building a Great Company (Pescadero, CA: K & S Ranch, 2012).
† Kent Beck et al., “Manifesto for Agile Software Development,” 2001,
http://agilemanifesto.org/. The history of the gathering that produced the manifesto is
recounted at Jim Highsmith, “History: The Agile Manifesto,” 2001,
http://agilemanifesto.org/history.html.
‡ Jon Kolko, “The Divisiveness of Design Thinking,” 2017,
https://www.jonkolko.com/writing/the-divisiveness-of-design-thinking. Kolko’s article is an
excellent look at the roots of design thinking, as well as a critique of its trivialization and
commercialization in some quarters.
§. Werner Vogels, “Working Backwards,” All Things Distributed, November 1, 2006,
https://www.allthingsdistributed.com/2006/11/working_backwards.html.
|| Kyle Evans, “Product Thinking vs. Project Thinking,” Medium, Product Coalition,
October 21, 2018, https://productcoalition.com/product-thinking-vs-project-thinking-
380692a2d4e.

Whatever methods it adopts or combines, every great digital

business has used experimentation to find its way to growth. It was
through constant testing and learning that Airbnb’s founders
validated their business model for a multisided market for travel
hospitality. From their very first test of their idea (in which they
themselves hosted guests on a weekend when San Francisco’s
hotels were fully booked) through countless iterations in one city
after another, they gradually validated what would draw users to the
site, what would earn the trust of both travelers and hosts, how the
company could comply with myriad local laws and regulations, and
how they could leverage network effects to fend off competition and
grow to a phenomenal scale.
Uncertainty in innovation can never be wished away or banished
by detailed case planning. Every digital winner knows that the only
way to reduce uncertainty is by learning directly through
experimentation in the market to convert hypotheses into facts.

Iterative Funding: What VCs Do

The second lever that managers can use to solve the challenge of
uncertainty is iterative funding. Digital start-ups use experimentation
to validate their path to growth; VC investors use iterative funding to
invest in those same start-ups. The whole point of VC investing is to
back new, highly uncertain ventures. VC firms handle the risk by
investing smaller amounts (at lower valuations) when a start-up is
new and its uncertainty is high, and increasing investment in later
rounds if the start-up is able to validate its business assumptions
(see figure 2.1).
Figure 2.1.
Typical VC investment stages

This approach is the exact opposite of how most established

businesses allocate capital. Recall that CNN spent a massive $300
million in its first year to launch CNN+, and it planned a similar
budget for each of the next three years. The traditional approach to
funding corporate innovation is to delay investment while completing
as much planning and analysis as possible. Then, if a top leader
decides, after much deliberation, to move ahead, the company
makes a large initial investment in hopes of increasing the odds of
success. Under digital-age conditions of high uncertainty, this
approach is doomed. Instead, to practice iterative funding,
businesses should follow four simple principles:

• Invest less when uncertainty is higher; invest more when

uncertainty is lower. If uncertainty is low (e.g., investing in
warehouses for your established e-commerce business), you
can afford to make a large investment from the beginning. But
the greater the uncertainty at the start of a project, the smaller
your initial investment should be. With new business models
and untested market ideas, you should start with as small an
investment as possible.
• Place more bets when uncertainty is high. In the early seed
stage, VC firms take stakes in many different start-ups. Later,
when uncertainty is reduced, they invest in a much smaller
number of ideas (those that have been validated through market
testing). If your company’s goal is to find a couple of big
innovations that move the needle on your entire business, start
by investing in a lot of high-uncertainty ideas, experiment, and
winnow them down based on what you learn.
• Know what you are spending for. When investing with low
uncertainty, you are spending money to make a financial return.
You goal is to “spend to earn.” But when you invest under high
uncertainty, your goal should be to “spend to learn.” Early
investments should not be directed toward launching a product
but to conduct small, cheap tests that reduce uncertainty. Even
 if you think you understand the customers’ needs and their
problem to be solved, your first goal is to test and validate.
• Accelerate spending as uncertainty diminishes. If
experimentation reduces the uncertainty of an idea, you should
ramp up quickly to much larger investments, for example, to
prepare a product for market launch. But be careful. Investment
increases should happen only at learning milestones, that is,
when critical business hypotheses are validated. Do not release
more money just because the calendar has rolled forward on
your budgeting.

When Facebook received its first seed round of funding led by

investor Peter Thiel, the service was just four months old, had no
revenue, and had been tested in a handful of college campuses.
Thiel invested $500,001 for a 10.2 percent stake, which gave the
company a valuation of $4.9 million. The following year, Facebook
had surpassed 1 million users and was generating revenue from
major advertisers like Mastercard. In its Series A round, Facebook
received $12.7 million and was valued at $98 million. The next year,
as Facebook spread to colleges, high schools, and corporate
campuses, its Series B round valued the company at $500 million.
The following year, when Facebook had launched on mobile, started
its Newsfeed, and opened to app developers, Microsoft led its Series
C round, valuing Facebook at $15 billion. The core idea of Facebook
did not change between its launch and its Series C round. But the
market value of that idea went from $5 million to $15 billion—even
though the company would still not turn a profit for two more years.
What changed between 2004 and 2007? The uncertainty of
Facebook’s business model.

Maximum Upside, Minimum Downside

The harsh truth of innovation is that most ideas that sound good on a
sketchpad turn out to be unworkable in the real world. Innovating
under uncertainty can work only with the use of both experimentation
and iterative funding. The two together will not give you better ideas
or a crystal ball to see the future. (You will still dream up innovations,
like a 3D smartphone, that seem amazing to you but will fail in the
real world.) What they will do is capture the upside of every idea that
works while minimizing the downside of ideas that don’t work.
The problem with CNN+ was not that it failed but that it cost $300
million to fail! It was actually a good idea to test, but it could have
been tested with $30,000. By combining the two levers of
experimentation and iterative funding, any business can master
innovation under uncertainty—cutting its losses early on ideas that
prove infeasible and doubling down on those with the highest chance
of success.
In the five steps of the DX Roadmap, we will learn how to tackle
uncertainty with tools and frameworks that can be applied by any
organization. We will introduce the Four Stages of Validation used to
test the assumptions behind any business model. We will apply them
with a visual tool to guide experimentation of your next innovation
from notepad sketch to global execution. We will also learn how to
apply iterative funding, drawing on pools of dedicated resources and
allocating them quickly and flexibly while using smart shutdowns to
avoid wasteful disasters.

Solving for Proximity

As we have seen, the other innovation challenge faced by every

established business is proximity—how to manage innovation far
from its core business. This is hard for many reasons, including
structural barriers (metrics, resources, and organizational design) as
well as mindset (vision, focus, and fear). Despite this, we see mature
digital firms, like Amazon, Apple, and Alibaba, and legacy
businesses, like Walmart, Mastercard and the New York Times,
pursuing innovation successfully beyond their core. As we explore
the DX Roadmap, I will provide more examples of companies that
are conquering the challenge of proximity. You can do the same in
your business by following a few essential principles:
• Focus on problems that play to your strengths. How does a
business find growth ideas outside its core and escape the trap
of strategic myopia? The answer is to focus on customer
problems to be solved rather than your current products or
industry. But once you broaden your lens in this way, how far
should you stretch beyond your current business? If Ford
Motors decides it is no longer a car company but rather a
mobility company, should it now build airplanes? Not
necessarily. Effective strategy looks not just at opportunities but
at where you hold a unique advantage over competitors. How
did Alibaba, as an online retail business, decide to move into
digital payments? It started with a customer problem to solve: its
own customers needed ways to pay each other on its
marketplace. But Alibaba also had a unique ability to solve this
problem by leveraging its scale, existing customer base, and
rich user data.
• Make it independent. Many opportunities for digital innovation
should be managed in your existing business units because
they directly involve your core business and cannot be pursued
outside it (think of Domino’s Pizza reinventing the delivery
experience of its restaurants). But when you pursue a new
venture that diverges from your core business (with different
customers, revenue, cost structure, or capabilities), that venture
should be managed independently from the core—at least in its
early stages. This independence may even include physical
distance, to escape interference and meddling. When Amazon
decided to pursue its idea for AWS, it made sense to run the
project independently from the e-commerce division. AWS was
started as an independent group of fifty-seven people led by
Jassy, with the majority of its hires from outside Amazon.20 A
team led by Pinkham set up shop in Cape Town, South Africa, to
build EC2, one of two initial AWS products.21 (That’s about as
far from Seattle headquarters as possible!)
• Keep an umbilical cord. As important as independence is, a
growth venture beyond the core business should never be left to
sink or swim entirely on its own. These ventures still need
 access to critical assets and resources from the parent firm. In
the early days of the Web, British grocery chain Tesco launched
Tesco.com, the first grocery e-commerce business to succeed in
the United Kingdom. CEO Terry Leahy set up the team as a
distinct operating unit because he knew Tesco’s culture was
ferociously competitive—even internally. He saw a real risk of a
visceral organ rejection by the company of the new dot-com
business unit. But despite its independence, the new venture’s
relationship to the core (physical store) business was critical.
Employees of Tesco.com needed to pick products off the
shelves for speedy delivery to nearby customers. So, retail
stores were given “shadow accounting” metrics that attributed to
them part of the profits from the online business. The online
business also had access to the core’s data sets and customer
lists to mine for prospects, and it shared the same outside
marketing agency. Chris Reid, who led Tesco.com at the time,
explained to me that his unit was run in many ways like a start-
up but with “a tight umbilical cord to the main business.”
• Manage by different rules. New ventures outside the core need
more than just separation into distinct units. These units need to
be managed by different rules—even if they are housed on the
same corporate campus as the core. They need to operate with
their own budget and a dedicated funding source. They need
their own metrics for success; their own talent pool, with many
hired from outside the firm; and their own executive
sponsorship, reporting, and oversight. Alphabet has used X
(formerly Google X) as a “moonshot lab” to explore ventures far
from its core business—like driverless cars, supply chain
software, and health care. X was set up to have metrics,
financing, and leadership that are all different from the core
business of search and digital advertising. The New York Times
established its Standalone Products and Ventures Group to
develop and launch new digital products beyond the news
subscription. The Group was set up on a separate floor of the
Times’s headquarters, with each new app or product assigned a
cross-functional team that applies product management
methods and works in a single conference room.
 • Plan where it will land. Any time a team is set up to start a
venture outside the core business, its sponsors should think
ahead about where they expect it to land. If the venture
succeeds, it should eventually either merge with the core or
become its own permanent business unit—in some cases,
triggering a redesign of the whole organization. AWS was set up
as its own business unit before its public launch in 2006; it has
continued to grow rapidly and runs as a separate unit to this
day. Retailers like Tesco who started dot-com teams in the early
days of e-commerce ultimately converged on an omnichannel
strategy. In most companies, this meant reorganizing by product
category or customer type. Online and physical store teams
were combined and given unified metrics (like customer lifetime
value) that prioritized total value to the firm. In managing
ventures outside your core, start with a plan but be ready to
adapt as you go. Efforts that need to be separate at the start
may be better combined as the market and the organization
mature.

In the five steps of the DX Roadmap, I will introduce tools and

frameworks to manage growth beyond the core in any business. I
will provide tools that will enable you to focus your strategy on
problems to solve and to avoid the trap of strategic myopia. We will
see how to define a future landscape and your unique right to win
and how to use those perspectives to pick which opportunities to
pursue. And I will introduce a model called the Three Paths to
Growth to apply the right governance for ventures both within and
beyond your core.

No company will survive in the digital era if it ignores the challenges

of innovation. Digital transformation cannot be just low-risk
investments with clear benchmarks and best practices. It cannot be
only about upgrading and digitizing the core. Digital transformation is
first and foremost about using digital tools to solve new problems
and drive new growth. This is why the DX Roadmap takes a holistic,
growth-focused approach—one that is integrated with your work on
strategy, innovation, governance, and culture as much as with your
IT planning.
In the five chapters ahead, we will see how to link digital vision to
strategy and innovation, how to focus on customers and problems to
solve, how to experiment with new ventures while applying
governance to grow them to scale, and how to support all of this with
the right capabilities and culture.
The table is set. It’s time to begin.
 3
Step 1: Define a Shared Vision

At the 2011 TED conference in Long Beach, California, William “Bill”

Clay Ford Jr. strode onto the stage to speak about a vision of the future
of transportation. Bill Ford was the executive chair and recent CEO of
Ford Motor Company. He had spent his whole life and career in the
automotive industry; he was the great-grandson of the company’s
founder, Henry Ford. Yet he began his talk with a provocative question:
what if the future of Ford Motor was not simply to make and sell more
cars? The presentation that followed was not a typical TED Talk but the
outline of a defining corporate vision from the chair of a leading global
corporation.
Bill Ford defined two great challenges facing the automotive
industry: environmental impact and a looming mobility crisis. On the
first, Ford Motor was already making progress toward electrification of
engines and other technologies aimed at developing zero-emissions
vehicles. On the second challenge, Ford pointed to the increasing
stresses of urbanization; a rising global middle class; and the threat of
global gridlock, from Beijing to Abu Dhabi, that could threaten the
freedom of mobility that his great-grandfather had championed in
starting the business. The existing solution—selling more vehicles and
building more roads—was insufficient for a future population of 8 billion
people, 75 percent of whom would be living in cities. The implications
were a stifling of economic growth and the inability to deliver food,
health care, and essential services effectively to communities around
the world.
The response Bill Ford laid out was to invest in building a smart
network of vehicles, roads, fueling stations, and public transport using
data and digital technologies to connect every component and create
new integrated solutions around travel, parking, payment, and safety.
Pointing to examples of innovation from Hong Kong to Masdar, he
described a future of connected vehicles that talk to each other in an
“integrated system that uses real-time data to optimize personal
mobility on a massive scale.” He committed to focusing Ford Motor on
pursuing this vision not just as an automaker but as a digitally driven
mobility services company.1
This vision would go on to shape the strategy of Ford Motor for the
next decade and beyond, even as that vision continued to evolve and
be refined. The vision has led to billions of dollars in new digital
investments and acquisitions and has shaped the choice of the next
three CEOs. It continues to shape the strategy of Ford today as it
pushes ahead into the digital future.

Why a Shared Vision Matters

Change is easy . . . in a crisis. During the midst of the COVID-19
pandemic, Lucy Kueng, a scholar of the media industry, remarked that
COVID provided a once-in-a-lifetime opportunity for legacy companies
to transform. “Organizations are unfrozen. People are expecting
change . . . There will never be a better time to tackle deeper changes
that need to happen.”2 In the context of an immediate crisis, everyone
in your business realizes that the status quo is untenable—and so
everyone is primed for action and for setting aside old ways of working.
But in the vast majority of cases, you will not be in this kind of crisis.
(Thank goodness!) The rest of the time, transformation is very hard.
Change is resisted. The status quo is the default, and employees’ bias
is toward more of the same. The unspoken assumption of planning is
that the future will look like the past. Success breeds complacency—
why change what has been working? As one senior executive told me,
“Our biggest obstacle for the future is that we have been so successful
for the first 79 years of the company.”
To generate urgency for change during a digital transformation (DX),
consultants will commonly urge you to “find your burning platform.” The
expression comes from Nokia, from a memo that CEO Stephen Elop
wrote to rally the company as its mobile phone business was being
disrupted by Apple’s iPhone and cheaper Android smartphones.3 But
did Elop’s burning platform memo succeed? No. Nokia fell from its
leadership position and eventually exited mobile phones completely.
The real lesson from Nokia is that a burning platform is not enough.
Negative urgency alone—proclaiming the sky is falling and digital
disruptors are just around the corner—is not a motivating story for
employees or shareholders. One major Latin American retailer I
advised was extremely focused on the threat posed by Amazon and
told employees they needed to change to survive. But the message
ignored the retailer’s history—decades of growth driven by customer-
centricity and business innovation—and how that might point the way
forward in their digital future. Negative urgency is helpful only when it is
paired with positive urgency: a vision of how digital transformation can
create new value, unlock new growth, and solve new problems.
As you craft a positive story of value creation, be mindful that
shareholder value is not enough. Explaining how your digital
transformation will drive new efficiencies and raise earnings per share
may inspire investors. But quarterly profits alone will not inspire
employees to take on the uncertainty and hard work of transforming
your organization. To inspire employees, leaders must show the impact
and meaning of their work.
To drive change, leaders must appeal to extrinsic motivation (ROI;
profits; cost savings; and earnings before interest, taxes, depreciation,
and amortization [EBITDA]), which motivates shareholders. But leaders
must also appeal to intrinsic motivation (value created for customers,
employees, partners, and society), which motivates employees.4 When
Bill Ford laid out his case for the digital future of Ford Motor, he gave it
intrinsic motivation by linking it to the company’s founding: “My great
grandfather, Henry Ford, really believed that the mission of the Ford
Motor Company was to make people’s lives better and make cars
affordable so that everyone could have them. Because he believed that
with mobility comes freedom and progress.”5 Bill Ford framed his vision
of a connected transportation future as a way to safeguard core values
of freedom and mobility for the next generation.
The first step of the Digital Transformation (DX) Roadmap is to
define a shared vision of the future of your business, one that will
inspire, align, and make the case for change. This vision must be
shared; that is, it must be known by everyone in the organization. And it
must be unique to your business. A generic commitment to “become
digital-first,” to “be recognized as a digital leader,” or to “future-proof our
business” will not suffice. Every business needs a vision that is specific
—that describes where your world is going, what role your organization
will play, and why.
The importance of a shared vision is well-established. Daniel
Goleman’s classic research on leadership found that the most powerful
of six leadership styles is when leaders “mobilize people toward a
vision.”6 More recently, McKinsey’s research has found that the biggest
single factor for DX success is having a “clear change story.”7
It is, therefore, shocking how many companies lack this kind of
shared vision. In my workshops with large organizations, participants
have consistently chosen this as one of their top barriers: “[we have] no
shared vision of our company’s digital future.” The incoming CEO of an
insurance firm recently shared with me, “Honestly, if you ask three
people what’s our vision for digital transformation in this business,
you’d get three different answers.” In a discussion with fifty executives
from a large biopharmaceutical firm, I was told that their CEO had
announced that they would become a “digital-first” company and had
hired a chief digital officer (CDO), just like their peers. But when I
pressed these divisional leaders on what digital meant for their
business, they confessed they were mystified and had hoped I would
explain it.
Defining a shared vision of the future is not easy. But without a clear
and compelling answer to the question, “Why must we change?,” any
DX will stumble. Table 3.1 shows some of the key symptoms of
success versus failure in the first step of the DX Roadmap.

Table 3.1.
What’s at Stake—Step 1: Vision
Symptoms of Failure: Symptoms of Success: Vision
Vision
• Employees fear change • Employees at every level
and lack a clear sense of understand the digital agenda
where the firm is going. and push it forward.
• Backing for digital • Support for digital
investments is weak from investments is strong from
investors, CFOs, and P&L investors, CFOs, and P&L
heads. heads.
• Digital initiatives are • Only digital initiatives with a
generic, following the competitive advantage receive
examples of peers. investment.
• Generic digital maturity • The business impact of digital
metrics are used to guide is clearly defined, with metrics
efforts. to measure and track results.
• The firm follows the • The firm leads the market,
market, reacts to alert to critical trends while
others, and is there is time to choose a
surprised by new course.
entrants.

What’s Ahead
In this chapter, we will see how any leader can define a shared vision
for the DX efforts of their organization, business unit, function, or team.
We will examine four essential elements of a strong shared vision.
Each element seeks to answer a different question:

• Future landscape—where do you see your world and your

business context going?
• Right to win—what are the unique strengths and limits of your
organization that will define the role you play?
• North Star impact—what impact do you seek to achieve in the long
term, and why?
• Business theory—how do you expect to capture value and recover
the investments you make for the future?

After examining each of these four elements and a range of case

studies, we will introduce a strategic planning tool, the Shared Vision
Map. This tool will help you to define each element of your shared
vision for the DX you are leading. Finally, we will learn why defining a
shared vision is so critical to making the shift from a top-down to a
bottom-up organization.

Future Landscape

The first element of a shared vision for your business is what I call your
future landscape, which is a description of where your world is going
and how the context of your business is changing. As Microsoft CEO
Satya Nadella describes it, a leader’s job is to “see the external
opportunities, and the internal capability and culture, and all of the
connections among them.” He goes on to explain, “It’s an art form, not
a science. And a leader will not always get it right. But the batting
average for how well a leader does this is going to define his or her
longevity in business.”8
Your future landscape should capture the most significant shifts in
the world of your customers, partners, and competition—and the
threats and opportunities that these pose for your business. Are you a
legacy business in a highly regulated industry (e.g., finance or health
care) where regulations are shifting, and new entrants are appearing
with digital-first business models that seek to “unbundle” your business
by offering a subset of the services you provide but without the level of
regulatory burden you face? Are you an industrial business that has
sunk huge investments into fixed assets (e.g., oil wells, shipping fleets,
or telecom networks) that used to provide barriers to entry by
competitors, but new start-ups are appearing that are happy to sit atop
your assets and build new business models based on data, predictive
analytics, and systems integration? You might be a professional
services firm (e.g., in advertising, auditing, or human resources) whose
business has been built on deep relationships with your business
clients. But the problems you used to solve for them are increasingly
handled by algorithms and AI—forcing you to keep reinventing your
business around new client needs, redefining the services you sell, and
transforming the talent of your workforce to keep up.
Understanding your future landscape—where your particular world
is going—will be critical to defining where and how you aim to compete
in the digital future, what markets you will serve, and where you will
hunt for growth.

Scanning the Future

As Sami Hassanyeh, CDO of AARP, one of the largest nonprofit

organizations in the United States, said to me, “The job of a CDO is to
ensure that culturally, technologically and strategically, your
organization is living up to the ever-changing demands of the
customer.” To keep up with change, leaders of every organization must
constantly be engaged in a process of scanning the environment, both
inside and outside their immediate industry. Any business must learn
from looking far beyond its traditional peers, competitors, and
customers if it hopes to understand its digital future.
In developing your future landscape, I advocate focusing on four
broad areas:

• Customers—Focus on understanding the changing behaviors,

expectations, and needs of your customers. As you do so, be
aware of influences outside your industry. Even if you sell
machinery to purchasing departments of large companies, the
 individuals in those departments are using Amazon, Google, and
Netflix every day, and their expectations (for customization, speed,
self-service experiences, and more) are shifting rapidly. You will
need to keep up! As you study the future of your customers, focus
both on your existing legacy customers (who built your business
and may still deliver the majority of current revenue) and on new
emerging customers (who will provide your growth and future
revenue). Very often, their behaviors and expectations are
markedly different and will require tailored strategies to manage a
split market.
• Technology—Identify and learn all you can about new technologies
that are shaping the experience of your customers, especially
technologies they use to discover, purchase, and use services like
yours. But focus also on technologies impacting your internal
business operations and the operations of your partners, supply
chain, and adjacent industries. As you study the technology
landscape, be sure to take a broad view of time and tech maturity.
Start with technologies that are actively being used now: How is
your manufacturing partner upgrading its factories? What social
media are your customers using? Then look to technologies that
are just starting to be used by a handful of early adopters. Tracking
start-ups in your industry is a great way to see new applications of
the latest technology. Last, keep an eye on research labs and
venture capital investing in technology not yet on the market to
spot the long-term trends that could reshape your industry.
• Competition—Take a broad view of competitors and partners as
you seek to understand the business ecosystem that will define
your future. Who are the new entrants in your industry? What new
products and new business models are being tested? What is the
pace of change and the level of competitive threat you face in
different parts of your business? As you analyze your future
competition, be sure to consider both symmetric competitors (your
traditional peers, with the same business model as you) and
asymmetric competitors (companies with a competing value
proposition but a different business model). In many cases,
asymmetric competitors will include your own closest business
partners (see figure 3.1).
Figure 3.1.
Symmetric versus asymmetric competitors for an automotive manufacturer
• Structural trends—Be sure to study major trends in the external
environment beyond business that may shape your future context.
Look at demographic trends—including population growth,
generational differences, urbanization, and the aging of
populations. Examine macroeconomic trends—including economic
development, globalization, and resource availability—and
government trends—including laws, regulations, tariffs, and
investments. Finally, look at any other trends—social,
environmental, health, climate, geopolitical—that you think may
have an impact on your business. If your business operates in
multiple markets, be sure to look for differences between them and
identify which trends have the biggest potential impact in each.

Your insights into all four of these areas should come from sources
inside and outside your firm—conversations with customers, meetings
with new start-ups and your existing partners, research reports, and
regular updates on the perspectives of frontline employees.

Tools for Scanning the Future

Four tools from my last book, The Digital Transformation Playbook, will
be helpful as you work to develop your future landscape.
The Value Proposition Roadmap (in chapter 6 of the Playbook) is an
indispensable tool for distilling knowledge about your customers’
evolving needs. First, this tool helps you segment your customers
based on their needs and the value they currently receive from your
business. Next, you examine each value element you deliver (are any
being replaced or disrupted?), and survey new technologies and trends
to identify new possible value elements. Finally, the tool guides you to
define a new value proposition to meet the future needs of each
customer segment.
The Competitive Value Train (in chapter 3 of the Playbook) helps
you understand your competitive ecosystem—the interplay of
competition and cooperation between your firm and the businesses
around you. I have used this tool with numerous executives to analyze
new digital entrants to their industries (insurance, retail, consumer
goods, etc.) and to understand the mix of competitive threat versus
collaborative opportunity that they pose.
In chapter 8 of the Playbook, you will find two tools for assessing
the threat of digital disruption to your business. The first tool, the
Disruptive Business Model Map, helps you determine whether a new
business model truly poses a disruptive threat by examining two sides
of each business model (the value proposition and the value network).
The second tool, the Disruptive Response Planner, helps you assess
your options in the face of a disruptive challenger. The tool examines
your disruptor on three dimensions to help you identify which of six
strategic responses is your best choice.
All these tools define digital disruption as a call to innovate and lead
toward your future. As Chris Reid, now an executive vice president
(EVP) at Mastercard, told me, “I think disruption oftentimes is a good
thing, as long as you’re agile, as long as you can predict it. You should
be able to figure out how do you either navigate around it, or how do
you take advantage of it.”
A clear view of your future landscape comes from continually
educating yourself about what is happening in the lives of your
customers, in the evolving landscape of digital technology, and among
other businesses that you must compete and cooperate with to
succeed.
 What a Future Landscape Looks Like

So what should a future landscape look like? The output from the work
above may be highly detailed or extremely succinct. What is important
is that it provides actionable guidance for you to move forward as you
think about the direction your business unit or company will take and
the opportunities it will pursue. The next sections describe three
examples: Merck Animal Health, BSH Home Appliances, and Acuity
Insurance.

MERCK ANIMAL HEALTH

At Merck Animal Health (known as MSD Animal Health outside the
United States and Canada), global CMO Fernando Riaza is focused on
understanding the future of a business that serves both pet owners and
agriculture companies raising livestock (beef, dairy, poultry, swine, and
fish). He sees dramatic changes in the consumer side of the business
as pets become integrated into our digital lives and equally big changes
in the agricultural side as well. The customers’ “path to purchase” for
animal medications is changing dramatically, relying less on buying
through veterinarians and more on direct digital sales. Rather than the
vet making the decision, customers are seeking their own digital
content and tools for decision making. Future opportunities are also
being shaped by technology trends in the agriculture industry, where
digital sensors are providing more and more real-time data for dairy
and meat businesses.

BSH HOME APPLIANCES

BSH Home Appliances is one of the world’s leading providers of home
appliances for cooking, dishwashing, laundry, and refrigeration; it sells
under the brand names Bosch, Siemens, Gaggenau, Thermador, Neff,
and others. The company’s future landscape starts with understanding
the changing lives of digitally connected consumers and how they use
apps, data, and smartphone-enabled services to meet their basic
needs around food and clothing. As a result, BSH is not only tracking
the digitally enabled products developed by competitor appliance
manufacturers (i.e., the latest digital fridge). It is also watching new
entrants like Uber Eats (for food delivery), recipe apps, online
influencers, and laundry service start-ups. This view of the future is
informing BSH’s expansion from appliances to new digital solutions, to
the cooking and cleaning needs of tomorrow’s customers.

ACUITY INSURANCE
I had the chance to work in depth with Ben Salzmann, CEO of Acuity
Insurance, as his leadership team developed a future landscape for
their property and casualty insurance business. Acuity operates in the
B2C market (personal lines) selling home and vehicle insurance to
individuals. But they also operate in the B2B market (commercial lines),
selling more complex insurance policies for small and midsize
businesses in sectors like construction, trucking, and manufacturing.
On the B2C side, Acuity sees a generational divide, with younger
customers seeking more convenience, customization, and self-service
options through their phones. On the B2B side, new customers include
growing sectors like eldercare and gig economy workers, and long-
standing customers are increasingly looking to Acuity for insights and
advice on their industry.
Technology trends are moving fast as well. Customers today expect
service across mobile, web, chat, and social media, and to be able to
file and track a claim with a swipe and a click on a smartphone app.
Popular home sensors like Nest and Ring are providing new sources of
data for property insurance. Robotics and automation are being
adopted by many of Acuity’s business customers—which has an impact
on which assets they need insured, as well as the worker’s
compensation insurance required for an increasingly skilled workforce.
Meanwhile, Acuity is tracking the development of autonomous vehicles.
In the future, if full autonomy becomes widespread, it will dramatically
change the insurance needs for vehicles—whether they are owned by
families, trucking companies, or other businesses.
The competitive landscape is shifting for Acuity as well. The firm is
working with many new partners, from digital ad platforms to online
marketplaces, to providers of data for underwriting and risk
assessment. Its industry has attracted VC investment in a host of new
start-ups with widely varying business models. We used the
Competitive Value Train to analyze over 100 new digital entrants into a
few categories—including aggregators, virtual agencies, value-added
partners, and digital direct writers—and to define the competitive threat
and opportunity for each. We also identified potential future threats,
such as the “manufacturer as insurer” business model.
Although the excitement of investors in the insurtech sector is
intense, it became clear that not all of these firms had a viable path to
profitability. Acuity found that its B2C business faces more risk than its
B2B business (where the complexity of insurance policies poses a
barrier to new entrants). And few insurtech firms are attempting to
disrupt insurers like Acuity (with most looking to partner with them
instead). The most immediate digital threat is to insurance agencies
that Acuity has historically partnered with. But in the long term, the
biggest threat to Acuity is if it fails to adapt fast enough to changing
customer needs.

Parmenides’ Fallacy

Perhaps the most important reason to describe your future landscape is

to be able to answer the question, What will happen if we do nothing?
This question is critical to avoiding what Philip Bobbitt calls
Parmenides’ fallacy: the mistake of comparing a new course of action
against your present state rather than what the future will look like if
you do nothing. (Parmenides’ fallacy is named after the Greek
philosopher who argued that all change was an illusion.)9 Too often,
decisions in large organizations are shaped by an unstated assumption
that the future will look like a continuation of the present. (One of the
more perverse arguments I have heard against DX is based on current
financial success: “We can’t afford to change; our EBITDA is too high!”)
Parmenides’ fallacy is so strong because of a series of well-known
cognitive biases: status quo bias, endowment effect, and omission bias.
Their net effect is that we tend to fear the risks of action while staying
blissfully blind to the risks of inaction. But in a time of rapid change, this
is a grave mistake. As John F. Kennedy is reported to have said, “There
are risks and costs to a program of action. But they are far less than the
long-range risks and costs of comfortable inaction.”10
 Right to Win

The second element of any shared vision is your right to win in the
future you see ahead. Why you? What gives your business a reason to
succeed in your future landscape? When Jim Hackett stepped into the
role of CEO for Ford Motor, he said that the biggest challenge facing
Ford’s employees was “to have everybody see the future . . . and
secondly, that it’s our right to win there. That we don’t have to cede that
to anybody. Tesla, or any of them. It’s our right to win.”11
To find your right to win, you will need to understand your unique
strengths as an organization, the qualities that set you apart from
others. You will also need to identify the most important limits or
constraints on your strategic choices. Where your future landscape
arises from external knowledge (via continuous learning about
customers, competitors, and forces outside your business), your right to
win will stem from deep internal knowledge about your own
organization.

Unique Advantages

Finding your right to win begins with knowing the distinct strengths of
your business to compete in the marketplace. What is your organization
uniquely good at? What assets or skills distinguish your business, add
value to your products or services, and give you a competitive edge? I
call these your unique advantages.
When Netflix filed for its initial public offering (IPO) in 2002, its
paperwork identified three unique advantages that distinguished the
firm: its subscriber base, its massive data sets on customers’ media
preferences, and its proven ability to deliver personalized
experiences.12 In the years since then, Netflix has leveraged these very
advantages as it tested a variety of business models to drive its
dramatic growth.
For Mastercard, its vast network of consumers and business
partners (both banks and merchants) is one key strength as it looks
toward the digital future. Another is its access to massive amounts of
economic data tied to individual commercial transactions.
 As Walmart focuses on growth in the digital era, it starts by looking
at its unique advantages. First among these is its store network. As
COO Jeff Shotts told me, “The biggest asset we have is 4,700 stores in
the US that are within 10 miles of 90 percent of the population in
America.” Although built for the company’s core retail business,
Walmart’s stores are now being leveraged for its online business too—
thanks to the 150,000 products, on average, in each store, ready to be
delivered to nearby customers. Another powerful asset is Walmart’s
vast data set of customer buying patterns, from 150 million shoppers in
its stores each week.
The Canadian Automobile Association (CAA) faces daunting
challenges to reinvent its business model given that many of the
traditional services provided to its members can now be found for free
via smartphone apps like Waze, Google Maps, and Uber. However, as
their President Tim Shearman explained to me, CAA still retains some
powerful strategic assets: a membership of 36 million; a treasure trove
of data from transactions those members make using CAA’s loyalty
program with a range of other businesses; and the CAA brand, which
was rated as the most trusted brand in Canada across every industry.
These are precisely the kinds of advantages that can be leveraged in
digital business model innovation.
Table 3.2 lists the unique advantages of all four companies: Netflix,
Mastercard, Walmart, and CAA.

Table 3.2.
Unique Advantages of Select Firms
Company Unique Advantages
Netflix • Subscriber base
• Data on customer media preferences
• Ability to deliver personalized
experiences
Mastercard • Network of business partners (banks
and merchants)
• Network of consumers (card users)
• Unparalleled amount of
actionable economic data from
transactions
Walmart • Retail stores
• Proximity to customers
• Data on buying patterns of 150
million shoppers
Canadian • Membership base
Automobile
Association (CAA)
• Data on transactions via CAA’s
loyalty programs
• Most trusted brand in the country,
across any industry

Every organization has its own unique strengths and advantages;

otherwise, it would no longer be in business. These can range from
physical assets to patents, technology, data, customer relationships,
brand reputation, strategic partnerships, employee skills, and more. As
you identify your unique advantages, it is essential that you take a
critical and skeptical eye. Whenever an advantage is proposed for your
list, ask yourself, how unique is this really? (Are we truly unparalleled in
this dimension? Or are we in the top 20 percent of our peers? Or in the
top 50 percent?) Next, ask, What is the competitive benefit that this
affords to our business? (Does this advantage reduce our costs? Does
it make our customers more loyal? Does it help us draw the best talent
from others?) Without an honest assessment of these two questions, I
have seen the discussion of unique advantage turn into a laundry list of
things that an organization is quite good at but that are simply “table
stakes” to operate a modern firm in their industry. These kinds of
qualities are certainly necessary, but they will never differentiate you
from the competition.

Putting Unique Advantages to Work

Knowing your unique advantages is critical because effective strategy

is not about spotting opportunities that could work for any business. It is
about finding opportunities that play to your strengths, where you will
have an “unfair” advantage over other firms who might attempt the
same thing. These opportunities are not just generically good ideas that
any business should pursue. They are opportunities where you have a
true right to win.
To leverage your unique advantages, look for how they may help
you excel in creating value for customers. Just as important, find where
they will help you capture value—achieving more profits than a
competitor would from the same business model.
As part of its ambition to become a fintech company with services
beyond credit card payments, Mastercard has pursued a variety of
digital strategies. Each is designed to take advantage of Mastercard’s
unique strengths as a global hub for commercial transactions. One
growth area has been in retail data analytics and insights. Mastercard
has launched a service that allows businesses to tap into
geographically mapped data on retail purchase patterns at the level of
a city block. (Users have included banks, real estate developers, and
even public policymakers looking to identify “food deserts” for
intervention.) Another new strategy for Mastercard has been
cybersecurity and digital identity authentication—using a range of new
technologies (geolocation, biometrics, and more) to verify identity and
thereby enable secure digital payments. Both its retail analytics and
digital identity efforts are advantaged by Mastercard’s existing
strengths—its unparalleled network of business partners and its access
to transactional data.
Another example comes from Intuit. When Rania Succar was
recruited to work at the firm, it was already a market leader providing
accounting, payroll, and payment tools to small and midsize enterprises
under the QuickBooks brand. Succar was tasked with pursuing a new
strategic opportunity for Intuit: providing credit for small businesses.
The venture, she explained to me, was rooted in a unique advantage:
“We believed at Intuit that we could solve small business lending in a
way that no one else has. We have access to more data about small
businesses—through what they put in QuickBooks—than anyone. And
the hypothesis was that we could leverage that data on behalf of small
businesses, to build best-in-class underwriting models and be able to
serve small businesses profitably, and dramatically increase the
access-to-capital rate.” After experimentation and validation, the
product launched as QuickBooks Capital, leveraging 26 billion data
points to train an algorithm that could offer loans of up to $100,000 to
small businesses in forty-nine states. Customers were delighted, with
90 percent saying it had a direct impact on the growth of their business.

RECIPROCAL ADVANTAGE
The most powerful strategies don’t just leverage the existing
advantages of the business to pursue new strategies. They pursue
what I call a strategy of reciprocal advantage—as one business model
succeeds, it both leverages and grows the core assets of the others, in
a virtuous cycle (see figure 3.2). This is much more than a simple
conglomerate that combines unconnected businesses (e.g., Samsung
Group operating in electronics, financial services, and shipbuilding) or
is vertically integrated within a single supply chain (e.g., an automaker
buying a tire company). Instead, a reciprocal advantage organization
combines business models that support each other’s growth by sharing
and improving the strategic assets that link each business (e.g., data,
algorithms, and customers).

Figure 3.2.
Reciprocal advantage

We can see this dynamic at work in the most successful businesses

of the digital era—Google, Apple, Amazon, and others. Google Maps,
for example, is an amazing product that the company is constantly
evolving and innovating to be more useful for customers. Yet Google
has done little to monetize Google Maps directly over the years. How
can the company afford to keep investing in the product? The answer is
that Google Maps not only leverages but also grows the unique
advantages of Google’s other business models—by capturing
tremendous amounts of location data and greatly expanding the time
that users spend logged into a Google service. Those other business
models (search, display advertising, YouTube, and more) are where the
company monetizes its advantages with high margins and tremendous
revenue (see figure 3.3).

Figure 3.3.
Google’s reciprocal advantage

Reciprocal advantage is an opportunity in the DX of legacy firms as

well. Walmart is pursuing this path as it looks at new digital strategies
(see figure 3.4). Its e-commerce business, Walmart.com, can leverage
the proximity of stores for delivery to customers as well as the stores’
sales data to predict product demand. But the website also expands
Walmart’s customer purchase data, with insights into browsing
behavior. And by designing for online purchases to be returned in-
store, the website also drives more traffic to the retail business. After
establishing its e-commerce business, Walmart expanded into an
online marketplace for third-party sellers. This strategy leveraged the
company’s existing online traffic. It also expanded the product selection
on Walmart.com while generating more data on customers’ search and
product interests. More recently, Walmart has expanded into two
promising new sectors—health care and finance—by launching
services inside their stores (Walmart Health clinics and Walmart
MoneyCenters, respectively). These new business models benefit from
the stores’ customer traffic as well as its data (e.g., in assessing a
customer’s creditworthiness). At the same time, these new businesses
benefit the stores by growing their value to the customer, increasing the
frequency of store visits, and generating even more data. As Shotts
explained, “We think about what is the ecosystem that we’re
building . . . It’s not only where do people spend their money, it’s where
do they spend their time, and how do you increase your value and
relevance in both?”

Figure 3.4.
Walmart’s reciprocal advantage

Strategic Constraints

Just as important as understanding your unique advantages is

understanding your strategic constraints. As you formulate new digital
strategies, it is critical to identify any guard rails you won’t touch, limits
that your business will have to work within, or places you won’t go (that
other firms might).
We are not speaking here about the challenges of process, culture,
talent, and technology, which the DX Roadmap is meant to address
and improve with time and iteration. Rather, we are looking at
constraints—like government regulation or legal agreements—that your
business will need to work within, even as it digitally matures, and that
anyone working in your organization needs to be cognizant of.
Common types of strategic constraints include the following:

• Ownership or legal structure—Are you a publicly traded company?

A private family business? State-owned? A franchise business?
The CAA is part of a global organization, the Fédération
Internationale de l’Automobile (FIA), which includes 246
automobile clubs in 146 countries, each facing very similar
sources of digital disruption. But FIA is incorporated as a very
small global organization, with budget and power residing in its
affiliated national chapters. This structure means that FIA itself
cannot dictate or even centrally manage a global digital strategy.
Instead, it must focus on shaping the thinking of its national
chapters, sharing their best practices, and coordinating their efforts
to collaborate.
• Partner agreements—This often takes the form of channel conflict
—when one company’s ambitions conflict with those of its sales
partners. Many firms I have advised have had to contend with
such conflicts as they consider new digital business models.
Others have been constrained by legal contracts with critical
suppliers that prohibit them from competing in certain product
areas or certain markets. Recall that CNN was unable to bundle its
most prized content (its daily cable news shows) into its CNN+
digital streaming service because this was prohibited by contracts
with its cable distribution partners.
• Legal regulation—This is a major strategic constraint for many
companies, especially in industries like health care and financial
services. Banks like Citibank and Chase must contend with a wide
range of regulations—around data privacy, know your customer
(KYC) compliance, security and more—as they consider new
digital strategies. Acuity’s strategy for growth by geographic
expansion is constrained by the different licensing requirements
for insurers in all fifty U.S. states. In many cases, industry
regulations are applied differently for new digital start-ups than for
legacy businesses, leading to an uneven playing field.
• Local infrastructure—Infrastructure, such as the availability of
broadband, strong transportation networks, or credit-scoring
institutions, may pose significant constraints on your new
 strategies and business models. The availability of essential raw
materials or a lack of technical standards within an industry can
also pose significant constraints. When Amazon launched in India,
its strategic constraints included government regulation but also
logistics hurdles in a nation that lacked the shipping partners,
highways, and street address system of more developed markets.
When Alibaba first moved into mobile payments (launching
Alipay), it was because the Chinese market lacked a well-
developed system for consumer billing, like the credit cards seen
in other countries.
• Other factors—Various other factors, such as a company’s social
mission, may pose strategic constraints. The New York Times
Company’s digital strategy is shaped by its commitment to its
journalistic mission (not just to a goal of creating content that
draws the most eyeballs). When Saudi Arabia’s National
Commercial Bank (NCB) began its DX, one major constraint was
its low employee turnover and commitment not to let go of staff
hired and trained to run a brick-and-mortar bank.

As you identify your constraints, recognize that they may have a

variety of implications for decision making. Sometimes, a constraint will
not rule out a strategy completely. Instead, that constraint can be
treated as a risk factor to manage. New strategies that touch on the
constraint will require thoughtful consideration of your risk appetite and
of why you might be willing to push against it.
In other cases, your business will want to establish clear strategic
redlines that you will not cross. For example, Acuity’s leadership has
declared that they will not pursue growth by acquiring other risk-bearing
entities (e.g., another insurer) because that would jeopardize Acuity’s
unique advantage of having exceptionally well-managed underwriting in
its “book” of policies.

North Star Impact

The third element of a shared vision for a business is what I call your
North Star impact, which is a statement of what you seek to achieve
over time. Steve Jobs famously referred to Apple and its employees as
seeking to “make a dent in the universe.”13 What impact do you seek to
make through your DX? Your answer, like the North Star, can give
guidance and direction to your efforts over time. As you consider your
own North Star impact, try to answer these questions:

• What impact do you seek to have on the world?

• What problems are you uniquely able to solve?
• Why would the world miss you if you disappeared?

Your North Star impact should stem from your external knowledge
of the world (future landscape) and internal knowledge of your
organization (right to win), as well as an understanding of its history
and founding. Your North Star impact should not simply be an inspiring
string of words. It should be a statement that shapes business
decisions for years to come—the investments you make, the people
you recruit to leadership positions, and the strategies you choose to
pursue.
Digital-native businesses are often known for defining a clear
statement of the impact they seek to make. Google has a long-stated
mission to “organize the world’s information.” Microsoft, with its long
history of building technology tools for others, has a mission “to
empower every person and every organization on the planet to achieve
more.” But we can also see examples from non-digital-native
businesses. Every business I have seen with a cohesive agenda for DX
has been guided by a very clear sense of the impact they want to
make. Table 3.3 shows four examples.

Table 3.3.
North Star Impact at a Company Level
Co North Star Impact for the Company’s DX
mpa
ny
For Meet the environmental and mobility needs of a
d growing, urbanizing planet with connected vehicles
Mot and transport systems.
or
Mas Power and protect secure commerce in the digital
terc world, across every device, partner, and platform.
ard
BS Improve the experience of the digitally connected
H home consumer in their cooking and cleaning needs.
Ho
me
Appl
ianc
es
Do Offer the ultimate pizza delivery experience for
min today’s digitally connected consumers.
o’s
Pizz
a

Two things stand out from the North Star impact statements given in
table 3.3. First, they are extremely ambitious. None are goals that will
be easily achieved and completed in the near term. Second, they
encompass both the evolving core and the newest parts of the
business. The goal speaks to everyone in every part of the
organization. This means that everyone has a stake in the company’s
future.
While these four examples look at the impact of DX for an entire
company, a North Star impact can be defined at any level of the
organization. For Merck Animal Health, one team may define the
impact of its DX for dairy farmers, while another defines the impact of
DX for pet owners. For Acuity Insurance, its Personal Lines division
would define the impact of DX for individuals and families with
insurance, while the Commercial Lines division would define the impact
for small businesses.

Why Versus What

The most important point to stress about any North Star impact is that it
seeks to answer the question Why? rather than the question What? It
should be a statement of what you hope to achieve, not what you hope
to do. It is about outcomes not activities. This distinction is often missed
by executives crafting what they believe to be mission statements or
corporate strategies. Adam Bryant, who has interviewed hundreds of
CEOs, has observed, “I’ve looked at a lot of strategy documents that go
to a high altitude and just describe what the company does, not what
they’re trying to achieve.”14
This is especially critical for DX. The focus of your North Star impact
should be to answer “why” you must transform before you get into the
eventual questions of “what” you will do. As I have written before, digital
transformation is not about technology. I see many companies who
cannot write a sentence about their DX efforts without leaping to terms
like “artificial intelligence,” “blockchain,” and “metaverse.” But before
you start thinking about specific technologies that you will deploy, you
must first identify the purpose for which you will be using them.
Domino’s Pizza, for example, is not using digital technology for its
own sake. This is why there has been no effort to 3D print its
pepperoni, create a blockchain for its mozzarella, or sell non-fungible
tokens (NFTs) for its garlic knots. While Domino’s considers itself “a
technology company as much as a pizza company,” all of its digital
efforts have been ruthlessly focused on the goal of delivering the
ultimate experience for pizza ordering and delivery—simple, seamless,
everywhere, fast, and stunningly convenient.
The most impressive Web3 start-ups I have seen are focused on
the customer problem they aim to solve. For example, Qikfox’s mission
is “to make the internet more trustworthy, safe and reliable for
consumers.” It says nothing on whether the start-up will use blockchain
or Javascript or some other technology to get there.
The goal of digital transformation should never be defined in terms
of the technology you will use. Nor should it be defined in terms of the
capabilities you will build (data analytics, machine learning, cloud
infrastructure, etc.) or the processes you will use (customer journeys,
agile squads, etc.). Instead, use your North Star impact to ensure that
your DX efforts are focused on their impact. When Olivier Delabroy,
vice president of marketing, describes the DX at industrial giant Air
Liquide, he says that they are “obsessed” with defining digital as an
enabler of value creation. “My goal for any transformation is that it is for
the business, by the business . . . Not digital for the sake of digital, but
to create value for customers, shareholders, and employees.”
 As you think about the impact you want to have, remember the
power of intrinsic motivation and root your change efforts in the value
that your DX will create for others—be they customers, partners, or
society at large.

ZOOMING OUT
As you define your own North Star impact, it may be helpful to step
back and redefine the category that you are in. I call this zooming out
because it typically involves stepping back from a narrow product
definition of your business to a broader definition based on the needs
that you serve.
In thinking about its digital future, Ford zoomed out from the
products it had previously made and focused on the challenges of
environmental sustainability and mobility for a growing, urbanizing
planet. Ford redefined itself from a car manufacturer to a mobility
services company—with connected automobiles (networked, electric,
autonomous) as a key component.
Longtime CEO Ajay Banga began Mastercard’s digital journey by
declaring that it was no longer a credit card company but a fintech
using technology to enable commerce via payments and beyond
payments. Mastercard zoomed out from the product category of “credit
cards” or even “payments” to focus on the commercial needs of
consumers and merchants in an always-on, mobile-connected world.
Of course, there are limits to zooming out. As one executive asked
me, “If Ford is a mobility company, does that mean they should
manufacture airplanes and compete with Boeing?” The key is to look to
your right to win as you rethink the parameters of your business. Only a
definition that matches your unique advantages and constraints will be
sustainable in the highly competitive digital environment.
Try zooming out by asking, “What business are we really in?”
Answering this question honestly can help any business to escape Ted
Levitt’s trap of strategic myopia and keep the focus on the problems it
aims to solve.

Definition of Success
A statement of your North Star impact should be ambitious and
qualitative in nature (“offer the ultimate pizza delivery experience,”
“power and protect digital commerce,” etc.). But it can be very helpful to
pair this kind of qualitative statement with one or more concrete
metrics, what I call a definition of success. The point here is not to
precisely measure everything you will do but to answer the simple
question, How will we know if we’re making major progress toward our
North Star impact?
Mastercard’s overarching goal is to transform from a legacy credit
card company to a fintech that powers secure commerce in the digital
world—and it has launched new business units focused on
cybersecurity and other services for the financial sector. To measure
progress toward its goal, the company has set two ambitious targets: to
grow its services business at twice the rate of its core credit card
business and to grow it to 40 percent of the firm’s total revenue.
Together, these two targets provide a powerful measure of success for
Mastercard’s DX.
Similarly, the New York Times has defined measurable targets for
the transformation of its business model from advertising- and analog-
first, to subscriber- and digital-first. The company’s first target was to
grow digital revenue to $800 million annually (a figure that would
sustain the Times’s journalism if the print edition went away
completely). The second was to reach 10 million subscriptions across
its news and non-news products.
Table 3.4 shows both these cases, the New York Times and
Mastercard. In the table, I refer to the qualitative statement of each
company’s North Star impact as its “objective” and the measurable
definition of success as its “key result(s).” These terms are borrowed
from the management method known as objectives and key results
(OKRs). Originating in Intel under CEO Andy Grove, the OKR method
was refined by John Doerr (who began his career at Intel); evangelized
by Doerr in his work as a VC investor and mentor; and eventually
embraced by countless organizations, from Google and Intuit to Under
Armour and the Gates Foundation. The key to the OKR method is to
distinguish between your objectives (the goals you seek to achieve)
and your key results (the measurable steps you believe will advance
each objective).15
Table 3.4.
Definition of Success for a North Star Impact
Co North Star Definition of Success (Key
mpa Impact Results)
ny (Objective)
Mas Power and • Grow new services
terc protect secure businesses to 40 percent of
ard commerce in the the firm’s revenue
digital world
• Grow services twice as fast
as core business in credit
cards
New Secure our • Double digital revenue to
York journalistic $800 million, in five years
Tim efforts in the (enough to sustain news
es digital economy reporting)
Co
mpa
ny
• Reach 10 million
subscriptions by 2025

Your North Star impact (objective) offers a qualitative description of

what you seek to achieve. It drives alignment by motivating everyone to
think about how they can contribute to that ambitious goal. By contrast,
your definition of success (key results) is quantified. It will never
capture everything you want to achieve, but it drives alignment
precisely because everyone can see the progress you are making
objectively.
There is tremendous power in aligning teams and individuals across
an organization with clear, ambitious, and measurable goals like this.
One example I have seen came from Google in the early days of its
YouTube business. In 2012, YouTube’s leadership set an audacious
goal to grow the number of hours of video watched on its platform by a
factor of ten, to reach 1 billion hours per day. This number seemed
astronomical at the start, but it became a rallying cry across the
business. Diverse teams each contributed to the goal by driving
innovations in everything from YouTube’s search and recommendation
algorithm to improving its data centers for better bandwidth, to adding
virtual reality and gaming videos for the first time, to improving the living
room experience of “casting” YouTube onto your television. As
YouTube CEO Susan Wojcicki explained, the point was for leaders to
say to everyone in the company, “This is the direction we want to go,
now tell us how you’re going to get there.”16
Another reason to craft a definition of success is to engage
everyone in the process of choosing metrics. One of the key principles
of the OKR method is that it is practiced at all levels of the organization,
from the CEO to the bottom of the organizational chart. Every team is
involved, constantly asking themselves, “what is our objective?” and
“how will we know if we are achieving it?”
I have long held that the most important part of measurement is
engaging in a thoughtful debate about what your metrics should be.
Done right, this process brings about strategic alignment and employee
empowerment. You will have already reaped 90 percent of the benefit
of your metrics—before you even track anything!

Business Theory

To quote social psychologist Kurt Lewin, “There is nothing so practical

as a good theory.”17 The fourth and last element of a shared vision is
thus a business theory: your hypothesis about how you expect to
recover the investments you will make for the future.18 Your hypothesis
should explain how you will not only create value for customers but
capture value in turn. Let me be clear: a business theory is a causal
theory that doing X will lead to Y, but it is not a business case, with
financial projections of specific outcomes at specific times. It is not a
quantitative model spelling out things like unit economics, gross
margins, or your precise point of profitability. Instead, a business theory
addresses broader, directional questions. Will your financial return
come from selling more, reducing costs, generating new sources of
revenue, or from some mix of these? Will your growth stem from
internal innovations, acquisitions, or a mix of both?
Some of the most famous business theories have been captured in
visual form. Walt Disney’s diagram (easily found on the internet but
guarded from reproduction here by Disney’s licensing strictures)
mapped out the business theory that would lead his company to
decades of profitable growth. Sketched in 1957 (two years after the
launch of Disneyland), the diagram visualized how each expansion
beyond Disney’s core business leverages, creates additional revenue
from, and reinvigorates the core. The sketch places Disney’s theatrical
films business—with its unique advantage of creative talent—in the
center of a web of crisscrossing, labeled arrows. The arrows link the
core to all other categories—theme parks, television, music, books,
magazines, comics, merchandise, and character licensing—and
interlink these ventures, showing how the core and beyond-the-core
categories cross-pollinate one another. For example, films “feed tunes
and talent” to music; music “helps ‘kick off’ new [and reissued] films”
and “keeps films in mind,” while the theme park generates “ideas for
albums” and “plugs movies.”19
Great digital-era businesses have also been built on simple
business theories. Amazon’s e-commerce was guided from the
beginning on a very specific theory. Jeff Bezos proclaimed it an “article
of faith” that keeping Amazon’s prices low would earn customer trust
and that customer trust would drive free cash flow over the long term.20
This theory was expanded when Amazon included third-party sellers on
its site for the first time. In some ways, this move appeared risky—
introducing competitors directly to Amazon’s customers. But Bezos had
a new business theory, which he called Amazon’s “virtuous cycle” (or
“flywheel”). The theory—which he sketched on a napkin (see figure 3.5)
—explained how third-party sellers would grow Amazon’s e-commerce
business through cycles of mutual benefit. Amazon’s low prices and
large selection had created an experience that attracted customers.
That customer traffic would attract third-party sellers to sell on Amazon.
Third-party sellers would not only pay a commission; they would further
expand the product selection on the site. They would also lower
Amazon’s cost structure—by increasing the use of fixed assets like
servers and warehouses—which would enable Amazon to lower its
prices even further. Lower prices and bigger product selection would
attract even more customers, driving growth in a virtuous cycle.21
Figure 3.5.
Amazon’s virtuous cycle business theory, circa 2001

A good business theory is essential for companies looking to invest

significant resources in DX. When Saudi Arabia’s NCB began its digital
transformation, CDO Omar Hashem knew that it would not be easy or
cheap.22 In order to justify the large investments NCB would be making
(in back-end technology, consumer-facing digital experiences, and
retraining its retail workforce), the bank needed a theory for how digital
investment would create value for the firm. Hashem and his CEO
started with two major insights about their business. First, NCB was
suffering from legacy IT systems whose bugs were causing a low
straight-through processing rate. This meant that far too many
transactions required manual intervention by bank staff each day.
Second, NCB had a large untapped opportunity to serve unbanked
customers beyond the urban areas of Saudi Arabia that the firm had
historically served. The untapped market was well known, but NCB
lacked the human resources to serve customers there.
 Based on this, NCB’s leaders developed a clear business theory for
DX: reducing errors in the bank’s straight-through processing would
make it possible to design a best-in-class banking app and free
employees to pursue new unbanked markets. At the same time, the
improved digital app experience would help them win those new
markets. The strategy took years to execute, but the results were clear:
processing errors declined, customer use of NCB’s apps rose, and the
redeployed workforce drove a major geographic expansion. Without a
clear theory for how DX would pay off, NCB might never have
sustained the investments it took to make all this happen.

Pick Your Value Drivers

How do you get started on your own business theory? One way is to
pick a few key value drivers for your DX. These value drivers are broad
categories where you expect digital to generate value for your
organization. I have seen several companies use the following value
drivers to think about their DX:

• Customer experience (CX)—This includes any financial benefit

from digital innovations that touch the customer: new products,
new services, improved experiences, or changes in marketing.
Typically, value is measured from customer acquisition, customer
retention, and average revenue per user.
• Operational excellence (OpEx)—This is any financial benefit from
digital innovations that improve your operations or reduce risk to
your business, including efficiencies in your workforce and supply
chain (e.g., through data, AI, automation, or other measures).
Typically, this is measured in cost reductions.
• New business models (NBMs)—The first two drivers describe
value from improving your current business, but this driver
encompasses value from any new business models that you bring
to market.

These three value drivers work well. However, you may want to
define different value drivers for your organization. One nonprofit
organization that I have advised defined its value drivers as impact
(societal value created in support of its mission), relevance (loyalty and
affinity among its millions of members, who are critical to the
organization’s work and financial health), and revenue (financial
measures linked to its different business models). Every digital initiative
at the nonprofit must define its contribution to one or more of these
drivers to be approved.
Once you have agreed on your value drivers for DX, the next
question is, Where do you expect the most value? Try to define a pie
chart with an expected mix of your value drivers. Do you expect digital
efforts to generate 50 percent of their value from OpEx, 40 percent
from CX, and 10 percent from NBMs? Or is the mix 20 percent, 30
percent, and 50 percent? This is not a budgeting process but a
conversation on priorities. Agreeing on expected value drivers will bring
clarity to everyone’s digital efforts. When Mario Pieper joined BSH as
its first CDO, the CEO made clear that his number-one focus was top-
line growth—with a target to double revenues while growing at twice
the rate of peers. “When I heard that, it was clear we were not talking
about using digital mostly for efficiencies in our production line,” Pieper
told me. “We would be looking at customer growth, at selling more
valuable products, yes. But we would also need new business models
like [digital] services that could expand us into new sources of
revenue.”

Benefits of a Business Theory

A clear business theory has many benefits. The first is alignment on the
results you are hoping for. Taking the time to reach agreement among
your top stakeholders on a business theory for DX is incredibly
important. It is much better to get all stakeholders aligned on the theory
for a modest investment than to secure a big budget for your DX
without everyone agreeing on what it is meant to deliver. Far too many
digital transformations begin with no shared agreement on how or when
investments are meant to pay for themselves.
Another benefit of a business theory is to guide resource allocation.
United Technologies Corporation’s (UTC) CDO Vince Campisi used his
own set of value drivers to think through the resource allocation that
made the most sense for his company’s DX efforts. Campisi defined
the value drivers for UTC’s digital efforts as “rebooting productivity”
(OpEx), redesigning the current customer experience (CX), and
pursuing new business models based on data and analytics from the
industrial equipment UTC sold (NBMs). When he assessed the
company’s existing efforts, he estimated that 95 percent of digital
resources were being spent on OpEx, 5 percent on CX, and 0 percent
on NBMs. After setting the business theory for UTC’s digital agenda, he
quickly rebalanced: OpEx was still the largest part, at 70 percent, but
the other two value drivers finally received enough resources to make a
difference.
The third benefit of a business theory is that it can point you toward
the right metrics (or key results) to assess your unique digital efforts.
For NCB, its business theory suggested early metrics as it reconfigured
IT systems (measuring the straight-through processing rate and how
much time employees spent fixing bugs). Later, the theory suggested
metrics to track business impact (customer expansion into new markets
and user adoption of NCB’s banking app). Amazon’s virtuous cycle
theory implies several metrics for tracking the health of its e-commerce
business: product selection, overhead costs, product pricing, and
customer traffic and retention.

BUSINESS THEORY AND SHAREHOLDER COMMUNICATIONS

For any publicly held company, a business theory has one more
important benefit: communicating your vision to shareholders. Any CEO
of a public company has a responsibility to communicate the
company’s shared vision to investors and explain how it will generate
financial returns. Every year since its IPO, Amazon’s CEO has
published a letter to shareholders that spells out the business theory for
the company and how it relates to that year’s investments.
When Disney CEO Bob Iger decided to make a huge strategic bet
on streaming media with the launch of Disney+, he knew he would
have to explain his business theory to investors. At the time, Disney
had been licensing its most prized content (classic movies from Disney,
Pixar, Star Wars, and Marvel) to Netflix and other digital streamers. But
those lucrative licensing agreements would have to end before Disney+
could launch with a library of exclusive content. So Iger laid out his
vision to investors and explained the thinking behind it. Disney’s unique
content was its greatest asset. For years, he said, Disney had been
“selling our nuclear weapons technology” to a competitor (Netflix). By
ending those agreements, Disney would be hit by a “substantial
decrease in our revenue” in the short term. But, as Iger explained, it
would “thrust us into a [streaming] business that is the most compelling
growth engine in the media today.”23 Shutting off a current, renewable
revenue stream is a risky move for any publicly traded company. Iger
was very clear that Disney+ would not turn a profit in its early years.
But when it launched in 2019, shareholders responded by raising
Disney’s stock price by 9 percent in two days.24
The aim of a good business theory is not to persuade every investor
of your vision for the company. It is to explain your vision so that
investors who will back it are drawn to your stock, while those who are
not in agreement can shift their holdings to other companies. Bezos
credits Warren Buffett for explaining this to him: “[Buffett] says, ‘You can
hold a rock concert, and that’s OK. You can hold a ballet, and that’s
OK. Just don’t hold a rock concert and advertise it as a ballet.’ The job
of the public company is to be clear about whether you’re holding a
ballet or a rock concert, and then investors can opt into that.”25

Tool: Shared Vision Map

We have now seen the key elements of a shared vision for the future
and why it is essential to any organization seeking DX. Our first
planning tool, the Shared Vision Map, is designed to help craft a shared
vision that is unique to your organization and that can align employees,
investors, and other stakeholders behind your DX (see figure 3.6) Let’s
briefly walk through each step of the tool to see how to apply the Map
in your business.
Figure 3.6.
The Shared Vision Map
 Set Your Level

Before getting started, the most important step is to set the level at
which you are trying to define a shared vision. Is this meant to be the
shared vision for a company-wide digital transformation? Or are you
tasked with leading DX efforts within a single business unit, or within a
function (like marketing, HR, or supply chain)? The Shared Vision Map
can be used at any of these levels or even at the level of an individual
team to define where its world is going and the role it will seek to play in
that future. Let’s start by picking the level for which you are defining a
shared vision.

1. Future Landscape

The first element of your shared vision will be your future landscape,
which describes where your world is going and how the context of your
business is changing. Be sure to involve as many perspectives as
possible in crafting this; seek out customers, business partners, outside
analysts, and diverse points of view from people at all levels within your
organization. Your goal is to synthesize these different perspectives to
define a rich landscape view of where your world is going. As you do
this, focus on describing the four areas of customers, technology,
competition, and structural trends.

CUSTOMERS

• Changing needs and expectations—How are your customers’

needs changing in the digital era? What new expectations do they
have from businesses like yours?
• Legacy versus growth customers—How do your long-standing
customers differ from the customers driving your current growth?
• Emerging customer segments—Are there any customer segments
you hope to target for the first time? What have you learned about
their preferences and needs?
 TECHNOLOGY

• Commercialized technology—Look to digital technologies that

have ample market adoption. Which of these is most relevant to
your business? Which have been adopted by your customers? By
your partners? What use cases from other industries could be
applied to your own?
• Viable technology—Look at technologies that are technically
proven but whose applications are not clear. Can any of them be
applied to solve specific customer or business problems in your
industry? Can you learn from another sector that has found a
viable use case?
• Emerging technology—Look at technology still in technical
development. Where are VCs and large technology firms
investing? Which technologies, if they were to become available in
the future, might have the biggest impact on your industry?

COMPETITION

• New products and services—What new products and services are

appearing in your industry? Which ones are coming soon but are
already generating interest?
• New entrants—Who are the new digital-era entrants into your
industry? Look at partners, disruptors, and substitutes.
• Competitive Value Train analysis—Apply the Competitive Value
Train to analyze new entrants: how they intersect your current
business model, and the mix of threat and opportunity they pose.
• Pace of change—How fast is change coming to different parts of
your business? Who is moving faster and slower in your industry?
• Threat levels—How severe is the threat of digital change to
different parts of your business? Look at different business units,
customer segments, and product lines.

STRUCTURAL TRENDS
Look at broader structural trends that may have an impact on your
operating environment, including demographics, talent, regulations,
supply chains, macroeconomics, and more. Which ones are or may
become key drivers of change for your industry? If you operate in
different markets globally, look at commonalities and differences in
these trends. Consider volatility: Are these trends likely to remain
steady, or do you need to prepare for possible dramatic shifts?

WHAT IF WE DO NOTHING?
Your future landscape provides a clear viewpoint on the most important
ways that your world is changing and the threats and opportunities to
your unique business. Your future landscape should also answer the
critical question, What happens if we do nothing?
Discuss the likely impact of trends on your business if you were to
simply “stay the course.” Consider posing the following question to
colleagues: If we change nothing—at what time in the future might we
no longer be in business? (There is no correct answer. The only wrong
answer is never.) If some of your team members think the threat of total
disruption is five years away, but others say thirty years, engage in a
robust discussion and come to a shared point of view.

2. Right to Win

The second element of any shared vision is your right to win in the
future you see. Where and how does your business have a unique
reason to succeed in the future?

UNIQUE ADVANTAGES
Start by generating a list of your organization’s advantages as it
competes against similar organizations. Think of your strengths and
advantages in any of the following areas:

• Assets—tangible assets and intangible assets

• Capabilities—skills, insights, training, or talent
 • Relationships—brand reputation, customer relationships, or
business partnerships
• Positioning—differentiation from peers in where you compete and
how you go to market
• Culture and process—employee culture, effective management,
speed to market
• Other—any strengths that don’t fit the above categories

For each advantage you have listed, define how it benefits your
organization. For example, you might define a unique advantage as
your “brand reputation as an innovator.” But how specifically does that
benefit your company? (Does it lead to higher customer retention? Help
you market new offerings? Attract top talent for your team?) If you
cannot identify a clear business benefit, strike that item from your list.
Score each of your remaining advantages on two dimensions:

• How unique they are

• How beneficial they are to your business

Now create a matrix of your advantages on the two dimensions.

Your important advantages will score at the top on uniqueness or
benefit. The most important will score at the top on both.

STRATEGIC CONSTRAINTS
Next, identify the constraints that will affect which strategic choices are
available to you. Consider each of these areas:

• Ownership or legal structure

• Partner relationships
• Social mission
• Regulation
• Local infrastructure
• Other industry constraints

For each constraint you have identified, ask: What limit will this
impose on your strategy? For example, do differences in regulations
mean that you will need to pilot digital strategies in certain markets first,
before attempting them in others?
Looking at all your constraints, create a list of your biggest strategic
risks. What is your risk appetite for each? (Are you willing to take on a
high amount of risk, a moderate amount, or no risk at all in this area for
the right opportunity?) Then ask what trade-offs would make you
consider assuming such a risk. (For example, are you willing to risk
greater regulatory scrutiny to expand your business into a new product
category or region?) Lastly, look at any constraints that seem inviolable:
strategic redlines that you would never cross. Spell out any specific
strategies that are completely off limits, and why. This will indicate
when this redline might be reconsidered should conditions change.

3. North Star Impact

The third element of your shared vision is your North Star impact—what
you seek to achieve over time and why.

FRAMING QUESTIONS
As you consider your North Star impact, try to answer these questions:

• What impact do we seek to have on the world?

• What problems are we uniquely able to solve?
• Why would the world miss us if we disappeared?

Try to reframe your business by zooming out, expanding your

definition of your business. Ask yourself these questions:

• What products or services do we deliver?

• What fundamental customer needs do these products serve?
• What business are we really in?

STATEMENT OF IMPACT
You are now ready to write a statement of the impact you seek to have.
As you write a North Star impact statement, make sure you focus on
why you do things and not on what you do as a business. Your
statement should describe outcomes and impact—not actions taken,
tools used, or products delivered.
What is the most important goal that you would seek to rally your
team, business unit, or company to achieve in its DX? Look to the
future: What new problems could digital technologies enable you to
solve that you couldn’t solve before? How might DX enable you to
serve a new type of customer? What problems have emerged in the
digital era that you may be able to address?
Once you have a draft statement of impact, try applying the
following tests:

• Is it ambitious (for example, not something you could complete in

three years)?
• Does it encompass all your work (both long-standing work and new
ventures)?
• Does it provide intrinsic motivation to employees (describing the
value they will create for customers, partners, or society)?

DEFINITION OF SUCCESS
Next, pair this impact statement with a few concrete metrics—your
“definition of success.” These metrics do not need to capture everything
that you will do in your DX, but they should be strong indicators that
you are making progress toward your North Star impact. Choose one to
three metrics that are clearly measurable and assign ambitious targets
that will take you years to achieve. These metrics should align your DX
work and provide evidence of its long-term value creation.

4. Business Theory

The last element of your shared vision is your business theory. Think of
the kinds of digital investments you anticipate making. How do you
think those investments will pay for themselves over time?
 VALUE DRIVERS
Try to choose three to five broad value drivers that capture the different
types of value your DX efforts will bring. Consider these common value
drivers: customer experience, operational excellence, and new
business models. Also consider your own formulation of the value
drivers that apply to your business. If you are a mission-driven
organization (e.g., a nonprofit, public sector, or nongovernment
organization), choose value drivers that define the impact you are
seeking from your DX. Create a pie chart with the expected mix of your
value drivers. What percentage of the value from digital do you expect
from each driver?

CAUSAL THEORY
Try to go beyond your value drivers and sketch a working theory of how
your investments will drive value within your firm. Look to examples like
the business theories of Disney, Amazon, and NCB.
What current barriers do you see to growth or profitability? What
new actions will you be taking? What do you believe will be the results
of those actions? Do you see a virtuous cycle where one positive
change will lead to another?
As you sketch a business theory for your DX, ask yourself if it will
pass the investor test. If you explained your theory to your CFO or in a
letter to shareholders, do you think they would be persuaded to give
you the resources to pursue your vision?

Vision from the Bottom Up

A shared vision is often thought of as coming from the top, but a shared
vision is not a tool meant for command-and-control leadership. In a
classic top-down organization, there is no need for a shared vision.
Employees do not need to understand the long-range thinking behind
their tasks. They don’t need to understand what mission or purpose
guides the organization (although that may improve morale). They just
need to be told what to do and what numbers to hit for their next
performance review.
 But a shared vision is essential to leading any organization in a truly
bottom-up fashion. Only when everyone is aligned around a vision of
where the business is going and why can individuals and teams be
given real ownership and accountability for decisions. This is what
unleashes the speed and adaptability at every level that are the
hallmarks of a bottom-up organization. As Bill Ford said, “A clear view
of the future—once you have that, you make decisions very rapidly.”26

A Vision at Every Level

We often think of a shared vision as something that sits at the level of

the entire organization, and that central shared vision is certainly
essential. But for real transformation to happen, every part of the
organization should define a shared vision at its own level. That means
a vision for each business unit, each function, and each team.
If you are leading DX for the marketing function of your business, for
example, it is imperative to define a future landscape for marketing:
Where do you see the future of your work as you look to new marketing
technologies, new customer expectations, and new marketing partners
and competitors? What new structural trends, such as data privacy
regulations, will affect you?
The marketing team should define the remaining elements of a
shared vision as well. What is the North Star impact that you seek to
achieve through DX of your marketing? What are the unique
advantages and strategic limits that shape your right to win? What is
your business theory for how DX in marketing will deliver financial
returns? What are the value drivers for digital in your part of the
business?

Cascading Up, Not Down

Whatever level you start from in defining your vision, what you do next
is critical. Typically, when I talk about a shared vision, executives will
say “I’ve got it! The leaders come up with this, and then we cascade it
down to the rest of the organization.” “Cascading down” means that a
plan is set at the top by leaders who tell their next reports, who tell
theirs, and so on, down the organizational chart. But that is not the way
transformation happens in cases of great DX success. I encourage
leaders to think instead about what I call cascading up.
The difference between cascading down and cascading up is in the
conversation that happens at each juncture. In a top-down
conversation, the leader says, “This is my goal,” and tells their direct
reports, “Here is what I need you to do to support this.” In a bottom-up
conversation, the leader says, “This is my goal,” and asks their direct
reports, “What do you think you should do to support this?” Ideas then
come from those who will do the work. Discussion ensues. The leader
must still sign off on the plan, but it is each person’s responsibility to
propose what role they should take and what goals they should
achieve. Once an agreement is reached, each of those direct reports
can have the same bottom-up conversation with those who report to
them.
Recall YouTube’s audacious goal to grow its total daily watch time
by a factor of ten. Leaders did not tell every team what they needed to
do to support the goal. They told them the goal and asked, “How are
you going to help us to achieve this? What metric and target do you
propose for your own work?”
As you share your vision for DX with others, keep in mind that your
role as a leader is not what it would be in a traditional top-down
organization. Leadership in the digital era consists of three jobs: define
a vision of where you are going and why, communicate and align
everyone to that vision, and enable others to act in support of that
vision.

No effective digital transformation can start without first aligning

everyone with a shared vision of the future. A shared vision provides
alignment of purpose and direction. It provides a shared understanding
of where your world is going and the role your business will play in the
digital future. A shared vision also provides the motivation for change.
For investors and executives managing P&Ls, your business theory
provides extrinsic motivation—an answer to how digital will generate
financial rewards. This will be essential to securing the long-term
resources for your transformation. For employees, your North Star
impact will provide intrinsic motivation—through a promise of value
created for customers, society, and others. This will be essential to
muster the sustained effort, will, and creativity for ongoing change. Both
your business theory and your North Star impact require that you have
a clear and compelling view of your future landscape and your unique
right to win.
A shared vision can shape the actions of a business for years.
Nearly a decade after Bill Ford’s TED speech, the company’s Chief
Transformation Officer Marcy Klevorn told me that “the themes in that
speech were a lot of my and my team’s inspiration for how we
prioritized work.” But a shared vision is also a work in progress. In the
DX Roadmap, you do not “finish” your shared vision and put it on the
shelf as you proceed to the next step. You continue to work at it,
deepen it, and renew it, even as you begin to use that vision as the
foundation for the next steps of the DX Roadmap. With the first draft in
hand, you are not finished with your shared vision, but you are ready to
begin your next step.
When starting a DX, many organizations are quickly overwhelmed
with ideas for digital projects and innovations, with no criteria to judge
them. A compelling shared vision—with a commitment to empowering
teams—will quickly inspire ideas from every level of an organization.
How do you choose among these ideas and avoid moving in 100
directions at once? In the next chapter, we will see how any leader can
define strategic priorities to pursue the goals of a shared vision. You
may think of these priorities as critical problems to be solved or as
opportunities for growth. Together, they will give focus to your digital
efforts—defining the scope of where you will compete and seek to
create value with your DX.
 4
Step 2: Pick the Problems That Matter Most

When Imran Haque stepped into a new role leading digital strategy for
Pfizer Animal Health, he was an ideal candidate. Haque had started his
career at a digital agency and founded a health-care start-up before
moving into the corporate world of pharmaceuticals. At Pfizer, he had
led a variety of technology projects: applying big data to the clinical
trials processes, revamping the analytics tool for safety and risk
management, and overseeing technology investments for worldwide
operations. Now Haque was tasked with a different role: defining a
digital growth strategy for the animal health business unit, which would
soon be spun out as a separate business named Zoetis.
Haque began by studying major trends that would define the digital
future of animal health. In the broader technology landscape, he
tracked the rise of the mobile web and social media, the growth of
online education, and trends in cloud computing, big data, and the
Internet of Things (IoT). The animal health industry had not been a
pioneer in digital innovation, but Haque was focused less on his peer
businesses than on the rapidly changing needs of his customers.
These included pet owners; agricultural businesses raising cattle,
swine, poultry, and fish; and the veterinarians who, up until now, were
both the sales channel and the key influencer of every purchase
decision. Among these customers, Haque observed a growing appetite
for animal health content, a nascent interest in e-commerce, and
evolving needs in animal husbandry, such as the growing importance of
diagnostics as farmers sought to spot disease earlier in their animals.
With a clear vision of the future, Haque’s next challenge was turning
this into a strategy for growth. He did this by looking for specific
customer problems to solve and business opportunities to leverage. He
developed a short list of strategic priorities for his business, including
(1) improve the customer’s digital path to purchase—from online
discovery to e-commerce, to loyalty programs; (2) provide online
learning and content for animal health professionals; (3) enter the
growing diagnostics market to detect and prevent illness in animals;
and (4) empower livestock management with digital tracking and
analytics. Where others in his industry were still focused on optimizing
their legacy operations and sales channels, Haque now had a clear set
of priorities for growth.
These strategic priorities led Haque and his team to launch a wide
range of digital innovations over the next several years. Some fell
under the umbrella of digital marketing—revamping the mobile, web,
and social media presence of a diverse family of product brands;
growing a digital loyalty program for the cattle industry; and launching
the industry’s first e-commerce channel. Others were opportunities for
new revenue streams, including e-learning training and certifications for
veterinarians, and paid content publishing in the form of wellness
reports targeting different customer segments. Still others involved new
business models. Zoetis launched a digitally powered services
business for animal husbandry, helping farmers with everything from
digital billing to verifying which pigs were impregnated. The firm moved
quickly into the diagnostics business through internal innovations and
several start-up acquisitions. And with new IoT ventures like Smart
Bow, Zoetis gave cattle owners the ability to track their herd using small
sensors on each animal that could measure and analyze its
movements, water and food consumption, and vital signs to detect
sickness, health, and the ideal timing for impregnation.
Zoetis’s digital innovations were widely adopted by customers and
built support for digital transformation (DX) over time throughout the
organization. During its first four years as a public company, and while
Haque served as head of digital business, Zoetis grew from a $13
billion market cap at launch to $60 billion. Throughout that time, Haque
led and supported a diverse range of digital initiatives. These initiatives
did not arise haphazardly but from a clear and evolving list of strategic
priorities for digital growth.

Why Priorities Matter

As we consider the role of strategy in DX, it is helpful to start with a

simple question—why does any organization need a strategy?
Management theory tells us that the purpose of strategy is to guide
decision making by defining priorities amid limited resources. Every
business must make critical trade-offs between options. Thus, strategy
is fundamentally about choices—defining the opportunities you will
focus on and, just as important, those you will not. As Michael Porter
observes, “The essence of strategy is choosing what not to do.”1
Defining strategic priorities might seem like an obvious early step in
DX. We can see how well it worked for Zoetis. But strategy is often
overlooked. As one seasoned CDO confided to me, “This is what
makes it so difficult to find peers to talk to [at other firms]. They haven’t
done that exercise. They went a different route: doing digital projects,
but without a strategic focus.”
 In too many organizations, DX operates in a strategic vacuum, as
little more than a collection of individual projects. Without clear priorities
for growth, it is extremely common for a business to find itself
overwhelmed with ideas for new digital projects. Lacking any clear
criteria for decision making, digital becomes hijacked by the latest shiny
new things from the technology world. Often, DX becomes limited to
cost cutting and optimization of legacy processes. At worst, digital may
be handed off to a dedicated team, separate from the day-to-day
challenges of the business. This small group focuses on blue-sky digital
ideas while the rest of the firm continues their work unchanged.
In the first step of the Digital Transformation (DX) Roadmap, your
task was to define a shared vision of the digital future for your firm. In
the second step of the Roadmap, you must build on that vision by
defining specific priorities for investment of your limited resources and
time. What are your biggest digital priorities for the future? Think of
these as problems to be solved for your customers and your business,
and as opportunities for your firm to create and capture new value.
Your conversation here should not be about technology (“we need
to leverage machine learning”) or digital skills (“we need people who
understand crypto”). It should be about priorities for growth and value
creation. You are not yet looking for a list of digital projects or
innovation ideas. Instead, you are defining, at the start, where you will
look for innovations—that is, the scope of ideas you will explore.
Instead of new digital products or solutions, think first about where you
will innovate and compete.
Picking the priorities that matter most to the future of your business
is not easy. But without a defined set of strategic priorities, any DX will
fail to achieve a lasting impact. Table 4.1 shows some of the key
symptoms of success versus failure in Step 2 of the DX Roadmap.

Table 4.1.
What’s at Stake—Step 2: Priorities
Symptoms of Failure: Symptoms of Success:
Priorities Priorities
• Digital transformation • Clear priorities provide
is a series of scattered direction to digital
 projects with no clear transformation across the
direction. organization.
• Digital efforts are defined • Digital efforts are defined
by the technologies they by the problems they solve
use. and opportunities they
pursue.
• Digital is focused solely on • Digital is focused on future
operations, cost cutting, and growth as well as improving
optimizing the current the current business.
business.
• A few people in the • Every department is
organization drive digital pursuing its own digital
while the rest stick to old ventures, with a backlog of
ways of working. ideas to try next.
• Transformation is • Transformation is linked to
disconnected from business the needs of the business
needs and loses support and gains support over time.
over time.

What’s Ahead

In this chapter, I will take you through Step 2 of the Roadmap. We will
see how any organization and any leader can define a focused set of
strategic priorities for their DX efforts. We will look through two lenses
—problems and opportunities—and see the power of each to help
define your strategic priorities. We will see how popular methods—such
as customer journey maps and customer interviews—can help to
identify valuable problems and opportunities. We will introduce two new
tools: problem/opportunity statements (to crystalize your strategic
priorities and spark ideas for new innovation) and the
Problem/Opportunity Matrix (to define strategic priorities at any level of
the organization). We will also learn why engaging everyone in the
strategy process is critical to making the shift from a top-down to a
bottom-up organization.

The Problem Lens

The first lens through which we must define our strategic priorities is
the problem lens. This entails thinking about strategy in terms of a set
of problems that we aim to solve rather than in terms of known
solutions that we can build. For example, rather than simply focusing
on ways to sell more cattle antibiotics, Zoetis chose to focus on a major
problem for farmers—how to track vital signs in a large herd in order to
detect illness and treat it earlier. Other problems that Zoetis chose
include veterinarians’ need for professional training and certification,
and pet owners’ need to shop more easily for products like medicated
shampoos.
In this chapter, I’ll guide you on using the problem lens effectively
for your DX. At the start, it’s important to note that much of modern
innovation practice is rooted in this perspective. In both agile and
product management, every cross-functional team is defined in terms
of a persistent problem that it is working to solve iteratively over time,
whether as part of launching a new product or optimizing a long-
standing business process. In design thinking, the problem lens
matches a key goal study and solve problems that are complex,
human-centered, and systems-based. Many of the most popular tools
in design thinking are focused on problem definition. Thomas Wedell-
Wedellsborg’s book What’s Your Problem? encapsulates much of the
best of this thinking.2
You may recognize the problem lens under many other names; an
innovation “problem” may be called a “pain point,” a customer’s “job to
be done,” or an “unmet need.” As Microsoft CEO Satya Nadella has
stressed, “The more we can invoke our ability to meet unmet,
unarticulated needs, that’s the source of innovation.”3
Whatever you call it, the problem lens requires a major shift for
traditional organizations. Product managers often refer to this shift as
“product thinking” or a “product mindset.” Never mind that these names
are a bit misleading; the point is that you should focus on the problem
you aim to solve rather than on the product you think you might build.4
This shift is aptly described in one of Amazon’s innovation principles as
“working backwards from customer needs.”5

Fall in Love with the Problem, Not the Solution

Focusing on customer problems is not easy. In most companies,
people love to start with the solution in mind: “Tell me what to build, and
I will tell you how we deliver it!” The natural tendency in an established
business is to focus on your own core competencies and the products
and services you deliver. Instead, you must frame everything from the
customer’s point of view rather than your own. It all begins with asking,
What is our customer’s need? How will we fill it? In a famous analogy
of Ted Levitt’s, if you are running a hardware store, you may think you
are selling a quarter-inch drill bit, but the customer is buying a quarter-
inch hole. Jeff Bezos applied the same reverse thinking to Amazon’s e-
commerce business, observing, “We don’t make money when we sell
things. We make money when we help customers make purchase
decisions.”6
This mental reversal requires humility and a willingness to temper
our enthusiasm for our initial ideas. The lean start-up method
recognizes that most entrepreneurs begin with a solution in mind—
some brilliant invention they want to bring into the world. Hence, the
method starts with talking to customers in order to validate what
problem your innovation would really solve.
Very often, the problem lens requires some coaxing. At United
Technologies Corporation’s (UTC) digital accelerator, managers from
different business units regularly bring in requests for a new digital
solution they would like built. The first job of the digital accelerator team
is to engage the manager in a workshop to define a clear problem
statement, including the customer experience or outcome the business
is trying to create. The business unit’s initial request is really “a problem
disguised as a solution,” says CDO Vince Campisi. “Ten out of ten
times, when an opportunity walks in our door—by the time it leaves, the
solution we will actually build looks a lot different.”7
Perhaps the biggest benefit of focusing on the problem is that it
helps you look more broadly at possible solutions to explore. Mario
Pieper explained how it helped his team at BSH Home Appliances to
look beyond incremental innovations—such as adding recipe screens
to its refrigerators—to envision wholly new ways to help customers
meet their kitchen needs. “If the pain point of somebody in the kitchen
is that he wants to eat, then of course you can cook, but you also can
order something,” Pieper observed. “The question is, should we also
discover innovations where we deliver food and do not use the kitchen?
This is exactly what we wanted to do—asking ourselves ‘what is the
best solution for the problem?’ and not ‘what product do we have and
what can we do with it?’ ”
As you pursue any effort at DX, it is critical to understand that
technology itself can become a distraction from strategy. Too often,
business managers become fixated on a specific new technology and
how they are going to use it rather than focusing on the problems they
are trying to solve (see the box “The Distraction of New Technology”).
For years, I have been telling executives that AI is not a strategy. The
same is true of blockchain, NFTs, Web3, or any other technology. But
what do you do if your organization has already allocated resources to
a dedicated team focused on a technology like AI or blockchain? A
common workaround I have seen is to ask that team to focus on a few
early use cases. In other words, pick a few problems you think could be
solved using this new capability and test to see if you can deliver
results. In this parlance, “use case” becomes a synonym for a business
problem—a substitute that seems palatable to those who think
technology first.

The Distraction of New Technology

New technology can exert a powerful gravitational pull, drawing

business leaders’ attention away from strategic thinking and the
focus on problems. When any new technology reaches the peak of
its hype cycle (think IoT in 2016, blockchain in 2019, metaverse in
2022), less passionate observers will describe it as “a solution in
search of a problem.” This description rings true, but it describes a
particular stage—before the attention shifts to solving real-world
problems.
Let me give an example. During the period 2012–2014, a series
of breakthroughs in artificial neural networks (using what is called
deep learning for pattern matching) sparked tremendous excitement
about the potential for AI. In a series of demonstrations,
technologists at Google demonstrated for the first time that AI
systems trained on vast amounts of data could detect pictures of
cats among images pulled from the Web. Other companies made
 similar dazzling demonstrations of how deep learning could
understand speech. This generated tremendous excitement about
the little-known field of AI. Suddenly, every executive with a digital
pedigree was expected to explain their “AI strategy.”
Despite a rush of venture capital, the new generation of AI
struggled at first to produce value. Early efforts to sell deep learning
as a raw capability to other companies failed in the market. Yes, AI
could spot a cat among a thousand images, but . . . so what? Within
a few years, however, things began to change. The real value of
deep learning emerged as new start-ups began to define common
business problems that it could address. Deep learning became
“productized” into numerous niche services: an algorithm trained to
detect fraudulent transactions for banks, an algorithm to spot defects
in concrete construction using photos from drones, an algorithm that
could hear when callers on your customer service line were
frustrated with your staff. By this time, the buzz had moved on to the
next new tech. But that too was a signal that deep learning had
finally started to create real value.

I have seen more than one company move its DX efforts out of its IT
function to ensure that DX is focused on customer problems—which
the business units know best. An executive leading digital strategy at
the oil and gas services business Schlumberger stressed to me the
importance of their moving digital out of IT, where it had started. “We
needed to take digital out of our software vertical and bring it into a
horizontal that cuts across [the organization… because] one of our key
tenets is for digital to solve customer problems.”
As you focus on digital innovation in your own business, remember
this enduring mantra of Silicon Valley entrepreneurs: “Fall in love with
the problem, not the solution.”

Whose Problem Is it?

Every problem has a stakeholder—the person or group of people who

are directly affected by it and will benefit from a solution. Commonly, we
think of problems where the stakeholder is a customer—whether an
end consumer (for a B2C business) or an enterprise customer (for a
B2B business). Zoetis’s digital strategy has focused on solving
problems for pet owners as well as commercial livestock businesses
and veterinarians. But it is also important to recognize that the problem
you are solving may belong to a stakeholder within your own company
—what we might think of as a “business problem.”
We can see examples of both customer problems and business
problems in the DX work at Air Liquide, global supplier of gases and
technologies to health-care and manufacturing companies. An
important customer problem that Air Liquide is focused on is helping
hospitals and manufacturers know when their tanks of gas (oxygen,
hydrogen, etc.) are running low so that they can be replaced before
running out. Solving this problem requires innovations in the tanks’
digital sensors, predictive analytics based on consumption, user alerts
and notifications, and seamless reordering. Every aspect of the solution
must be designed and tested to ensure that it works for Air Liquide’s
customers in the context of their daily work.
Air Liquide is also working on a business problem in the same
division: how to predict customer churn. Solving this means predicting
which customers are at greatest risk of unsubscribing from its service in
the next thirty days. Solutions involve predictive models based on
usage patterns across the entire customer base, being careful to avoid
statistical false positives (such as when a small business lowers its
usage during a vacation or seasonal slowdown). In this case, the
problem’s stakeholder is internal—Air Liquide’s marketing team in
charge of current accounts and customer retention. Any solution to the
problem must be designed for that team, be adopted by that team, and
deliver results in the context of their daily workflow.
Of course, it is important to recognize that trying to understand one
stakeholder’s problem will sometimes lead you to uncover a problem
that touches a different stakeholder. For example, I know a software-
as-a-service (SaaS) company that was very successful at selling a new
software solution but struggled with a business problem of high
customer churn. By examining not just who was unsubscribing but also
why, they discovered an unaddressed customer problem: users found
the software exceedingly difficult to use. After a few months of paying
for it without reaping its benefits, customers unsubscribed. By
addressing the external customers’ problem (with a better training
program to accompany new sales), the SaaS business solved its own
business problem of customer churn.

The Limits of the Problem Lens

The problem lens is powerful for defining strategic priorities in any

business, but it is not without its limits. Too much dependence on
defining problems can lead you to focus exclusively on your current
customers and fine-tuned improvements to the business you know best
—your current core. In practice, I have seen how many design thinking
tools are more likely to uncover problems within your current business
than to point you toward new opportunities outside your comfort zone.
How can we overcome this inherent bias of the problem lens? We
can start by pressing more deeply into our customer insight work. The
popular five whys method uses repetition to seek the deeper underlying
motivations or needs behind a customer’s actions. And we can return
to Levitt’s “strategic myopia” question that pushes you to reconsider the
boundaries of your enterprise—what business are we really in? In
addition, we can expand our search for strategic priorities by deploying
a second lens to give us another perspective on the search for growth:
the opportunity lens.

The Opportunity Lens

The second lens through which we can define our strategic priorities is
the opportunity lens. This entails thinking about strategy in terms of
new ways to create value, both for customers and the business. The
power of the opportunity lens is that it can push you to look beyond
your current business and think about strategy more expansively.
One of Zoetis’s strategic priorities for its DX was to enter the
growing diagnostics category in animal health. Zoetis had no role in
diagnostics when the company was first spun off from Pfizer. Entering
this market was not a solution to any current problem, yet it was a
compelling opportunity to create value.
In the digital era, many of the biggest breakthrough products have
come from pursuing a clearly defined opportunity to create new value.
When Apple began work on the iPod, Steve Jobs’s vision was to give
the customer “1,000 songs in your pocket.” When Amazon began work
on what would become the Kindle, its ambition was to offer “every book
ever printed, in any language, all available in less than 60 seconds.”8
These were each incredible strategic opportunities to pursue, but they
were not exactly pressing problems where Apple’s or Amazon’s
customers were crying out for a solution. When looking to create
something truly new to market, boundary-pushing ideas are more likely
to be discovered with an opportunity lens.
Another way of thinking about this comes from the world of venture
capital (VC). When looking at start-ups to invest in, VC Kevin Fong
would famously ask entrepreneurs, “Are you selling painkillers or
vitamins?”9 In other words, are you creating a product that will solve a
pressing problem for the customer, or will it give them a better life in
ways they may not yet know they want? The common investing wisdom
is that “painkillers” have better odds of success. It is easier to market
the product because customers understand its value immediately and
see why they would want it. On the other hand, there may be a bigger
upside if a start-up can succeed with a “vitamin,” a breakthrough
product no one was asking for because they had not yet imagined it.
The two lenses are also helpful when thinking about your value
proposition for an established product or business. When defining the
benefits you give to customers, try asking two questions: What current
frustrations or pain points are we alleviating? What new delights are we
providing? (I refer to both as value elements. Alex Osterwalder calls
these two types “pain relievers” and “gain creators.”10)
As we make use of the opportunity lens, it is important to remember
what a strategic opportunity is and what it is not. A strategic opportunity
is not just a business metric (“grow top-line revenue 20 percent this
year”). Nor is it a solution you have already decided on (“build a virtual
reality [VR] headset with longer battery life”). Rather, it is a focused
idea about the value your business could create and where it could
compete (“create a VR meeting experience for designers that is better
than being in the same conference room”).

Whose Opportunity Is It?

Just as with problems, we can think of strategic opportunities from the
point of view of the customer or the business.
Customer opportunities focus on ways to create new value for a
specific customer. Think of the vision statements for the iPod and the
Kindle. When Uber began business, it defined an opportunity to offer
“transportation as reliable as running water, everywhere for
everyone.”11 Whereas customer problems address an evident point of
friction or pain, customer opportunities produce an unexpected benefit
or delight. This quality of unexpected delight is critical to theories of
customer satisfaction. The Kano model, for example, distinguishes
product benefits that the customer values, demands, and expects
versus benefits called “delighters,” which are unexpected but create the
greatest excitement among customers when delivered.12
Business opportunities, on the other hand, focus on new growth and
expansion for the business. Zoetis, for example, identified an
opportunity to expand into the growing diagnostics business. Recall
YouTube’s ambition to achieve 1 billion hours watched per day—
another clearly defined opportunity. Amazon Web Services (AWS)
began as an opportunity for the retailer to generate revenue from its
new IT infrastructure by renting it to other companies as a service.

Defining an Opportunity

I have seen four powerful ways that companies define strategic

priorities with the opportunity lens:

• Customer delight statement—Try describing an unexpected and

delightful experience to create for the customer. Bezos’s dream for
the Kindle was “every book ever printed, in any language, all
available in less than 60 seconds.”
• Attractive market with a right to win—Find a large or growing
market to enter, one where your business has a built-in advantage.
For Zoetis the opportunity to “enter the diagnostics business”
meant selling to an existing base of customers, one that it was
already serving with treatments and vaccines.
 • New capabilities with clear application to your business—Look for
new skills or emerging technologies that are relevant to an
important part of your business model. Amazon chose to invest in
robotics because of a specific application: picking products off
warehouse shelves. More recently, drugmakers have invested in
machine learning because of its specific application to key steps in
the drug discovery process.
• 10x stretch goal—Try envisioning value creation at scale, not just
incremental improvements. “Grow watch time to 1 billion hours a
day” was a huge goal for YouTube and a compelling business
opportunity. One of Google’s core principles is to “think 10x, not 10
percent.”13 Similarly, the Gates Foundation defines ambitious
goals for public health by interviewing doctors and pushing them to
think big, asking, “What would you do if you had unlimited
resources?”14

When defining opportunities, push yourself to think big about what

matters most. Remember, strategy is about defining your top priorities
for growth, which must rise above all others.

Two Complementary Lenses

It is important to realize that problems and opportunities are two

powerful and complementary lenses for looking at strategy. In fact, a
problem and an opportunity may just be two ways of describing the
same idea. For Zoetis, entering the diagnostics market was a clear
business opportunity. But it could also be described as a customer
problem for farmers—how to detect their animals’ disease sooner to
treat it better.
What begins as a broad growth opportunity will often evolve into
more specific problems to be solved. Zoetis’s entry into diagnostics was
a huge opportunity for the business to capitalize on its position in the
marketplace. But pursuing that opportunity meant choosing which
customers to serve first. Who among its customers had the biggest
unmet need to detect animal disease? What precise problems could
Zoetis help them solve first? This is a typical evolution: from defining an
opportunity to identifying key stakeholders, to understanding their
biggest unmet needs or problems to be solved.
I often advise companies to begin their strategy process with the
problem lens and then to expand their thinking with the opportunity
lens. The best opportunities, in time, will lead back to new problems to
solve.

Tool: Problem/Opportunity (P/O) Statements

Now that we have a clear understanding of why strategic priorities are

important and how they can be identified through the complementary
lenses of problems and opportunities, let’s get started with the first tool
in Step 2 of the DX Roadmap: the problem/opportunity (P/O) statement.
Whether you are using the problem lens, the opportunity lens, or
both, you want to end up with one list of strategic priorities. Each
priority should be summarized in a concise statement that offers clear
guidance and inspiration for a range of different possible solutions. I
call these problem/opportunity statements, or P/Os, for short.
Table 4.2 lists examples of P/O statements that spell out strategic
priorities for the DX Roadmaps for four companies. In each case, the
statement describes a clear strategic opportunity or problem for the
business or its customers.

Table 4.2.
Problem/Opportunity Statements for DX Strategy
Com Problem/Opportunity Statements for DX Strategy
pany
Zoet • Improve the customer’s digital path to purchase—
is from online discovery to e-commerce, to loyalty.
• Provide online learning and content for animal
health professionals.
• Enter the growing diagnostics market to detect and
prevent illness in animals.
• Empower livestock management with digital
tracking and analytics.
Mast • Provide financial inclusion to unbanked
erca communities.
rd
• Solve enterprise needs for cybersecurity via digital
identity authentication.
• Harness our retail transaction data for
analytics and insights.
• Deliver innovation as a service for partners in the
financial services sector.
Air • Leverage the data in our physical assets—from
Liqu manufacturing plants to gas cylinders—to improve
ide operations and unlock new value.
• Empower and connect with our customers across
every channel of communication—from app to web,
to phone call, to sales representative.
• Collaborate with an expanding ecosystem—of
employees, partners, and new start-ups—via new
digital business models and tools for collaboration.
Acui • Offer new insurance policies to commercial
ty customers for the business risks of the digital era.
• Use new data sources in our underwriting and
marketing models to offer the right customer the right
coverage at the right price.
• Provide seamless, omnichannel customer service
and claims.
• Sell insurance directly to buyers wishing to
purchase online rather than through a
traditional intermediary.

Notice that each P/O statement provides a clear description of

where the company will seek to create and capture value, without
deciding in advance the type of solutions it will build. When defining
your own strategy, remember to focus on the “what” and not yet the
“how.” Define the opportunities or problems you will address but hold
off on describing the solutions you may use to address them.
When Amazon began work on what would become its Kindle e-book
platform, the company did not even attempt to specify what the solution
would look like in terms of hardware and software. It defined the
opportunity: “every book ever printed, in any language, all available in
less than 60 seconds.” As Kindle succeeded over several years, the
solution changed significantly, evolving from an e-reader hardware
device to a cloud-based library of books created by both traditional
publishers and self-publishers and accessible on almost any
smartphone, tablet, computer, or e-reader.

P/O Statements at Different Levels

Strategic priorities should be set not just for the entire enterprise but at
other levels as well. P/Os can be used to define the strategy of a single
function (e.g., marketing or human resources) or a specific business
unit (like Pfizer’s animal health unit). P/Os can be defined for an
individual team to clarify strategy for whatever part of the business it
manages. P/Os can even be defined for a specific decision or event—
such as an acquisition of another business—to define the strategic
opportunities that it poses.
For an example of P/Os at different levels of the same organization,
let’s look at the DX of Walmart in table 4.3. At an enterprise level, we
can see three major P/Os for Walmart: define the e-commerce future of
grocery purchases, leverage retail stores to win the last mile of online-
to-offline commerce, and reinvent the interaction of humans and AI in
the retail workplace. Further down in the table, at a channel level
(stores versus e-commerce), P/Os can describe Walmart’s strategic
priorities for each channel. For Walmart stores, one P/O is to “ensure
availability on the shelf when a customer walks in the door to find a
specific product” (an area where Walmart has used robotics
successfully). For Walmart.com, one P/O is to “improve our customer’s
purchase frequency online.” For omnichannel efforts to link the two
channels, an important P/O is to “leverage our store employees to
enable faster delivery of online orders to customers nearby.” Further
down still, at a team level, we can see how specific P/Os guide
Walmart’s DX. The team working on the Walmart mobile app is focused
on solving a number of customer problems, including “improv[ing] the
customer’s returns experience in-store.” This P/O led to the addition of
a Mobile Express Returns feature within the app that allows customers
to scan their paper receipt with their phone, select which items they
want to return, drop off those products in a fast-track lane at the store
using a quick-response (QR) code, and receive a refund within a day.
The final solution reduced customers’ in-store return time by 74
percent.15

Table 4.3.Walmart Problem/Opportunity Statements at Different Levels

Level Scope Problem/Opportunity Statements
of the of
Orga Strate
nizati gy
on
Enter Walm • Define the e-commerce future of
prise art, grocery purchases.
Inc.
• Leverage retail stores to win the last
mile of online-to-offline commerce.
• Reinvent the interaction of humans
and AI in the retail workplace.
Chan Walm • Ensure availability on the shelf when
nel art a customer walks in the door to find a
stores specific product.
Chan Walm • Improve our customer’s purchase
nel art.co frequency online.
m
Chan Omni • Leverage our store employees to
nel chann enable faster delivery of online orders
el to customers nearby.
Team Walm • Improve the customer’s returns
art experience in-store.
mobil
e app

A highly diversified Latin American business that I have advised has

applied the P/O approach to prioritize its DX at three levels. First, each
operating unit focused on strategic opportunities for digital within its
own business (these were in very different industries). Second, the
corporate team focused on strategic opportunities across the
businesses—by linking their data and capabilities to generate new
insights and value. Third, the company’s venture team identified
opportunities outside its existing industry mix—to pursue by
acquisitions, joint ventures, and investing in digital start-ups.

Statements That Spark Ideas

A great P/O statement is meant to spark new ideas for innovation. This
is particularly clear in innovation challenges or hackathons, which can
involve internal employees or outside partners, start-ups, and
customers. At the start of any innovation challenge, the problem-to-be-
solved is declared, and a reward—for example, a cash prize, start-up
investment, or a chance to work at the business—is promised to
whoever comes up with the best solution.
In order to generate useful ideas, it is essential to have a well-
defined P/O statement. Stephen Liguori, the former head of global
innovation at General Electric (GE), told me of their first innovation
challenge: the company simply asked for participants to submit
“innovative clean energy solutions.” They were quickly overwhelmed.
“We got 77,000 entries for five judges to review!” says Liguori. The
submissions included ideas like “put a bunch of electric eels in a
swimming pool with power cables.” In subsequent challenges, GE
offered much clearer problem statements, working with engineers to
add technical detail. In a challenge to explore 3D printing for jet engine
parts, Liguori said, “We received 155 entries, of which 125 blew away
anything we had figured out by that point!”
Clarity, detail, and focus are essential to any good P/O statement.
But grammar can matter too. One easy way to improve a P/O
statement is simply to turn it into a question. The benefit of writing your
P/O as a question is that it will force you to truly focus on the problem,
not the solution. One of my favorite techniques is to begin each P/O
with the words “how might we.” For example, the first Zoetis P/O
statement in table 4.2 would be rewritten as, “How might we improve
the customer’s digital path to purchase—from online discovery to e-
commerce, to loyalty?” “How might we” is a popular phrase in
innovation and design; it was coined by Min Basadur while he was
working at Proctor & Gamble in the 1970s. Today it is widely used at
firms like Google, Facebook, IDEO, and the Cooper-Hewitt National
Design Museum.16 By using “how might we,” you quickly step back
from any presumed or foreordained solutions (e.g., thinking only of the
loyalty program for veterinarians that your marketing teams are already
working on), and open your thinking to more wide-ranging innovations.
A great question is worth a thousand answers.

EIGHT QUALITIES OF GREAT P/O STATEMENTS

From my own observations and coaching of leaders on strategy across
diverse settings, I have identified eight key principles of an effective
P/O statement:

1. It is posed as the question “How might we. . . ?” (“How might we

provide financial inclusion to unbanked communities?”)
2. It focuses on an important problem or opportunity to create value.
(“Improve the customer’s digital path to purchase—from online
discovery to e-commerce, to loyalty.”)
3. It takes the point of view of the customer or the business. If the
P/O statement is for a customer, use language the customer
would use (“works well on my current phone”); if it is for the
business, phrase it in business terms (“upsell customers on our
latest data plans”).
4. It focuses on the desired outcome (“improve tracking in our supply
chain”) and avoids suggesting a solution (not “create a
blockchain ledger to track our supply chain”).
5. It is open enough to inspire different choices. (“Enter the market for
diagnostics,” not “acquire diagnostics company X.”)
6. It is narrow enough to give helpful guidance. (“Improve the in-store
product returns experience,” not “create a better user
experience.”)
7. It avoids compounds. (Instead of “reduce shipping errors and
increase speed,” write one P/O to “reduce shipping errors” and
another to “increase shipping speed.”)
8. It includes a measurable definition of success. (“Reduce employee
onboarding time by 75 percent.”)
 The real test of a P/O statement is the inspiration it sparks in those
who read it. Remember that a good P/O will help you to fall in love with
the problem, not the solution.

From One P/O to Many Ventures

As you begin to develop a short list of strategic priorities for your own
team, business unit, or enterprise, it is important to understand the link
here between strategy and innovation. Put simply: each P/O statement
should generate many different innovation ideas. If every P/O poses a
question (“How might we?”), that single question should generate
numerous possible answers. Those answers—new products, services,
processes, business models—each suggest a different possible
solution to that same problem or opportunity.
We can see this illustrated in the case of the NextGen Cup
Challenge, a moon-shot innovation project launched jointly by
Starbucks and McDonald’s with the support of design firm IDEO. The
two sponsor businesses alone produce billions of paper cups each
year, and in order to hold a customer’s hot coffee durably, these paper
cups are coated with a plastic lining that prevents recycling. Both
companies are looking for innovative ideas to reduce the environmental
impact of all those cups. Rather than submitting a request for proposal
(RFP) for cup manufacturers to build a specific solution, they
announced an open innovation challenge, with a clear P/O statement to
solve the problem of coffee cup waste.
The NextGen Cup Challenge has attracted submissions from
hundreds of teams proposing a wide range of solutions to the same
P/O statement. CupClub adapted ideas from bike sharing, with an
innovative plan for reusable cups that could be dropped off in specially
marked recycling bins. Muuse proposed putting QR codes on reusable
cups to enable tracking, collection, cleaning, and reuse. Colombier
Group focused on the dimension of recycling instead of reuse, aiming
to replace the plastic layer with a water-based coating that is recyclable
or compostable. The success of these ideas will require iterative tests
and prototypes in the real world to validate technological feasibility,
human behavior and adoption, and economics at scale. To begin that
process of validation, winning teams were each given $1 million to test
and attempt to commercialize their ideas.17
The NextGen Cup Challenge demonstrates the goal of any good
P/O statement: to generate a variety of possible innovations, or what I
call growth ventures. A growth venture can take many forms. It could
be a process innovation, a new product or service, or even a new
business model. I use the term “growth ventures” to underline that
these initiatives should always be defined in terms of growth. Any
digital innovation at this step should be defined as creating value and
capturing value for the firm. Be careful that your own growth ventures
are not defined in terms of technology (“invest in cloud computing”), or
skill building (“upskill our marketers on social media”). That kind of
capability building is critical, but it will come in Step 5 of the DX
Roadmap, when you have already begun to explore and test a variety
of possible innovations sparked by your P/Os.
Table 4.4 shows how a single P/O can lead to many possible
venture ideas for a company. At Walmart, teams innovating on the P/O
of “How might we provide convenient online grocery ordering for
customers?” have pursued ideas that range from an annual
membership fee for grocery delivery to a free buy-online-pickup-at-
store (BOPS) experience, to partnering with the delivery app
DoorDash, to having Walmart employees drop off packages on their
drive home.

Table 4.4.Each Problem/Opportunity (P/O) Leads to Multiple Possible

Ventures
Com P/O Different Possible Growth
pany Ventures for That P/O
Wal How might we • Unlimited grocery
mart provide convenient delivery with annual
online grocery membership.
ordering for
customers?
• Buy-online-pickup-at-
store (BOPS) experience,
with no charges.
• Partner with the delivery
 app DoorDash.
• Walmart employees
drop off packages on their
drive home.
New How might we reach • Launch a daily news
York audiences seeking podcast that interviews
Time an audio experience our reporters on top
s of journalism? stories they are covering.
Com
pany
• Launch a mix of opinion
podcasts on different
themes, as part of our
Opinion department.
• Acquire Serial—a
producer of popular
investigative nonfiction
podcasts.
• Acquire Audm—a start-
up that hires voice actors
to read long-form
journalism from leading
magazines.
Citib How might we • Online platform to
ank increase economic connect investors with
vitality for opportunity zones
underserved investment.
individuals and
communities?
• Digital tool for job
seekers, using market
data to evaluate career
paths and develop skills.
• Invest in fintech
start-ups offering
peer-to-peer
 lending or
microcredit.
Star How might we • Create a sharing system
buck reduce the impact of for reusable cups
s coffee cup waste? dropped off in recycling
and bins.
McD
onal
d’s
• Put QR codes on
reusable cups to enable
tracking, collection,
cleaning, and reuse.
• Replace the cups’
plastic lining with a water-
based coating that is
recyclable or
compostable.

A major opportunity for DX at the New York Times Company has

been to use digital audio to engage new audiences. Its different
ventures have included news podcasts with Times reporters talking
about the biggest stories of the day, opinion podcasts from the writers
at its Opinion desk, acquiring a successful podcast producer, and
buying a start-up that produces audio versions of articles from leading
magazines.
At Citibank, one key P/O for digital has been “How might we
increase economic vitality for underserved individuals and
communities?” This has led to multiple innovations, including an online
platform connecting investors to opportunity zone investments, and a
data-driven tool to help job seekers evaluate career paths and develop
new skills. At the same time, Citibank’s venture investing arm can
contribute to this same P/O by investing in start-ups that offer their own
solutions to economic empowerment.
Our goal, then, for the strategy process is that a few P/Os will
generate a venture backlog—a prioritized list of new growth venture
ideas that are aligned with your strategic priorities and your shared
vision. The appropriate team, unit, or division can then begin validating
and testing those venture ideas and investing in those that work best. (I
am borrowing the term “backlog” from its use and meaning in agile
software development.) We will see more about the use of a venture
backlog in Step 4 of the DX Roadmap, the topic of chapter 6.

Effective Tools to Identify P/Os

Where do great P/O statements come from? First, they should emerge
from the shared vision that you defined in Step 1 of the DX Roadmap.
There are also some helpful tools commonly used in customer insight
research. And I particularly recommend using four of the strategy tools I
developed in The Digital Transformation Playbook. Let’s look briefly at
how you can bring these tools together to identify valuable P/O
statements for any business.

Shared Vision

The first place we can look to identify valuable P/Os is in the shared
vision that we defined in Step 1 of the DX Roadmap. This includes our
future landscape, right to win, North Star impact, and business theory.
Each of these four elements should assist us in identifying P/Os.

FUTURE LANDSCAPE
Your future landscape is a shared view of how your business context is
evolving based on insights about your customers, new technology, the
competitive ecosystem, and broader structural trends in the economy.
These are precisely the kinds of insights that should point you to new
problems and opportunities for your business. Let’s look at some
examples:

• Zoetis’s P/O to “Empower livestock management with digital

tracking and analytics” arose from trends in big data and IoT, as
well as understanding how midsize farms were still struggling to
adopt these technologies successfully.
 • Walmart’s P/O to “Define the e-commerce future of grocery
purchase” arose from seeing a customer need in grocery delivery
that competitors had still not solved, leaving more space for
innovation than other categories of e-commerce.
• Acuity’s P/O of “Offer new insurance policies to commercial
customers for the business risks of the digital era” was spurred by
trends toward robotics and automation within manufacturing firms,
the growing gig economy, and customers’ vulnerability to
cybersecurity risk.

RIGHT TO WIN
P/Os should also arise from your right to win—your definition of the
unique advantages and limits of your own organization. As Michael
Porter says, “Strategy is about being different. It means deliberately
choosing a different set of activities to deliver a unique mix of value.”18
By linking your strategy to your right to win, you can pursue
opportunities where you have a unique advantage.

• Intuit used the unique advantage of its data—payroll and tax data
shared by customers using its existing products—to identify an
opportunity to provide credit for small businesses unserved by
other lenders.
• Walmart used its own unique advantages—its retail store network
and its shopper data—to identify opportunities to enter consumer
health care and financial services.

NORTH STAR IMPACT

Your North Star impact—the long-term impact you seek to achieve—
should also be a source of guidance in identifying P/Os.

• Mastercard’s North Star impact, to “power and protect secure

commerce in the digital world” helped point it toward the P/O of
solving enterprise needs for digital identity authentication.
• Ford Motor’s North Star impact, defined by Bill Ford, is to “meet the
environmental and mobility needs of a growing, urbanizing planet.”
 Several P/Os emerged from this, including those for autonomous
vehicles and fleet management, electric vehicles and charging
networks, ride-sharing and urban mobility solutions, and in-vehicle
transactions and payments.

BUSINESS THEORY
Your business theory—explaining how you expect to capture value
from your investments in the future—can point you to specific P/Os for
your business.

• Walt Disney’s business theory map from 1957 clearly identified

P/Os in theme parks, merchandise, television and music,
character licensing, and books, magazines, and comics.
• Amazon’s theory of the virtuous cycle of product selection and
customer growth pointed the company to the P/O of working with
third-party sellers.

Customer Insights Tools

In addition to your shared vision, I have found that many popular

methods for customer insights research can be useful in identifying
P/Os. Three tools used in lean start-up, agile, design thinking, and
product management can be particularly helpful in uncovering customer
problems to solve and opportunities for customer delight. Let’s look at
each.

GET OUT OF THE BUILDING

The spirit of lean start-up is perhaps best captured by Steve Blank’s
imperative to “get out of the building”19—to stop creating business
plans in your office and engage directly with the customer to learn from
them. In fact, the first stage of the lean start-up process of “customer
development” starts with simple customer interviews—no prototypes,
no MVPs, no slide decks—just talking with the customer on their own
turf.
 UTC’s digital team spends regular time in the field—from Las Vegas
to Shanghai, to Turin—talking to industrial customers to gauge their
interest in the digital innovations it is considering. At Zoetis, the digital
team goes on regular customer ride-alongs, such as spending a day
with a veterinarian to learn what is happening in their business and how
industry needs are changing. Observing a day in the life of a customer
will yield far more insights than a survey or focus group.

PR/FAQ
Another powerful tool for identifying and defining P/Os is what Amazon
calls a press release/frequently asked questions (PR/FAQ). This tool
was developed as a way of instilling the philosophy of “working
backward from customer needs.” In other words, don’t start with the
product you intend to sell; start with the impact you want to create for
your customer. The PR/FAQ comes in two parts: a press release that
announces the imagined product as if it were ready to launch, and an
FAQ section that answers additional questions. The press release is
less than a page long. It should name the product in a way that the
customer will understand. It then describes the benefits the customer
will gain and the problems that it will solve. It concludes with quotes
from the company and a hypothetical customer describing the value
they are receiving from the new product. The FAQ follows, and it can
be up to five pages long. It answers a list of expected questions from
the customer (How will the product work? What will it cost?) and also
from the business (What about market size, economics, technical
feasibility, business partners?).20

CUSTOMER JOURNEY MAPPING

Customer journey mapping is a popular research method widely used
in both design thinking and agile. The customer’s current experience of
your business is defined as a journey, broken into a sequence of stages
(e.g., from prepurchase to postpurchase). Research is conducted at
each stage to understand the customer’s behavior (actions taken,
touch points used), and experiences (their goals, motivations,
questions, feelings, pain points). These findings are used to identify
strategic opportunities for improvement.
Customer journey mapping can be a powerful tool to identify P/Os
for your current business, especially customer problems to solve. But
take note: this tool has become so popular that I have met companies
who believed the purpose of their DX was to “map customer journeys.”
Don’t mistake the process for the goal! The purpose of DX is, and
remains, value creation, which begins with a great P/O statement.

Digital Strategy Tools

Several strategy tools in my last book, The Digital Transformation

Playbook, are designed to help any business identify strategic
opportunities, that is, P/O statements. Four tools in particular can be
helpful to you:

• Customer network behaviors—In the chapter on customer strategy,

I describe five customer behaviors—the desires to access,
engage, customize, connect, and collaborate—that influence us in
the digital age. The tool explains how to uncover opportunities to
create value (in products, services, and customer experiences) by
designing around these core behaviors of customers.
• Data value templates—In the chapter on data strategy, I describe
four templates for value creation from data—insights, targeting,
personalization, and context. The accompanying tool shows how
any business can identify opportunities for value creation by
applying these templates to its existing data assets and taking
steps to grow those assets over time.
• Competitive Value Train—In the chapter on competitive strategy, I
introduce the Competitive Value Train (also described in chapter 3
of this book). By applying this tool to any part of your current
business, you can identify the upstream and downstream partners
that help you deliver value to your ultimate customer, and the
value exchange and leverage that exists at each point of
interaction. With those insights, the Competitive Value Train can
help identify new business opportunities that give you more
leverage in your ecosystem—such as increasing the uniqueness
 of your role, extending your business further upstream, or
extending it downstream to a direct customer relationship.
• Value Proposition Roadmap—In the chapter on value proposition
strategy, I introduce the Value Proposition Roadmap (also
described in chapter 3 of this book). By analyzing your current
value proposition and its declining and growing elements of value,
this tool will help you identify strategic P/Os that will increase your
future value for each of your customer segments.

Tool: The Problem/Opportunity (P/O) Matrix

We are now ready to introduce another new tool, the

Problem/Opportunity (P/O) Matrix. The purpose of this tool is not to
generate a single new P/O. Rather, it is to organize and clarify a short
list of the most important P/Os for any business, division, unit, function,
or team. As such, it is best used after initial work with some of the tools
described in the prior section.
The Problem/Opportunity Matrix in figure 4.1 has two dimensions:
problem versus opportunity (our two lenses for defining strategy), and
customer versus business (the points-of-view of external versus
internal stakeholders). The resulting matrix is composed of four
quadrants—customer problems, business problems, customer
opportunities, and business opportunities. Each of these four quadrants
offers a different means of identifying a strategic priority for your
business. But remember that the same P/O can be written in different
ways (e.g., as a customer problem or a business opportunity) and that
P/Os evolve over time. The point of the P/O Matrix is not to select the
“correct” quadrant for any given strategy. Rather, use the four
quadrants to help you spot different priorities for growth that you might
otherwise miss.
Figure 4.1.
The Problem/Opportunity (P/O) Matrix

I developed the Problem/Opportunity Matrix while working with

numerous organizations to define the most important strategic priorities
in their DXs. Let’s briefly walk through each step of the tool to see how
to apply the Matrix in your own organization.

1. Pick Your Level and Your Customers

First, start by deciding which level of your business you are defining
strategic priorities for. Is it for a product team? A particular sales
channel? A business unit or geography? A function like marketing,
finance, or HR? The entire organization? The P/O Matrix can be
applied at any level of your enterprise. So, before you begin, you need
to choose the organizational unit you will focus on.
Next, identify your most important customers. These are your key
stakeholders outside your chosen unit. You will typically have multiple
customer types. Recall that Zoetis’s customers include pet owners;
farmers raising cattle, swine, and other animals; and the veterinarians
who serve each of them. If you are using the Matrix to set strategy for
an internal function (like HR or finance), your customers may include
stakeholders in other parts of your company. For an HR division,
“customers” could include company employees as well as colleges
where you recruit talent. For a supply chain division, “customers” could
include teams or business lines whose products you deliver.

2. Write P/Os for Each Quadrant

Now that you know your organizational unit and the customers you are
serving, it is time to start writing down P/Os for each of the four
quadrants of the Matrix.

• Customer problems—For the first quadrant, look at each of your

customers that you identified in step 1. For each one, what are
their biggest, most enduring problems that you might help them to
solve? What is their most pressing unmet need?
• Business problems—In this quadrant, ask: What are the most
pressing pain points currently facing your part of the business?
(Again, think of the organizational unit you chose in step 1.) For
each problem, who is the key stakeholder (for example, a
marketing manager or HR director) that will need to use and adopt
any solution?
• Customer opportunities—Look again at each of your customers
identified in step 1. What unexpected value, gain, or benefit could
you create for them? Try writing a “customer delight statement” for
each customer—much like the Kindle’s promise to offer “every
book ever printed, in any language, all available in less than 60
seconds.”
• Business opportunities—In this quadrant, ask, Where could my
unit expand to create and capture new value for the business? Try
to identify an attractive growth market where you have a clear right
to win. Look for a new capability with a specific application to your
business. (For example, instead of “use machine learning,” write
“use machine learning for drug discovery.”) Try writing a 10x
 stretch goal—for something you normally try to improve 10 percent
but where there might be an opportunity to improve by 10x.

Within each quadrant, identify as many interesting P/Os as possible.

Aim for at least three or four per quadrant.
As you write each P/O statement, think back to the eight qualities of
great P/Os identified earlier in this chapter, and follow these guidelines:

• Make it a question—using Min Basadur’s words: “how might we.”

• Avoid compound sentences—if your P/O has the word “and” in it,
split it into two ideas!
• Never give the “how”—specify the outcome you want, not the
means. A good P/O is not a solution masquerading as a problem.
• Write with a clear point of view—if you are writing a customer
problem, don’t use business jargon; use the language the
customer would use.

Remember that a given P/O could be written differently to fit in more

than one quadrant. If one of your P/Os doesn’t seem crisp or
compelling enough, try rewriting a problem as an opportunity (or vice
versa), or flipping the perspective from the business to the customer (or
vice versa).

3. Pick Your Keepers and Combine

In the next step, you should review all the P/Os you have written and
choose only the most important P/Os to keep. Each one should be
focused on solving a real problem or creating significant value. Each
should clearly matter to a specific customer (external) or the business
(internal). The P/Os should be interesting as well, sparking new
thoughts about how you might create and capture value. Then, take
your chosen P/Os out of the four quadrants and combine them into a
single list. You should now have one list of P/O statements written as
questions (“how might we…?”).

4. Refine Each P/O

Take some time to discuss and refine each of the P/Os on your list. For
each P/O, spell out again who the key stakeholder is that must use and
benefit from any eventual solution.
Be sure each P/O is narrow enough to give focus to a team. For
example, do not write “How might we attract more customers?” Instead,
write “How might we engage audiences who prefer audio journalism?”
Be sure each P/O is broad enough so that it can be answered in
multiple ways: not “How might we create a daily news podcast that
interviews reporters on the morning’s top story?” but “How might we
use audio to engage readers with our news coverage?”

5. Brainstorm Test

Now it is time to give your P/Os what I call the brainstorm test. Look at
each of your refined P/O statements and try to brainstorm multiple
ideas for solutions—that is, different answers to the same question. As
you do, try to find ideas that are genuinely different from one another;
that is, they take a different approach to solving the same problem or to
capturing the same opportunity.
This step will reveal if a P/O will be genuinely helpful to future
innovation efforts. If you can think of only one solution idea for a P/O,
seek the input of others to broaden your thinking. If you still can see
only one solution, your P/O is probably a “solution masquerading as a
problem.”

6. Success Metrics

In the last step, you will try to define success metrics for each of your
P/Os. Ask yourself, How would we use data to choose the best solution
among the ones we just brainstormed and others that may be proposed
in the future? For each P/O still on your list, define a metric (or two or
three) that would distinguish the best solution for that
problem/opportunity.
If you have done your work right, you should now feel willing to
assign a team of high-value employees to work on any one of your
P/Os. And you should be willing to give that team wide latitude to
explore new innovations, knowing they will test and validate them with
your success metrics as their guide.

Strategy from the Bottom Up

The approach to strategy that we have seen in this chapter is a

departure from how many established businesses operate. At many
organizations, “strategy” is an annual agenda item for a select group.
Senior executives take a few weeks each year to spend time in
meetings or perhaps at a retreat to update the strategy for the firm. The
resulting strategy is a document—a deliverable, like a budget, that must
be handed in once a year—that is meant to trickle down to the rest of
the organization.
But this model completely fails the challenges of the digital era, with
its constant change and need for adaptation. For any business today,
strategy must happen not just at the top but at every level of the
organization. And strategy must become not just an annual product but
an ongoing, continuous process.

Strategy at Every Level

As we have seen, strategic priorities (P/Os) are critically important at

every level of the organization. Whether you’re leading DX for the firm,
division, function, or team—every organizational unit should be focused
on identifying its most important problems to solve and opportunities to
create new value. When an organization makes that shift, people at all
levels of the business will be engaged in strategy.
At the energy services firm Schlumberger, every business unit head
is expected to develop their own digital strategy, drawing on guidance
and insights from the corporate level digital strategy team. In its 2020
Group report on digital innovation, the New York Times laid out a goal
that every “desk” (a group of journalists covering an area of news)
needs to have its own strategy statement—defining its customers
(target audience), value proposition (what it will and will not cover),
competition (and how it will distinguish itself from them), metrics (what
success looks like), and operations (what skills it needs to develop and
how it will interact with other desks or departments).21
Any company’s overall enterprise strategy should shape the work of
every other part of the organization. But again, this influence should
happen by the process I call cascading up. A senior leader does not tell
her reports, “My strategic priorities are X, so yours should be Y.”
Instead, she should say, “My strategic priorities are X,” and then ask,
“What do you think your team’s strategic priorities should be to support
this?” Through cascading up, company P/Os are linked to strategic
priorities at every level and in every department.

Strategy as Ongoing Conversation

Defining your strategic priorities is a continuous, ongoing process.

Every leader and team should regularly review its P/Os to consider and
incorporate new learning. A strategic priority may wax or wane in
importance for many reasons: customer interest may fade, new
competitors may enter the same space and reduce the likelihood of
profit, an unexpected event like the COVID pandemic may scramble
your time frames or add new external constraints.
Strategy must be reimagined—not as a deliverable but as an
ongoing process that everyone contributes to. Donald Sull describes
strategy as a series of conversations that need to happen at every level
of an organization, in an ongoing, iterative fashion.22 This approach
can be seen in the DX of UTC. Steve Serra, who leads UTC’s digital
accelerator, describes how its teams have regular meetings with the
presidents of UTC’s business units to understand what is top-of-mind in
their business, to show them emerging trends and examples from other
markets, and to identify strategic opportunities that are of the greatest
importance to their business.
A powerful shift in mindset will happen as everyone gets involved in
these strategy conversations. At every level of the organization, people
will start to take a problem-defining mindset. Their every task, project,
or workflow will begin with a step back to ask, “Why am I doing this?
What am I solving for?” As everyone begins to make that shift—as they
start to think of new ways to solve the problems that matter most to
their own work—their creativity will be unleashed.
 Teams and Leaders

The results of doing strategy continuously and from the bottom up will
be clear. Every team will be guided by its own shared vision defined for
its particular division or unit. Every team will have its own set of
strategic priorities—perhaps three to seven P/Os—that guide its own
work and are aligned with the priorities of the whole company. And
every team should have its own venture backlog—a list of perhaps ten
to thirty ideas for innovations to pursue—that changes regularly but is
aligned to its strategic P/Os.
As strategy becomes more bottom up, leadership changes as well.
The role of a leader is no longer to think of the right answers but rather
to frame the right questions.
Recall the three jobs of a leader from chapter 3. These translate into
three tasks that any leader should focus on in relation to strategy. First,
leaders must define the most important problems and opportunities that
will guide their team’s efforts. Second, they must communicate these in
clear P/Os and work to align everyone in their team. Finally, they must
empower others as they test new ideas and pursue new solutions to
these shared strategic goals.

For digital transformation to succeed, you must first define where it will
compete and seek to create value. Without priorities, any DX will
become a series of scattered projects that are disconnected from the
needs of the business and easily hijacked by the hype of new
technologies. In Step 2 of the DX Roadmap, you saw how any
organization can define its strategic priorities and link DX to a clear
agenda for growth. You saw the power of thinking in both problems and
opportunities, and how various strategy tools can help define P/Os for
your own business. And you saw the link between strategy and
innovation—why you must first define the problems that matter most
before seeking ideas for solutions.
Every strategy is a work in progress, but with a good first draft in
place, you can now begin the next step of the DX Roadmap. With P/Os
identified and ideas for new digital innovations sparked by them, you
are ready to begin turning those ideas into real-world growth ventures.
As we embark on Step 3 of the DX Roadmap, we will grapple with
the challenge of uncertainty identified in chapter 2. Humility and
experimentation will be key. To turn your new venture ideas into
engines of growth for your business, you don’t need a crystal ball. You
need a clear process to rapidly validate any idea, test its business
assumptions, and discover if it will deliver results at scale in the real
world. In chapter 5, we will see how to validate new ventures
throughout your organization and start to create real value through DX.
 5
Step 3: Validate New Ventures

When I visited Walmart Labs in Silicon Valley to meet with COO Jeff
Shotts, his company was engaged in a broad digital transformation
(DX). Walmart was pursuing numerous innovations at the same time,
ranging from operational improvements to new customer experiences,
to new digital business models.
 One recent innovation was Jetblack, an invitation-only service for
conversational commerce. Jetblack allowed customers to text a
shopping request—for example, “I need a birthday gift for a ten-year-
old boy”—and, with a few exchanges back and forth, have the perfect
product delivered within a day. In another innovation, robots that
roamed the aisles of Walmart’s retail stores cleaned the floors while
using machine vision to scan shelves for products that needed
restocking. Another robotics solution was being developed in Walmart’s
fulfillment centers to expedite pick-pack-ship for a selection of millions
of products offered on its website. Meanwhile, Walmart’s plans for
online grocery ordering were gearing up for a broad national rollout.
With each digital innovation, Walmart’s teams were careful not to
“fall in love with the solution.” Instead, in teams across the company,
Walmart was engaged in a constant cycle of rapid experimentation—
testing in the marketplace, often with quick and simple prototypes, to
learn what would and would not work in the real world. Much of this
learning was about customers. What were their biggest unmet needs?
Which features mattered most to them? How could Walmart create the
right customer experience? Other learning related to operations. What
were the technology requirements for a given solution? What security
or data privacy issues needed to be addressed? How scalable was this
idea in practice? Still other experiments focused on economics: running
tests on pricing, operating costs, and the value to the firm of benefits
like additional customer data. As Shotts explained to me, with every
innovation, Walmart must “try to figure out how can we scale an offer
that’s good for the customer, while still meeting some economic
thresholds that we have.” This means being open to learning as you
test and validate new ideas in the market—and staying flexible on your
next steps.
One example of such learning came from Walmart’s robots, where
market testing led to different decisions for different use cases. The
robotic floor sweepers performed well at first, saving labor costs on
restocking of store shelves. So they were rolled out from a few
locations to a national test in 10 percent of Walmart stores. But
ultimately, the program was shut down when the company found it
could achieve comparable results by improving human performance.1
In Walmart’s fulfillment centers, however, robots have remained key to
the online business model. In fact, they found a new role as Walmart
began to build much smaller warehouses, each one adjacent to a
traditional store. Robots in these small warehouses worked in tandem
with human employees who walked the aisles of stores to pull the right
products for same-day delivery. Testing at a store in New Hampshire
showed this hybrid solution led to more product availability, faster order
fulfillment, and greater utilization of Walmart’s warehouse space. Those
proven results led to a much bigger investment and rollout.2
Not every innovation moves forward after testing. Shotts pointed to
the example of Jetblack, the service using text messages to provide
concierge shopping. Customers loved it. “We’re seeing that we have
launched a value proposition that resonates with customers enough
that they’ll stop using the competition [Amazon]. But we can’t stop
there. We have to figure out, how would we scale this offering for 200
million people in the US—and can we do it profitably?” Jetblack’s early
trial was in select urban markets. It relied on a human-powered back
end that would never scale but allowed Walmart to test what user
experience would shift customer behavior. But delighting the customer
was not enough. The goal had been to run Jetblack on machine-
learning algorithms, but the technology was not ready to take over from
human agents. Scaling up Jetblack would be impossible. So Walmart
took what it had learned and chose to wind down the project.
At the same time, Walmart’s experiments in online grocery ordering
proved there was real customer demand and multiple paths to meet it
at scale. One key unknown was the customer’s willingness to pay for
grocery delivery. “It’s hard to make money on a $25 basket if you are
delivering it for free,” Shotts explained. “I’m not opposed to testing that.
But scaling it is a real challenge.” Walmart ran numerous experiments,
testing different price points for paid delivery, free delivery with a
minimum basket size, and an annual membership model like Amazon
Prime. With Walmart’s low prices and razor-thin margins, finding the
right formula was essential to drive customer adoption while being
financially sustainable. Ultimately, the firm launched a nationwide
service called Walmart+, where membership gave you free one- or two-
day delivery of online products, plus same-day delivery of groceries
from a store. Grocery delivery was also offered without a membership,
priced at $7.95 per delivery.
Testing and learning does not end when a product or service
launches, though. As Walmart’s online grocery service grew in market,
the company continued to learn and adapt to customers’ changing
behaviors. Widespread lockdowns during the COVID-19 pandemic led
to a huge surge in demand for what Walmart called “click and collect”—
ordering groceries online, then driving to the store where an employee
would bring out your bags and put them in the trunk of your car.3 As
lockdowns eased, Walmart discovered a surprising new customer wish
—to have a delivery person not only bring groceries to their home but
come inside and put them away. The service, dubbed Walmart InHome,
comes with greater personalization, additional security measures, and
a premium price for customers.4

Why Experimentation Matters

In describing the New York Times Company’s remarkable DX during his

tenure, executive editor Dean Baquet confessed, “We have no idea
what the future of journalism is going to look like. When we started
podcasts, we had no idea. We take risks, we screw up, we try stuff . . .
we don’t know what’s going to stick.”5
Transformative leaders share Baquet’s humility. They recognize
from the start that any digital idea they have is just a set of untested
hypotheses—about customers, competition, operations, profits,
technology, and more. They approach every idea ready to learn
because they know what they don’t know. This is why the Four
Religions—lean start-up, design thinking, agile, and product
management—are each built around rapid cycles of testing and
experimentation. Each one starts by recognizing the uncertainty of any
new venture.
Many businesses leaders mistakenly think that the key to innovation
is coming up with good ideas. But ideas are rarely a source of
competitive advantage. The real challenge of innovation is to become
great at validation: learning as early and cheaply as possible which of
your ideas could work and how. The mistaken focus on ideation comes
in part from traditional innovation theory that put all its attention on
where ideas come from. But modern innovation theories give far less
attention to ideation; instead, they focus on the best methods for
validating and improving on ideas (by identifying their assumptions,
talking to customers, designing MVPs, gathering data, and adapting
based on learning). Great ideas are made, not born.
In the first two steps of the Digital Transformation (DX) Roadmap,
you worked to define a shared vision of the future, chose strategic
priorities (your P/Os), and began to generate ideas for new ventures
based on those P/Os. In the third step of the DX Roadmap, your goal is
to rapidly test these new ventures to validate which ones may work and
how.
The best digital businesses succeed because they learn rapidly
from customers and the market, and use that learning to revise, pivot,
and adjust an idea until it reaches the right combination of factors for
liftoff. The challenge for Walmart and any business pursuing DX is to
become good at doing this for different ideas across the company.
Growth ventures may take many shapes: new products, new customer
experiences, new marketing strategies, or entirely new business
models. To build even one venture that has an impact on your
business, you will need to test many ideas, shut some down, and
change others, repeatedly, in response to testing and feedback. Doing
this requires important shifts in how companies think about and
manage innovation:

• Think like a scientist—Avoid debates based solely on opinion.

Instead, ground every strategic decision in data. Use data from
your own experiments, not preexisting third-party data. Every
experiment should begin with a theory and aim to validate
something specific—about your customer, your technology, the
problem you think you are solving, and so on. Think in hypotheses.
And don’t design tests to prove you are right; instead, think like a
scientist and design tests that could prove your hypothesis is
wrong. If your venture keeps passing those tests, you know you
are on to something!
• Get to market sooner to learn faster—Push to get your ideas into
customers’ hands in the real world as soon as possible. Only six
months after beginning work on Intuit’s small-business credit
product, Rania Succar’s team launched a streamlined version in a
limited release in the state of Georgia. It felt incredibly early, but it
proved incredibly valuable. “We learned that everything we
expected to happen in the product . . . it was 100 percent
 different,” she explained to me. Her team had to pivot immediately
and change the scope of the product to respond to the features
that mattered most to drive customer demand. “What I learned
from that experience is that you have to get to market super fast,
to know what you don’t know. Because your roadmap is probably
all wrong, based on assumptions you’re thinking about in your
head. So you’ve got to push teams to get to market quickly.” As
Steve Blank puts it: no business model survives contact with the
customer.6 Remember that all true learning comes from the
customer—not from benchmarks or experts. So the only way
forward is to go to customers directly, as early and often as
possible.
• Accelerate your learning—Once you have started to document
your assumptions and test your ideas with MVPs, the next step is
to accelerate the whole process. “Test in minutes, not months” is
an axiom of design thinking. Agile methods like Scrum set their
cadence around a unit of work called a sprint, which includes a
cycle of ideation, design, writing code, publishing it, and measuring
response. Sprints are short, from one to four weeks. When
Jonathan Becher was building SAP Digital, he found the most
important metric for his start-up within an enterprise was not return
on investment (ROI) but speed measured as time-to-market for
each new innovation, feature, or experiment.7 Experimentation is a
constant cycle of learning and adaptation. Each test will shed new
light on the opportunities and weaknesses of your idea, which in
turn will lead to new ideation and new questions to test. Speed up
this learning cycle.

Innovating through rapid experimentation may sound simple in

theory, but it is hard in practice. I have seen many established
organizations hire agile coaches or conduct design thinking workshops
but then struggle to change the way they operate. Leadership remains
focused on choosing a few big ideas. Detailed analysis and business
cases are expected before any decisions are made. Once a solution is
chosen, everyone is locked into delivering it according to plan. Some
genuinely good ideas may emerge, but they advance slowly compared
to digital-native competitors. In these organizations, DX produces a
stream of projects that make good press releases but never deliver a
sustained impact on business fundamentals.
Becoming a truly experiment-driven company requires deep change
in the practices, habits, and mindset of most organizations. But it is the
only path to effective DX. Table 5.1 shows some of the key symptoms
of success versus failure in Step 3 of the DX Roadmap.

Table 5.1.
What’s at Stake—Step 3: Experimentation
Symptoms of Failure: Symptoms of Success:
Experimentation Experimentation
• Innovation is focused on • Innovation is focused on
coming up with a few great testing many ideas to learn
ideas. which work best.
• Decisions are made based • Decisions are made based
on business cases, third-party on experimentation and
data, and expert opinion. learning from the customer.
• Once they start a • Teams stay focused on the
project, teams are problem but flexible on the
committed to building the solution.
solution in full.
• Failures are costly, so the • Failures are cheap, so
fear of risk is high. there is a bias toward risk
taking.
• Good ideas move slowly • Good ideas grow fast and
and don’t seem to move the deliver business value at
needle on the business. scale.

What’s Ahead

In this chapter, we will see how any team in any organization can
rapidly validate new ventures and drive new digital growth. I will provide
a new model, the Four Stages of Validation, to organize the process of
continuous learning for any new venture. You will learn how to use
MVPs to test your ideas in the market, and you will learn the difference
between illustrative and functional MVPs. I will also provide a new tool,
the Rogers Growth Navigator, to guide any new venture through the
Four Stages of Validation, and from the earliest napkin sketch to global
delivery at scale. Finally, you will learn why experimentation at every
level and in every function and department of your business is critical to
becoming a bottom-up organization.
Before we get to the Four Stages of Validation, let’s look at two
elements that are often misused or misunderstood: MVPs and metrics.
Each is essential to the Four Stages of Validation.

MVPs That Accelerate Learning

One of the most important but most misunderstood ideas in innovation

theory is the MVP, or minimum viable product. Let me offer my own
definition: an MVP is a minimal artifact designed to test a business
assumption. An MVP can be as simple as a napkin sketch that you
show to a prospective client across a table or as complex as a beta
version of a video game released to early customers for feedback. The
term “MVP” was coined in the 1980s by Frank Robinson and
popularized by Steve Blank, Eric Ries, and others as a central element
of the lean start-up method.8 The main confusion around MVPs comes
from the word “product” in the name. As Blank and Ries have stressed,
an MVP need not be a product at all—in fact, it is often much less than
a working prototype.
The key to a great MVP is two things: minimal cost (in time as well
as money) and maximum learning. The “lean” in lean start-up is
because your goal is not just to learn what does or doesn’t work in your
business venture but to spend the absolute minimum time and
resources to learn it.9 Stop spending months building a complex
prototype just to validate if customers are interested in your product—
you could likely learn that in days with a simple online test! In fact, a
rougher MVP is often better for generating insightful customer
feedback. As IDEO’s Joe Brown told me, “If you show the customer a
polished looking prototype, they will see only its flaws. If you show them
a very rough one, they’ll see its potential.”
One famously rough MVP was used by Marc Lore and Vinit Bharara
to test their vision of an online store focused on baby products,
Diapers.com. For an initial MVP, the two entrepreneurs announced an
offer—overnight diaper delivery—on their and their wives’ personal
Facebook pages. By morning they had received 240 orders, which they
scrambled to fulfill by driving their family minivans to nearby stores and
sending packages from their local UPS Store. This bare-bones
experiment was no test of the operations, costs, or profit margins of a
scalable business, but the founders got their first real-world data on
customer demand, product preference, and price point, all in just
twenty-four hours. Five years later, after phenomenal growth, they sold
their start-up for over $500 million.10

Illustrative Versus Functional MVPs

In many organizations, terms like “MVP,” “prototype,” and “proof of

concept” are used without a clear definition or purpose. The result is
innovation where MVPs are too few and not at all minimal—they
consume far too much time and resources and yield inconclusive data.
By contrast, effective innovators have a clear understanding of the
different types of MVPs they are using and why.
I have found it useful to think of two broad types of MVPs:
illustrative MVPs and functional MVPs. Each type has its place, helping
you learn different things at different stages of validating a new venture.

ILLUSTRATIVE MVPS
Illustrative MVPs illustrate the benefits, features, and design of your
proposed solution, but they do not yet deliver those benefits to the
customer. An illustrative MVP could be a napkin sketch that you show a
prospective customer to gauge their reaction. It may be a static wire
frame showing screenshots of a planned digital experience or an
interactive wire frame that includes scrolling, buttons to click, and
simulated feedback with fake data. An illustrative MVP for a service
could be a video that shows what that service will look like in the life of
the customer. For a physical product, it could be a prototype made of
clay that a customer can hold in their hands and imagine putting to use.
You may hear other names for an illustrative MVP, such as a low-fidelity
MVP (per Steve Blank and Bob Dorf), a pretotype (per Alberto Savoia),
or a proof of concept (POC) in product management.
 The point of an illustrative MVP is to give the customer something
tangible to react to rather than just listening to you describe your
planned innovation. As customers interact with the illustrative MVP, be
sure to watch and listen carefully for what they notice or miss, what
they don’t understand, and what questions they ask.

FUNCTIONAL MVPS
A functional MVP is a version of your innovation that is limited in scope
but delivers your essential value proposition to the customer within their
real work or life context. A functional MVP really does work. Any data it
uses is real (not placeholder data). It delivers value and solves a
problem for the customer. And, critically, it must be used by the
customer in the real-world context that the innovation is meant for
(whether their work setting or personal life). A functional MVP is
sometimes called a high-fidelity MVP (per Blank and Dorf) or a
prototype (although to engineers, that term often implies a one-off
product built with the complete set of features, which is not what we
want here).
Note that a functional MVP is not a complete version of your product
or service. It should be limited in scope, with only the minimum features
for use (reflecting the goal to “get to market sooner”). It is provided only
to limited customers at first—whether by invitation only, in a limited
location, or for a limited time. It will often rely on manual operations that
are not scalable for a final release.
Rania Succar uses the metaphor of hamster wheels to describe
how Intuit employees manually processed data for the first test of
Intuit’s loans product for small businesses. She knew that the process
would need to be automated before the product could be launched
widely. The first MVP for Diapers.com followed the same approach.
Spending all night driving to stores, repackaging diapers, and mailing
them by hand was not meant to be a scalable operation. But it gave
customers the experience of overnight diaper delivery, and it generated
invaluable insights about how customers would use the service in real
life.

Not One MVP but Many

Remember that a successful venture team does not produce one MVP
but many of them! A single innovation should go through several
illustrative MVPs (from napkin sketches to robust wire frames) to clarify
and validate what you are planning to build. When you get to functional
MVPs, this iteration continues. Succar told me, “We launched seven
versions of our product over three months during the test flight, while
we were inviting customers to use it and comment on the interface,
features, and more.” There is no standard number of MVPs to be
made. But the more you make, the faster you can accelerate your pace
of learning.
Testing multiple MVPs promotes flexibility in your team. Lean start-
up guru Bob Dorf says, “Your early MVPs should be cast in Jell-O”—
that is, infinitely malleable. You want your team to have an unwavering
commitment to the problem you are solving, matched by an open-
minded flexibility on what the best final solution will be. Eric Ries
describes this as a willingness to pivot. Every entrepreneur starts with a
solution in mind, but those who succeed do so by constantly modifying
their ideas as they get new feedback from customers. Always be ready
to listen to the customer and to respond.
Every good MVP is designed to answer specific questions that
matter most to its venture at that moment in time. One of my favorite
examples is from the early days of Netflix, when the company was
planning to launch a DVD-by-mail rental service. Cofounders Marc
Randolph and Reed Hastings mailed a single disc to themselves (a
recording of Patsy Cline’s greatest hits), using a pink greeting card
envelope and the local post office in Santa Cruz, California. This simple
test was critical. Although no customers were involved, it allowed them
to validate three things: Could they use the U.S. Postal Service for
delivery? How fast would delivery be? Could DVDs be mailed in paper
envelopes without breaking? The success of that test led them to press
ahead with an idea that turned out to be revolutionary.11
Always remember: every MVP is a creative leap. It must be
designed around the needs of that particular moment in your innovation
journey. To design your next MVP, ask yourself these three questions:

1. What is the one thing I need to learn next about my venture?

2. What data would be most helpful to my learning that?
 3. What is the quickest, cheapest, simplest test that I can run to get
that data?

Remember: minimum cost, minimum time, and maximum learning.

Let your creativity do the rest.

Metrics that Matter

As you take an iterative approach to MVPs, you will want to do the

same with metrics. Keep in mind these guiding principles:

• Measure what matters now—At any moment in the life of your

venture, the most important metric you should focus on will vary.
The thing you need to learn one week may be which type of
customer is most interested in your solution. But a week later, it
may be which features will convince your free-trial users to
become paying customers. Focus on what you need to validate
next and choose the metrics that match. Your metrics will change
rapidly as you move ahead.
• Be skeptical—Beware of vanity metrics—data that is easy to
collect and makes your idea look good but doesn’t yield critical
meaning for your business. Also, don’t believe everything a
customer tells you they like, value, or would do. Customer
feedback is most meaningful when they have “skin in the game”;
that is, they are investing money or time, which proves their
interest is genuine.
• Balance behavioral and psychological data—It is a common maxim
in experimentation to prioritize behavioral data—measuring what
customers do rather than what they say. But psychological data is
essential as well: what does the customer care about, remember,
attend to, think about? This type of data will help you understand
their needs and point you to the design changes that will drive your
next change in customer behavior. Look to balance both qualitative
and quantitative methods.
• Match lagging and leading metrics—Lagging metrics measure
current business results, including revenue, traffic to your site or
store, and other key performance indicators (KPIs). Leading
 metrics are data that are predictive of future results. You may find
that how often a customer uses your service (twice a month, or
twice an hour) is predictive of whether they quit in the next thirty
days. Customer complaints or word of mouth may be leading
metrics too. Seeing the future by knowing the causes of growth
versus decline is powerful! So always look for leading metrics that
predict your future KPIs.
• Focus on a few metrics at a time—Keep your list short. Give laser-
focus to three to six key variables that matter most to your venture
at that moment. To find them, remember the design principles of
MVPs: start with what you most need to learn next and then find
metrics to match. There will always be other metrics that seem
important too, but you can focus on them in your next test.

The Four Stages of Validation

With any new growth venture, the path from exciting idea to success at
scale is incredibly daunting. So many things could go wrong. As Succar
told me, “Whenever we start a new venture, the amount of uncertainty
is crippling.” Any new venture faces a range of questions, such as:

• Who is my customer?
• How big is the market opportunity?
• Who is my competition?
• What’s my competitive advantage?
• What price should I charge?
• Should I charge a flat price, charge on usage, or charge a
membership fee?
• What features should I build first?
• Can I deliver the experience I’m promising?
• Do I have the necessary skills or intellectual property (IP)?
• Will the technology work?
• Will partners agree to work with me?
• Where do I find my first customers?
• What channels should I use to market to them?
 • What are my costs of doing business?
• Am I solving the right problem?
• Does anyone really care?

The possible experiments you could run and the MVPs that you
could build will be just as numerous. For the leaders I meet, the most
perplexing challenge is where to start. We know we must take an
experiment-driven approach. But where do we begin? How can we
organize all this testing and learning? What I have seen across scores
of companies is that teams desperately need a guide to navigate from
their very first market tests to driving growth at scale.
Based on my own years of advising teams and studying successful
new ventures, I have developed a framework to define the sequence of
effective innovation. I call this framework the Four Stages of Validation
(see figure 5.1).

Figure 5.1.
The Four Stages of Validation

Each of these stages seeks to validate (i.e., prove or disprove) a

different aspect of your innovation. The Four Stages of Validation test
four things—the problem (you think you are solving), the solution (you
think will address it), the product (you think the customer will use), and
the business (you think you can build from this innovation). Each of
these stages is focused on answering a single, essential question (see
table 5.2). Every test, MVP, and customer interview should be designed
to yield data that helps to answer one of these four fundamental
questions. Only by answering all four questions can you validate
whether a new venture will create value in the real world.

Table 5.2.
The Four Stages of Validation and Key Questions to Answer
Validation Key Question to Answer
Stage
1. Problem Are we focused on a genuine problem for an
validation actual customer?
2. Solution Does the customer see value in our proposed
validation solution?
3. Product Can we deliver a solution that customers use?
validation
4. Business Can we capture sufficient value from this
validation innovation?

Sequenced and Overlapping Stages

You must understand two critical aspects of the Four Stages of

Validation before you can put them to work.
The first critical aspect of the four stages is their sequence (shown
in Figure 5.1). This is where corporate innovation frequently runs into
trouble. In most large organizations, teams rush to start on the later
stages—focusing on the product, its features, or the business case—
before they validate the problem being solved and who (if anyone) has
that problem. For example:

• In finance-driven firms, we see a rush to start with stage 4:

business validation: “Show me the business case first. Then we
can approve a budget to spend on market testing or validation.”
• In engineering-driven firms, we see the urge to start with stage 3:
product validation: “We know the solution we need. Let’s build a
prototype as a proof of concept to see if it can be done. You can
show it to customers just as soon as we build a complete product
for them to try.”
• In marketing-driven firms, we see a greater focus on the customer
and an inclination to start with stage 2: solution validation: “We
have a great idea that came out of our strategic brainstorm. Let’s
mock up some wire frames to quickly get customer feedback!”

All these approaches are mistaken! Each starts the process of

validation in the wrong place. By contrast, every effective innovation
process I have ever seen begins with stage 1—validating the problem
you hope to solve with the customer you think it will matter to—and
then moves sequentially through the stages.
 The second critical aspect of the Four Stages of Validation is that
they are overlapping. The sequence in figure 5.2 (problem > solution >
product > business) marks only the beginning of each stage. That is,
you begin to validate your problem before you start to validate your
solution; you begin to validate the solution before you start to validate a
working product; you begin to validate your product before you start to
validate your business. The arrows circling back in the figure show that
each stage continues even as others begin.

Figure 5.2.
The overlap of the Four Stages of Validation

The Four Stages of Validation is not a waterfall or stage-gate

process where success means that you conclude one stage then move
on to the next. In fact, each stage is never finished. Even as you test
the finances of your business (stage 4) and move to a public launch of
your product, you will continue to learn and revise your understanding
of the problem you are solving (stage 1), the features you need next for
your solution (stage 2), and how you can deliver them for the right
customer use cases (stage 3). Eventually, you will be validating all four
stages simultaneously. By the time Intuit had a test flight of its loans
product in the hands of early customers, Succar’s team was dealing
with issues and questions on every level—how to underwrite loans
accurately (product), stimulate customer demand with the right features
(solution), revise revenue projections (business), and more.
 For any innovation to deliver value to your business, you must
validate all four stages. Real innovation means more than just proving
that customers want your solution (product-market fit). It means
validating the entire business model. This is why the Four Stages of
Validation does not stop with your first functional MVP in the hands of
early customers; it takes you all the way through usage, delivery, costs,
revenue, and path to profit. As Chris Reid told me, “That’s the Product
Manager’s job. In my view, you’re accountable for the value chain of
your solution: a cost base that’s competitive, developing a value prop,
how do I operate it competitively, and how do I sell it competitively. I
just think that last piece, in a digital environment, is where some people
pause and don’t give enough thought.”
To see how all this works in the real world of growing a new venture,
let’s look at each of the Four Stages of Validation in detail. For each
stage, we will identify the best use of MVPs, the key metrics to track,
and the biggest threats to your innovation’s success.

Stage 1: Problem Validation

The first stage of validation for any new venture is problem validation.
In this stage, the critical question to answer is, Are we focused on a
genuine problem for an actual customer? Answering this question
begins with defining the problem you think you are solving and then
talking to real customers to find out if and how that problem truly
matters to their life or work.
I cannot stress enough how important it is to begin your innovation
journey here, validating your problem and your customer first. It is
tempting to rush to work on designing your solution but, as Bob Dorf
regularly reminds me, “Most new businesses die from lack of
customers!” This is why Citibank’s Discovery 10X process always
begins with problem validation. As Chief Innovation Officer Vanessa
Colella explained to me, every Citibank venture begins with an in-depth
effort “to try and get at what is it that clients want to do differently. We
use a variety of techniques to validate this very early on, before
anything has been built.”
Problem validation focuses on understanding and confirming the
problem you are solving, the unmet needs of the customer, the context
in which they experience that problem, and its urgency for them. This
stage also focuses on learning as much as you can about who the
customer is, where you might find them, and which customers care the
most about your problem. Your goal should be to find a highly
motivated segment of customers, called early adopters, who will be
willing to try your earliest solution before anyone else. (Hint: If no
customer cares passionately about what you are solving, you should
move on to solving another problem!)
If you do identify an urgent problem, you also need to learn what
customers are currently doing to address it. This can range from
makeshift solutions to simple resignation (i.e., “We just live with it”). We
can think of these as existing alternatives to any solution you may seek
to develop. It is important to learn what these alternatives are, how
satisfactory they are to the customer, and where they fall short.

Know Who the Customer Is

Problem validation starts and ends with listening to the customer. To

find your first customer, ask: Whose problem do we think we are
solving with this growth venture? Who would actually use any solution
we develop? Be careful that you don’t mistakenly label the company
you are selling to as your customer. The customer is always a person—
whether a consumer archetype (“tech-savvy new parents”) or a
business job role (“plant supervisor”).
If you are developing an internal innovation, your customer will work
at your own company. Think of Air Liquide’s problem of predicting
customer churn for its gas canister business. In order to validate this
business problem in stage 1, Air Liquide’s team must talk to the
marketing employee who will use the solution in their day-to-day work.
Any innovation that uses a platform business model will serve
multiple customers, each with their own needs. Think of Uber’s
platform, which connects drivers and riders, or Amazon Marketplace,
which connects third-party sellers with customers. Uber needs an app
that is successful with drivers just as much as it needs an app that
succeeds with riders. If you have a multisided business model, you
must validate your innovation with each type of customer it touches.
 B2B innovations typically serve multiple customers as well. If you
are creating an enterprise SaaS solution, your customers will include
every stakeholder in the client’s company whose needs must be met in
order for your innovation to succeed. These could include the users of
your new SaaS service, the IT supervisor who must approve it for their
intranet, the procurement manager who must agree to your
subscription terms, and the finance officer who must approve the
budgetary expense.
In problem validation, every customer matters! Make sure you know
who your customers are and that you listen to each to validate their
unmet needs.

Stage 1: No MVPs

Each of the Four Stages of Validation uses MVPs differently. In this first
stage of problem validation, it is actually best to avoid using any MVPs
at all. The reason: at this stage, you want to keep the focus entirely on
understanding the customer’s problem and not on your proposed
solution.
Instead of an MVP, use in-depth observational interviews to learn
from customers. Often called problem interviews, these are the starting
point for customer discovery in lean start-up. In design thinking, this is
the ethnographic and qualitative research done early on to capture the
experience and voice of the customer. Whatever you call it, the goal is
to uncover the customer’s own stories, language, and experience in
order to understand their needs and motivations.
The most important thing to remember in problem interviews is that
these are conversations and not sales pitches. You should refrain
entirely from talking about your own proposed solution. Instead, focus
on listening and observing. Here are a few key tips for problem
interviews:

• Use open-ended questions—Your goal is to get the customer

talking. Never ask a yes or no question.
• Shut up and listen!—Say as little as possible. Be patient and give
your customer time to elaborate.
 • Interview and observe in context—If possible, talk to the customer
in their own home or workplace to see where and how the problem
takes place.
• Look for emotion—Your goal is to focus on the most extreme pain
points and needs of the customer. Always listen for whatever
triggers an emotional reaction.
• Probe for specifics—Follow up with when, where, how, and why
questions. Ask the customer what they are currently doing about
the problem and how well that is or isn’t working.
• Don’t talk about your solution—This is not a sales conversation!
Don’t pitch your idea. Instead, postpone saying anything about
your proposed solution until future conversations.

Stage 1: Metrics and Learning

Your metrics for problem validation will be the simplest among the four
stages because your validation here is mostly qualitative, not
quantitative. It is critical, however, to set goals for your team members
and push them on some important numbers. Your most important
metric is the sheer number of customers that you speak with. As you
identify different customer segments, make sure to track the number of
interviews you conduct with each segment. Track measures of depth,
such as number of site visits to a customer’s home or workplace, pages
of interviewer notes, verbatim text from customers, and interview
videos. You should also track the number of high-interest customers
you speak to in interviews—those who express great interest in your
problem and might become early adopters.

GOLD STANDARD
In problem validation, the gold standard of learning (i.e., the best
possible outcome) is to find a few lead users who are not only
obsessed with your problem but will show you the hacks or prototypes
they have built as workarounds and then ask you to make something
similar for them to purchase. The term “lead users” comes from Eric
von Hippel, who first observed that many of the best innovations come
from watching the industrious efforts of hypermotivated customers.12

TYPICAL LEARNING
It is more common in the problem validation stage that you will learn
that your customer or problem definition is not quite right. Your intended
customer may face a somewhat different problem than you imagined,
or the problem you are focused on may matter but to a different
customer. The faster you learn this, the faster you can iterate and
refocus your innovation on its best opportunity for success.

Stage 1: Threats

As you begin work on problem validation by engaging customers in

problem interviews, look for the following red flags—common indicators
that your planned venture is not off to a good start:

• No problem found—You can’t identify a pressing customer problem

you are solving (which means you have a “solution in search of a
problem”).
• Low priority problem—Customers nod their heads when you
mention the problem, but no one shows great urgency to solve it.
(Dorf says that, unless it is among a customer’s “top five
problems,” there is no business to be built solving it.)
• Status quo is good enough—The customer is satisfied with the
existing alternatives for addressing the problem. They may not be
perfect, but they are good enough for the customer.

In the problem validation stage, your key competitive threat is

customer inertia. Does your customer care enough about the problem
you identified to change their behavior? In many cases, the answer is
no. That is why it is critical to examine existing alternatives. If the
customer is satisfied enough with their current workaround (i.e.,
muddling along with their current state of affairs), then even the most
ingenious new product will fail.
 Stage 2: Solution Validation

Once you have started to validate the problem you are solving with a
group of real-world customers for whom it is truly urgent, you are ready
to begin work on the second stage of validation: solution validation. In
stage 2, the critical question to answer is, Does the customer see value
in our proposed solution? This means testing whether the customer
understands your innovation idea, cares about the features and
benefits you plan to deliver, and is motivated and ready to change their
behavior to adopt it.
This stage requires defining and validating your value proposition:
What are the pains you will relieve and the gains you will provide to
your customer? What is the job you will help them get done? You need
to list each value element that you are planning, and for each one,
validate if the customer really cares.
You will also be validating the design of your solution—that is, how
you intend to deliver your benefits to the customer (Face-to-face or self-
service? immersive experience or clean-and-simple? Twelve-month
training program or a library of online resources?). This also means
drafting and validating a feature roadmap: What features do you need
to launch with, and which ones are the next highest priority for the
customer?
At this stage, you will also seek to measure and quantify customer
demand. How many customers really want your product? How badly do
they want it, and what features are they willing to pay for? Successful
proof of customer demand for a new innovation is commonly referred to
as “product-market fit.”13

Stage 2: MVPs

In stage 2: solution validation, we will make extensive use of illustrative

MVPs. As explained earlier in this chapter, these are MVPs that
illustrate the features, benefits, and design of the proposed solution
without yet delivering those benefits to the customer. An illustrative
MVP can be extremely rudimentary (a hand-drawn sketch), or relatively
elaborate (a wire frame of a planned digital service showing different
interactive screens with simulated data).
 Air Liquide’s Olivier Delabroy stresses the simplicity of an illustrative
MVP (or proof of concept, as Air Liquide calls it) and how it is used at
his industrial firm: “It can be super simple—a mockup or a simple
screen. You just want to check that the end user will understand its
value and use it.” He also stresses using minimal time and resources:
“Build something super cheap that you can put in the trash the next
day. The only expectation is to validate that it creates value.” This kind
of validating of design and user experience can be done virtually, but it
often benefits from in-person testing—where you can observe how the
customer interacts with the MVP, hear what questions they ask, and
see what features they request or ignore.
Customer demand is best validated with behavioral data. Don’t ask
if the customer will buy. Instead, give them a buy button and see who
puts their money where their mouth is. Advertise your planned
innovation to different audiences with ads describing different benefits
or features, and measure who clicks and which descriptions they
respond to. This kind of test can be done quickly with low-cost
advertising in search engines and social media. For a paid product,
your goal should be to see if someone will provide their credit card
number to preorder. (The crowdfunding service Kickstarter is built
entirely around this idea—collecting customer’s paid deposits in
advance for an unmade new product—as a way of learning if the
market really wants it.) In other cases where you don’t plan to charge
the user (e.g., an advertising-based product or an internal innovation
for employees), you can measure how many customers are willing to
create an account or enter their email address to be notified when the
innovation goes live.
Building such simple MVPs rapidly and in constant interaction with
the customer is a big change for companies with a traditional
engineering culture. “This is not in the DNA of our industry,” Delabroy
explains. “We were used to only going to see the customers when we
were 200 percent confident the product was working. Our mindset is
now progressing fast.” Unlearning these kinds of habits is essential if
you want to accelerate learning at the early stages of your innovation
process.

Stage 2: Metrics and Learning

Metrics are critical to effective solution validation. At this stage, you
should be measuring the priority of the benefits you are promising and
the features you intend to build. Which benefits matter most to which
customers? What features do they want you to prioritize in your product
roadmap? In addition, stage 2 metrics should focus on measures of
customer demand. Depending on your innovation, these measures
could include online registrations, app downloads, product trials,
customer purchase orders, or even cash deposits. Remember, you are
not yet delivering an actual working solution to any customers (that
comes in stage 3). But if you can illustrate the expected solution well
enough, you should see measurable demand from customers who are
ready to start paying or using your innovation.
Lastly, don’t just believe the customer when they tell you what they
want! The best test of how much credence to give a customer’s
response is to look at how much “skin in the game” they have (one of
Alberto Savoia’s favorite points).14 The more time, effort, or money your
customer commits, the more you can trust what they are telling you.
This is why a cash deposit is a much better signal of customer demand
than an email registration.

GOLD STANDARD
The gold standard of solution validation is for customers to give you a
cash deposit for your forthcoming product—like Tesla collecting $1,000
deposits for its Model 3 more than a year before the first car was
built.15 For a B2B innovation, the equivalent is a customer providing a
signed purchase order, with the required delivery specifications and an
agreed price if you deliver on time.

TYPICAL LEARNING
It is more common at this stage that you will learn that many of your
planned features or benefits don’t really matter to your customer, while
other features that you hadn’t prioritized will be essential for them to
even consider your innovation. The key is that you want to learn this
now, before you build your first working product! This avoids the huge
waste of time and resources that can be spent engineering a working
prototype without first validating what your customers truly want and
why.

Stage 2: Threats

As you use illustrative MVPs and validate customer interest in your

proposed solution, look for the following red flags—common indicators
that you have not achieved product-market fit:

• No urgent demand—You receive polite interest in your solution

(“That’s very interesting. Please keep in touch and let us know
when you launch”). But there are no requests to register, use, or
buy (“How soon can you have something ready?” or “We’d love to
be part of your first pilot”).
• No killer feature—When you test your value proposition, no single
feature or benefit pops out as driving customer demand. Many of
your benefits seem nice to the customer, but none are enough to
motivate behavior change.
• No advantage over existing solutions—Once you describe your
planned solution, customers identify a comparable solution in the
market, and they don’t see a compelling reason why they must
choose yours instead.

At the solution validation stage, your key competitive threat is

existing solutions. As soon as you propose a specific solution to your
customer, they will view it in comparison to an existing solution already
in the marketplace (whether you agree with that comparison or not). Be
sure to validate. What existing solutions do customers compare yours
to? Does your solution have any point of differentiation or advantage in
the customer’s mind? If not, you are setting yourself up for a costly,
losing battle.

Stage 3: Product Validation

In start-up culture, there is a huge focus on achieving product-market fit

(the goal of stage 2). And in many enterprises, innovation teams think
that, once they discover a value proposition that customers are
clamoring for, they can hand off their venture to marketing, sales, and
supply chain to build the product and scale it in the market. But the
innovation team’s job is only half done! Once you have validated that a
promising segment of customers sees value in your planned solution
and is eager to try it, you are ready to begin the third stage of
validation: product validation.
In this stage, the critical question to be answered is, Can we deliver
a solution that customers use? This means testing two things at the
same time. The first is how the customer uses your innovation in the
real world—that is, in their day-to-day work or life context (not
sequestered in a test lab at your offices). The second is what your
business must do to deliver the essential value of your innovation to the
customer (what tasks must be done, what resources are needed, what
partners will be critical, etc.).

Use and Delivery

Much of stage 3 is focused on validating usage by the customer—what

are the different use cases (or contexts) in which customers use your
solution? For each use case, what is the customer’s journey from start
to finish? You will never know how customers will use your product until
you put it in their hands.
Susie Lonie was working for telco Vodafone and its Kenyan
subsidiary Safaricom when her team developed a new mobile phone
app called M-Pesa. The app allowed small payments to be made and
received by anyone with a mobile account on their network. The
company’s plan was that M-Pesa would be a tool to support the
microfinance sector of Kenya’s economy, facilitating small loan
repayments by entrepreneurs who lacked a bank account. But once M-
Pesa was in the hands of its earliest customers, they started to find
many more uses for it. Business owners used it for security while
traveling, depositing their cash into the digital wallet before departing
on a trip and withdrawing it when they reached their destination. City
workers used M-Pesa to send remittances to their families in distant
villages. One customer used it to text money to her stranded husband
to catch a bus home. As customers found more and more uses for M-
Pesa, the world’s first mobile wallet took off, quickly dominating the
economy of Kenya. In other countries, it has been used to fight
corruption (one government switched to paying its police force digitally
to ensure everyone got their full salary) and to bring gray-market
transactions into the formal economy.16 With any truly novel innovation,
the early adopters will teach you the real use of your product. So get an
early functional MVP into their hands as soon as you can and find out
the real potential of your innovation and what features it truly needs.
The other major area of validation in stage 3 is how your business
will deliver your solution to the customer. This includes validating what
assets and skills you will need, what activities your business will need
to do well, and what should be handled by external partners. It also
means validating the feasibility of your technology and your ability to
meet any requirements related to risk, regulation, or security. These
kinds of operational questions can only be answered by delivering your
value proposition to at least a few customers in the real world.
When restaurant chain Panera first developed a suite of new digital
services for ordering (including in-store kiosks, a mobile app, and
delivery drivers), it didn’t launch them across all its hundreds of stores.
Instead, it tested the new digital-ordering experience for months in a
single store in Braintree, Massachusetts. The results showed that
customers loved the new digital features, but they caused operational
issues, including food preparation errors and new bottlenecks of
customer traffic. Only after these issues were resolved was the new
customer experience launched at scale and to great success.17
Customer usage and product delivery are often critically intertwined.
Before heading down a particular path for delivery, be sure to validate
your customer’s willingness to use it. Chris Reid of Mastercard
explained how customer willingness matters when designing digital
solutions for finance. “Even before you pick up a pen on your product’s
design, or hand it over to a Scrum team, you may say, ‘This has to fit
into someone’s app.’ So, will we deliver this by a simple API
[application programming interface]? Or is it an SDK [software
development kit] integration, which is a heavier lift?” Your business may
have a clear preference between these two technical routes, but if the
customer is unwilling to use your preferred model, there is no point in
going forward with it. As Reid said, “If you haven’t thought that through,
there is a significant risk that even a great product won’t be successful.”
 Lastly, for any innovation to succeed in the long-term, you will need
some kind of competitive moat to prevent others from doing the same
thing just as well. As part of product validation, you need to verify that
you have some unique advantage over other firms that will help you
deliver this innovation.

Stage 3: MVPs

In the product validation stage, illustrative MVPs have only limited use,
such as showing a customer different possible use cases for your
product to learn which ones are most relevant to test.
Instead, almost all your learning in stage 3 will come from functional
MVPs, which deliver your essential value proposition to the customer
within their real work or life context. Your earliest functional MVP should
be a bare-bones product with the absolute minimal features.
(Remember the purpose of a functional MVP is not to deliver a finished
product but to deliver learning!) First, use this MVP to learn if your
innovation really does create value in the life of your customer. Second,
learn how the customer uses it. (When and where do they use it?
Which features gets used versus ignored? Is the user interface
confusing or obvious?) Third, use the MVP to learn what you must do
to deliver this innovation. (What issues must be addressed in terms of
supply chain, compliance, customer service, etc.?)
For the sake of speed and rapid learning, your first functional MVP
may work in only a single limited use case. It may take technical
shortcuts and rely on work being done manually in the background.
(I’ve heard this called a “Wizard of Oz” MVP, or, by Eric Ries, a
“concierge” MVP.18) It will likely have only a limited release for a few
users to try in the real world (perhaps just to selected employees).
Remember that speed of learning is your top priority. As LinkedIn
founder Reid Hoffman said, “If you are not embarrassed by the first
version of your product, you’ve launched too late.”19
As product validation continues, you should evolve from a
makeshift, often jerry-rigged functional MVP to a much more scalable
product or service. Each iterative MVP should add more features based
on what users request. It should work in a wider range of use cases.
The back end should become more robust—more automated, more
secure, using more data, and able to serve more customers. This
iterative process will guide your product validation from a limited
release (invitation only or a local test market) to a true public launch.

Stage: 3 Metrics and Learning

During product validation, many of your most important metrics will

revolve around customer adoption and usage. You want to measure not
just how many customers try your innovation once but what happens
next. Do they keep using it? How frequently? In what locations and at
what times of day? What features are most popular? Do they
recommend your innovation to others?
One critical metric to track will be growth in total number of users
over time. Just as important, however, is to track specific customer
cohorts (small groups who start at the same time with the same feature
set) so you can measure their repeat usage (stickiness), their
satisfaction, and their referral rates or advocacy (such as Net Promoter
Score). It is also critical to measure different customer segments,
keeping an eye out for differences between them. Do businesses in
one industry make more frequent use of your product than those from
another sector? That kind of insight will be critical to scaling your
product.
In addition to customer usage, you should be tracking metrics
related to your own delivery and operations. How many errors do you
see in your product or service? How reliably can you deliver it on time
and to the right customer? How fast and well are you delivering to
different customers and in different use cases? What growth rate are
you able to sustain in rolling out your product? All these points will be
critical to improving operations and validating that you can scale your
current product.

GOLD STANDARD
The gold standard in product validation is for customers to adopt your
product, keep using it, and attract others with word of mouth—leading
to exponential growth in your number of customers. At the same time,
customer data and feedback provide clarity about which use cases are
most promising and which features are most important. Your operations
prove to be robust and easily scalable.

TYPICAL LEARNING
It is more common at this stage that you will learn that the customer
uses your innovation in very different ways than you planned. You may
learn that the technical uncertainty around your innovation is greater or
less than you expected, that your delivery time line will be longer or
shorter than you thought, or that tasks you planned to do yourself are
better off outsourced (or vice versa). All this learning will put you in a
much better position to deliver your innovation at scale in the real
world.

Stage 3: Threats

As you use functional MVPs to validate customer usage and delivery of

your product or service, look for the following red flags:

• Usage doesn’t follow interest—Customers may like your innovation

in theory, but they don’t use it when it is offered. Or you may see
strong initial trial by customers but little repeat usage, or rapid
churn.
• Too hard to deliver—You can’t find a path to deliver a solution
customers will use. The current technology may not be ready. You
may lack essential skills, assets, or distribution partners, or you
may be unable to overcome regulatory hurdles.
• Too easy to deliver—Your solution works well, but competitors can
easily deliver the same thing as well or better than you can.

In the product validation stage, your key competitive threat is

copycat products launched by other companies once your innovation
proves popular. If other firms copy your product, will you have any
unique advantage (exclusive partners, unique data or IP, network
effects) that allows you to deliver it better or more cheaply than they
do? Imitation may be the sincerest form of flattery (per Oscar Wilde),
but it can crush many a first-mover in business.
 Stage 4: Business Validation

Once you have started to validate how customers will use your solution
and how you will deliver it, you are ready to begin work on the fourth
stage of validation: business validation. In this final stage, the critical
question to be answered is, Can we capture sufficient value from this
innovation? For any venture to succeed, it must deliver not just value to
customers but also value for your business. That means proving not
just product-market fit and use by the customer but ROI for your firm.
A historical example may illuminate this point. Perhaps no individual
is more associated with innovation than Thomas Edison, and the
invention he is most known for is the electric light bulb. But the
incandescent bulb itself was invented by English chemist Humphry
Davy and others. What Edison contributed on top of that technical
breakthrough was a dogged search for an economic model that could
make electricity the dominant technology for lighting the home. Only by
making the economics and pricing work could Edison persuade
average consumers to switch from kerosene lamps to an electric future.
Without the right economic innovation, the technical innovation would
mean nothing to people’s lives.
One of the economic hurdles that Edison recognized was the high
cost of copper, which was used to transmit electricity from power
stations to the home. Through experimentation, Edison discovered that
you could reduce the amount of copper needed if the electric current
were run at a higher voltage. But this voltage would burn out every light
bulb unless filaments were designed with a much higher resistance.
This ran counter to the thinking of the bulb’s engineers (high-resistance
filaments wasted energy), but it was critical in bringing down
transmission costs and ultimately lowering prices enough to trigger
customer adoption. In studying Edison’s notebooks, historian Thomas
Hughes found that they held as many notes on economic experiments
as scientific ones.20
To succeed with any innovation, we must all learn, like Edison, to
focus on testing and perfecting our economic models just as much as
our products and value propositions. Our fourth stage, business
validation, requires that for any new growth venture, we must test and
learn the answers to several questions:
 • How will your business capture value from your innovation?—
Value may come from new revenue. It may come from reducing
costs or operational risks. Some innovations generate
nonmonetary value, for example, data or customer relationships
that you monetize elsewhere. For a nonprofit, value may be
measured in terms of impact on your core mission.
• What is your customer lifetime value (CLV)?—Innovations often
create value by customer acquisition, by improving customer
retention, or by expanding the average revenue you earn per user.
Which of these will be affected by your innovation?
• What is the cost structure of your innovation?—Costs may include
marketing, operations, and outside partners, as well as ongoing
research and development (R&D) to maintain the competitiveness
of your innovation. It is critical to understand which costs are
variable versus fixed (e.g., the same cost whether you serve ten
customers or 10,000) and what economies of scale you might
achieve if you grow.
• What is your path to profit?—The goal of innovation should always
be net value creation. Getting there requires a formula for when
value capture will surpass costs. Even if your innovation is certain
to be profitable, you need to know when and also the maximum
upside. Based on margins and market size, is your innovation a $1
million opportunity or a $1 billion opportunity? The answer may be
critical in deciding how far to pursue it.

As you validate the economics of any innovation, be careful not to

leverage false economies to exaggerate its promise. For enterprise
innovations, it may be possible to get your company’s sales team to
promote your product to existing customers or leverage other
relationships to accelerate your venture’s initial growth. But help like
this, which can’t be scaled indefinitely, can mask the true economics of
what you are building. It is better to find out early on what it would cost
to deliver all aspects of your new venture out of your own budget rather
than be surprised later that a promising idea turns untenable when it
has to run on its own steam.
One more warning on business validation: I often meet executives
who are told to do stage 4 first. That is, to build a business case for the
ROI of their new venture before they have validated the problem they
are solving, the customer they will serve, the solution they are planning,
or the product they will deliver. It is important to realize that any
business case built before those three stages of validation have begun
is pure fiction. Don’t get boxed into delivering fake data at stage 4! If
your current bureaucracy will not release any innovation funding
without a business case, don’t ask for it. Instead, ask for a small budget
for “market research.” (No one expects an ROI from market research.)
Use this money to start your early stages of validation, and then come
back to business validation (and the bigger budget request) once you
have some validation of what you are even trying to build.

Stage 4: MVPs

At the business validation stage, illustrative MVPs will be of limited use

for gathering data. You could show customers different offers and ask
their willingness to pay. You may be able to estimate your customer
acquisition costs by “dry testing” different marketing channels (i.e.,
promoting a product you don’t yet have with ads placed in different
media).
Almost all your learning in stage 4, however, will come from
functional MVPs—versions of your innovation that deliver your
essential value proposition to customers in their real-life contexts. A
functional MVP is the only way to start learning about your true
operating costs and path to profitability. You may think you can predict
your revenue by surveying customers on the price they would pay, but
you will never know your true price point until you start to deliver value
to customers. After they try your product, they may demand a price cut.
Believe it or not, they may ask to pay you more! (I have seen this with
entrepreneurs whose first customers desperately wanted them to stay
in business.) With a functional MVP, you will learn how many customers
return your product or ask for a refund and whether you can grow your
revenue by offering a volume discount.
In practice, business validation will require tests on a wide range of
financial drivers. For example, just validating your marketing costs may
require several tests—of advertising channels, conversion rates,
customer churn, and word of mouth. Your earliest MVPs will give only
partial information on these drivers. Diapers.com’s first MVP measured
how much an average customer might spend in one week on diapers
by mail but that was just one factor in estimating revenue. The Netflix
post office test measured the cost of shipping one disc to one
customer, but shipping costs would also be shaped by how frequently
customers mailed back their DVDs each month.
As you continue down the path of business validation, you should
progress from tests of specific financial drivers (weekly revenue,
shipping costs, etc.), to broader tests of total profitability. In the case of
Panera’s digital redesign of the customer’s order experience, this
meant validating the total impact on store profit based on numerous
changes in costs, revenue, and customer behavior from its digital
innovations. As your own functional MVPs become more complete, the
economic data they validate will become richer as well.

Stage 4: Metrics and Learning

In business validation, all the metrics you focus on will be financial

metrics. These will include measures relating to revenue (price point,
total revenue, top-line growth, etc.), as well as costs (variable and
fixed) and net profit. Your key financial metrics will depend on your
business model. Retailers focus on metrics like sales per square foot.
Subscription businesses focus on average revenue per user (ARPU)
and CLV. In ventures where your value comes from risk reduction,
financial metrics should focus on underwriting and the cost of
reinsurance. For innovations that serve internal customers in your
enterprise, the simplest financial metric is often to determine what you
would pay an outside vendor for this same outcome or what are you
currently paying.
Nonprofits are often less skilled in developing financial models, but
it is no less critical for them to validate investment in new innovations.
The Gates Foundation uses disability-adjusted life years (DALYs) as a
metric to compare the impact of wide-ranging investments in global
health. Nonprofits should ask of any innovation, What does it cost us to
achieve a comparable outcome (e.g., lives saved or children educated)
through our current means? Does this innovation deliver the same
impact with fewer resources? For every innovation, the organization
must ask whether this is the most effective use of its resources.
 GOLD STANDARD
The gold standard in business validation is to capture value, or turn a
profit, from your very first users—that is, to have a self-funding project
from the beginning. In the start-up world, this path is known as
bootstrapping—when you are able to grow a new business without the
help of venture capital.

TYPICAL LEARNING
It is more common at this stage of validation that you will learn that
costs, revenue, and profitability are all different than what you
imagined. You may find that the customers you were focused on are
not the most profitable ones to build a business on (but others may be).
You may find that what you thought was a new revenue project is
actually better run as a customer retention project or a cost-savings
project. A great solution to a real problem can usually capture value for
your business somewhere, but it might not be in the place you were
originally looking! The sooner you learn any of this, the sooner you can
find out what the path to profitability might be for your innovation and
whether the opportunity is big enough that you want to continue down
this path, or if you should shift your focus to another innovation idea.

Stage 4: Threats

As you use functional MVPs to validate the business value of your new
venture, look for the following red flags:

• No value capture—Customers like your offering but they won’t pay

for it, and you can’t find another value stream for your firm.
• No path to profit—You are capturing some value, but your costs
are too high, and they won’t go down enough to break even as you
scale.
• Too small a prize—Even with a margin of profit, you may find the
maximum upside of your innovation is not large enough to merit a
continued focus by your organization. As an innovation manager at
 a large firm once told me, “The worst thing I can bring our leaders
is a $10 million idea.”

At the business validation stage, your key competitive threat is other

investment opportunities. Every venture comes with an opportunity cost
—you are pursuing it instead of something else. In many enterprises,
the CFO will calculate an internal rate of return (IRR), to reflect the
financial return the firm can get by spending money on safe
investments (e.g., paying down debt or investing in infrastructure). The
classic CFO rejoinder to any proposed innovation is, “Sounds great.
But can your ROI beat my IRR?” At stage 4, you should ask the same
thing. Even if your venture grows to scale, will there be better places to
invest the firm’s resources?

Recap: Metrics and Competition

To conclude our tour of the Four Stages of Validation, let’s recap what
we’ve learned by summarizing the metrics we use at each stage and
the competitive threats that we face. Table 5.3 recaps the important
quantitative metrics in each of the Four Stages of Validation. It is true
that much of the learning from customer interviews and MVPs will be
qualitative (the questions customers ask, the problems they bring up,
the emotion of their responses, etc.). But quantitative metrics will clarify
many of your biggest insights, and they will help you to measure your
team’s progress in validating any venture.

Table 5.3.
Quantitative Metrics by Stage of Validation
Stage of Types of Sample Metrics
Validation Metrics
1. Customer • Number of interviews
Problem interview
validation metrics
• Customers spoken to in
each user segment
• Number of site visits
• Hours of conversation
 recorded
• Verbatim transcripts
• Early adopters identified
2. Customer • Adoption rate or
Solution demand conversion of leads
validation metrics
• Number of online
registrations
• Requests to pilot the
product
• Product trial or
downloads
• Customer deposits or
purchase orders
3. Use and • Total number of users
Product operations
validation metrics
• Growth in the number of
users
• User satisfaction
• Repeat use/stickiness
• Customer referral
rate
• Net Promoter Score
• Operational accuracy
• Operational downtime
• Speed of delivery
4. Financial • Price point
Business metrics
validation
• Total revenue
• Customer acquisition
cost
• Retention or churn rate
• Cost to serve (marginal
versus fixed)
 • Profit margin or ROI
• ARPU
• Customer lifetime value
• Total profit versus loss

Table 5.4 lists the key competitive threats you will address in each
of the Four Stages of Validation. Each stage must address a different
key competitor, and each stage poses a different competitive question.
Only if validation provides a clear yes answer to all four questions will
your innovation be ready to implement at scale.

Table 5.4.
Competitive Analysis in the Four Stages of Validation
Stag Key Key Competitive Question
e of Competi
Valid tor
ation
1. Custom Does the customer care enough
Prob er about this problem to change their
lem inertia behavior?
valid
atio
n
2. Existing Does your solution have a compelling
Solu solution advantage versus comparable
tion s solutions?
valid
atio
n
3. Copycat If other firms copy your product, will
Prod products you have any unique advantages in
uct delivering it?
valid
atio
n
4. Other If this innovation succeeds, will it
Busi investm deliver a better return than other
ness ent known opportunities?
valid opportu
atio nities
n

Tool: The Rogers Growth Navigator

We are now ready to introduce our next tool, the Rogers Growth
Navigator. I developed the Navigator over years of work with innovation
teams to address their most critical area of confusion: how to sequence
and organize business experimentation. The Rogers Growth Navigator
ends this confusion by enabling you to visually map your progress
through the Four Stages of Validation (see figure 5.3).
Figure 5.3.
The Rogers Growth Navigator

The Navigator is designed to guide growth ventures of any kind—

whether a disruptive new product, a redesigned customer experience,
or a plan to optimize your business operations. It works for large
enterprises and start-ups alike. And it works for the full life cycle of
innovation—from whiteboard sketch to operation at a global scale.
 The Rogers Growth Navigator captures three things: your current
business assumptions, your experimental learning to date, and what
you need to test and learn next. By combining all of this in one diagram,
the Navigator provides a single, shared view of the venture, which you
can use to align both your team and your sponsors. As you test and
learn, your Navigator will go through many revisions. Keeping copies of
each weekly iteration allows you to capture your thinking over time.
Let’s briefly walk through each block of the Rogers Growth
Navigator to see how you can use it to capture learning across all Four
Stages of Validation for your next growth venture.

1. Problem Validation

The first two blocks of the Rogers Growth Navigator will capture your
hypotheses and learning from stage 1: problem validation.

PROBLEM
In this first block of the Navigator, you will capture everything about the
problem that your innovation is meant to solve, including:

• Problem definition—What exactly is the problem that you are trying

to solve? Why does this problem matter? How urgent is this
problem? (Is it among anyone’s “top five” problems?)
• Existing alternatives—What existing alternatives are being used to
address this problem? Even if you think your solution will be the
first of its kind, customers are clearly taking some approach to the
problem currently (even just to mitigate it). What is that alternative
approach?
• Unmet needs—What are the shortcomings of these alternatives?
What needs do they leave unmet among the customers you are
trying to serve?

CUSTOMER
In this block, you will capture your learning about the customer who
faces the problem you are solving. Begin by breaking your customers
into distinct segments who might use your innovation in different ways
or for different reasons:

• Addressable segments—Who are the different segments of

customers with this problem? Focus on distinct groups that you
could reach or address separately. For consumers, describe each
segment in terms of demographics, psychographics, and
technographics. For business customers, describe the type of
organization and the role of the person who will use or make
decisions about your innovation.
• Total addressable market (TAM)—For each segment, what is the
total addressable market? That is, how many people or companies
are there that currently match your description? How might this
change if you create a compelling enough offering?
• Early adopters—Every successful innovation is adopted first by a
handful of customers who are exceptionally motivated. Who will
your early adopters be? Why does the problem you are solving
matter more to them than to other customers? Why will they be
willing to try an early solution or MVP?

2. Solution Validation

The next two blocks of the Rogers Growth Navigator are where you will
capture your hypotheses and learning from stage 2: solution validation.

VALUE PROPOSITION
In this block, define the value that you will provide the customer. It is
critical to avoid talking about product features and instead to describe
the benefits of your venture from the point of view of the customer:

• Value elements—List all the benefits you will provide to the

customer. Include as many as possible, and always describe them
from the customer’s point of view. First, think of current pain points
your innovation will solve for the customer, and then list ways it will
 delight the customer and enhance their experience. (The concept
of value elements is explained in detail in The Digital
Transformation Playbook.)
• Job to be done—This powerful concept comes from the writing of
Clayton Christensen and Michael Raynor.21 The idea is to describe
the benefit of any innovation by asking what it enables the
customer to do or to achieve. Describe this “job” as something that
truly matters to the customer.

SOLUTION
In this block, define the product or service that you will use to deliver
your value proposition to the customer. Now you should switch from
describing the customer’s experience to describing your product and its
features:

• Delivery and design—How will you deliver the benefits of your

value proposition to the customer? For a product, what will it look
like and how will it work? For a service or process innovation,
when, where, and how will it be delivered?
• Competitive differentiation—What existing solutions will the
customer compare this innovation to? What difference will motivate
customers to pick your solution over these competitors or to switch
if they are already using these competitors? Are customers “locked
in” to using those competitors, or will it be easy to get them to
switch?
• Feature roadmap—Which features have you built so far, if any?
Which features do you need to add next, and in what order? Which
existing features could you cut? Which planned features can be
delayed?

Top-Line Summary

The aim of this block is to capture the essence of your venture in a

single sentence. Think of this as the elevator pitch you would use to sell
your idea to a potential sponsor or investor.
 A good top-line summary uses a single sentence to capture the
most essential points from the first four blocks of your Navigator. At the
beginning of your journey, this will be a statement of your vision for the
new venture. As you validate, you will revise the summary to capture
the most essential things you learn from stages 1 and 2. I recommend
using the following template for your top-line summary:

• For [Customer]
• Who [Problem],
• [your innovation’s name]
• Is a [Solution]
• That [Value Proposition].

For example, a top-line summary for Walmart’s InHome service

might be: “For shoppers who want affordable groceries plus time-
saving and convenience, Walmart InHome is an annual membership
program that lets them order affordable, fresh groceries online and
have them delivered right into their home’s fridge and cabinets by a
safe and trusted Walmart agent.”
As you write your top-line summary, be sure to test it on people who
are not fully familiar with your venture. Avoid jargon. Ask yourself if your
top-line summary would be clear to an average customer. (Better yet,
use it on a customer and see if it is clear.)

Metrics That Matter Now

In this block, identify three to six key metrics that matter most to your
innovation at its current stage of development. In doing so, draw on
every block of the Navigator, encompassing all Four Stages of
Validation. Think of lagging and leading metrics. Focus on customer
metrics with “skin in the game.” And remember that the metrics that
matter most should be constantly evolving. Identify what matters to
your venture right now.

3. Product Validation
The next four blocks of the Rogers Growth Navigator will capture your
hypotheses and learning in stage 3: product validation.

USAGE
In this block, you want to capture learning about the customer’s usage
of your innovation:

• Customer use cases—In what different situations do customers

use this innovation? List as many use cases as possible,
describing the context of each one and how it shapes the
customer’s usage.
• Customer journeys—In each use case, how does the customer
actually use the innovation? Describe the sequential steps of a
typical customer’s experience, including when, where, and how
they use the innovation. Which features do they use and why?

DELIVERY
In this block, you want to capture learning about the back-end
operations required to support your venture:

• Business activities—What does the business need to do in order to

deliver your innovation to the customer in all these use cases?
Focus on your repeated activities and those that you must do
exceptionally well for the venture to work.
• Technical requirements—What technology standards must your
innovation meet to work reliably and effectively? What other
technologies must it integrate with? What standards of
interoperability must it meet? How will your solution meet these
requirements?
• Compliance and risk—What legal requirements must your
innovation adhere to? What company rules for compliance or risk
management must you meet? How will you ensure your solutions
meets all of them?

CAPABILITIES
Use this block to define the capabilities that will be needed to deliver
this solution, the partners that will support you, and the roles they will
play. Remember that few innovations will be delivered solely by your
own organization.

• Key assets—What tangible assets are needed, such as

infrastructure, real estate, or manufacturing materials? What
intangible assets, including intellectual property, data, software
code, and brand reputation, are needed?
• Key skills—What does your business need to be skilled at in order
to deliver this innovation and to excel at it?
• External partners—Who do you need to partner with to help your
business deliver this innovation? What essential activities will
these partners perform on your behalf? What assets or skills will
they provide so you don’t need to?

RIGHT TO WIN
If your venture is to succeed in the long term, there will need to be
some kind of barrier or competitive moat that prevents competitors from
doing an equal or better job of delivering the same thing.

• Unique advantages—What unique skills, assets, relationships, or

other advantages do you hold that will enable you to deliver this
innovation? For ideas, look back at our discussion of unique
advantages in chapter 3.
• Competitive benefits—How exactly do these advantages help you
deliver this innovation better than others? For example, do they
lower your costs, lock in an established customer base, or provide
a uniquely better customer experience?
• Deepen the moat—Look to the future as you consider your
competitive moat. How will you reinforce your unique advantages
for the future? Or how can you invest to develop a competitive
moat that you currently lack?

4. Business Validation
The final four blocks of the Rogers Growth Navigator are used to
capture your hypotheses and learning in stage 4: business validation.

VALUE CAPTURE
Here, you need to define the value captured by your own organization
from this innovation, including any revenue streams or other streams of
value to your business:

• Revenue model—Will you charge customers for this innovation? If

so, how? Options include sale, license, rental, or fee for
performance. What price is the customer willing to pay, and how
can you influence this? Are you charging a fixed price or a sliding
scale, or using a freemium model (with free versus paid versions)?
• Paying customers—Who pays you? Is it the user or someone else
in their organization? For a multisided business model, which
customer type pays? If your innovation grows, could you add
revenue streams from another customer type, for example, from
an advertiser seeking to reach your primary customer?
• Cost/risk savings—Will your innovation deliver cost savings to your
business? Will it reduce a significant risk for your business? If so,
can you quantify the financial value?
• Nonmonetary value—Will your business capture value in other
ways, for example, data, or new relationships? How might you
monetize these assets? What would you pay for them from a third
party? If your innovation delivers value toward a nonprofit mission,
how can you measure this? What would you need to spend
elsewhere to achieve an equivalent impact?

CLV
Use this block to define the CLV for your innovation—a measure of the
total profit you can expect to earn per customer. CLV is calculated
based on average customer revenue, profit margin, and life span (i.e.,
time before they churn). To maximize your CLV, focus on four aspects
of customer value:
 • CLV by segment—How does CLV vary by customer segment?
Which customers have the highest lifetime value and why? Which
segments should you avoid due to their low value?
• Acquisition—What is your cost to acquire customers (CAC)? How
can you reduce this cost? If CAC is less than CLV (your goal), how
much and how fast can you spend to acquire more customers?
• Retention—What is your current customer churn rate? What
triggers, influences, or predicts churn in an individual customer?
How can you retain customers longer? What will that cost?
• Expansion—How can you expand revenue or profit margin—or
both—per customer, for example, upgrade their plan, increase
their purchase frequency, sell additional products, or encourage
referrals to new customers?

COST STRUCTURE
Use this block to capture all your costs. These will include your costs to
acquire and retain customers (look at the CLV block of your Navigator),
the costs to deliver your solution (look at your delivery and capabilities
blocks), and your ongoing costs of innovation (look at your value
proposition, solution, and right-to-win blocks). To understand your cost
structure, categorize all these costs into three types:

• Fixed costs—Which of your costs are roughly the same no matter

how many customers you serve?
• Marginal costs—Which costs are variable and proportional to your
number of customers? Think of these as the extra costs (from
acquisition to delivery) for each new customer.
• Scale economies—Which of your costs are variable but will cost
less per customer as you grow in scale?

PATH TO PROFIT
In this block, bring together costs and revenue to project the net value
of your innovation and to judge whether it is worth pursuing:
 • Profit formula—Combine your cost structure and your value
capture to create a formula for the net loss or net profit as your
venture scales in size. At what point will you break even? What
should the profit margin of this venture be at full scale? (The
interplay of costs, revenue, and scale are described as your
business’s unit economics.)
• Time frame—When do you aim to achieve a net financial return on
this venture? Is it meant to be a near-term revenue boost or cost-
cutting effort? Or is it a long-term investment in growth?
• Maximum upside—What share of your total addressable market
can you realistically capture? What would that mean in total profit
for your firm if you succeed? Is that opportunity large enough to
continue to pursue this innovation?

Tips on Using the Navigator

At the start, each block of your Rogers Growth Navigator will contain
only hypotheses and assumptions about your new venture. (Any third-
party data you might have found is just another hypothesis that will
need to be tested in the real world to see if it holds true for your
venture.) Remember, all learning comes directly from the customer!
I advise teams to fill only the top six blocks of the Navigator when
they begin. Once you have made progress on stage 1 and stage 2
validation—and learned something about your problem, customer,
value proposition, and solution—you can start to write down your
hypotheses for the bottom eight blocks of the Navigator.
As you test and learn with your team, you will gradually replace the
untested hypotheses you wrote in the Navigator with validated facts
you have learned in the real world. As you do, use color coding to
distinguish hypotheses from facts (I recommend writing hypotheses in
red and validated facts in green). Choose a third color (e.g., blue) for
the hypotheses you need to validate next. This use of color will help
your team visually track your progress.
Keep copies of your Navigator’s weekly iterations to capture your
thinking over time. You may choose to add supplemental documents
digging into a block in more detail (e.g., a detailed map of your
customer journey or your CLV calculation per segment). But the
Navigator itself will remain your single-view summary—both of your
current understanding of your business model and of what needs to be
tested and learned next.
In some cases, your venture may use a platform business model
(one that facilitates a value exchange between two or more distinct
types of customers).22 If so, you should complete a separate Growth
Navigator for each customer type. For example, for Uber, you would
create one Navigator for drivers and one for passengers. Each
Navigator will define that customer’s problems, your value proposition
to them, their customer journey, and so on. A few blocks may repeat
unchanged between your Navigators, but it is critical that you validate
your business model for each distinct customer type. I also suggest
using the Platform Business Model Map (a tool from The Digital
Transformation Playbook) to analyze the exchange of value between
customers and through your business.

Validation Never Ends

For any innovation, growth venture, or new business—the process of

validation never ends. As Bob Dorf says of running a start-up,
“Validation only stops when you sell your business for millions, or the
furniture for pennies!” As we have seen, there is a sequence to the
Four Stages of Validation, but each stage repeats iteratively. None of
the stages—problem, solution, product, or business—is ever “done.”
Any successful venture will make progress in validating at stage 1,
2, 3, and even 4. Yet we still need to keep going back to revisit, and
revalidate, every stage. Why does this happen? Why, for example,
would you need to continue problem or solution validation once you are
already in the market with a product that customers are using and
paying for? There are many reasons why an innovation will need to
circle back and iterate on earlier stages. These are some of the most
common:

• Feature rollout—Your current customers are actively using your

product (stage 3). Now you are ready to build the next items on
your feature roadmap (which you validated in stage 2). Before you
 build anything, be sure you go back to your customers to
revalidate: do they want the same features next, now that they’re
actually using your solution every day?
• Unexpected business problems—Is growth stalling? Customer
churn increasing? Profit margins starting to narrow? No matter the
problem your growing venture may face, the best way to identify
the root cause and address it will be to go back to the customer
and validate from stage 1 onward, with continuous learning.
• Scaling operations and delivery—If your new venture is
succeeding and is profitable (stage 4), you will want to scale up
operations in order to expand into new markets and new use
cases. This may require rebuilding the back end of your product in
a number of ways: more automation and less human oversight;
more data from more sources; more robust IT architecture with
greater scale, reliability, and speed; different sales and marketing
channels that can scale to higher volumes; different partners
(versus doing it all yourself); more security because bigger
ventures attract more malicious actors; more compliance,
especially if new markets bring new regulations. Each of these
changes will require stage 3 validation to test and learn what
works best to deliver a robust and scalable product.
• Unexpected changes in the environment—Whenever competitors
enter the market or change their product or pricing, you will need
to revalidate whether your offer is still differentiated in the eyes of
the customer (stage 2). At other times, your customers’ needs will
change. A former student of mine was leading digital sales for
Brazil’s leading bank, Itaú Unibanco, when COVID struck. His
customers’ banking needs changed overnight due to quarantine,
requiring a rapid redesign of the bank’s mobile app. This meant
going back to stages 1, 2, and 3—revalidating the customer’s new
problems, the features that would help, and the best product
design to deliver them.
• New market, new customer—As any new venture grows, it will find
itself shifting to serve different customers. As Geoffrey Moore
explains in his book Crossing the Chasm, there is often a major
shift when a venture expands from its early adopters to the
broader customer population.23 You may also change customers
 because you are expanding into a new market or geographical
location. Successful teams focus on understanding each unique
customer segment and continually repeating the Four Stages of
Validation. For every customer segment, be sure you understand
that customer’s problem, their best solution, their use cases and
delivery needs, and how to price and capture value from them.
• Finding your ideal customer—Once an innovation achieves
product-market fit and adoption, it is no longer enough for you to
sell to any customer that has your problem and will buy your
product. Instead, you need to focus on validating and learning
which of your customers will drive repeatable, scalable, profitable
growth. Any business leader should be asking themselves, Are we
focused on the right customers to scale this business? For a good
example, see the box “Optimizely Revalidates for a New
Customer.”

Optimizely Revalidates for a New Customer

When Jay Larson arrived as the new CEO of Optimizely, the start-up
was eight years old and ready to push for its next stage of growth.
Optimizely was founded to give businesses the tools to run data-
driven experiments—to test features, design, pricing, and
personalization within websites and communications.
In its early days as a SaaS start-up, growth was all that mattered
to Optimizely, and every customer was a good customer. By the time
Larson arrived, they had over 5,000 business customers, ranging
from small local businesses (whom Larson dubbed “Joe’s House of
Wicker”) to global brands like Nike, Best Buy, and the Wall Street
Journal. Larson quickly realized that not all customers were equal,
and Optimizely would need to focus on the right customers in order
to reach its next stage of growth.
Larson’s team focused on learning which business customers
were getting the most value from Optimizely’s tools. These turned
out to be businesses with lots of customer interactions (providing
more data) and business models like e-commerce or subscriptions
where experimentation drove measurable change in revenue.
Further analysis found that these customers were already spending
over $50,000 a year with Optimizely. Firms that spent less were
getting less benefit from Optimizely and had much lower retention (a
key metric for any SaaS business).
Larson decided to change the target market. Larson told his team
there would be no more sales to “Joe’s House of Wicker,” and
implemented a $50,000 floor for all new sales. Small-business
clients were told they would need to pay more or find another
product. Meanwhile, Optimizely shifted its focus to sell to companies
that fit the highest-value customer profile.
As Optimizely homed in on its most valuable customers, it
redesigned its products and services to meet their particular needs.
Optimizely launched a new full-stack product that allowed clients to
run experiments on their own software code, integrating much more
of their operations than just websites and communications. In
customer interviews, Larson’s team discovered that the biggest
barrier to clients’ success was not a lack of experimentation tools but
the challenge of building a more data-driven culture. Firms saw the
value of experimentation, but they had to unlearn old habits of
decision making. To address this need, Optimizely established a new
customer success division focused on consulting services to help
enterprise clients become more effective experimenters.
These changes made a huge impact on Optimizely’s customers.
Nike, for example, had started as a client when its marketing
department was looking for a workaround to do website testing
without going through their own IT department. Now, Nike began to
use Optimizely’s full-stack solution to run experiments not just on
websites but also on its apps and services like Nike Run Club.
Working with the customer success team, Nike broke down barriers
between its own silos and brought teams together to look at e-
commerce and understand individual customers’ buy flow. Nike even
used Optimizely to experiment on its own vendors, measure the ROI
it was getting from each, and push them to work harder for Nike.
Larson also decided to revalidate Optimizely’s pricing and
revenue model. In examining customer performance data, he found
a financial disconnect. The best clients, like the Wall Street Journal,
were measuring revenue gains over $50 million a year thanks to
Optimizely, but they paid less than 1 percent of that in fees. In
 response, Optimizely began to price its services differently,
sometimes based on a split of their clients’ revenue gain, and thus
capture more value for the firm.
The results of refocusing and revalidating the business model on
the most important customers were dramatic. Optimizely raised the
average size of its new customers from $40,000 to $200,000. They
eliminated customers with the lowest value and highest turnover
rates. And they successfully pushed upmarket into the enterprise
clients who could deliver sustainable growth into the future.

Validation for Every Stakeholder

As you continue to grow a new venture or business, strive to do

validation for every stakeholder, treating each one as a customer
whose needs must be met. In an interconnected world, your innovation
will only succeed if you can bring the right partners to the table.
When Edison prepared his electric lighting business for the market,
he did not focus only on the user of light bulbs at work or home. He
focused on the whole ecosystem of electric bulbs, meters, power lines,
stations, and generators—and on the individuals and organizations
behind all of it. Edison aligned with industry partners on the technical
standard for electrical current. He ran his first pilots near Wall Street to
attract the interest of investors. And he used his fame and reputation to
lobby support from regulators and overcome resistance from the
lamplighter’s union.24
As you scale your own venture, look to the external partners in your
Rogers Growth Navigator and the allies that will be most crucial to your
success. Include upstream suppliers and downstream business
partners, technology standards and data suppliers, as well as policy
makers and other influencers. A student of mine leading a consumer
health start-up found it was equally important to validate and solve the
top problems of her big-box retail partner as it was to validate and
problem-solve for the end user of her product.
If your venture is facing resistance from your own organization, try
applying the Four Stages of Validation on the inside. Treat every
internal stakeholder—whether compliance, the sales force, or the CDO
—as a customer. Then do validation on them: Who are they (i.e., job
title and role)? What’s their problem (i.e., the root cause why they may
not support your venture)? What’s your solution (i.e., how can you
address their concerns to get them on the same page)? Yes, this
means interviews with internal stakeholders, just like any external
customer. This is particularly critical in stage 3 and stage 4 validation as
you test how you will deliver your innovation and do it profitably. Always
remember: Validation is not a stage of innovation. Validation is
innovation.

Experimentation from the Bottom Up

Embedding this approach to innovation will require a change in how

many leaders see their role. Innovation is no longer about a few
important people coming up with, or signing off on, “the big idea.”
Instead, it is about lots of people coming up with lots of ideas and using
a repeatable process to test and learn which of them could work and
how. Effective leaders understand that ideas are easy, but validation is
hard.
Experimentation also demands a different approach to decision
making. In many organizations, decisions are made based on seniority
and experience. In Silicon Valley, this is called decision making by the
HIPPOs (highest paid person’s opinion). To support a culture of
experimentation, leaders need to refrain from opinion-based decision
making and instead task others to experiment and find the right data.
As Optimizely’s vice president of customer success, Jennifer Ruth, told
me: “If you have a leadership team who are just all gut instinct and
believe that no one else in the world knows better than they do, then
experimentation is not going to be successful.”
Leaders must work hard to instill a culture of experimentation. This
means encouraging others to place bets and take smart risks, to spend
a little to learn a lot. It means celebrating “smart failures” like Walmart’s
Jetblack or its robot floor sweepers. In The Digital Transformation
Playbook, I offered a four-part test for smart failures: Did you learn
something important? Did you apply that learning to change your
strategy? Did you share your learning in your organization? Did you fail
as early and cheaply as possible? Leaders must make a point to
celebrate failures that meet this four-part test.
Effective leaders focus on instilling the right mindset within their
teams. They continually communicate and live experimentation’s key
principles: Don’t debate; validate. Think like a scientist. Get to market
sooner to learn faster. Spend as little as possible to learn as much as
you can. Test in minutes not months. All learning comes from the
customer. Commit to the problem; be flexible on the solution.
Just like strategy, experimentation should happen at all levels of the
organization. It is not the case that one team “does strategy” and
another team “does innovation” for the business. As we saw in chapter
4, strategy cascades up—every team hones its strategic P/Os in
support of the P/Os of the teams above them. Every team should then
be developing venture ideas based on its strategy and iteratively
validating those ventures in the market.
The methods I have laid out for you—illustrative and functional
MVPs, staged metrics, the Four Stages of Validation, the Rogers
Growth Navigator—are all designed to work at every level of the
organization. The same process of validation works whether you are
building an entirely new business, revamping an existing product, or
updating an internal process. Experimentation and continuous learning
should be applied not just to product development but to every function:
marketing, sales, human resources, risk management, supply chain,
and so on. Of course, this is not what we see in many organizations.
Innovation is often handled by a dedicated team in a special “island of
innovation,” exempt from the normal rules of business. As we’ll see in
chapter 6, experimentation from the bottom up requires a very different
approach to managing growth throughout the company.

At the start of a DX, no business, no matter how innovative, can know

which digital products, services, or business models will succeed in the
real world, delivering value to customers and capturing value in turn.
Without experimentation, the greatest leadership commitment to “go
digital” will end up in costly failures, like CNN+, which lost $300 million
launching an idea pushed from the top. In Step 3 of the DX Roadmap,
we have seen how any organization can avoid this fate by learning to
validate new ventures to learn which ideas may work and how. You
learned how to test your assumptions with iterative MVPs, each one
designed to answer a specific business question. And you learned how
to use the Four Stages of Validation to guide any venture on its path
from an idea on the page to a business at scale.
As you start to apply the process of experimentation to digital
ventures in your own business, you are now ready to begin the next
step of the DX Roadmap: repeating this process at scale across your
entire organization. Major challenges must be addressed. Which
ventures should be funded? Who will decide when to shut some down?
What rules should be waived for new ventures at the start? How will the
successes be handed off to the core business? What will you do with
digital ventures that don’t fit your current organizational structure?
To scale the process of experimentation across your business, you
will need a clear approach to managing resources and people across a
portfolio of ventures aimed at different strategic opportunities. You will
need a governance model that embraces digital innovations inside your
core and beyond it, with both low uncertainty and high uncertainty. In
chapter 6, we will see how to manage growth at scale so that digital
transformation touches every aspect of your business.
 6
Step 4: Manage Growth at Scale

When Vanessa Colella was appointed Chief Innovation Officer for

Citibank, her mission was to establish a new group called Citi Ventures
to help drive innovation across the global company of 200,000
employees. The challenge was daunting: to help a legacy bank keep
up with the ferocious pace of change being driven by emerging
technologies, changing consumer needs, and the venture capital
flooding into fintech start-ups.
Colella’s team worked with the rest of the business to identify a set
of problem/opportunity (P/O) statements that mattered most to the
future growth of the bank—with themes ranging from cybersecurity to
distributed payments, to adapting banking products to suit changing
social trends. But the essential question Colella faced was, How do we
pursue these strategic priorities across such a large corporation? What
would be the best structure and governance for new ventures so we
can move fast like a start-up but achieve scale and impact within a
huge legacy enterprise? Should new ventures be managed inside the
business units? In a separate lab? Outside the business entirely by
investing in start-ups?
Colella decided from the outset that it was critical not to try to move
ahead without the core business itself. “The era of innovation as a
separate unit that then hands things over to the business is gone,”
Colella told me. Instead, Citi Ventures aimed to partner extremely
closely with the business to ensure that its innovation process was
deeply embedded, and it involved the very bankers who work and
interact daily with Citibank’s clients.
The first step was to establish D10X, an innovation accelerator for
Citibank’s existing business units. Its ambition is to support radical,
transformative innovations that are still related to Citibank’s core
businesses. There are two D10X accelerators, one each for the Global
Consumer Bank and the Institutional Clients Group. As Colella
explains, “Employees in corporations like Citi also have fabulous ideas
of how they can better serve their clients.”1 They just need a repeatable
process and set of rules to turn those ideas into innovation with an
impact.
That process starts with a pitch, just like a start-up seeking
investment from a VC. Citibank employees of all ranks (from interns to
senior vice presidents) pitch their innovation ideas to a panel of
decision makers called a growth board. Ideas deemed promising are
entered into the D10X program with a very small amount of seed
funding. Employees are placed in small multifunctional teams, coached
by a set of entrepreneurs-in-residence, and given a very short time
frame in which to validate their ideas with clients to learn if they
address a genuine market need. Rather than building a product, the
initial focus is on validating the customer’s problem. At the end of that
short sprint, the team returns to the growth board to present its data,
and—if the team thinks that data show a genuine opportunity—to pitch
again for additional funding and time. These pitch sessions, called Deal
Days, happen on a rolling basis. Teams return to them repeatedly, just
like an independent start-up going back to investors for repeated
rounds of investment as it proves the case for its business.
At any given time, 100 of these internal start-up teams may be at
work in various stages of D10X. Teams from the consumer banking
side have launched new customer experiences to transform budgeting,
account management, and personal debt. On the institutional banking
side, D10X projects have included CitiConnect for Blockchain, in
partnership with Nasdaq, and a virtual proxy-voting platform for
investors called Proxymity, which was so successful it was spun off as
a separate public company.
But D10X is not the only approach to driving digital innovation at
Citi. A division called Citi Ventures Studio was launched to pursue
innovation outside the scope of Citibank’s core business—where
employees might not otherwise look to innovate. Citi Ventures Studio
was started by Valla Vakili, a seasoned entrepreneur Colella hired to
bring ideas from lean start-up and design thinking into the bank. One of
Citi Ventures Studio’s first ventures was Worthi, a free online tool that
uses labor market data to provide personalized career development
tools so individuals can explore new jobs, estimate salaries, and
develop skills that match market needs. Another venture was City
Builder, a data-driven platform to support investments in U.S.
opportunity zones by bringing together investors, fund and wealth
managers (at Citibank or elsewhere), and cities that are seeking
investment to drive economic renewal.
Other programs at Citi Ventures look beyond Citibank’s employees
to spark innovation. The Citi University Partnerships in Innovation and
Discovery (CUPID) program uses hackathons to engage students from
leading universities in innovation efforts across Citibank. Citi Venture
Investing invests directly in the start-up ecosystem, funding early-stage
fintechs that have achieved product-market fit and are operating in one
of a few focus areas for Citibank (payments, fraud, machine learning,
customer experience, etc.). Investments have included highly
successful fintechs like Docusign, which later had an IPO on the
Nasdaq, and Honey, which sold to PayPal for $4 billion.
Despite the wide-ranging approaches that fit under Citi Ventures,
Colella is clear about her team’s mission: “We don’t run innovation.”
Rather, they enable it across the enterprise. Citi Ventures is comprised
of less than 100 people, but its influence is far-reaching. Many
thousands of Citibank employees have participated in D10X, Citi
Ventures Studio, CUPID, and other programs. Senior leaders from
every business unit are actively involved in growth boards, overseeing
a new approach to innovation. The mandate of Citi Ventures is not just
to launch new ventures but to change the culture and practice of
innovation in the whole enterprise.

Why Governance Matters

Rapid, iterative, customer-centric innovation is possible in large

companies—but not if you carry over your traditional ways of working.
Innovative ventures need innovative governance.
When BASF launched its Onono lab in São Paulo, Brazil, its
mission was to accelerate innovation for the global chemical firm
through rapid collaboration with partners and start-ups. But Onono’s
director Antonio Lacerda was told that the lab would have to follow the
same data policies designed to secure the company’s entire cloud
infrastructure. As Lacerda explained to me, that would have made it
impossible to partner quickly and nimbly with new start-ups. So, before
launching Onono, he expended significant political capital to arrange an
exception: a “sandbox” of separate data was created for the Onono
team, with special permission to share APIs with new partners. Lacerda
was able to customize governance to support Onono, but digital
transformation (DX) cannot rely on a series of ad hoc decisions and
waivers to business rules granted by higher-ups. It needs new
management practices that scale and repeat.
In Step 4 of the Digital Transformation (DX) Roadmap, you take on
the challenge of managing growth at scale. In Step 3, your task was to
validate, adapt, and grow a single digital venture. Now your goal is to
grow not just one venture but a portfolio of ventures serving a range of
strategic priorities. In Step 4, you expand your focus to managing
growth across an enterprise.
Designing repeatable processes for innovation is essential for the
growth of any established business. Yet it is incredibly hard. Every
talented executive I meet bears the scars of corporate innovation
struggles. In too many organizations, new ventures are green-lit based
on a single executive sponsor. Once started, ventures move slowly,
managed by teams that sit in traditional silos. Resource allocation is
slow too, and promising projects wait weeks or months for their next
round of approvals. Because each project is backed by an influential
executive, no one wants to shut it down, even when it shows little
promise. Meanwhile, risk aversion leads businesses to fund only their
low-hanging fruit—incremental improvements in the core that bring a
guaranteed, quick ROI. This path will never lead to DX. Instead, you
need governance that embraces uncertainty and supports growth both
within and beyond the core. Table 6.1 shows some of the key
symptoms of success versus failure in Step 4 of the DX Roadmap.

Table 6.1.
What’s at Stake—Step 4: Governance
Symptoms of Failure: Symptoms of Success:
Governance Governance
• A top executive must • Established structures
personally approve any new provide resources and
innovation. governance for innovation.
• New ventures move slowly, • New ventures move fast,
led by traditional teams in led by highly independent,
functional silos. multifunctional teams.
• Allocating resources to new • Resource allocation
ventures is slowed by the happens quickly through
annual budgeting cycle. iterative funding.
• Innovation is limited to a • A steady pipeline of
few big projects, which are innovations is managed with
hard to shut down once they smart shutdowns to free up
are started. resources.
• The only ventures to • Governance supports
gain support are low-risk ventures with low and high
 innovations in the core uncertainty, both in the core
business. and beyond.

The challenge for Citibank, BASF, and any business seeking growth
is to pursue a variety of new ventures in different business units and
functions at the same time. These must include some ventures in the
core and beyond the core, and low-risk ventures as well as highly
uncertain ones. Each venture must be supported by the right
governance model so that it can succeed. This means developing not
one but a mix of different structures—like Citibank’s mix of internal
accelerator, external investing, university partnerships, and innovation
studio—designed to manage different growth opportunities.
For each structure, governance rules must be carefully designed to
address several issues. The first is oversight. Who approves new
projects? To whom do they report? And who shuts them down? Next is
funding. How will you allocate resources across ventures and avoid
“comparing apples to oranges” (where a long-term bet on growth must
compete for funding with a critical infrastructure project)? How will you
ensure funding is iterative rather than locked into annual budgeting?
Equally critical are people. Who will decide that a talented executive will
be pulled out of your core business to work on a new venture that has
yet to turn a profit? How will teams be formed with the right
multifunctional skills? Governance must also include metrics. How will
you measure the progress of new ventures? How will you assess
ventures that have different time horizons or different levels of
uncertainty? There is also the crucial management of compliance. How
will you help new ventures move fast while respecting safety,
regulations, risk, and integration with existing technology? All these
issues will be addressed in Step 4 of the DX Roadmap as we introduce
the elements and tools you need for successful governance.

What’s Ahead

In this chapter, we will see how any organization can design

governance models to drive digital innovation across the enterprise. We
will examine six essential elements to managing growth at scale:
 • Teams and boards—How to ensure that they work together to
accelerate innovation and allocate resources across a portfolio of
ventures.
• Green-lighting—How to approve and start new ventures with
minimal deliberation and minimal investment.
• Iterative funding—How to match investment to level of uncertainty,
shift resources rapidly, and accelerate when the time is right.
• Smart shutdowns—How to manage your pipeline of ventures
systematically and free up resources by shutting innovations down
smartly.
• Three paths to growth—How to manage ventures that have
different levels of uncertainty as well as ventures near and far from
your core.
• Innovation structures—How to set up different structures (labs,
hackathons, venture funds, and more) that provide pools of
resources and tailored governance for different types of ventures.

After examining each of these six elements, we will introduce a

strategic planning tool, the Corporate Innovation Stack—to help you to
define the rules for your innovation structures, boards, and teams.
Finally, you will learn why growth at scale requires a bottom-up
approach that redefines the roles of every team and every leader.

Teams and Boards

For iterative, experiment-driven innovation to work at scale inside any

organization, we require two different groups of people. The first group
are the innovation teams: those who do the work of innovation by
talking to customers, building iterative MVPs, running experiments to
validate business models, and bringing a successful venture to market.
The second group are the sponsors or, more specifically, the innovation
boards: those who approve new projects, allocate funding, and oversee
the progress of teams. To manage growth at scale, it is critical to
understand the roles of teams and boards, and the management
processes each will need to succeed.
 Innovation Teams

There is a saying in Silicon Valley that “leaders vote with head count.”
As a veteran product manager explained to me, tech giants like Google
and Meta have plenty of money to throw at new ideas. Their scarcest
resource is good talent; so that is what matters most to a team. At
Amazon, the clearest sign that Jeff Bezos was committed to a new
venture was the caliber of people he would peel off existing parts of the
business to work on the newer one.
Any team that is working on a new venture will, of course, need to
be well versed in the mindset and methodologies of iterative
experimentation, which we saw in chapter 5. But incredible talent and
the best practices of lean start-up, agile, design thinking, and product
management will not, by themselves, bring success.
For any team that sits within an enterprise, the rules governing that
team are critical to its function. Many readers will be familiar with the
idea of the small multifunctional team, which is central to the practice of
agile and product management. But size and composition are only part
of what sets up a team for success. By studying innovation in digital
natives like Amazon and Google, and digital transformers like Citibank
and Walmart, I have identified five essential pillars of team governance.
Great innovation teams are:

• Small—Margaret Mead famously said, “Never doubt that a small

group of thoughtful, committed citizens can change the world;
indeed, it’s the only thing that ever has.”2 Citibank’s Vakili shares
the same view: “I believe in the outsize productivity of small teams,
if they can be equipped properly.” Why keep teams small?
Research by J. Richard Hackman and others has shown that
smaller teams coordinate, communicate, and make decisions
much faster.3 And as we know, speed is essential to innovation.
Small teams are foundational to agile methods like Scrum, which
demand a rapid cadence of short sprints in which every team must
deliver new working code, test and learn, and adjust priorities. At
Amazon, innovation teams are called two-pizza teams because
each one must be small enough to be fed by two pizzas (maximum
of eight people). In traditional enterprises, I have heard objections
to the idea of such small teams, with executives insisting their
 project is too important to move forward without the involvement of
a dozen stakeholders or more. But I have never seen a team over
the two-pizza limit succeed in rapid innovation.
• Multifunctional—Great innovation teams have diverse members
who cut across functional silos (e.g., marketing, engineering, and
design). Each team should contain among its members all the
essential skills needed to do its work. Teams can then move fast
because they are self-sufficient. Instead of constantly waiting for a
report, data, or support from another department before taking the
next step in a project, a multifunctional team should be able to act
entirely on its own. The precise mix of team roles will depend on
the organization. At Procter & Gamble, a typical team combines
marketing, consumer insights, design, and R&D. At Walmart, a
nine-person product team typically has six software developers. In
start-ups, any team is usually multifunctional from the start, but for
most large organizations, multifunctional teams are a big shift from
the typical functional silos.
• Single-threaded—The best innovation teams have all their
members dedicated full-time to the team’s work. This is the model
of classic agile teams, and it is how design firms like IDEO work,
with every team member responsible for only one venture at a
time. The term “single-threaded” comes from Amazon. It is a
metaphor borrowed from computer science (for an application that
executes only one part of a program at a time). At a minimum, any
innovation team’s leader must be single-threaded. They cannot be
splitting their work week between the venture team and other
projects. Leading the team is their full responsibility. Single-
threaded leaders are the norm in a start-up (as are single-threaded
teams, as soon as there is funding to pay salaries). Embracing
single-threaded leadership is a challenge for most large
organizations where every manager is committed to several
projects.
• Autonomous—Innovation teams should have clear decision rights
that give them the authority to work under their own direction.
Teams should not need to get approval on their work from anyone
outside the team—whether it is on product design, what tests to
run next, or which customers to pursue. Autonomy also means
there are no prohibitions on contracting resources from outside the
 company. As one innovation team told me, “We don’t work with our
company’s shared services IT at all. We’re outside that ecosystem
for a reason.” Complete autonomy is to be expected in any start-
up, where investors are not managers, and board oversight is
never day-to-day. But team autonomy will be a radical shift for
large organizations that are wedded to top-down management.
• Accountable—Every innovation team must be accountable for the
results of its work. This is what allows it to maintain total autonomy
over how it achieves those results. Team accountability flows from
two factors. The first is a clear definition of success—which is
defined in terms of outcomes, not deliverables. That definition may
include concrete metrics as well as qualitative principles, and it
must be agreed on with leadership before the team’s work begins.
The second requirement is transparency. At any time, the team’s
results must be visible to anyone inside or outside the team. Every
test run, every MVP built, and every metric tracked should be
visible to anyone in the company. At Walmart Labs, transparent
tools display each product team’s members, what they are working
on, and every piece of code published. Accountability and
transparency are inevitable in a start-up (everyone knows when
your marketing campaign is failing to acquire new customers). But
this is a dramatic departure from corporate work. In most large
organizations, accountability is diffuse and shaped as much by
politics as by outcomes.

Table 6.2 summarizes these five pillars and shows their sharp
contrast with the business-as-usual (BAU) management of teams in the
predigital era. Too often BAU teams are sprawling, siloed, fractional,
micromanaged, and political—whereas great innovation teams are
small, multifunctional, single-threaded, autonomous, and accountable.
This gulf between teams is not the result of different people but of
different governance. Innovation teams have a different mandate than
the functional teams in a legacy organization. BAU functional teams are
designed either to execute a project with a fixed deliverable (e.g.,
“migrate our customer database to our new cloud service provider”) or
to carry out an ongoing function (e.g., “run marketing for our Southeast
Asia market”). In contrast, innovation teams are designed either to
pursue a new growth venture or to innovate for a persistent problem for
a key customer or stakeholder. Team form follows team function.

Table 6.2.
The Five Pillars of Innovation Teams Versus BAU Teams
Innovati To pursue a BAU To
on team new growth function execute a
venture or al team temporar
innovate for a y project
persistent or carry
problem out an
ongoing
function
Small • Fewer than ten Sprawli • Many
people on a ng team
team members
with
varying
levels of
involvem
ent
• Rapid cadence •
of Communi
communications cation
and work bottlenec
ks are
frequent
Multifun • Team Siloed • All
ctional members from members
different silos from a
single
function
or silo
• Have all • Work
essential skills depends
on the team on the
skills of
others
Single- • Team leader is Fraction •
threade committed 100 al Everyone
d percent is
working
on the
team
part-time,
juggling
other
priorities
• Team
members have
few or no other
assignments
Autono • Team owns all Microm •
mous decision making anaged Approval
between s are
funding required
milestones for all
major
decisions
(many
people
can say
no)
Account • Clear definition Political •
able of success Deliverab
le is
clear, but
success
is
debatabl
e
• Transparent • Access
metrics to metrics
is
restricted
• Team is solely • Many
 responsible people
are
responsib
le in
theory,
but none
in
practice

Growth Boards

The other critical group for managing innovation at scale is the

managers who will allocate funds and oversee the work of teams. One
of the first questions I ask executives looking to scale their digital efforts
is, Who sponsors innovation at your organization? The most common
answer is that new ventures are approved by a single sponsor. One or
more executives may play this role and use their clout within the
organization to support a new digital venture they deem strategically
important and promising. The problem with the single-sponsor model is
that it is inherently ad hoc, with decisions based on the instincts and
judgment of different individuals. It also runs the risk of sustaining pet
projects. Once a sponsor puts their name behind a project, it is very
hard for them to let it die, no matter what market validation shows about
its prospects.
The most successful model for sponsoring corporate innovation is
the board. In the board model, a group convenes and deliberates
together to decide whether to sponsor various possible innovation
ventures—much like a group of VC investors listening to pitches from
start-ups, or the judges of a hackathon picking a few winners out of
dozens of teams. One board can sponsor multiple innovation teams at
the same time. (The single-sponsor model, by contrast, is usually “one
executive backs one team,” which makes it extremely hard to withdraw
support.) Diversity, agility, and impartiality make the board model
inherently better at managing innovation as a repeatable process.
Corporate innovation boards are often dubbed growth boards, a
term that Stephen Liguori coined at General Electric (GE) in 2014 to
describe a small group that oversees funding decisions for internal
innovation. Eric Ries, who was an adviser to Liguori at the time, helped
popularize the term, which has been adopted at Citibank, Procter &
Gamble, and elsewhere.4 Other names for the same idea include
venture board and growth council. Whatever the name, any corporate
innovation board has four critical jobs:

• Green-lighting new ventures—The board must choose which

ventures to green-light with an initial investment and an agreed
definition of success. This initial investment may be an equity
stake in an outside start-up or the seed funding for an internal
team to begin validating its idea. When approving internal teams,
the board must allocate head count and other resources, as well
as budget.
• Iterative funding—Funding ventures is completely different from
funding departments or operating units; it must be given iteratively,
based on data. The board’s second job is to review each team’s
progress regularly and decide whether to release the next tranche
of resources to that team or to shut down the project and free up
team members to work on other priorities.
• Strategic guidance—Between funding cycles, the board should
meet regularly with each team to review each venture’s progress.
As the team learns from iterative testing and considers next steps
to advance its innovation, the board should provide strategic
advice.
• Liaising and advocacy—The board must also act as the liaison
between the venture team and the rest of the parent organization,
making introductions, connecting the team to company resources,
and removing organizational obstacles whenever needed.

Having the right people on a growth board is essential to its

success. As one business leader remarked to me, “You don’t want a
bunch of MBAs. You need people who think like VCs or entrepreneurs.”
Having observed growth boards in a variety of companies, I
recommend the following characteristics for the best board
composition:

• Small—Effective boards are small (no more than eight people),

following the two-pizza rule that successful teams do.
 • Heterodox—To support innovation, a board must be able to
challenge company orthodoxy, advocate a long-term view, and
bring in ideas from outside the industry. The best boards combine
internal stakeholders from different divisions or business units, and
at least one member with an external perspective. Experience in
entrepreneurship or venture capital is a huge benefit.
• Capable—Teams need a lot of help from their board, so it should
include members with knowledge of the market, topical expertise,
and clout within the organization. If the board is to be an effective
champion for its teams, it must wield real influence in the parent
company.
• Engaged—Boards must meet frequently so that feedback and
funding can be truly iterative. Every six or eight weeks is
preferable, and quarterly meetings are the absolute minimum.
Board members must be actively involved during and between
meetings, whether in person or remotely. A golden rule for
meetings is, If you don’t attend, you can’t vote.

The seniority of board members is important to get right. At least

some board members must be senior enough to have real clout in the
organization, but no members should be so senior that they are too
busy to make time for board meetings. I have seen organizations that
try designating the top executives of the entire company as their
venture board to oversee innovations. Too often, these executives want
to be the ones who green-light new ventures, but they simply don’t
have the bandwidth to track and counsel each team’s progress. In
some cases, top executives may be too far removed from the market to
judge the team’s work as well as others less senior.
Defining decision rights is critical for boards to work effectively with
the teams they oversee. We can think of the roles of teams and boards
as analogous to the roles of independent start-ups and their VC
investors. Each side needs to have very clearly defined authority:

• Team decision rights—The innovation team, like a start-up, has

autonomy, that is, complete control of its venture between funding
milestones. While the board may provide input and advice, the
team holds decision rights on what customers to pursue, what
MVPs to develop, and what tests to run.
 • Board decision rights—The innovation board, like a VC investor,
has complete funding authority for each team in its portfolio. The
board’s decisions should be made in conversation with the team,
as part of an open and lively debate, but the decisions remain with
the board. Other senior executives, including the CEO, may advise
and provide input to ventures, but they cannot vote on or overrule
the board’s investment decisions.

The board model is superior to the single-sponsor model for several

reasons. The benefit to the team is that a board provides multiple
coaches and advocates who can each offer the team different
perspectives, technical expertise, and relationships inside and outside
the company—more than any single sponsor can. The benefit to the
company is that a board is designed to allocate resources based on
results (i.e., how a venture validates and performs in the market) rather
than based on personal conviction or pride—being unwilling to admit
the idea you green-lit is not fit to continue. Where the single-sponsor
model is inherently ad hoc, personal, and political, boards provide a
sponsorship model that is repeatable, rational, and strategic.

Green-Lighting

The first critical process where boards and teams work together is
green-lighting, where new ventures are approved for their very first
round of validation. Approval of a venture comes with an initial release
of resources from the board (typically including time, money, and
people). The best green-lighting practices minimize the initial
investment to each team and maximize the number of ideas that are
approved to be tested.
In green-lighting, it is important to resist the urge to try to pick the
“best” ideas among those submitted. First, you truly have no way of
knowing which ideas will work. That knowledge can be learned only
through validation (so don’t try to guess!). Second, successful ventures
often emerge from ideas that are initially flawed but evolve in response
to testing, feedback, and iterative learning. As Colella explained to me,
“Particularly in large corporations, people tend to think in buckets of
good versus bad ideas. But . . . good ideas are often cloaked inside a
bad idea!” Instead of trying to guess which ideas will work, I
recommend judging ideas based on problem definition, strategic fit, and
team mindset. When hearing first-round pitches for D10X, Citibank’s
growth boards listen for the problem to be solved, whether there is
some unique take or insight into that problem, and whether the team
has the right mix of passion plus an assumption of total ignorance
about the future success of their idea.
Your goal should be to fund as many promising ideas (i.e., strategic,
well-defined, and with the right team) as possible in the first round.
Some companies use an open-door approach to green-lighting
employee ideas. Google’s famous “20 percent time” policy allows
engineers to use 20 percent of their paid time to start working on any
idea that piques their curiosity and see where early exploration might
lead. At Adobe, any employee with an innovation idea can request a
“Kickbox” that includes a $1,000 credit card to fund the very first round
of testing of their idea. The box includes guidance on a five-step
process to validate their idea before choosing whether to pitch it to
senior management. The only requirement is that employees share the
results of what they learn.5
The key to green-lighting many innovation ideas is to build a fast,
cheap, and effective validation process. This must bring in the voice of
the customer to rapidly test whether the venture is focused on a
genuine problem (i.e., stage 1: problem validation). At Citibank,
ventures often begin with a two- or three-day workshop in which
employees explore a strategic P/O statement and have a chance to
develop their own innovation ideas in a rapid, iterative fashion with
actual customers. This approach allows Citibank to test hundreds of
venture ideas within a program like D10X. As you increase the speed
and drive down the cost of your early validation, you can afford to test
and pursue more and more possible ideas for growth.
New ventures should be started with minimal deliberation and the
smallest possible investment. But that initial investment should be
focused entirely on learning—spending not to make a product but to
test the hypotheses of the business model. If those early tests are
promising (most often they are not), later stages can follow. Only after
multiple rounds of learning and validation (or adapting the strategy until
it begins to see validation), should any significant investment be made.
 Iterative Funding

The next critical process for managing innovation is iterative funding,

the process by which boards allocate resources to growth ventures
after they have been green-lit. Iterative funding is designed to be
extremely agile and responsive to market validation. To see how this
works at scale in an enterprise, you need to look at investing under
uncertainty, the role of learning, and how iterative funding differs from
traditional budgeting.

Uncertainty, Net Present Value, and Option Value

Every new venture begins with a degree of uncertainty—in market

demand, technological maturity, the willingness of partners, the
approval of regulators, and so on. To master iterative funding, you must
understand how that uncertainty shapes two kinds of value. Net present
value is the value of a financial return on your investment over time.
Option value is the value that comes from your right to take future
action. Both kinds of value are critical to investing in innovation (see the
box “Option Value Explained”).

Option Value Explained

Rita McGrath defines option value as “the right, but not the
obligation, to make a future decision.”* Investing in strategic options,
or “real options,” refers to any investment that grants the opportunity
to make a later strategic decision once more is known.
The value of options was first observed in financial markets—
where an investor can pay for the right to buy or sell an asset at a
future date at an agreed-on price. These agreements are extremely
valuable but not in the traditional sense. They yield no direct return
to the investor; instead, they create value by providing the option for
a future purchase or sale if conditions are favorable.
A more familiar example can be seen in airplane tickets. Many
airlines sell the same seat on the same flight at two different fares—
often called economy versus economy flexible. The only difference is
 that the second fare includes the option to cancel and get your
money back. The flexible fare costs more, even though the seat,
meal service, and luggage allowance are all identical. What you are
paying extra for is pure option value—the option, should you choose
it, of not boarding the plane. Not surprisingly, the price of that option
(the spread between the fares) is highest when the flight is still
weeks away. The price difference narrows in the final days before
the flight, when the option value (and the likelihood that you will
change your plans) is lowest.
One more example can be seen in the game of poker. Each
player, after receiving their first few cards, is asked to make an initial
bet called an ante. Only if they do so will they receive their remaining
cards and have the chance to bet further and thus the chance to win
the pot of money wagered. That initial ante is pure option value—you
pay for the option to bet again later, when holding your complete
hand. Only in the final round of betting is there any present value,
that is, the chance to make money.
In the VC funds that back Silicon Valley start-ups, the
combination of option and present value is critical. For any VC, an
early-stage investment in a start-up is viewed as mostly option value:
the VC invests for the right to follow with more investment if the start-
up turns out to have a viable business model. Only in later rounds of
investment do VCs start to judge their portfolio companies on
measures of present value—such as the start-up’s revenue,
customer acquisition cost, and operating margin.

* Rita McGrath has shed great insight on the importance of option value to innovation under
high uncertainty. See Rita Gunther McGrath, “Falling Forward: Real Options Reasoning and
Entrepreneurial Failure,” Academy of Management Review, 24, no. 1 (January 1999): 13–
30. Also, Rita McGrath, “A New Approach to Innovation Investment,” Harvard Business
Review, March 25, 2008, https://hbr.org/2008/03/a-new-approach-to-innovation-i.

Some innovations begin with low uncertainty, for example, using a

known solution to solve an existing problem in your current business,
with good benchmarks and clear metrics for success. Low-uncertainty
investments are best judged in terms of their net present value
(typically, ROI). This is the same logic used for investing in day-to-day
business operations. If a project requires $50,000, it should be
approved only if it will yield more value than putting that same cash to
use elsewhere in the firm. Traditional budgeting is based entirely on
present value and on a premise of predictable financial results.
However, many ventures that you will want to invest in will begin
under high uncertainty. They may start with a poorly defined problem,
unclear metrics, and no good benchmarks. They may require
establishing relationships with new partners or serving new customers.
In these cases, there are no predictable results! For innovations under
high uncertainty, financial management based on present value is
practically impossible. Investing under uncertainty does make sense,
but only when it is understood as an investment in option value. When
a growth board provides initial funding to an innovation team, it is not
investing for ROI. It is investing to learn while retaining the right to
invest further if testing shows promise in the market. The goal of your
initial investment should be to validate an important assumption (e.g.,
“is this a genuine problem for a real customer?”) and bring data back to
the board as quickly as possible. Then the board members can decide
whether to exercise the strategic option to pursue the venture further.
That opportunity to take future action is the essence of option value.
To help visualize the relationship of uncertainty to option value and
present value, I have developed a model called the uncertainty curve of
innovation, which is shown in figure 6.1. The horizontal axis shows the
level of uncertainty of a business venture. Within the same company,
different innovations begin with more or less uncertainty. For example,
a new digital business model applying an untested technology in a
volatile market would start far to the left, at the point of maximum
uncertainty. By contrast, an innovation that optimizes the operations of
your current business using a well-known technology would start further
to the right, with much less uncertainty.
Figure 6.1.
The uncertainty curve of innovation

The vertical axis in the figure shows the level of investment that is
appropriate for a venture. The curved line shows the relationship
between investment in a venture and the venture’s uncertainty; as we
follow the curve from left to right, we see that it bends upward. On the
far left, when the uncertainty of a venture is at a maximum, the firm is
investing purely in option value. In that case, the size of the investment
should be very small. On the far right, when a venture’s uncertainty is
at a minimum, the firm is investing entirely in present value. In that
case, the size of the investment can be quite large.

How Learning Shapes Funding

The whole point of validation is that uncertainty is not fixed because

experimentation drives down uncertainty through learning. Figure 6.2
shows that a novel innovation will typically start as a “big idea” with
great uncertainty (think of the brainstorming session where the Amazon
Fire Phone was first conceived). The role of validation is to reduce that
uncertainty with iterative experiments—things like customer interviews,
wire frames shown to customers, and iterative prototypes or MVPs.
Each of these tests, if designed right, will yield new insights that
validate or invalidate crucial aspects of the planned innovation. At the
bottom of Figure 6.2, you can see the arrow of learning pointing to the
right. Only by validated learning can a business move a venture from
high uncertainty to low uncertainty.

Figure 6.2.
Learning and the uncertainty curve of innovation

Look next at the uncertainty curve in figure 6.3, starting from the left
side of the figure. A start-up in its first funding round will ask VC
investors for only a very small initial investment because the business
is so uncertain and any funds given are at maximum risk. Similarly,
corporate teams working on uncertain ventures should be granted only
a very small operating budget, limited time, and limited head count in
their first round. Citibank’s D10X grants as little as $2,000 to spend on
initial validation. If the team invests those resources effectively in
learning, it will reduce the venture’s uncertainty—moving to the right in
figure 6.3. The value of the venture will shift from pure option value to
more present value, and the size of each funding round should follow
the curve up. This is why each round of VC fundraising for a successful
start-up will increase dramatically in size. As the start-up’s business
model is validated, the risk of failure declines, and the size of
investment will grow exponentially.
Figure 6.3.
Investment and the uncertainty curve of innovation

Many corporate managers tell me how they struggle to get budgets

approved for innovation. My advice is always to try asking for less
money. Then use what you get on early experiments to validate if there
is a profitable opportunity. If validation comes back positive, you will
then be in a position to ask for the budget you originally wanted.
The uncertainty curve of innovation captures two final insights about
managing innovation:

1. For high-uncertainty ventures, milestones should be pegged not to

time but to validation—Notice that time does not appear
anywhere in the uncertainty curve of innovation. In the traditional
planning-oriented approach to management, time is dominant.
Time appears on the horizontal axis in every Gantt chart and
project management tool. When managing under uncertainty,
however, a key mindset shift is to let go of arbitrary milestones of
time and instead manage according to milestones of validation.
2. In times of rapid change and uncertainty, speed of learning is your
organization’s greatest competitive advantage—By mastering the
process of experimentation, your firm will be able to validate,
test, and learn much faster. By learning faster than your
 competitors, you can pursue the same strategy with less
uncertainty. Thus, you can invest resources sooner, at less risk,
and with a higher chance of success. In the digital era, those who
learn fastest will always win.

Iterative Funding in Practice

The practice of iterative funding for growth ventures is based on the VC

approach to financing start-ups. Colella explains how this works at
Citibank: “Just like in a VC, [our] employees will test their ideas,
validate the markets, and if they get positive results, they’ll come back
to the growth board and pitch again for additional funds.”6 This iterative
process is quite different from traditional BAU budgeting in large
enterprises, as shown in table 6.3.

Table 6.3.
BAU Funding Versus Iterative Funding
BAU Funding Iterative Venture
Funding
Slow, big start Quick, small start
Long budgeting cycles Short funding cycles
Decision based on executive Decision based on
opinion validation
Incremental growth Exponential growth

Let’s look briefly at each of the key differences between BAU

funding and an iterative funding process:

• Slow, big start versus quick, small start—In traditional corporate

budgeting, a new project will be granted a large initial budget (to
show commitment to the project), but only after a long business
case analysis—striving (misguidedly) to assess the chances of an
uncertain new venture through third-party data and modeling. With
iterative funding, the approach is the opposite. New ventures are
given a tiny initial budget based on a quick assessment that asks
only whether the opportunity is well defined and strategically
relevant to the enterprise.
 • Long budgeting cycles versus short funding cycles—The next
difference is in frequency. In traditional corporate budgeting,
projects and departments are funded annually through a complex
process that takes months. A promising new venture can wind up
waiting over a year to get resources for a four-week test. Iterative
venture funding is made in short cycles of one to three months,
providing a team with thirty to ninety days of resources before it
must return and argue for further support.
• Decision based on executive opinion versus decision based on
validation—The third change is in how funding decisions are
made. In traditional budgeting, decisions are based on the
opinions of senior executives—which can be influenced by
personal conviction (pet projects) or the persuasive talents of team
members. Instead, each round of funding should be decided
entirely on a venture’s validation data. As we saw in chapter 5, the
key metrics will change as validation progresses. So, with each
round of funding, the board must agree with the team on what data
they need to bring to their next review. The Rogers Growth
Navigator can be used to guide the discussion of what has been
tried and learned, and what the team must validate next if the
board is to continue funding.
• Incremental growth versus exponential growth—The rate of
change in funding is quite different between the two funding
processes. In most organizations, the default budget in the next
cycle is an incremental change to the prior budget (e.g., “last year
plus 3 percent”). With iterative funding, if validation is successful
and the venture continues, the size of each investment round
should grow exponentially. Human resources may also increase—
either by adding head count or shifting employees from part-time
to full-time on the project.

Any iterative funding process requires a great deal of flexibility.

Boards must be ready to ramp up investment quickly in the ventures
that prove themselves in the market. If a team’s testing is extremely
positive, it may request to meet with the board earlier for its next
funding review, to accelerate the pace. For a board to be flexible, it will
need a pool of resources that is funded up front (e.g., for a year). The
board can then allocate those resources judiciously across a portfolio of
projects. Within that portfolio, only similar innovations should compete
for funds, for example, only high-risk innovations within a given
business unit, or only innovations outside the core. (We will see
examples of this kind of resource pool later in the chapter.) Fund the
portfolio first, and then the ventures.
At some point, as a venture scales, the type of budget that funds it
may need to shift, along with who makes the funding decision. In
industries with complex physical products, a budget bump typically
occurs when a team moves from testing with illustrative MVPs to
functional MVPs, which cost much more. At Air Liquide, this happens
whenever a team moves from building wire frames to building a
working industrial product with live data. “It does not have to be a lot of
money. Something in the low $100,000s range, significant enough to
make it count,” explains Olivier Delabroy, vice president marketing at
Airgas, an Air Liquide Company. At Air Liquide, the source of funds
changes then from an operating budget overseen by the CDO to a
capital budget disbursed by an innovation board.
Any large successful corporate venture will eventually graduate
from funding by the innovation board. If a thriving venture needs a $100
million infusion after a couple of years to reach its next level, this will
likely go to the company’s executive board to approve. Likewise, senior
leaders may weigh in on any decision to expand from a minority stake
in a start-up to buying it outright.

Smart Shutdowns

Of course, not every review will conclude with a decision to continue

funding. One of the classic problems that bedevil corporate innovation
is that companies learn how to start new projects but not how to stop
them. For innovation to deliver results, firms must be ready to exit
projects that prove unsuccessful or are insufficiently aligned with
strategy. As Stephen Dunbar-Johnson, president, international of The
New York Times Company, told me, “It’s easy to start new things. The
really hard part is shutting them down!”
Shutting down ventures smartly—that is, systematically and
regularly—is a critical job for growth boards. Every time a board meets
for an iterative funding review, the question must be, Do we fund this
venture further or shut it down? Whichever decision is made, it should
be data driven. Table 6.4 shows common test results at each stage of
validation that indicate serious trouble for a venture. Any one of the
results listed in the table is a clear signal to either pivot (fundamentally
change course) or end an innovation project and free up its resources.

Table 6.4.
Signals to Shut Down a Venture at Each Stage of Validation
Va Common Signals to Pivot or Shut Down
lid
ati
on
St
ag
e
1. • No problem—You can’t identify a customer problem
Pr you are solving (i.e., you have a solution in search of a
ob problem).
le
m
val
id
ati
on
• Low priority—Customers recognize the problem, but it
does not make their top five list of priorities.
• Problem solved—Customers are satisfied with
existing alternatives for addressing the problem.
2. • No urgent demand—You receive polite praise for your
So solution (“that would be nice”) but no requests to
luti register, use, or buy.
on
val
id
ati
on
• No killer feature—None of your proposed benefits are
enough to motivate behavior change by the customers.
 • No competitive advantage—Customers don’t see a
reason why they must choose your solution over
existing solutions.
3. • Low usage—Customers don’t use your solution when
Pr offered or discontinue use after the initial trial.
od
uc
t
val
id
ati
on
• Too hard—You have no clear path to deliver a
solution customers will use (e.g., technology is not
ready, regulation won’t permit it, or you lack essential
IP)
• Too easy—Your solution works, but competitors can
easily deliver the same thing as well or better than you.
4. • No value capture—Customers like your offering but
Bu they won’t pay for it, and you can’t find another value
sin stream for your firm.
es
s
val
id
ati
on
• No path to profit—You are capturing value, but costs
are too high and won’t go down enough to break even
as you scale.
• Too small a prize—The maximum upside is not large
enough to merit a continued focus by your
organization.

Overcoming the Barriers to Shutting Down

At legacy firms, the biggest barrier to shutting down innovation projects
is often cultural. There is often an aversion to admitting failure and an
irrational feeling that failure of any kind poses too much risk. As
Citibank’s Colella puts it, “Most people in large companies associate
failure with something that has economic consequences or bad
consequences for clients or somehow impairs the safety and
soundness of the system.” But failure in testing new ideas, if managed
appropriately, is not risky at all—and it is much less risky than not
pursuing innovation. According to Colella, “If we’re testing an idea or a
prototype and a client tells us this doesn’t solve a problem they have,
and they wouldn’t buy it . . . we haven’t lost anything other than the time
and effort putting together the idea.”7
By contrast, there are very real costs to the firm if your teams do not
shut down ventures quickly and smartly. Without this discipline, your
innovation will lack focus, your resources will be spread too thin, and
you will run out of bandwidth for new experiments. You will be stuck
with “zombie projects”—unsuccessful ventures that never shut down
and continue siphoning off resources. As an executive at Axel Springer
media group describes it, “You have millions of little projects going on
forever and limiting your ability to try new things out . . . You need those
people, money, to spend on other projects.”8 At Johnson & Johnson, an
entire new series of innovations was funded by evaluating the existing
portfolio and shutting down projects that no longer matched the
company’s updated strategy.9
Shutting down projects will become easier only if you make it a
routine decision. Innovation boards with a regular calendar of funding
reviews will make a huge difference. In GE’s oil and gas division,
projects were rarely shut down before its board was instituted. As soon
as the board began, it easily shut down 20 percent of existing projects
in its first ninety-day cycle. As the board and teams became focused on
aligning to strategy, this rose to 50 percent of new ventures shutting
down within sixty days.10 At media giant Schibsted, the goal is to “take
something out” whenever you are “putting something new in” the
development pipeline. When a project comes up for review, set a high
bar for yourself by asking, Why shouldn’t we shut this venture down?
The following five practices are essential to achieving smart
shutdowns in any organization:
 1. Plan a pipeline with survival rates—Many corporate innovation
programs find a survival rate as low as 50 percent, or even 30
percent, is typical in their first funding review—after a bright shiny
idea has contact with real customers. In subsequent rounds,
survival rates typically increase. Understanding your survival
rates at different stages of validation allows you to plan a pipeline
for the future. For example, if a growth board is expected to help
launch three or four new ventures to market within a year, it
needs to be sure to plant enough seeds at the start to have high
odds for success. An executive at the Washington Post
described this shift in thinking after the paper was bought by Jeff
Bezos: “Before Jeff we were very cautious . . . We would do the
absolute safest play because we would maybe develop one big
new product a year, and it had to succeed. Now we’ll do a ton of
products and the majority of them won’t succeed, but we’ll figure
out many ways of how not to do things . . . so it’s leaning into
risk.”11 Figure 6.4 illustrates this approach, showing a pipeline
plan for a typical hackathon, with survival rates for each
milestone. A pipeline plan could also show validation milestones:
problem validation, solution validation, product validation, and
business validation.

Figure 6.4.
Sample pipeline for an innovation hackathon program

2. Use your backlog to reassign swiftly—Each innovation board

should maintain its own venture backlog, a ranked list of ideas for
ventures that have been approved but not yet begun. Using this
backlog in your review process will make shutdowns much
 easier. The point of a shutdown is not simply to kill a failing idea
but to free the team and its resources to work on a more
promising idea from the backlog. When you shutter a project,
quickly reassign members to the best next idea. In many cases,
your next step may be to refocus that team on a different solution
to the same problem.
3. Extract value from shutdowns—When you decide to shut down a
project, look to extract as much value as possible, whether
financial value, option value, or strategic learning. In some cases,
a company may be able to sell the venture to investors or
another business. When Walmart’s Vudu streaming video service
was no longer a strong strategic fit, Walmart spun it off and sold it
to media giant Comcast. Sometimes a venture will be shut down
because it is promising but not yet workable at scale. By
shrinking your investment, you may maintain the venture as a
hedge against future options. After the failure of Google Glass as
a consumer product, the company did not pull the plug. Instead,
it shrank the initiative to an enterprise-only device (focused on
applications on factory floors), where Google quietly continued to
develop augmented reality technology.12 Sometimes only a full
shutdown makes sense, and the key value to extract is the
learning gained from experimentation. When Amazon shut its
unsuccessful Amazon Auctions and zShops services, it applied
the lessons it learned for the subsequent launch of Amazon
Marketplace to great success.
4. Share learning widely—Sharing what you learn from shutdowns is
one of the key principles of smart failure, but it is the hardest one
to follow. Most companies prefer to look away from projects that
didn’t work out. As the New York Times Company’s 2014 report
explained, “When we do shut down projects, the decisions are
made quietly and rarely discussed, to protect the reputations of
the people who ran them. As a result, lessons are forgotten, and
the staffers involved become more risk averse.”13 Overcoming
this reluctance was essential at the New York Times: lessons
learned from its failed Times Select initiative paved the way to
the paper’s turnaround of its business model. Sharing learning
from failed ventures is even more important as organizations
become larger and more decentralized. The German affiliate of
 Fédération Internationale de l’Automobile (FIA) ran an innovation
lab where eight of ten projects were killed in a single year. Its
biggest win? Sharing those results with other FIA affiliates
around the world who were struggling with the same challenges
in their own markets.
5. Distinguish people from projects—This is a final critical piece to
building a culture that accepts and learns from failure. A strong
review process will hold teams accountable for their results. But
you should be careful not to associate a failed project with the
merit of the individuals who worked on it. Those same team
members could achieve tremendous success for you in their next
project. Google’s Susan Wojcicki helped launch two innovation
projects within a year of each other: Google Answers and
AdSense. Both faced significant risks. The first was shut down as
a failure but provided lessons that were applied to future Google
products. The second became one of Google’s most profitable
products of all time.14 Be sure to encourage your innovators to
keep working on their next idea!

To manage your venture pipeline, you will want to track the rate of
failure at different stages. But to improve results, you should focus on
measuring the quality of your failures. Six months after every shutdown,
conduct a review applying the four-part test for smart failure: Did you
fail as early and cheaply as possible? Did you learn from the failure?
Did you apply the learning to strategy? Did you share the learning with
others? The best-run innovation efforts track and evaluate failures
against such criteria, and they recognize (and even reward) the best
failures to show everyone how failure can be done well.
If your shutdowns are working, you will begin to see volunteering.
When teams are truly focused on learning through validation, they will
often suggest their own shutdown to the board, reporting, “Here’s what
we have learned and why we recommend shutting down now.” This is
what happened at GE Oil and Gas once the board process was
established. Similarly, at Citibank, employees report to their D10X
board, “It’s with a heavy heart but a strong conscience that I
recommend that you kill my project because here’s what we learned
when we went outside the building and worked with our clients . . .
They don’t really need this.”15 Those same employees will soon return
to your board with another venture idea. It could be your next big
breakthrough.

Three Paths to Growth

Many leaders feel comfortable focusing only on innovations that have

low uncertainty and are closely aligned with their core business. But
such a narrow focus will close off many of the biggest opportunities for
growth in the digital era. To succeed in DX, any business must be able
to manage innovations with varying degrees of uncertainty and at
varying distances from the core. This means overcoming the
challenges of both uncertainty and proximity, which we saw in chapter
2. Together, these two challenges point toward three different paths to
growth—each with its own rich opportunities and its own management
challenges (see figure 6.5).

Figure 6.5.
Uncertainty, proximity, and the three paths to growth

(Note: There is no fourth quadrant for “far from the core” plus “low uncertainty” because any
innovation far from your core will involve great uncertainty for your organization to execute.)

Three Paths Defined

Let’s take a close look at each of these three paths to growth to ensure
that we can adapt our governance to support ventures to scale on each
path. A summary is provided in figure 6.6.

Figure 6.6.
The three paths to growth

PATH 1 VENTURES
Path 1 (P1) ventures are innovations within your core—that is, they
improve or solve a problem for an existing business unit or division. In
addition, P1 ventures have low enough uncertainty that they can be
effectively managed within existing business units and functions (e.g.,
marketing, HR, finance).
P1 innovations tend to address a known problem, have an easily
agreed-upon metric for judging performance, and rely on established
technical solutions. They are clearly doable with the skills and tools of
your own organization or your partners. Still, these straightforward
innovations can provide a lot of tangible value to your business, even if
they are just fixing known problems or helping you catch up with peer
competitors.
While P1 ventures do not deliver the kind of exponential or
disruptive innovations that draw admiration in Silicon Valley, they are an
essential part of the healthy growth of any mature business. And, yes,
that includes digital titans like Google, Amazon, and Alibaba.

PATH 2 VENTURES
Path 2 (P2) ventures are also innovations to your core business, but
they involve too much uncertainty to be managed effectively by the
business units alone. P2 innovations may involve changes to your
customer experience, your value proposition, or the delivery model of
your current business. It is often unclear exactly what you should build,
whether customers will adopt it, how it will generate a financial return,
or whether your organization can even deliver it.
P2 ventures are critical to the continued growth of any established
business in a rapidly changing environment. Every P2 venture
addresses a problem or opportunity that is directly related to the core
business. But each venture requires relentless validation, prototyping,
and discovery before it yields a solution that customers will adopt and
that your business can deliver profitably.

PATH 3 VENTURES
Path 3 (P3) ventures are innovation opportunities that do not fit within
the current core business of your firm. They typically serve new
customers, use a new revenue model, or carry a different cost structure
than your existing business.
P3 ventures may compete with the core, directly threatening it with
replacement or cannibalization. Or they may pose opportunities in a
whole new industry, serving different customers. P3 ventures pose
great uncertainty and difficulty in management precisely because they
do not fit within your normal operations. Yet P3 innovation cannot be
ignored. Every truly successful business in the digital era has pursued
P3 ventures by looking beyond narrow definitions of its products,
customer base, or industry.
Recall the case of Amazon Web Services (AWS), a classic P3
innovation. When AWS began, Amazon was a pure consumer retail
business. The new B2B cloud-computing service was a radically
different business model, with a completely different type of customer,
revenue model, and sales process. In time, AWS grew to become the
biggest source of profits for the entire company.
Table 6.5 provides examples of ventures from each of the three
paths to growth for companies in different industries.
Table 6.5.
Examples of P1, P2, and P3 Ventures for Different Industries
Indust P1 P2 P3 Ventures:
ry Ventures Ventures: Outside the Core
: In the In the Business Plus
Core Core High Uncertainty
Busines Business
s Plus Plus High
Low Uncertaint
Uncertai y
nty
News Use Personaliz Offer stand-alone
paper truck e user non-news
route notification subscriptions
optimizat s and (e.g., crossword
ion recommen or cooking apps).
software dations
to cut through
delivery the mobile
costs of app.
print
newspap
ers.
Teach Offer weekly
editors to news series
design licensed to radio
articles or streaming
with data video.
and
interactive
elements
to engage
readers in
complex
topics.
Offer paid online
courses or live
events.
Retail Use new Create an Create a mobile-
bank data to omnichann only app that
predict el targets a new
custome customer generation of
r churn experience customers with a
and that links value proposition
optimize branches, of managing and
promotio ATMs, sharing your data,
ns to website, goals, and
maintain and mobile money.
custome app.
r loyalty.
Insura Use new Create a Offer direct-to-
nce sources mobile app policyholder
comp of data for claims insurance that is
any in filing and sold without
underwri tracking. independent
ting agents.
models
to
predict
risk.
Establish
online
communiti
es for
small-
business
owners.
Fashi Shift Grow e- Launch a new
on paid commerce brand sold
brand advertisi distribution exclusively direct-
ng mix with to-consumer via
to existing an app.
include and new
more retail
digital partners.
 platform
s.
Add 3D Partner Launch a
tools for with Apple subscription
product or Android giftbox service.
display OS to
on design and
websites brand a
. luxury
smartwatc
h.
Auto Use AI Incorporat Create an urban
manuf in e new ride-sharing
acture factories predictive network for cars,
r to detect safety scooters, and
manufac measures bicycles.
turing for drivers
flaws (digital eye
faster tracking,
and etc.).
more
cheaply.
Connect
in-car data
to the
owner’s
app to
track their
carbon
footprint.
Physi Use Offer Build an online
cal predictiv preorderin marketplace that
retaile e g online brings together
r analytics for pickup product brands,
for in-store. curators, and
selecting service providers.
new
store
 locations
.
Use Create a
robots mobile app
for experience
scanning for loyalty
store rewards
shelves and in-
to record store
inventor payment.
y.
Telco Use AI Provide Offer a home IoT
for cybersecur and wearable
market ity tools for device service to
segment small- monitor
ation to business customers’ elderly
target customers parents at home.
custome already
rs with using
the right broadband
offer at and phone
the right packages.
time.
Provide a mobile
payment platform
for consumers
whether or not
they use our
phone service.

Challenges of Each Path

As we have seen, all three paths hold the potential for growth for any
business. At the same time, each path faces unique challenges for
effective management and governance.
 P1 CHALLENGES
P1 innovation is the easiest path in some ways because it faces neither
the challenge of uncertainty nor proximity. But there are two common
traps in managing P1. The first trap is to pursue only P1—innovation
aligned to your core business, and with low uncertainty. While this
approach may feel comfortable and low-risk, it will close off many of the
biggest opportunities for growth for your firm.
The other mistake is the opposite: to neglect P1 innovation and
focus exclusively on big-idea innovations. A business must maintain a
steady stream of P1 innovations in its digital pipeline—incremental
innovations that are guided by strategy and that yield returns to the
bottom line. Beware of anyone who wants to take innovation out of the
hands of the business units. They may recite a skunkworks theory that
mature businesses are incapable of innovation and that it must be left
to separate teams of fast-moving iconoclasts. This is flatly wrong. As
the director of an innovation lab at a global financial services company
told me, “We don’t own innovation. We have innovation assets we bring
to the rest of the organization. But if we want to be an innovative
company, we can’t pretend that there’s a place called Labs that owns
innovation.”

P2 CHALLENGES
P2 is inherently more difficult because it involves greater uncertainty.
Many companies try to pursue P2 ventures within their core business
units, but these units frequently lack the needed skills in iterative
experimentation. As a result, they will make too many assumptions,
reach for an “obvious” solution to every problem, and then rush to build
it.
Other companies try to take P2 innovation out of the hands of the
core so it can be managed by innovation experts. In these companies,
the core business is encouraged to deliver a list of projects for a digital
innovation team that develops solutions entirely on their own, before
handing them off to the business. This “build it and throw it over the
wall” approach leads almost inevitably to disappointment. Even if the
outside team discovers a great solution, it will struggle to be accepted
and implemented by the core business it was meant to serve. The
result is a kind of organ tissue rejection that ends in failure.
Other governance challenges for P2 innovation relate to funding.
The core business almost always lacks a process for iterative
budgeting, which is essential for innovating under uncertainty. In
addition, if the core is asked to pay for P2 ventures, it will underinvest in
them—precisely because they are risky and unlikely to pay off in the
near term (when the core is held accountable for quarterly results). But
if the core is not asked to pay anything toward a P2 venture, it will have
no “skin in the game” for an innovation that is meant to work in its own
business unit. That sets up the venture for weak sponsorship and lack
of adoption later.

P3 CHALLENGES
The first challenge facing P3 ventures is that they do not fit anywhere
within the existing organizational structure. Without a logical home in
any existing business unit, they lack a power center. As a result, these
innovations will struggle to attract sponsorship and support. If the
organization is focused exclusively on current customers and the
metrics of its current business, a P3 opportunity that addresses a
different market will be ignored or neglected entirely.
If the central company leadership does choose to sponsor a P3
venture, it will often be derided as a distraction from the real work of the
company. Investors famously chided Jeff Bezos to give up his AWS
project and get back to focusing on his retail business. A 2006 cover of
BusinessWeek declared, “Bezos wants to run your business with his
Web technology. Wall Street wishes he would just mind the store.”16
In addition, P3 ventures often breed resentment or backlash in the
organization. Those working hard in the core business complain about
the attention given to these new, unproven ideas: “Hey! We’re the ones
making the money that funds you with your experiments!” In other
cases, a P3 venture faces active resistance if it is perceived as
cannibalizing sales for the core business.

Governance for Each Path

Despite the challenges discussed above, all three paths lead to
tremendous growth for the companies that manage them right. But
succeeding in each path means adapting your governance model for
each path. Table 6.6 provides a summary of how to manage each path
to meet its particular challenges. Let’s look at the governance of each
path in more detail.

Table 6.6.
Governance for the Three Paths to Growth
Gove P1: In P2: In the P3: Outside the
rnanc the Core Core Business
e Core Business Plus High
Busine Plus High Uncertainty
ss Plus Uncertainty
Low
Uncert
ainty
Sum Inside Partner with Outside the core
mary the the core
core
Starti Inside Separate Separate unit,
ng the unit, in close with loose ties to
point core partnership the company
busine with the core
ss
Team Standa Multifunction Multifunctional
s rd al innovation innovation teams
functio teams
nal (or
matrix)
teams
Work Standa Iterative Iterative
rd experimentati experimentation
plannin on
g
Metri Standa Staged Staged validation
cs rd validation metrics
busine metrics
 ss
metrics
Fundi Standa Iterative Iterative funding
ng rd funding approved by a
budgeti approved by growth board
ng a growth
proces board
s
Spon Funde Funded Funded entirely
sorsh d partly by the by a separate
ip entirely core to start, budget
by the later funded
core entirely by
the core
Traje Starts Handed off to Starts outside the
ctory in the the core after core, eventually
core, key merges with or
stays validation becomes a new
in the milestones business unit
core

P1: INSIDE THE CORE

A P1 innovation should be managed inside the core using your
standard functional (or matrix) teams—for example, it is run by the IT
department, the marketing department, or a local operations team in a
geographic business unit. P1 innovations can be managed with the
standard metrics and planning that are well established in the core
business. Because their uncertainty is low, P1 innovations can be
funded by traditional budgeting based on net present value. All funds
for P1 ventures should come from the core business, out of normal
operating budgets.

P2: PARTNER WITH THE CORE

A P2 innovation should be managed within a dedicated innovation unit
working in close partnership with the core business unit that the
innovation is meant to serve. That starts with defining the goals of the
venture with sponsors from the core who will ultimately own the project.
“First, you have to align everyone,” says Olivier Delabroy. “You have to
put the business in the driver’s seat and have a committed plan . . . If
you want to scale your ideas, you need, from day one, to embark with
the business.”
After that initial alignment, the business must continue to partner
with the innovation team throughout testing and validation. Work is
done by a multifunctional innovation team, including members from the
core alongside experts trained in iterative experimentation. Funds are
disbursed by a growth board, following iterative funding practices, to
account for the high uncertainty of the venture. To ensure buy-in, any
P2 innovation must be financially sponsored by the division that will
eventually own it. The funding may be subsidized at the early stages
(e.g., from a central innovation budget), but the sponsoring division
must have “skin in the game” from the start.
After uncertainty has been reduced (typically based on milestones
of problem validation and solution validation), a decision is made
whether the venture should continue. If the decision is yes, the venture
will be handed off entirely to the business division that sponsored it to
scale and launch it in the market. At that point, the business unit takes
over all funding for the venture and will reap all profits from it.

P3: OUTSIDE THE CORE

A P3 innovation should be managed outside the core in a separate
unit, much like an independent start-up that the parent company
provides with seed capital (in some cases, that may actually be what
you do). A P3 venture should be set up under the sponsorship of
corporate leadership but with maximum independence from the core
business. This setup allows it to attract entrepreneurial talent and gives
it the chance to compete directly with your own business, if that is truly
necessary. At the same time, a P3 venture should never be 100
percent separate from the organization. Where needed, there should
be a mandate for resource sharing and collaboration—including
support such as access to data, branding, supply chain partners, or a
talent pipeline.
 P3 innovations must be managed for high uncertainty, just like P2
innovations. This means a small multifunctional team that applies
iterative experimentation to rapidly validate and adapt to market
learning. And it means iterative funding by a dedicated growth board.
P3 projects should be funded entirely by the parent company and not
out of the operating budgets of any existing business units. Eventually,
if a P3 venture is successful, a leadership decision must be made
either to merge it with an existing business unit or to establish it as a
new permanent division of the company.

PLAN WHERE TO LAND

Whatever its path, every new venture that succeeds eventually ends up
in the core: P1 starts in the core; P2 hands off to the core; P3 joins or
becomes a new unit in the core. Planning early for this eventual home
is critical. As Mario Pieper at BSH Home Appliances explained to me,
“You need to know not just the business model, but also have an idea
for a landing spot—somebody in the organization whom you have to
work with.”

THE NEED FOR ALL THREE PATHS

Managing all three paths to growth is essential for long-term growth in
the digital era. Many companies mistakenly try to manage innovation
on only two paths (see the box “Why Dual Transformation Falls Short”).
Only with separate governance models for all three paths, however,
can your business achieve real and continuous transformation.

Why Dual Transformation Falls Short

I have seen many companies attempt a dual-path approach to

transformation, which is given various names, including
champion/challenger, battleship/speedboats, or (in the banking
sector) run the bank/change the bank. In each case, the first term
refers to the core business and the second to an independent unit
focused beyond it. In the boat metaphor, the battleship is the legacy
business, which can only be turned slowly toward a digital direction.
The speedboats are small teams that move swiftly in pursuit of
digital opportunities because they are untethered to the core.
James G. March provided early thinking on the tension between
exploiting the core and exploring beyond it.* His ideas were built
upon by Charles A. O’Reilly and Michael L. Tushman’s writing on the
ambidextrous organization. † Although their work identifies an
important problem—what I describe as the challenge of proximity—
the two-part solution they offer is still incomplete.
In practice, I have seen the dual-path approach to transformation
lead to frustration. In many companies, the core languishes, with its
organization and culture unchanged because it is given an excuse to
move slower. Meanwhile, the independent teams generate many
ideas but struggle to scale them to have a real impact. These teams
drift farther from the main business over time and lose leadership
support.
There is a reason for this trouble. Even when executed well, a
dual approach is simply a model for managing P1 (battleship) and
P3 (speedboat) innovation. But for most businesses, P2 is where the
most valuable growth happens! That’s right: most digital growth
comes from P2—new ventures that are related to your core business
and customers but involve a significant amount of uncertainty. In a
dual-path organization, these ventures are managed as if they were
either P1 or P3. Either case is a recipe for trouble.
When you treat a P2 innovation as if it were P1, you send it to the
core to die. You treat an uncertain opportunity as if it were a well-
defined, everyday business project. That means running it within an
existing division and applying traditional project management,
business case analysis, and budgeting. In the best case, these P2
innovations are underfunded (because they can’t prove an ironclad
case for their profitability) and are left to wither on the vine. In the
worst case, they are approved and overfunded, and become
expensive and embarrassing digital failures when initial planning
turns out to be riddled with assumptions.
When you manage a P2 innovation as if it were P3, you spin it
out on its own to die. You treat a new innovation that is inextricably
linked to your current business as if it were a stand-alone start-up.
That means handing it off to an independent team that operates
 without the core business. These teams often generate promising
business ideas, but their ideas inevitably fail when they have to
integrate with the core business—whether it is due to resentment
(“not invented here” syndrome), a failure to understand the needs of
the business (“our customer would love that, but it will bankrupt our
P&L”), or some other source of misalignment (“this doesn’t match
our product roadmap”). The result is innovations that never scale to
have a meaningful impact on the goals of the firm.

* James G. March, “Exploration and Exploitation in Organizational Learning,” Organization

Science 2, no. 1, Special Issue: Papers in Honor of (and by) James G. March (1991): 71–
87.
† Charles A. O’Reilly and Michael Tushman. Lead and Disrupt: How to Solve the
Innovator’s Dilemma (Stanford, CA: Stanford Business Books, 2021).

Governance, Not Ideation

When I introduce the three paths to executives of large enterprises, a

certain misplaced enthusiasm often arises along with a particular
question: Should we use the three paths as a springboard for
generating new ideas? That is, should managers start out by
envisioning P1 ventures, P2 ventures, and P3 ventures? The answer is
no. The three paths to growth is a model for innovation governance, not
for generating innovation ideas.
As we saw in chapter 4, idea generation is a strategic process. For
new ventures to have the greatest chance of delivering real value, we
must begin by looking at customers and the business and identifying
the most important problems and opportunities for each. Use the
strategy tools in chapter 4 to guide your search for innovation ideas.
Use the three paths once you have a new venture in mind to pursue.
Only then are you ready to identify the governance model that will give
that new venture its best shot at success.

Innovation Structures
Too many organizations start innovating without the resources and
governance in place to carry ideas through to scale. They approve new
projects or organize hackathons with no support structure to build on
the ideas they generate. For innovation to drive growth at scale,
businesses must establish innovation structures that bring teams and
boards together and provide them with the resources and management
they need to succeed.
I define an innovation structure as a pool of funding and talent for
innovation, with defined sponsorship and governance rules. This pool
of dedicated resources, funded in advance, allows a growth board to
fund a portfolio of either P2 or P3 ventures iteratively—shutting down
some ventures and accelerating investment in others—without needing
to pause work constantly to seek funding or staff for the teams that
board oversees.
An innovation structure is where P2 and P3 innovation come to life
inside an organization. (P1 innovations are managed within the core
itself, so they do not need a separate structure.) Innovation structures
come in many forms and with even more names. Some of the most
common types are the following:

• Digital accelerators—Also called centers of excellence or digital

factories, these structures are set up to accelerate the
development of innovations within the company’s core business.
• Innovation labs—Also called innovation studios, these structures
focus on launching new ventures outside the company’s current
core business. Teams are granted autonomy to move as quickly
and as independently as possible.
• Innovation challenges—Often called hackathons, these contests
solicit ideas from a wide pool of participants who may have access
to a shared data set or tool base. A challenge may be open to
employees, business partners, university students, or the public.
After an initial round with many participants, a few winning teams
receive funding for iterative testing and validation.
• Start-up incubators—Also called start-up accelerators, these focus
on partnering with and growing external start-ups that are relevant
to the company’s strategy. Direct investment may happen, but the
focus is on collaboration.
 • Corporate venture capital—Corporate venture funds invest in
external start-ups, managing a portfolio of investments over time
and taking an equity stake, much like a traditional VC fund.
• Mergers and acquisitions (M&A) teams—M&A is a major part of
the digital strategy of many firms, where acquisitions bring in new
digital business models as well as digital talent.

As these examples show, an innovation structure may be internal

(utilizing only company employees) or external (partnering, acquiring,
or investing in outside start-ups), or the structure may bring together
internal efforts with outside support and partners.

Governance and Design Matter

Beware of setting up any innovation structure without giving careful

attention to its governance and design. Providing innovation resources
without governance will lead teams to run amok. I have heard
companies lament over setting up an innovation lab that is funded by
the central headquarters but does not have clear operating rules.
These efforts typically serve solely as a corporate branding exercise
(“Look, we have a digital innovation lab you can visit!”). No ventures
emerge that deliver real value to the business. This leads quickly to
resentment from the core business, and the lab is ultimately shut down.
Equally bad is an innovation structure with a clear purpose but
whose design does not match its mission. One disastrous example of
this was GE Digital, a unit set up by GE in 2015 to pioneer a completely
new business model—a software operating system called Predix to
power the machines of the industrial world. This was a classic P3
innovation mandate: a new unproven business model outside anything
GE had done before. It should have called for a lean and independent
structure: a small, dedicated team that experimented, validated, and
found product-market fit before seeking to grow. Instead, it was
assigned to a legacy division that provided IT services to GE’s core
business units (aviation, power, transportation, etc.). The legacy
division was given the new mandate for business model reinvention
(“build Predix!”), but it was still led by its prior CEO, and it still retained
its old responsibilities to internal GE customers. As a result, GE Digital
launched with a huge staff (1,700 employees within its first year) and
massive overhead costs. With a quarterly P&L and revenue demands
to meet, it was forced to focus almost entirely on serving internal
customers rather than pursuing new markets. The result was
predictable: GE Digital utterly failed at its reason for being—to build
Predix as a new business model to drive GE’s future growth.17
Before launching any innovation structure, it is critical to align on a
few key elements:

• Mandate—Any stand-alone innovation structure, beyond the

business units themselves, must have a clear reason for being.
This rationale should include the benefits that the structure is
meant to provide to the rest of the organization, and the strategic
problems and opportunities it is meant to pursue.
• P2 or P3—It is critical to decide if the structure is focused on P2 or
P3. Is the structure meant to support innovation within the core
business or beyond it? If you don’t know this answer from the start,
it will be impossible to define clear rules on who sponsors projects,
who holds decision rights, and what (if any) are the roles and
obligations of business units in the core.
• Funding source—Who will fund the structure? If it is focused on
P3, its funding must come from the central headquarters. But if it is
focused on P2, its funding should be split: you will need to decide
how much is paid by relevant business units and how much by
headquarters. The innovation structure should be funded up front
on an annual basis. Then its growth boards can allocate resources
flexibly across different venture teams.
• Goals and metrics—It is critical to decide in advance what success
will look like so everyone has the same expectations. If you don’t
agree on goals at the outset, your structure will be hobbled by
conflicting expectations (e.g., an innovation lab thinks it is
investing in ten-year bets, but it is funded by a committee that
expects new products to launch in year one). Choose metrics to
match your goals. A P2 structure could be measured on the impact
of its innovations after they are handed off to the business and
scaled up. A P3 structure might measure results like a venture
 capital fund—tracking its financial return on a portfolio of bets that
pay off over several years.

Most importantly, you should start small and expect each innovation
structure to evolve as you learn what works best and as the needs of
your business change. Ford’s Smart Mobility unit (a P3 structure
pursuing new business models beyond car ownership) started with just
twelve people. BASF’s Onono lab intentionally began with only two full-
time employees because its P2 mission was to spark innovation by
bringing together business unit leaders with start-ups and customers.
The last thing Antonio Lacerda wanted was for Onono to head off
pursuing innovation ideas on its own. He explained, “With only the two
of us full-time, we knew that nothing could happen at Onono without
the involvement and support of our business units.”

Spotlight on Two Innovation Structures

The following two cases provide examples of smart governance and

design in innovation structures from companies in two very different
industries: industrial manufacturing and financial services.

P2 Spotlight: United Technology Corporation’s Digital Accelerator

United Technologies Corporation (UTC) launched an innovation

structure in Brooklyn, New York, called the Digital Accelerator to
enable digital innovation in its four core business units. Vince
Campisi, the company’s CIO and CDO, knew the firm was adept at
using technology to increase productivity within its operations. But
he saw it struggle to harness the data in its industrial products to add
value for customers and transform the user experience. Campisi
brought in Steve Serra, formerly an outside innovation consultant, to
establish UTC’s Digital Accelerator to support this P2 innovation in
the core business.
Given its mission, Serra knew that no venture should start in the
Digital Accelerator without explicit sponsorship approval from the
core business. As a P2 innovation structure, the Digital Accelerator
uses a split funding model: 20 percent of its budget comes from
corporate seed funding, and the rest is from the four business units
themselves, with each unit funding a particular “studio” supporting its
ventures.
Venture teams are comprised of innovation experts working full-
time at the Digital Accelerator, combined with representatives from
the core business. Campisi explains: “When a business unit
president shows up with their staff at our Digital Accelerator, our first
question to them is, ‘What are your priorities, your business
objectives and outcomes?’ ” Those objectives are quickly translated
into a problem statement, which kicks off a process of rapid
experimentation. If validation shows there is a real business
opportunity, the team moves ahead quickly with MVPs and
prototyping.
Every venture begins with a plan for its handoff back to the
business unit. Early stages of validation are done at the Digital
Accelerator, but when product-market fit is found, the innovation is
transferred to the business. Serra pointed to the example of a digital
product designed to monitor the health of an aircraft engine for the
company’s Pratt & Whitney business unit. The Digital Accelerator
worked with Pratt & Whitney to develop the first functional MVP and
put it into real-world use with customers in a limited test. “We
released the software to our first customers and saw how they used
it. We trialed it with scores of users, not thousands.” The team
waited until they had three business customers and until they “knew
what was working, what was not, and what additional things
customers wanted” so they could define a product roadmap. At that
point, Serra explained, they were ready to move ownership to the
business, which would continue to refine the product, scale it, and
integrate it into their operations.
For an accelerator like UTC’s, the right talent is critical, as is the
balance of newcomers and people with long experience in the parent
company. As Serra told me, “More outsiders mean you will move
faster. More insiders mean you can integrate faster.” UTC placed the
Digital Accelerator in Brooklyn to tap a prime talent market while
staying on the flight path between the company’s two headquarters
in Connecticut and North Carolina.
The vision for the Digital Accelerator was dynamic from the start.
In addition to seeding innovations, it was meant to change the
culture of innovation at the core. Campisi stressed, “We can’t do
enough highlighting of the pioneers who were willing to make the
leap of faith with us—to show how digital is a vehicle for solving
important needs for customers.” Since its launch, the Digital
Accelerator has continued to evolve, adapting to the changing needs
of UTC, which was later rebranded as Raytheon Technologies after
a merger. Digital Accelerator employees moved into the business
units they had previously supported, further driving innovation in the
core. Meanwhile, the focus of the Digital Accelerator shifted from
customer-experience design to data-driven value creation.

P3 Spotlight: Mastercard’s Start Path

One of Mastercard’s innovation structures is Start Path, which was

designed to seek growth opportunities beyond Mastercard’s core
with the help of fintech start-ups. Partnering with start-ups is a
popular tactic for legacy companies seeking to drive digital
innovation, but without a clear focus, it rarely yields lasting results.
Mastercard developed Start Path with a clear focus: to learn about
the fast-moving fintech landscape and build partnerships early—both
for its own core business and for the global network of banks and
merchants that are Mastercard’s customers.
Start Path is a free, one-year accelerator program for fintech
start-ups around the world. Mastercard’s learning about emerging
trends and opportunities is fed each year by seeing the pitches of
2,000 applicants who seek to join Start Path. That learning continues
with the forty start-ups who are accepted into the program, as
Mastercard helps nurture them through their next stages of market
testing and validation.
Mastercard is careful to define the growth stage of start-ups it will
consider for Start Path. Because its goal is to learn from businesses
that already have product-market fit, Mastercard makes a conscious
decision not to accept applications from start-ups at the idea-on-a-
napkin stage. Instead, it focuses on finding what it calls real actual
businesses building interesting tech (RABBITs).
These start-ups already have many options for raising capital.
So, rather than investing in every Start Path member, Mastercard
focuses on the commercial benefits it can connect them to, which
 means matching each start-up with the banks and merchants in
Mastercard’s network. By introducing start-ups to this top-tier client
list, Mastercard has become the preferred partner for some of the
world’s best fintech start-ups. At the same time, Mastercard offers
access to its Start Path ecosystem as a value-added service to its
own business customers around the world.
Two case examples of well-designed innovation structures are
described in the box “Spotlight on Two Innovation Structures.”

One Company, Multiple Structures

It is important to recognize that there is no single perfect innovation

structure. Too often, I have seen executives convinced there is a
magical silver bullet for innovation in their business (whether a digital
factory or a moon-shot lab like Google’s). Your goal should not be to
pick one structure but to develop multiple structures to pursue different
growth opportunities. The most innovative companies that I have seen
all use multiple innovation structures to support P2 and P3 while
continuing to pursue P1 innovation in their core. Table 6.7 lists
examples of companies that use a combination of innovation structures
to drive growth and transformation across their business.

Table 6.7.
Firms with Different Structures for Managing P2 and P3 Innovations
Com Structures for P2 Structures for P3
pany
Citib D10X—This internal Studio—This lab
ank accelerator solicits incubates new digital
venture ideas from ventures outside the
employees in current business that
Citibank’s consumer focus on Citibank’s
and institutional strategic P/Os.
businesses, pairs Ventures include Worthi
them with mentors, and City Builder.
and helps them build
digital innovations
 within the core.
Projects have
included Proxymity
and CitiConnect for
Blockchain.
CUPID—The Venture Investing—
hackathon program This fund invests in
engages students external start-ups that
from leading have product-market fit,
universities around offer strategic value to
the world to support Citibank, and can
ongoing innovation benefit from Citibank’s
projects across global commercial
Citibank. relationships. Exits
include Honey and
Docusign.
Air Digital Fabs for i-Lab—This innovation
Liqu business units— lab explores digital
ide Four teams focus on business models in
accelerating DX in Air areas new to the
Liquide’s four business, including
business lines. energy stewardship,
Ventures include a home health care, and
predictive air pollution reduction.
maintenance data
project within the
Large Industry line.
Digital Fabs for M&A—Acquisitions
business functions have been used to
—Two teams focus on expand into new
digitizing the global product areas (such as
finance function and biogas) and to acquire
the HR function. digital capabilities (such
as Alizent, providing
IoT solutions across the
firm).
ALIAD—This
corporate VC fund
 invests in start-ups
that the core business
partners with as a
customer. The
portfolio focuses on
sustainable energy,
AI, and IoT—with
start-ups like Plug
Power and
Avenisense.
BSH Sprinter Model— Company Builder
Hom BSH Digital partners Model—This model is
e closely with the core used to launch new
Appl business to test and ventures whose
ianc rapidly scale digital business model is
es innovations that distinct from BSH’s
leverage assets of the core business.
core. Examples include We
Wash, a digital service
for community laundry,
with IoT-connected
machines for urban
dwellers.
BSH Start-Up Strategic Venturing—
Kitchen—BSH BSH surveys the
partners with start-ups market when a new
(Series A and B) digital venture is
looking to work with proposed to see if a
the home appliances solution already exists.
industry. After an early If a match is found,
pilot-test, BSH BSH explores options
leverages the scale of to acquire, invest, or
its core business to partner. Examples
accelerate the growth include Chefling (AI for
of these partners. pantry management)
and Kitchen Stories
(recipe app).
BSH Future Home
 Accelerator—This
start-up accelerator
runs in partnership with
Techstars. Its goal is to
grow the ecosystem of
start-ups focused on
future home living
(cooking, AI assistants,
sleep management,
etc.).

Tool: The Corporate Innovation Stack

We are now ready to introduce our next strategic tool, the Corporate
Innovation Stack. The purpose of this tool is to manage innovation at
scale in any established business. To do this, we need to define
governance rules for innovation at three different layers (see figure
6.7):

• Innovation structures (e.g., digital accelerators, innovation labs, or

corporate VC funds), which provide resources and oversight for P2
and P3 innovation.
• Innovation boards (or growth boards), which green-light, advise,
and provide iterative funding to a portfolio of different growth
ventures.
• Innovation teams (small and highly independent), which do the
work of ideating, validating, and growing a single venture through
rapid experimentation.
Figure 6.7.
Three layers of the Corporate Innovation Stack

The tool is comprised of three different “charters” for the three

different levels (a charter for each structure, each board, and each
team). Each charter is meant to define governance and rules for that
level of the innovation stack before its work can begin (see figure 6.8).
Figure 6.8.
The Corporate Innovation Stack

1. Structure’s Charter

Step 1 of the Corporate Innovation Stack focuses on crafting a charter

for each innovation structure. The charter spells out its mission and the
processes that will guide its operations. This charter must be defined
for each P2 or P3 innovation structure before any resources are given
to it.
 STRUCTURE: MISSION

• Mandate—Why is this innovation structure needed? How will it

benefit the organization, and how should it influence the
organization over time? What strategic P/Os does it hope to
address?
• P2 versus P3—Is this structure focused on P2 (innovation in the
core) or P3 (beyond the core)? If it is focused on P2, which of your
core business units or functions will it support? How will it partner
with them?
• Goals and metrics—How do you define success for this innovation
structure? What is its time frame? What metrics will you use to
measure its long-term impact? What metrics will measure its near-
term progress?
• Resources—For a P2 structure, what share of the initial funding
will come from the business units and how will that be budgeted?
For a P3 structure, how much funding will it receive from the
central organization and how will it be budgeted? What other
resources will the organization provide (access to data, to
customers, to channel partners, etc.)?
• Leadership—Who oversees the innovation structure? To whom do
they report?

STRUCTURE: PROCESS

• Talent pool—Will this structure’s innovation be pursued by internal

teams, external teams, or a collaboration between both? What mix
of skills will you need? What is the minimum head count (for
venture teams plus administration)?
• Recruiting—What proportion of its staff should be new hires versus
internal recruits? Will internal recruits be permanent or on short
rotations to work in the structure? Will you procure external talent
through partners (consultants, universities, etc.) or M&A?
• Independence—What sandboxes will the structure be given for
compliance (security, risk management, data access, regulation,
etc.)? What independence will it have from normal corporate
 functions such as IT (going outside the existing tech stack) and HR
(different hiring and compensation)?
• Learning—What mechanisms will be used to capture learning
(both successes and failures)? How will that learning be shared
across the organization? How will shutdowns be evaluated and
smart failures celebrated?

2. Board’s Charter

Step 2 of the Corporate Innovation Stack focuses on crafting a charter

for each growth board. This charter defines the members the board will
need and the process of their work. The charter must be created by the
innovation structure before a growth board’s members are recruited or
resources are given to it. Depending on a structure’s size, it may use
one or more boards.

BOARD: MEMBERS

• Composition—How many members will serve on each board?

What mix of seniority will you seek? Where will members come
from inside the organization? (Any P2 board must represent the
business unit its ventures will support. P3 boards should draw from
across the enterprise for a diverse perspective.) Will you recruit
any board members from outside the company?
• Abilities—How will you ensure that board members understand the
principles of iterative validation and funding? What other skills will
you seek in your members (subject-matter expertise, VC or
entrepreneurial experience, internal influence in the business,
etc.)?
• Responsibilities—How frequently will the board meet? A regular
cadence is important, and meeting every four to twelve weeks is
preferable. How will you define board members’ key
responsibilities? What does the board need to report to the
structure’s leaders (e.g., after each funding decision)?
 BOARD: PROCESS

• Pipeline—What milestones will define every venture’s progress?

What percentage of ventures do you expect to drop off at each
milestone? What is the expected time from green-lighting until a
venture is handed off to grow on its own?
• Portfolio—How many ventures will the board oversee at one time?
How many successful ventures do you want at the end of your
pipeline? Given attrition, how many ventures should you have at
the start? Will you use a venture backlog—approving ventures in
advance, so they are ready to launch when resources are ready?
• Resources—For internal teams, what resources will you allocate
(operating budget, head count, fractional time per person, etc.)?
For outside teams, what funding and other resources will you
provide? For start-ups, what investment stake will you seek?
• Iterative funding—What criteria will you use to green-light a new
venture? What is the size of a typical first-funding round? What
criteria will you use at each milestone to decide whether to
continue to fund? How quickly will funding grow at each
milestone? How will you track shutdowns and measure the quality
of “smart failures”?
• Handoffs—For P2, when do you hand off a venture to the
sponsoring business unit? What are the key validation milestones?
For P3, when does a successful venture graduate from the board
and go elsewhere for continued oversight or funding?

3. Team’s Charter

Step 3 of the Corporate Innovation Stack focuses on crafting a charter

for each innovation team that defines the roles of the team’s members
and the process of their work. Every team must have a charter, created
with and approved by the board, before the team is green-lit to begin
work.

TEAM: MEMBERS
 • Composition—What size should the team be? What skills will its
members need for the team to work completely independently?
For P2 structures, what members of the team will come from the
core business unit?
• Commitment—Will the team’s leader(s) be single-threaded to the
venture? Will other team members be single-threaded? If they are
not full-time at the start, when will they shift to a full-time
commitment?
• Incentives—What kind of upside stake (e.g., equity or performance
bonus) will team members be given in the success of their
venture? How will you ensure that team members have a viable
career path in your organization?

TEAM: PROCESS

• P/O statement—What problem or opportunity is the team

committed to? Who is the key stakeholder (a customer or
someone in the business)? What outcome are you seeking? How
will achieving this outcome create value for the firm (e.g., revenue,
cost savings, or another value driver)? What tenets should serve
as guiding principles for the team’s work?18
• Metrics—What metrics will you use to measure success? What
guardrail metrics will you use to avoid unintended consequences
or risks? How will you share these measures transparently across
your organization?
• Experimentation methods—What cadence will the team use (e.g.,
daily stand-up meetings and biweekly sprints)? What artifacts will
the team use (e.g., Rogers Growth Navigator, kanban boards, user
stories)? How many illustrative versus functional MVPs will you
expect to build? What innovation methods (e.g., lean start-up,
Scrum) will the team draw on?
• Decision rights—What can the team decide without any approval
from the growth board or any other external party?

Remember that any good innovation structure should evolve to

meet the changing needs of the business. A board or a team, learning
from experience, may discover ways to improve how it works. All three
charters—for structures, boards, and teams—should therefore be
revisited from time to time to update them for continued success.

Governance from the Bottom Up

The model of governance we have seen throughout this chapter—

innovation teams, boards, and structures—represents a shift from top-
down management toward a more bottom-up organization.
The governance of innovation teams is designed to support bottom-
up decision making and autonomy. These teams’ small size and
multifunctional skill set allow for independent action. Transparent
metrics and a clear definition of success allow each team to hold
decision rights over its work, without sign-off from other parties. An
example can be seen in the governance of Amazon’s two-pizza teams,
which are allowed to operate with complete autonomy once an
agreement is reached on what they are seeking to accomplish. As
longtime Amazon executive David Glick explained to me, a team’s
metrics, sometimes called its fitness function, provide “a mathematical
description of what you’re trying to optimize.” Before any staff member
is assigned to a new team, these metrics are hammered out in a
discussion between the team leader and top management. In the early
days, “Jeff [Bezos] used to review every single fitness function,” Glick
recalled. Once an agreement is made on the metrics, they are tracked,
shared transparently, and used to judge the team’s work and hold its
leadership accountable.
The governance of innovation boards is designed to support a more
bottom-up approach to leadership. Digital-native businesses focus on
pushing decision making down the organizational chart, with leaders
making far fewer decisions on the day-to-day work of those below
them. Pushing decision making down is precisely what growth boards
are designed to do. Rather than micromanage teams pursuing
innovation, boards are designed to empower them. They do this in
three important ways. They ensure that teams have resources—
budgets, head count, and the right mix of multifunctional skills. They
align processes—metrics, decision rights, compliance rules—to help
rather than hinder the work of teams. And they advocate for teams
internally and connect them to others outside the firm.
 The governance of innovation structures is designed to engage
people at every level of the organization and in every function. Whether
in a hackathon or a program like Citibank’s D10X, these structures can
surface and accelerate growth ideas pitched by anyone, from senior
vice presidents to the newest junior hire. Meanwhile, structures like
Mastercard’s Start Path bring in insights and ideas from outside that
can point the business in new directions. Remember that P2 innovation
structures should not only support your business units with new
products and commercial innovation. They should also support your
functional silos—HR, marketing, customer service, supply chain, and
so on—to solve problems in their work and create value for the firm.

For DX to deliver lasting impact to any organization, it must involve

more than a few teams operating in isolated pockets of innovation.
Without new governance models to allocate resources and manage
new ventures, the potential for growth will always fall short. In Step 4 of
the DX Roadmap, we have seen how any organization can design its
governance to manage growth at scale. We have seen how teams,
boards, and structures can scale experimentation across the
enterprise. We have learned how to manage the three paths to growth
to pursue opportunities of varied uncertainty within your core business
and beyond it. And we have seen how iterative funding and smart
shutdowns are essential to investing wisely and maintaining a pipeline
of innovations.
With a shared vision, strategic priorities, a process for validating
new ventures, and a governance model to manage growth at scale,
you are now ready to turn to the final step of the DX Roadmap and
grow your capabilities. As you began work on your first digital ventures,
you will likely have discovered some gaps in the capabilities of your
organization. These gaps may include your technology and data
infrastructure, your employee talent and skills base, and the culture and
mindset of your organization.
To build a strong foundation for the future of your business, you will
need to invest in the digital capabilities of your organization. In the next
chapter, you will learn how to define the capabilities that will be most
critical to your digital future and grow them to deliver on the promise of
transformation.
 7
Step 5: Grow Tech, Talent, and Culture

When the first of Volkswagen’s ID.3 electric cars rolled off the assembly
line in Zwickau, Germany, CEO Herbert Diess was joined by Angela
Merkel, who was chancellor of Germany at the time, to commemorate
the milestone. After five years and $50 billion in development, the
German auto industry’s answer to Tesla had finally arrived. But the
following spring, Volkswagen’s top executives were forced to postpone
the launch of the ID.3 to customers. The car’s software was rife with
bugs, and many of its digital features—including a cutting-edge heads-
up display—were not working properly. When the car finally did launch
in the fall, its first 50,000 customers were told they would have to return
to the dealership for a software update as soon as it was ready. These
updates were meant to happen over the air, but Volkswagen decided
that the operating system was not yet safe enough because it might be
exposed to online hackers. As the ID.3’s leader, Thomas Ulbrich,
explained, “Updating the vehicle’s core software is a complex process
and we have to make sure at any time that our vehicles are safe.”1 The
ID.3’s rollout made painfully clear that the company’s digital capabilities
were not yet prepared to deliver on its strategy.
As the largest automaker in the world, the Volkswagen Group has
long prided itself on its skill in hardware engineering and design—
reflected in brands ranging from VW and Audi to Porsche and
Lamborghini. But as Volkswagen shifts from gas-powered to electric
vehicles and works on future autonomous vehicles, the relative
importance of hardware and software to the business has shifted. The
capabilities that define a great car company are different in the digital
era.
These new capabilities start with digital technology. Traditional gas-
powered cars were built with software but only for running secondary
functions like heating, maps, and entertainment. These simple
applications were coded onto separate chips implanted in parts
throughout the car. They were easily outsourced to suppliers and
plugged in when the vehicle was assembled. But with the shift to
electric vehicles, software is now in the driver’s seat. It runs the entire
powertrain, brakes, battery, and lights—the most critical systems of any
car. The hundreds of applications in an electric vehicle cannot be built
as separate widgets; they must be part of an integrated operating
system. Just as important, the software must be continually updated
over the life of the vehicle, just like a smartphone’s operating system
and apps. Volkswagen’s ambition is to build an entirely new software
platform called VW.os 2.0 that will be used by every vehicle brand from
VW to Porsche, to Skoda. But to get there, the company will need more
than just new technology.
 To pursue its strategy, Volkswagen also needs a transformation of
its talent. Volkswagen and its peers have long outsourced their IT
needs to suppliers. “Over the past 20 years, the auto industry became
more integrators than developers,” observes Alexander Hitzinger, a
former member of Volkswagen’s board.2 This kept costs down but
required giving up control. To achieve its electric vehicle strategy,
Volkswagen set a five-year goal to shift from 10 percent of its vehicle
software being built in-house to 60 percent.3 That meant insourcing
deep technical knowledge that formerly resided in partners. It also
meant bringing together thousands of programmers previously
scattered across Volkswagen’s divisions to work in a centralized
fashion. Dirk Hilgenberg was hired from competitor BMW to lead this
centralized software unit—a new company under the Volkswagen
Group named CARIAD.
For Volkswagen to succeed, however, it needs more than digital
technology and talent; it needs a digital culture. As Hilgenberg sees it,
the company’s biggest challenge is the mindset of its people. “The
global transformation of the industry will take roughly 10 years—with or
without Volkswagen,” Diess declared in a LinkedIn post. As CEO, he
pushed back against the complacency he saw in senior executives.
Rather than grading themselves on traditional measures where VW
excelled, he pushed them to look at new metrics—such as battery
range and advanced computing—where it was far behind.4 In addition,
CARIAD’s software teams need a more agile culture, like a tech
company. To build VW.os 2.0, they must adopt collaborative ways of
working with technology partners like Continental. Tech companies like
Diconium have been acquired not just for their technical talent but for
the culture they bring. To attract even more talent, Volkswagen needs a
culture that enables teams to innovate around customer needs and
continuously ship new software. In Diess’s words, “a culture which is
customer-oriented, fast and agile.”5
Shifting the technology, talent, and culture of Volkswagen is
essential to its digital transformation (DX). But it will not be easy, and it
will not happen overnight. As Danny Shapiro, VP of automotive at
chipmaker Nvidia, says, “You can’t just flip a switch and be a software
company.”6
 Why Capabilities Matter

In Amazon’s early days, when it was still known as a website selling

books and household goods, Jeff Bezos remarked to employees that
“Amazon is not a retailer. We’re a software company.” He went on to
explain, “Our business is not what’s in the brown boxes. It’s the
software that sends the brown boxes on their way.”7 One of the
questions I hear legacy businesses struggle with in their DX is some
variation on this: “Are we a car company? Or are we a tech company
that sells cars?” Whatever you call yourself, what matters is whether
you have the capabilities that are most critical to your strategy for the
digital era.
In every industry, the defining capabilities of a great company are
shifting. As every business pursues its own DX, these capabilities will
distinguish tomorrow’s winners from its also-rans: new technologies,
new skills, and new cultural mindsets. We can see the interplay of
these different kinds of capabilities in the model of a modular
organization. Modular architecture has been a major trend in IT
throughout the digital era. Its software is composed of loosely coupled,
independent pieces of code called microservices. Each one acts as an
independent building block that performs a discrete business function
and can be updated without risk to the rest of the system. This provides
technical benefits of resilience and scalability. But the biggest impact of
modular architecture is how it affects the organization itself. By turning
the work of a business into individual components, modular architecture
allows teams to operate with much greater independence.
As companies like Walmart have migrated to modular IT systems,
they have transformed their organizations with independent teams that
innovate much more rapidly. Research by Mark J. Greeven, Howard
Yu, and Jialu Shan has shown the widespread benefits of modular
architecture in reducing organizational complexity.8 In a modular
organization, getting what you need from another department becomes
a self-service process. The extreme degree of this can be seen in
Haier, a manufacturer of home electronics, from refrigerators and
washing machines to televisions. Taking advantage of modular IT, Haier
has reorganized its entire company into 4,000 microenterprises (MEs)
of ten to fifteen people. Some MEs deliver final products to consumers;
others provide internal services like staffing, product design, or
manufacturing to other MEs. Each one operates with independence
and autonomy, and all coordination is managed through an internal
cloud-based platform.9 The culture of Haier matches this new
organizational model as well: it is bottom-up, risk taking, and
collaborative, with each team having a clear sense of ownership of its
work.
In the fifth step of the Digital Transformation (DX) Roadmap, your
objective is to grow the right mix of technology, talent, and culture to
support the digital growth trajectory of your business. The challenge for
Volkswagen, Haier, and any established business is to ensure that its
capabilities match the ambition of its digital strategy.
Why not start the DX Roadmap with growing capabilities as the first
step? Many consultants propose a DX agenda that focuses first on
capability building—such as moving to the cloud, establishing data
governance, and hiring engineers. But building capabilities is not a
strategy! The truth is that there is no generic digital organization and no
universal blueprint for the capabilities every business will need. Only
when you have some clarity on your vision for the future and have
started to pursue your first digital ventures will it become clear which
capabilities are most important to build first.
Growing your digital capabilities can be the hardest, longest, and
most costly work of any transformation effort, but the investment is
essential. Without it, progress will be thwarted by inflexible software,
incomplete data sets, and systems that can’t communicate with each
other or with partners. Your workforce won’t have the skills needed to
build and grow your digital ventures. Your teams will be held back by a
culture that is rigid instead of flexible, siloed instead of collaborative,
and tentative instead of confident about taking risks. If these deficits are
left unaddressed, every new digital innovation you pursue will be limited
in its impact.
Building the right digital capabilities for your business is not easy,
but it is the only way for transformation to deliver real change in the
long term. Table 7.1 shows some of the key symptoms of success
versus failure in Step 5 of the DX Roadmap.

Table 7.1.
What’s at Stake—Step 5: Capabilities
Symptoms of Failure: Symptoms of Success:
Capabilities Capabilities
• Inflexible IT systems • Modular IT systems
reinforce silos and limit integrate across the
collaboration. organization and with outside
partners.
• Data is contradictory, • Data provides a single
incomplete, and source of truth to managers
inaccessible to managers in across the company.
real time.
• Centralized IT governance • IT governance provides
causes bottlenecks for new oversight while keeping
projects. innovation in the hands of the
business.
• Employees lack digital • Employees can build and
skills, so digital projects iterate digital solutions
must be outsourced. themselves.
• A top-down culture and • An empowering culture and
bureaucracy stifle processes help employees
employees, breeding drive bottom-up change.
cynicism and inertia.

What’s Ahead

In this chapter, we will see how any organization can build the
foundation it needs for its unique digital future. We will examine three
essential types of capabilities for the digital era:

• Technology—including IT infrastructure, data assets, and

governance systems.
• Talent—including technical skills like data science and
nontechnical skills like design thinking.
• Culture—the mindsets and norms that shape day-to-day behavior
throughout the firm.

The chapter will introduce two new tools: the Tech and Talent Map,
which identifies and closes gaps in your technology and skills, and the
Culture-Process Map, which defines, communicates, and enables the
right culture at scale in your organization. Finally, we will see how the
right capabilities are critical to bottom-up change and transformation.

Technology

The right technology capabilities are essential to any digital strategy,

but many legacy businesses are held back by their limits in this area.
Among the symptoms are software that is slow to adapt, data that
cannot be accessed in real time, technology silos that match
organizational silos, and systems that don’t talk to each other or outside
partners. As you pursue your digital strategy, it is essential that you
work closely with your CIO to assess your current technology
capabilities and identify your most critical gaps. Specifically, I see three
main areas that organizations need to assess: IT infrastructure, data
assets, and tech and data governance.

IT Infrastructure

Any organization must constantly assess its technology infrastructure to

identify the capabilities it needs to support its future strategy. This may
include decisions about cloud computing—for example, whether to use
public, private, or hybrid cloud models or on-premise solutions to meet
the needs of the business. It includes system architecture, such as
microservices and application programming interfaces (APIs) to
connect to different parts of the business and outside partners. Data
storage, using models like data lakes or data warehouses, is crucial to
storing data for effective use and retrieval. Just as important are the
applications that run key business processes—for example, pricing and
inventory for a retailer, content and subscriptions for a publisher, and
customer analytics for almost any business. Having the right IT
infrastructure for your particular needs is essential to your DX success.
Recall the example of the National Commercial Bank (NCB) from
chapter 3. NCB’s strategy to become the top digital bank in Saudi
Arabia was to build a best-in-class mobile banking experience and thus
free client-facing employees to drive an expansion into unbanked
markets. The biggest obstacle to this strategy was the bank’s legacy IT,
which operated on a snarl of 160 different systems with a complex web
of integration. Multiple points of integration failure called for constant
intervention by back-office personnel. In a multiyear effort, NCB rebuilt
its core banking technology in a modern architecture of just fifteen
systems supporting seamless digital operations. The result was a
reliable, best-in-class customer experience that made NCB’s banking
app number one in the country. Customers shifted over 98 percent of
their transactions to digital self-service, driving cost savings and top-
line growth for the bank.

TECHNICAL DEBT
NCB’s story is a clear illustration of the concept of technical debt.
Technical debt is any future cost to the business caused by suboptimal
technology.10 The term applies to both software and hardware; it can
range from poor HTML code that slows down a webpage to
systemwide deficits in networking, data integration, or cybersecurity.
Technical debt has many causes, from deferred maintenance of aging
systems to changing technology standards, to poor initial design. It can
even be the result of an intentional decision to “move fast and fix later.”
CTO Fiona Tan explained to me that Walmart may delay integrating
new business acquisitions into its existing tech stack if its strategy is to
prioritize growth and speed-to-market of the new business in the near
term.
Left unaddressed, the costs of technical debt are many. It saps
resources, diverting IT budgets into maintenance rather than supporting
new growth. It slows down business teams and entire organizations.
Most important, technical debt impedes strategy whenever
infrastructure hinders innovation, as experienced by Volkswagen and
NCB.
A common mistake is for companies to fix infrastructure problems
with patches and workarounds that sustain inflexible legacy systems.
Just like a loan with interest, however, unaddressed technical debt will
grow over time. Instead of patching over the underlying problems, a
strong IT organization insists on periodically “paying down” technical
debt to improve the efficiency and agility of the business. Paying down
technical debt is hard. It means spending resources without gaining any
new products or functionality. But the benefit of rebuilding is to make
your IT faster, more reliable, more secure, more flexible to updates, and
better able to integrate with other systems. This process, called
refactoring, takes time and investment, and the payoff is not immediate.
But as NCB learned in refactoring its IT systems, it is essential to future
growth.

MONOLITHIC VERSUS MODULAR IT

One of the key types of IT refactoring today is a shift from monolithic to
modular architecture. In a traditional monolithic architecture, the
software—whether an app, a website, or a bank’s entire retail operation
—is built as one integrated program. This can work well on a small
scale, for example, when quickly building an MVP. But as a monolithic
system grows, it becomes increasingly rigid and inflexible. To change
one part of an application, you must test the impact on the whole
system or risk bringing it down. Updates of any kind become slow and
painstaking. This is why enterprise computing systems from the 1990s
and 2000s are famously hard to customize or adapt to changing
business needs.
In modular architecture, the same software is rebuilt as a set of
modules called microservices. Each of these software modules
communicates with the others through an automated interface called an
API. Within a single company’s architecture, hundreds or thousands of
microservices can interact this way, each one managed and developed
by a single team.
When Amazon.com started, its architecture was monolithic. The
entire website ran as a single, integrated piece of software. After a few
years of rapid expansion, the site had grown to millions of lines of code
whose limitations were impacting growth. As CTO Werner Vogels
explained, “Whenever we wanted to add a new feature or product for
our customers, like video streaming, we had to edit and rewrite vast
amounts of code on an application that we’d designed specifically for
our first product—the bookstore.”11 In 2002, Amazon began a shift to a
new architecture. According to legend, Bezos issued a memo
mandating that henceforth, all teams would have to write their code in
microservices that would communicate with each other and with the
outside world through APIs.12 The transition took several years, but it
laid the foundation for what became Amazon Web Services (AWS), and
it served as a pioneering example to other businesses.
Modular architecture hosted in the cloud brings numerous benefits:
it is more scalable, secure, and flexible. But perhaps its most important
impact is on organizational speed and flexibility. Under Amazon’s
original architecture, if one team wanted to add a line of products to the
store website, it needed to coordinate with numerous other teams
(marketing, warehousing, web design, etc.) through meetings, phone
calls, and emails. With microservices, everything one team needs from
another can happen through an automated software interface. Vogels
describes the impact: “We were able to innovate for our customers at a
much faster rate, and we’ve gone from deploying dozens of feature
deployments each year to millions, as Amazon has grown.”13
Today, a shift from a legacy monolithic computing system to a
modular architecture in the cloud is increasingly seen as essential.
Years ago, Netflix transitioned its entire website to a modular
architecture, hosted on Amazon’s cloud-computing service.14 Microsoft,
Google, and others now compete with AWS in providing these services.
Modular computing is now used in all kinds of legacy businesses, from
large enterprises like Walmart to smaller players like Acuity that have
built their own technology stack and APIs.

Data Assets

The second area of tech capability critical to any DX is data. As

explored in The Digital Transformation Playbook, data is now a key
strategic asset for any business. Every organization needs a strategy
for how it will invest in and grow its data asset over time. This data
asset should include customer data (e.g., individual customers’ profiles
and behaviors, including purchase, usage, and other interactions). It
should include data on the firm’s internal operations, people, and
assets (e.g., supply chain data, inventory records, employee
information, and more). And it should include data from all its products
and services—both underlying service data (e.g., mapping data for a
navigation app, or financial data for a brokerage service), as well as
product use data (which features are being used when, where, and
how).
 If it doesn’t capture data effectively and grow data as an asset, any
organization’s DX will stumble. The New York Times Company found its
early digital efforts were hampered by a failure to capture essential
data, including data on archived articles going back over 100 years.
Deficiencies in the data “tagging” of past articles made it impossible for
teams to find and leverage content easily from this enormous trove.
The Times watched digital-native publishers like the Huffington Post do
a better job repackaging New York Times content—using historic Times
stories to engage readers around events like the death of a political
figure or the release of a movie. Meanwhile, the Times also struggled to
automate the sale of its historic photos and spent years trying to create
a useful recipe database. Eventually, with strong investment in its data
assets, the Times began to reap rewards—from enhanced journalism
to new subscriber services, to increased traffic from outside channels.
Adding structured data to its cooking recipes, for example, increased
traffic to the Times from search engines by 52 percent.15
In the digital era, data is generated at an unprecedented rate and
from proliferating sources—including social media, mobile devices, and
the web of sensors that make up the Internet of Things (IoT). But they
are also being generated by new business models, like software as a
service (SaaS), which allows companies to observe customers’ use of
their digital products directly over time. Growing your data asset is not
just a matter of putting cookies on your website or purchasing data from
third parties. Rather, it must be part of the planning and design of your
products and services themselves.
One global energy utility I have advised is rethinking its operations
after realizing that it was collecting data on electricity use only at the
level of power meters—when what it wants is individual customer-use
data. As Nike shifted from being a brand selling products through brick-
and-mortar channels to an omnichannel brand that sells more and
more directly to the consumer, data has become central to its success.
From the design of its apps to its newer subscription business models,
to its partnerships with physical retailers, Nike plans at every stage how
best to gather and deploy data about its customers, products, and
business.

Tech and Data Governance

The third area of tech capabilities that is essential to DX is governance
systems for both your data and your technology assets. Any
governance system must address the needs of internal stakeholders
(different business units and functions) as well as external stakeholders
(customers, regulators, business partners, etc.). Whether defining the
rules for data sets or tech infrastructure, any governance model should
address a few key issues:

• Access—who accesses what, under which conditions, and with

what permissions and constraints.
• Integration—how data and software connect across technical
systems and organizational silos.
• Quality—how the integrity of data, software, and other assets is
assessed and improved continually.
• Security—how assets are protected from malevolent actors and
other risks, inside and out.

Yana Walker, a former student of mine, led the design of one such
governance model at Bristol Myers Squibb. Its purpose was to integrate
the data from three different divisions: manufacturing, quality
management, and regulatory affairs. The project had technical goals—
consolidating data from the divisions, harmonizing and standardizing
the data, and ensuring data quality (with a focus on remediation of
aging and incomplete data sets). The project also had organizational
goals—providing access and transparency to the right stakeholders,
ensuring data security, and ensuring regulatory compliance for every
country where the data’s servers were housed. The governance system
was implemented ultimately in two parts. An automated system was
built to manage the data, following more than sixty business rules to
handle most issues. And a data governance council was established—
with representatives from each division—to address inconsistencies
and questions that fell outside the business rules.
Governance systems enable the kind of data sharing and
collaboration that are essential to DX. An executive at the Sony Group
explained to me how, by linking customer data across its different
divisions, Sony gained powerful market insights and built better
predictive models. But data integration can only happen with effective
data governance. Integrating all of Sony’s data would never have been
allowed if it meant, for example, that Sony Pictures could use all the
individual-customer data from Sony’s PlayStation Network to bombard
gamers with emails promoting its next movie release. Before data
sharing can happen, rules on access, quality, and security are
essential.
Another critical function of governance is to help large organizations
establish a common set of data on which to make decisions—what is
commonly called a single source of truth. One of Walmart’s key digital
investments has been to create a single view of the customer through
data and make it available as a service to managers across the
company. Shared KPIs are incredibly powerful in aligning an
organization around a strategy, but they hinge on everyone agreeing on
the same shared data. Unless internal stakeholders trust that data, they
will use their own or start collecting data elsewhere. Shared, trusted
data is essential to organizational alignment.
One more critical aspect of governance is how much to centralize
key assets and capabilities. In too many organizations, I have seen an
overcentralized model, where all projects must go through a central IT
division located at a distant headquarters. The result is innovation rigor
mortis. The desire for strong governance must be balanced with the
need to keep innovation in the hands of the business, where it is
closest to the customer. The best governance models I have seen
manage to strike a balance. Data is stored in more than one location
but synchronized across the entire business. New areas of technical
expertise (such as machine learning) may temporarily sit in a central
unit but are then pushed out to the businesses. And software
applications remain in the hands of local business units but with a bias
toward funding applications that can be reused across the organization.

Build Versus Buy

One of the most common questions around technology in DX is this:

Should we build the capability we need (e.g., a new microservices
architecture) with our technology teams, or should we procure it from
an outside partner via technology purchase, license, or SaaS? The
choice between these two options is commonly referred to as “build
versus buy.”16
 In recent decades, there was a strong tendency toward “buy”—as
companies outsourced technology to partners and focused their
employees on industry-specific work. At most firms, technology was
deemed outside their core competencies and treated as a cost center.
IT departments evolved into vendor management units. In the digital
era, we have seen a big shift in thinking as technology capabilities have
become core to every business strategy. Companies like Volkswagen
discovered the limitations of outsourcing 90 percent of their software as
they face digital-native rivals like Tesla. Jeff Lawson, the cofounder of
Twilio, reframes the choice for incumbents as “build versus die.”17 In
other words, no company will survive long term if it does not develop
the capabilities to build its essential technologies.
A “build” strategy has several inherent advantages. It provides
greater control and the ability to customize a solution to the specific
needs of your business. PepsiCo’s chief strategy and transformation
officer Athina Kanioura found that even the best software it purchased
was too generic—lacking enough business-specific data and alignment
with PepsiCo’s business goals. So Kanioura followed a “build” strategy
and established digital hubs for the company in Dallas and Barcelona
to bring more capabilities in-house.18 A “build” strategy also provides a
greater ability to integrate with partners. At Acuity Insurance, CEO Ben
Salzmann has invested for years in building the company’s core IT
systems and APIs to allow fast and seamless integration with any
insurtech partner—what Salzmann calls “nimbleocity.” He gives an
example: “Google came to work with us because our peers will take
months and months to do anything with them. We can partner with any
technology vendor in two weeks.” One more benefit of a “build” strategy
is that it allows the business to retain ownership of technology IP. This
in turn generates greater profits from any technology that is important
to competitive advantage.
A “buy” strategy can have compelling advantages as well. If
technology solutions already exist in the market, deploying these
(rather than “reinventing the wheel”) can allow a firm to get to market
much faster and at much less cost. A “buy” strategy may bring less risk
—particularly if the company can try the solution in a pilot before
committing to it. And it may bring much less maintenance cost, for
example, in the case of a SaaS solution. So there are inherent trade-
offs to the build versus buy decision. You can’t and shouldn’t try to do it
all yourself.
Therefore, it makes sense to use both “build” and “buy” strategies
for at least some of your tech capabilities. My friend Anand Birje,
president of digital business at global services firm HCL Technologies,
suggests that you start by asking which technology capabilities are
most important to your competitive differentiation. If a technology will
enable you to compete and command a premium in the market, you
should build it yourself in a microservices architecture that you own. On
the other hand, if a technology is a simple cost of doing business, then
you should look for a solution that you can purchase on a flexible SaaS
basis where you pay only for what you use. Birje calls this
“composable” versus “consumable” tech.
Finally, be sure to assess the technical uncertainty around any
solution you are considering. Is this something—like public cloud
infrastructure—where solutions are widely deployed in the marketplace,
with ample third-party research on their costs and benefits? Or is this a
new, cutting-edge area where solutions will use unproven technology?
If technical uncertainty is high, pursue a “try before you buy” approach
—whether this means a limited early deployment with a service
provider (before an ongoing contract), or a pilot project with a tech
start-up (before buying them or building your own version).

Staging the Journey

Whether you build or buy, none of this work happens overnight. Paying
down technical debt on your infrastructure, building your data assets,
and establishing good governance models all take time. There is no
amount of funding or leadership commitment that will let you snap your
fingers and get these things done overnight. But be careful: as you
proceed down a long road of capability building, your organization’s
needs and priorities will inevitably change over time. Given this reality,
growing your tech capabilities should be planned as a multistage
journey, with flexibility to adapt as you go. Think of the organization as
a train in which you are rebuilding one car at a time, applying temporary
patches and duct tape to other cars until you get to them, all while the
train continues down the tracks.
 NCB, for example, took three years to build its new core systems
from retail to corporate banking and another two years to migrate to
them while running the business. Volkswagen has planned a five-year
evolution of its VW.os operating system, moving in stages from version
1.0 (a limited release, with open source code and software from outside
vendors) to 1.1 (used by more vehicles), to an ultimate 2.0 (with
advanced features and deployed across all the company’s brands).19
At Bristol Myers Squibb, Walker’s team built its data governance
model through a carefully staged journey. This began with six months
of developing a prototype to demonstrate to leadership that there would
be enough data to answer important business questions. In the next
phase, a robust test environment was deployed to test the logic of
business rules for the new system. Final migration happened with a
“post-go-live” period where the old data processes were run in parallel
with the new integrated one—to sift through numerous scenarios and
gain full confidence in the new system’s resilience and reliability. The
whole process took over a year. But the timing largely depended on the
number of systems being integrated. Walker’s advice? “Start early, start
early, start early!”
As you plan your journey, remember to put a premium on
modularity. A more modular design will allow for more flexibility and
quicker updates as your needs continue to change.

Talent

As important as technology is to any digital strategy, having employees

with the right capabilities is just as critical. No organization can build its
own apps and infrastructure if it lacks skilled engineers. No business
can pursue a direct-to-consumer strategy if it lacks people with digital
marketing skills. It cannot reorganize into small, multifunctional teams if
it lacks training in product management and agile methodologies. And
no business can become a data-driven organization if it lacks skills in
data science and analytics.
As you pursue DX, it is essential that you work closely with HR
leaders to assess your current talent capabilities and to identify critical
gaps. Skills to look for include technical and nontechnical skills in
different combinations and different parts of the organization. Closing
your talent gaps requires managing the entire talent life cycle—from
hiring and acquiring talent to training your workforce, retaining them,
and exiting those who are not aligned with your strategy while
minimizing attrition among those who are.

Technical and Nontechnical Skills

Any new strategy for the digital era is likely to demand that your
business grow its capabilities in several technical skills areas, including
software engineering skills with differing focuses on applications,
platforms, networks, and more. They also include skills in data science
and business analytics, and skills in emerging and fast-changing fields
such as machine learning and cybersecurity.
As Johnson & Johnson’s (J&J) vice president of human resources,
enterprise technology, Juliana Nunes is responsible for digital talent. It’s
her job to ensure that J&J has the skills it needs as the company
rapidly shifts strategy, products, and services for the digital era. Nunes
has partnered with IBM to assess the current technical skills within
J&J’s global workforce—using machine learning to analyze each
employee’s digital footprint and infer what skills they have and at what
level of maturity. Among the technical skills they found that J&J needs
to grow are data science, cybersecurity, and intelligent automation. “We
know you can’t predict the future,” Nunes says, but with the right skills
“we believe we can be better positioned for it.”
Any organization also needs to assess which nontechnical skills it
will need to support its DX efforts. Nontechnical skills typically include
innovation skills—such as training in product management, agile
methods such as Scrum, design thinking, or lean start-up. It may
include new go-to-market skills—such as digital marketing, e-
commerce, online sales, and channel management. Depending on the
problems they are solving, your multifunctional teams will need skills
from other disciplines, too—such as communications and design
(storytelling, graphic design, user experience [UX], etc.) and the social
sciences (economics, sociology, psychology, anthropology, etc.).
Amazon is known for hiring PhD economists and spreading them
across two-pizza teams throughout the company to aid in running
controlled experiments at the level where work is being done fastest.20
When Imran Haque was CDO at the global shipping firm CMA CGM,
one of his first initiatives was to design a new digital product serving the
needs of small and midsize shipping businesses. But CMA CGM had
never sold to these kinds of customers in the past, and so it lacked any
relevant sales experience. One of the keys to pursuing the new
strategy was first to acquire skills in digital marketing to help reach this
very different audience.

Different Levels and Combinations

Addressing your talent needs will require more than just a shopping list
of skills in broad categories. Many leaders speak about the total
number of programmers they are targeting to hire for their DX. (Recall
from chapter 1 the New York Times’s publisher boasting that they have
“more journalists who can write code” than any other newspaper.) But
what matters more is which specific kinds of programmers you have
and need in various specialized fields (network engineering versus
application programming, etc.).
It is equally important to recognize that your organization will require
different people with different skill levels in the same domain. For
example, you may need fifty expert data scientists to support your
digital efforts—but what about businesspeople who know how to talk
with these data scientists? You will likely need far more managers in
other functions (marketing, finance, operations, etc.) to acquire a basic
knowledge of data science so that they can collaborate effectively with
these scientists.
In a large organization, your talent needs are likely to vary
considerably across different business units and functions. When J&J
analyzed its own talent needs, it found several areas like data science
that were critical to teams across the company. But other skills were
needed only in specific business units—such as e-commerce skills for
its consumer business, and robotics skills for its medical device
business.
Many organizations find what is important are combinations of
different skills, including both technical and nontechnical. Nunes at J&J
spoke to me of the value of combining data science skills with deep
experience in the health-care sector—sometimes from different
members of a team but ideally in the same person. Many digital-native
companies like Meta look to hire product managers who are “T-shaped”
thinkers: people with real depth in one domain but an ability to
collaborate across several others.21

The Talent Life Cycle

Addressing your talent needs is about more than just hiring new people
with digital-era skills. It is just as critical to ensure that they have an
attractive future and career path in your organization. In its early days
of DX, the New York Times Company found that many of its best digital
talent hires were quick to leave the company. As one of them
explained, “I looked around the organization and saw the plum jobs . . .
going to people with little experience in digital.” Another departing talent
observed, “When it takes 20 months to build one thing, your skill set
becomes less about innovation and more about navigating
bureaucracy . . . If there’s no leadership role to aspire to, staying too
long becomes risky.”22 Thus, the challenge is not simply to find the right
people and plug them into your various departments. Growing the right
capabilities means managing the entire talent life cycle: hire, acquire,
train, retain, exit, and partner (see figure 7.1).
Figure 7.1.
The talent life cycle

HIRE
Recruiting new talent is critical to bridging talent gaps. But one common
mistake I see is to focus on star hires—bringing in a few leaders from
Google or Amazon and expecting them to transform your teams from
above. DX typically requires a critical mass of new talent. (One
McKinsey analysis estimated that traditional automakers need to grow
their software talent by 300 percent to meet the shifts in their
industry.23)
How can you compete with skilled digital talent in such high
demand? One key is to broaden your geographic lens. Instead of
headquartering your digital operations in Silicon Valley, you may opt to
build it in a tier-two or tier-three market where you can be a top-choice
employer for technical talent. Another key is to be willing to hire from
outside your industry. Nunes says of J&J, “We knew that, in order for us
to bring the skills that we need, we needed to look beyond the
traditional health care companies.” Today, that means J&J hires from
tech firms, telcos, financial services, and beyond.

ACQUIRE
Another important way to bridge talent gaps is through buying another
firm that has people with the capabilities you need—what is called an
acqui-hire. This strategy can be effective for speeding up a big shift in
skills, for example, McDonald’s acquisition of the AI firm Dynamic Yield
with its 200 employees.24 The value of such an acquisition is often in
the firm’s talent even more than the value of its revenues, customer
base, or proprietary technology.
For large firms seeking to accelerate the growth of digital talent, a
larger anchor acquisition may be more effective than a series of smaller
acquisitions. Bringing onboard a critical mass of hundreds or even
thousands of employees with digital skills and a digital mindset can
jump-start a broader change in the entire firm. After the anchor
acquisition is fully assimilated, it should be easier to integrate smaller
acquisitions of digital-native talent.25

TRAIN
The next step in bridging your talent gaps is training. It is essential to
develop and grow the capabilities of the people already in your
organization. Training is commonly thought of in terms of upskilling and
reskilling. Upskilling refers to building on the skills an employee has in
their current role (think of your long-time marketer learning new digital
tools and tactics). For an employee whose old role is disappearing,
reskilling means giving them the foundational skills for a new job (think
of a customer service representative whose job is replaced by bots and
is training for a new role in sales). Reskilling programs are vital to
employee morale, but they depend on the individual employee’s
willingness to embrace a bigger change.
 RETAIN
It is essential that your organization also have a strategy to hold onto
employees with critical capabilities. At J&J, Nunes asks: “How do I
create the best growth experience so that people are going to be willing
to stay here with us? Top talents are recruited by other companies all
the time. So how do I create an experience for them so they feel they
are continuously growing?”
Daniel Pink’s model for intrinsic motivation of employees is apt. He
stresses three elements: autonomy (the opportunity to direct your own
work), mastery (the opportunity to learn and grow through your work),
and purpose (the opportunity to benefit others through your work).26
Each of these elements is critical to employee retention. As Nunes
observes: “We need to create an environment for our talent to flourish.
If they feel that they are in the middle of a bureaucracy that won’t allow
them to do what they want to do, they are going to leave.”

EXIT
No employee will stay with your firm forever. I have seen many leaders
struggle with employee exits as part of DX. Sometimes the challenge is
orchestrating the departure of senior executives who are no longer
aligned. In other cases, it is shifting resources spent on one type of
employee toward hiring more of another type.
The first kind of exits to plan for are voluntary exits—how much
attrition do you want? Turnover rates that are too high can be costly.
But attrition can also become too low, making it difficult to bring in new
talent. Beware of creating “golden handcuffs” for employees, where no
one leaves because they are incentivized to stay for life. As Philipp
Wohland, chief people and transformation officer at Virgin Media O2,
explained to me, “During transformation, incentive systems that cause
a ‘playing to stay’ mentality are counterproductive. Instead, you want to
enable a ‘playing to win’ spirit.” Be sure your own incentives reward the
outcomes you are seeking rather than rewarding employee tenure. The
second kind of exits to plan for are involuntary exits—whether targeted
removals, across-the-board buyouts (which may be paired with layoffs
to reach a target), or reorganization (where an entire unit is dissolved,
but its staff are allowed to apply for positions in a redesigned unit).
 Perhaps the most important factor in employee exits is alignment.
Leaders need to focus on communicating where they are going as a
business and why, how current employees can be a part of that
journey, and what it will require from them. In my observation of
organizations going through major transformations, the highest turnover
rates happen at the most senior levels (as much as 70 percent leaving),
and these exits are largely voluntary. When a new direction is
articulated clearly, employees who have been around longer are often
less willing to change and will instead opt to take a buyout offer.

PARTNER
Effective leaders grow talent not just with their employees but with an
ecosystem of partners. You should make build versus buy decisions for
your talent, just like you do for tech capabilities. Ask yourself, “Which of
these needs should I fill in-house, and which can be better filled with
outside partners?”
The best partner relationships should not just compensate for skills
you lack on the inside. They should support each stage of your talent
life cycle. This can include training (great partners work with your
employees and help grow their skills in the process). It may lead to
hiring (recruiting individuals from an outside partner as permanent
hires). In some cases, a great partner can become the target of an
acqui-hire.
Great digital-era organizations design themselves to be open and
permeable. They fluidly deploy teams that combine internal hires and
outside partners. They cultivate networks of past employees whom they
continue to work with in the future. And they encourage individuals to
move across the perimeter of the organization.

Tool: The Tech and Talent Map

Our next tool is the Tech and Talent Map (see figure 7.2). The purpose
of this tool is to help any organization develop the technology and
talent capabilities it needs to thrive in the digital era. The tool guides
leaders through assessing the capabilities they need for the future,
identifying gaps and prioritizing them, and devising strategies to bridge
those gaps over time. The Tech and Talent Map can be applied at the
enterprise level, but it can just as easily be applied to an individual
business unit or department. It can even be used to assess the
capability needs of a single team. Regardless of the scope of its
application, the process is the same.

Figure 7.2.
The Tech and Talent Map

1. Technology Capabilities

Step 1 of the Tech and Talent Map focuses on technology needed to

support the DX of your organization, business unit, or team. It should
be completed in close partnership with your IT leadership, who can
speak with expertise on subjects like computing infrastructure, security,
or data integrity; as well as business and functional stakeholders, who
will best understand the needs around decision making, customer
experience, legal compliance, and so on.

ASSESSMENT
Begin with an assessment of your future technology needs to support
your evolving digital strategy. Be sure to focus on these elements:

• Tech infrastructure—Do you have plans for a shift to cloud

computing? What kind of architecture and APIs will support your
business growth? Should data be stored in data lakes or data
warehouses? What kind of applications do you need to run your
core business processes?
• Data assets—What kind of data do you need in order to support
your business strategy? What data do you need on customers?
What data do you need from your operations? What data do you
need to power your products, and what data should they be
capturing?
• Tech and data governance—What systems will you need to
manage access to your data and IT infrastructure and to ensure its
security, quality, and integration?

Next, you should conduct a gap analysis, comparing your future

needs with your current tech capabilities. Identify your biggest gaps in
each of the above areas. As you do so, be specific about where your
current capabilities fall short. For instance, do not state that you want
just “better IT infrastructure.” Instead, state, “We need more robust
APIs that connect with our sales channel partners.”
Prioritize each of these capability gaps. Again, be sure to gather
input from all your relevant stakeholders (IT, business, and functional). I
suggest asking each stakeholder to rate the gaps with a score of 1 to 5,
where 1 indicates a severe lack of capability holding back the business,
and 5 indicates an area needing only minor improvement. By
combining these ratings, you can create a prioritized list of tech
capabilities to grow for the future.

BRIDGE PLAN
Next, you will need to develop a plan for how you will bridge gaps in
your tech capabilities. This requires consideration of a build versus buy
approach:

• Build—Which capabilities are strategic, that is, they give your

business a unique advantage in the market? How can you best
build those yourself, with a flexible architecture and IP that you
own?
• Buy—Which capabilities are purely operational? Can you find a
solution to buy that will allow you to perform as well as your peers?
How can you ensure flexibility to customize or change your
solutions in the future?
• Test first—How much technical uncertainty is there in each
solution? Is the technology well established with clear performance
benchmarks? If not, how can you test it in a limited deployment
before committing to use it across your business?

You should also plan for how you will stage your journey to close
the gaps in your technology. Tech capabilities take time to build or buy,
and the business must keep operating while the new capabilities are
put in place. As you stage your journey, focus on these considerations:

• Business impact—Which investments are most urgent to your

business needs? How should you invest in technical debt today to
avoid costly impacts on your business tomorrow?
• Iterative process—How can you use MVPs to validate the business
problems you aim to solve and whether stakeholders will use your
solutions?
• Modular development—Can you break a complex solution into
smaller pieces, following the agile imperative to continuously
deploy increments of working code?
• Parallel tracks—How will you plan for a period of migration where
old and new systems run in parallel and thus ensure resilience in
the real world?

2. Talent Capabilities
Step 2 of the Tech and Talent Map focuses on the talent needed to
support the DX of your organization, business unit, or team. It should
be conducted in close partnership with HR leaders as well as business
and functional leaders, who will each have different insights into the
talent needs of your organization.

ASSESSMENT
Begin with an assessment of the technical and nontechnical skills that
are most essential to your DX. As you assess your future needs, look
broadly and include these areas:

• Software engineering—For different areas such as application

development, networks, infrastructure, security, and hardware.
• Data and analytics—Including data science, data integrity, decision
systems, and business analytics.
• Emerging technologies—Such as machine learning, robotics,
cybersecurity, and automation.
• Innovation skills—In methodologies like product management,
Scrum, design thinking, or lean start-up.
• Go-to-market skills—In e-commerce, digital advertising, influencer
marketing, and so on, to support entry into new markets.
• Other domains—For example, UX, storytelling, market research,
and economics for your multifunctional teams.

Next, you should conduct a gap analysis—comparing your future

needs with your current talent. Work with HR, using both quantitative
and qualitative data, to define your current skills and identify your
biggest gaps in each of the above areas.

• Specific skills—What particular skills are you most lacking (not

“coders” but the specific kinds of developers, and not “data skills”
but “machine-learning expertise”)?
• Distribution—Which skills are needed across the organization?
Which skills are needed only in particular business units or
departments?
 • Skill level—Where do you need high expertise in a particular skill?
Where do you need more general familiarity and basic knowledge?

Next, you should prioritize each of the talent gaps you have
identified. Ask each of your stakeholders to rate the gaps that they
have insight into with a score of 1 to 5, where 1 indicates a severe
talent deficit holding back the business, and 5 indicates an area
needing only minor improvement. Combine their ratings to create a
prioritized list of talent needs that you can work with human resources
to grow for the future.

BRIDGE PLAN
You need to develop a plan to bridge these gaps in your talent and
support your future growth. Your plan should encompass each of the
six steps of the talent life cycle:

• Hire—How will you recruit new talent to meet your capability

needs? Where will you look (company, industry, geographic
location), and how will you attract people to work with you?
• Acquire—Can an acquisition help close your talent gaps? If so,
what kind of firm might you target? How will you manage its
integration to accelerate the change you need and not lose what is
best in the acquired company?
• Train—How can you best grow the capabilities of your current
people? What upskilling can benefit them in their current roles?
Where can reskilling enable you to redeploy talented people
whose current roles are winding down?
• Retain—How should you engage your best talent so they don’t
walk out the door? Do they have a compelling career path with
you? How can you ensure their work has autonomy, mastery, and
purpose?
• Exit—How much turnover is healthy for the pace of change you are
seeking? Is your voluntary attrition too high, or is it too low?
Should you realign incentives to avoid the “golden handcuffs” that
keep the wrong people in place? What combination of buyouts,
 layoffs, or reorganization is required to align resources with the
talent you need most?
• Partner—How will you look beyond your team or organization for
your talent needs? Which partners can support your strategy?
How can you best work with them, learn from them, hire from
them, or possibly acquire them? How will you design the best
ecosystem of talent inside and outside to help your business
thrive?

Culture

Growing the right capabilities for DX cannot stop with technology and
talent. It is just as critical to include culture. The right organizational
culture will be essential to the long-term success of any DX effort.
Having studied digital efforts at scores of businesses around the world,
I am convinced that it is impossible to achieve a lasting DX without
culture change. Fortunately, I have seen effective leaders who prove
that growing a digital-ready culture is possible in even the most long-
established business.

Aftenposten’s Journey

When I met Espen Egil Hansen, he was a rising star at Schibsted, a

Scandinavian media conglomerate aggressively steering its portfolio of
news and advertising businesses to adapt to the digital era. Hansen
had recently led a digital start-up for Schibsted called VG Nett, where
he built a thriving and profitable online business. For his next
challenge, Hansen was tapped to transform one of the oldest and most
cherished news institutions in northern Europe, the Norwegian
newspaper Aftenposten. When Hansen took control as the new CEO
and editor-in-chief, he knew that he faced huge cultural barriers to
change. He was dealing with a proud institution with a history dating
back to 1860; he knew that it had a strong sense of tradition and what
Hansen called “one of the most conservative cultures anywhere—like a
church!” Hansen knew that transforming the business of Aftenposten
for the digital era could only happen if key changes were made to its
culture. Specifically, Aftenposten needed a culture that was:

• Entrepreneurial—With a bottom-up mindset where everyone

contributed ideas.
• Collaborative—Where employees worked across the traditional
barrier of editorial versus business and across silos of different
reporting desks.
• Data-driven—Where decisions were made based on data and
testing rather than on the verdicts of the highest-ranking editors.
• Customer-focused—Where work centered on the customer’s
evolving digital experience rather than Aftenposten’s historic print
product.

On his first day as CEO, Hansen attended the editors’ morning

meeting. Aftenposten’s most senior editors gathered to review the prior
day’s paper and discuss what lessons should be learned. As Hansen
recalled to me, “I’m sure this meeting has been in place more or less
the same since 1860. I listened only. Then, at the end of that meeting, I
said ‘This has been very nice. From tomorrow we will try something
new.’ ” After 150 years, he canceled the editor’s morning meeting. The
next day, he introduced a new meeting that would start each day going
forward. Everyone was invited—the entire editorial team and every
single reporter but also the advertising sales team, the subscriptions
team, the technologists, and even the janitors. Everyone was asked to
be part of an open and frank discussion of what Aftenposten did well
the day before and to share their ideas on what could make it better.
Other changes quickly followed. To foster collaboration across
traditional roles, Hansen introduced multifunctional teams that
combined journalists and technologists and gave them a mandate to
collaborate with readers and try new forms of data-driven journalism.
He took advantage of Aftenposten’s long-planned building move to
design a space that would support a more collaborative culture. In the
old building, employees had been spread throughout ten floors, with
offices clustered in the corners. “We had 40 corners, and it was like 40
subcultures at Aftenposten,” Hansen told me. For the new building, he
pushed to bring everyone onto two large floors and to create a central
space on each one—not only for morning meetings but also as the only
location for coffee machines and trash receptacles. “The whole idea
was to fuel spontaneous meetings of different people in the
organization. It was not only symbolic—it really works!”
To support a shift to data-driven decisions, Hansen instituted a new
model where metrics from each department were distributed openly
and transparently among all employees. To push everyone to focus on
the customer’s digital experience, he stipulated that news stories
shown in meetings should appear in their mobile screen format—to
remind the editors (many who started in print) that this was how most
readers read their stories.
When Hansen stepped down after six years, Aftenposten’s business
was radically transformed. “We turned the business model upside down
—from 80 percent advertising revenue to 80 percent subscriptions. Not
many companies survive that kind of change,” he explained. But just as
important, the culture was changed too. “We have gone from being a
business that was a monopoly in effect, back to market competition.
We have become a data-driven organization from not being one.” And
the old hierarchical mindset had given way as well: “Very little at
Aftenposten today is decided from the top.” Hansen stressed when he
shared what he had learned from the DX, “The hard part is not getting
the technology right but changing our culture.”

Culture in DX

The centrality of culture is not unique to Aftenposten’s DX. At the New

York Times Company, culture was the Achilles heel of the paper’s early
digital efforts. Its infamous Innovation Report diagnosed that its
transformation was being held back by a risk-averse mindset in the
newsroom, a culture of silos, a woefully outdated focus on the print
edition, work that was organized around old traditions rather than the
changing habits of readers, and a failure to share and learn from
mistakes. It was only when the Times’s culture started to change that
its transformation began to truly show results.
It is impossible to drive real DX without building the right culture. If a
business is driven only by financial results and lacks a customer-
focused mindset, it will never spot the next customer problem to be
solved. If its teams rely on seniority rather than data to make decisions,
efforts to “test and learn” will simply become “test to confirm.” If its
people are afraid to fail or admit mistakes, they will never be able to
iterate, pivot, and adapt their innovations.
Even the biggest changes to process and governance will fail if they
are not accompanied by changes in culture. Dániel Nőthig, who
coaches large organizations on adopting agile software development,
described to me how many clients want to focus only on the mechanics
of agile. They diligently adopt all the right processes (daily scrum
meetings, two-week sprints, kanban boards, etc.). But the essential
mindsets of agile—to be iterative, self-directed, problem-focused, and
customer-centric—are never communicated or embraced. Nőthig calls
this, “Let’s do the process, but not the culture.” In all his coaching
experience, he has never seen it work.
Digital-native businesses around the world are obsessed with the
importance of culture to their success. If you spend time in Silicon
Valley, you will quickly find that leaders there talk a lot more about their
company culture than about their technology. When Satya Nadella
assumed the role of Microsoft CEO, he was chosen by the board to
implement a dramatic shift in strategy—from a servers-based offering
to a cloud-based one, and from Windows-only to partnering with any
operating system. Yet when he assumed leadership, Nadella spoke
most passionately about his efforts to change the culture of Microsoft.
As Nadella told shareholders a year into leading the turnaround, “Our
ability to change our culture is the leading indicator of our future
success.”27
Microsoft is not alone. All the dominant digital titans that arose in the
internet era—Amazon, Alphabet, Meta, Tesla, and Netflix—exhibit a
similarly intense focus on culture. Each company has its own
codification of its guiding principles, and each goes to great lengths to
tie its stated culture to the actual day-to-day workings of the company.

Culture as Behavior

What is culture? The word has grown increasingly popular in business,

but its meaning is often fuzzy and ill defined. Discussions of
organizational culture often lapse into vague and imprecise language
untethered to clear action or business needs. For years, this vagueness
made me skeptical of the term “culture,” but my thinking changed as I
studied organizations where culture had become a powerful lever for
transformation and growth. I saw that each of these organizations
defined their culture in terms of everyday things that people do.
One famous definition of organizational culture comes from Herb
Kelleher, cofounder and CEO of Southwest Airlines: “Culture is what
people do when no one is looking.”28 Kelleher’s wonderfully concise
definition captures two ideas: that culture is behavior (“what people do”)
and that it is established by a mindset and norms rather than rules
(“when no one is looking”). As one executive put it to me, “Our
company’s culture is the behaviors that we revert to by default.”
How do the people in your organization interact and collaborate?
What do people in your organization spend the most time on? How do
they make decisions and run meetings? The Silicon Valley venture
capitalist Ben Horowitz wrote an entire book about culture, which he
titled What You Do Is Who You Are. Culture is all about our daily habits,
interactions, and how we prioritize what gets done first. As Horowitz
points out, the typical corporate mission statement offers no guidance
on the kinds of decisions that employees make every day—yet these
small choices will define your organization and determine its ultimate
success or failure.29
How can you define the culture you want and make sure that culture
takes hold? In early start-ups, culture is defined by the founders and
their initial hires. As organizations grow, culture may be shaped by how
the founders talk about the business to employees and how they model
behavior. But what happens as a company grows larger, and the
founding CEO can no longer stand on a chair and address the entire
organization at the start of each morning? How do we shape culture in
a large and complex organization?
To reinforce and embed culture at scale, you need to examine
process. As Vince Campisi, CDO of United Technologies Corporation
(UTC), told me, “In large organizations, process codifies culture.” When
a leader tells me about their vision and expresses frustration that their
organization is not moving in that direction, I always ask them about
process: What are you doing to change what you measure, what you
reward, and what you ask people to do daily?
 Building a customer-centric culture is impossible if all your metrics
are based on short-term financial performance and not on the value
gained by your customers. A collaborative culture is impossible if your
organization is designed in functional silos, and your procurement office
takes eight months to approve a two-week pilot with a partner.
Entrepreneurial thinking and risk taking will be seen merely as
buzzwords if your bonuses and incentives punish employees for project
shutdowns and reward them for predictable results. For culture change
to take hold, processes must be carefully designed to support
transformation and not impede it.
In studying leaders like Hansen who transformed their
organization’s culture, I have observed a common pattern. Leaders
who shape culture do three things:

1. They define the culture they seek in a clear set of principles and
the behaviors they demand.
2. They communicate that culture with stories, symbols, and actions
that bring it to life.
3. They enable that culture with processes that support the right
behaviors.

Define Culture in Principles and Behaviors

Effective leaders define the culture they seek by consciously crafting

their organization’s principles. Google has its Innovation Pillars30 and
its “Ten things we know to be true.”31 Tesla has six leadership
principles, which are taught to all new employees in its training
programs.32 Southwest Airlines encourages employees to “Live the
Southwest Way” and “Work the Southwest Way,” clearly identifying the
components of that ethos (Warrior Spirit, Servant’s Heart, FunLUVing
Attitude, etc.).33 And as Ford Motor transforms for the digital era, it has
defined its “Five Rules of the Road,” including “Respect Knowledge
Over Hierarchy” and “Solve the Problem.”
A clear set of cultural principles describes the behaviors that
everyone in your organization should aspire to in their day-to-day
actions, interactions, and decisions. Well-written principles provide a
shared touchstone and a common vernacular among employees. They
are a means to hold each other accountable. Two of the best examples
I have seen are from Amazon and Netflix.
Amazon is famous for its sixteen Leadership Principles, which
define the unique culture of the company. These principles—which
include “Customer Obsession,” “Think Big,” and “Bias for Action”—have
been refined over the years and are published on the company’s jobs
website.34 Each principle is paired with a brief description of what it
means in everyday work. But the Leadership Principles are not just
published to be read. They are brought up in daily meetings when
decisions are debated on any aspect of the business. The principles
are “deeply embedded into the culture and the lens through which
candidates, employees, and teams are evaluated and rewarded,” in the
words of one former employee. “Meet an Amazonian and I guarantee
they can name them from memory and give an example of
demonstrating them in their time there.”35
Netflix’s thinking on its culture was made famous by a 124-slide
PowerPoint deck posted publicly by CEO Reed Hastings and viewed
online more than 1 million times. That early manifesto described the
nine behavioral traits that Netflix looked for in hiring and promoting its
people. At turns humorous, philosophical, and practical, it explained
why “brilliant jerks” would not be tolerated, laid out how employee
compensation should be calculated, and described a Keeper Test
(asking, “which of your direct reports would you fight for if they received
an outside job offer?”).36 Netflix’s latest statement of its cultural
principles, at over 4,000 words, is full of stories about the principles that
define Netflix’s culture and what these principles look like in action. The
principle of “Freedom and Responsibility,” for example, ties employees’
creativity to the company’s practice of minimal rules and oversight: “In
general, we believe freedom and rapid recovery are better than trying
to prevent error . . . [O]ur biggest threat over time is lack of
innovation . . . We are always on guard if too much error prevention
hinders inventive, creative work.”37
As you start to define your cultural principles, focus on the following
eight qualities, all part of the best cultural principles:

• Shared—Known by everyone in the organization.

 • Discussed—Referenced as a part of day-to-day discussions,
debates, and work.
• Actionable—Giving guidance in decisions, especially when
priorities compete and trade-offs must be made.
• Nuanced—Defined with some complexity as to what they mean for
your particular organization and what they look like in action.
• Dialectical—Existing in tension with each other, including slight
contradictions that can lead to healthy debate rather than settled
dogma.
• Relatable—Described in language that resonates with employees
from different geographic locations, genders, ages, identities, and
backgrounds.
• Tribal—Helping to define who you are and to instill a sense of
group identity: this is who we are; this is what makes us unique.
• Intrinsic—Defined in service of something beyond profit, whether
for the benefit of customers, employees, or society at large.

During times of change, an organization’s culture may be described

best in terms of a shift “from X to Y.” After stepping up as CEO at
Microsoft, Nadella spoke about changing the organization from a “know
it all” culture, where employees gained status through proving their
expertise, to a “learn it all” culture, where people are driven by curiosity
and humility to pursue constant learning and personal growth.
Think about your organizational culture. What is your culture today
—that is, how do your people currently behave? What ideas or
principles guide that behavior? Is that behavior in line with your shared
vision of your digital future? If not, how does your culture need to
change?
Table 7.2 lists seven shifts that I have observed most often in
organizations seeking to realign their culture to support their DX. If you
work in a legacy firm today, one or more of these profound cultural
shifts will likely be vital to your success in the digital era.

Table 7.2.
Common Cultural Shifts During Digital Transformation
From Predigital Culture . . . To Digital Culture
• From expert driven . . . • To data driven
• From siloed . . . • To collaborative
• From cautious . . . • To risk taking
• From planning is everything . . . • To experimentation
• From top down . . . • To bottom up
• From committee led . . . • To ownership
• From solution focused . . . • To customer
obsessed

Communicate Culture in Stories, Symbols, and Symbolic

Actions

Effective leaders do more than define the culture they want; they
constantly communicate what that culture is and why it matters. They
do so in a way that captivates employees’ thinking so that the culture is
embraced over time in habits and actions. As leaders communicate
about culture, their most powerful tools are not diagrams, graphs, or
bullet points. Instead, they communicate best through stories, symbols,
and symbolic actions. These are the best tools to shape how people
think and behave in organizations.

STORIES
Human brains have evolved to process stories, and thus stories are
essential to leading culture change. As the Nobel Prize–winning
psychologist Daniel Kahneman has said, “No one ever made a decision
because of a number. They need a story.”38 Aristotle knew that only a
good narrative—rather than facts or rote instructions—can
communicate complex meaning and connections in a way that humans
will readily and lastingly absorb.
Stories can shape organizational culture in many ways. They may
tie a cultural principle to the company’s guiding mission or root it in the
history of the firm. They may illustrate the consequences of failing to
practice a cultural principle and inspire greater vigilance. Or they may
spotlight unsung heroes whose day-to-day behaviors embody the
culture to which the business aspires.
 Nadella has used stories in all these ways to reinforce culture
change at Microsoft. He underscores the company’s commitment to
empowering others by telling the story of Microsoft’s very first product:
a BASIC interpreter for the Altair. In the 1970s, this was a critical tool
for the emerging community of computer hobbyists. As Nadella
explains, “Our mission of empowering every person and every
organization on the planet to achieve more is really a look back to the
very creation of Microsoft.”39 To underscore the impact of Microsoft’s
work today, Nadella used a Super Bowl ad about a very niche product,
an Xbox Adaptive Controller for disabled gamers. The device, which
came out of a hackathon, became a passion project for its team, which
spent a year designing the packaging so that kids with no hands could
open it themselves. On watching the Super Bowl ad, one engineer
commented, “I genuinely have never felt so much pride saying, ‘I work
at Microsoft.’ ”40 As Nadella pushes employees toward a growth
mindset, he often retells the story of his company’s missteps in mobile
and search, two huge opportunities that Microsoft failed to pursue early
enough. His message: stay humble and never stop learning.

SYMBOLS AND SYMBOLIC ACTIONS

Symbols can also communicate the desired culture of an organization.
Objects, words, and actions that illustrate an idea can be powerful
vessels for shared meaning. By choosing the right symbols, leaders
can continually remind an organization of the cultural norms it seeks to
uphold.
These symbols sometimes take the form of a ritual whose power
comes from repetition and shared experience throughout the company.
Amazon has a tradition of designating one empty chair at every
meeting to represent the customer. The chair serves as a reminder to
everyone to argue for the perspective of the customer in their
discussion—further ingraining the Amazon principle of “customer
obsession.” Under CEO Jim Hackett, Ford Motor held a Viking funeral
whenever an innovation project was canceled to commemorate the
team’s valiant efforts. This sent a clear signal that failed ventures are
still admirable—vital to learning and to the future of the business.
 Chosen carefully, language can be a powerful symbol—with
meaning carried by the words we use and those we avoid. At the
Washington Post newspaper, executive editor Marty Baron declared
that no job titles would be allowed with the word “digital” in them.
Baron’s point was subtle: he and his team talked constantly about
“digital” when discussing the strategy of the business. But by allowing it
in no one’s job title, he signaled that digital was now everyone’s job—
not the specialty of a specific division or group of new hires.41
Physical objects that appeal to our sense of sight and sound can
also be powerful symbols. As founder of Blinds.com, Jay Steinfeld
wanted to push his employees to take risks and not be afraid of making
mistakes as they worked to grow the business. He installed five-foot-tall
test tubes in the office and instructed employees to drop in a clear
marble every time they tried a new tactic to reach customers. If a test
proved successful, a colored marble was also added; if it failed, the
clear marble was left by itself. As Steinfeld kept pushing employees to
experiment, a pattern emerged for all to see: there were many more
clear marbles (attempts) for each colored one (success). The message:
take risks and don’t expect to get it right every time.42
A leader’s actions can be powerful symbols as well when chosen to
be visible and surprising and to demonstrate the culture you are
seeking. Recall how, on his very first day as CEO at Aftenposten,
Hansen canceled the 150-year tradition of a senior editors’ meeting.
That move was meant to cause a ruckus—and communicate to
everyone that change was coming fast to that history-laden institution.
Walt Disney was famous for picking up trash whenever he walked
around Disney theme parks.43 That simple action spoke louder than
any speech about the importance of everyone’s care in shaping the
customer experience. When Hackett instituted Viking funerals for
innovation projects at Ford Motor, he made sure to preside over them
himself, in full Viking helmet and dress.

Enable Culture with Processes

Defining your culture and communicating it to everyone in the

organization is critical, but it is still not enough to drive change.
Effective leaders enable the culture they seek by ensuring that
processes support—rather than hinder—the right employee behaviors.
This becomes increasingly important as organizations grow larger and
more complex. Leaders must continually look at the processes used to
run their business every day and ask, Are these processes helping or
hindering the culture we want?
When aligning processes with culture, I suggest starting with what I
call the big three: incentives, metrics, and resource allocation. In 1975,
Stephen Kerr published his classic management article, “On the Folly
of Rewarding A While Hoping for B.”44 The article identifies the greatest
source of internal resistance to most change efforts: employees are
rewarded for behaviors you want them to stop doing (defending legacy
operating models, lobbying for political decisions, etc.) and punished for
the behaviors you want (taking smart risks, pursuing new areas of
growth, etc.). The good news is that aligning incentives with the
behavior you seek can have a dramatic positive effect on culture.
Culture change should also be baked into the metrics companies
use to define, measure, and guide their work. This includes business-
level metrics and KPIs, as well as metrics for work at the team level. If
you want your employees to be more customer-centric, for example,
are you measuring how much time they spend talking to customers?
Resource allocation is equally critical to driving culture change. If
you want a culture of speed and flexibility, ask yourself, Is your
financing frozen in an annual budgeting cycle, or is it allocated more
frequently? Just as critical is how you allocate head count and shared
services (e.g., IT support, marketing campaigns, or legal counsel).
Guidelines for easily accessing outside resources can also prevent
bottlenecks and keep teams operating the way you wish.
The big three—incentives, metrics, and resourcing—are essential to
scaling culture, but they are only the start. Many other processes shape
culture as well: criteria for hiring and promoting talent; reporting lines,
approval processes, and accountability; communications and how
meetings are run; employee access to data and analytics tools; as well
as the design of office space and guidelines on hybrid and remote
work. All these processes and more can reinforce or undermine the
culture you are seeking. Leaders must thoughtfully review every
process that touches, constrains, or encourages employee behaviors if
they want to reshape culture at scale.
 We can see many examples of how process enables culture in
successful DX. Aftenposten enabled a more collaborative and data-
driven culture with its all-hands meetings, cross-functional teams,
transparent data, and redesigned office space new employees:
One of Ford Motor’s key cultural principles is “Solve the Problem.”
As Marcy Klevorn, chief transformation officer, explained to me, this
principle means that if you see something is not working, “you don’t
pass the problem to somebody else; it stops with you.” One process
Klevorn established to support this in Ford Motor Company’s culture
was a daily meeting called “Office Hours,” led by a cross-functional
group of leaders appointed by Klevorn. “Every morning they gathered
for an hour and anyone around the organization could bring a problem
to them. My only rule was the problem had to be solved in that hour. I
said, ‘As long as you don’t break any laws, I don’t care what you do to
solve that problem, I will back you 100 percent—even if I don’t agree
with you. Whatever you did to solve their problem, don’t let them leave
without a solution to get them to move forward. And I will back you!’
They even called me, ‘Are you sure?’ Yes, just go do it. I believe you
guys will make great solutions. And guess what, of course, they did!”
Digital-native companies also offer powerful examples of using
processes to enable culture. To emphasize the importance of speed in
action at Tesla, it was declared that meetings should have as few
attendees as possible. To encourage this, Tesla instituted a rule that
any employee should leave in the middle of any meeting as soon as
they conclude that their presence is not adding value. While shocking
at first, this behavior soon came to be seen as the normal way to run a
meeting.45 As part of Netflix’s culture of “freedom and responsibility,”
the company operates with minimal controls on contract signing and
expenses, leaving managers free to use judgment rather than wait for
sign-offs. It applies this same principle to transparency around
company documents. “Nearly every document is fully open for anyone
to read and comment on, and everything is cross-linked. Memos on . . .
every strategy decision, on every competitor, and on every product
feature test are open for all employees to read.”46
Amazon is particularly famous for using process to ingrain its unique
culture. We have seen how its two-pizza teams with clearly designed
metrics drive a culture of agility and accountability, and how its API
mandate ensures flexibility across the business. In addition, Amazon’s
culture of “customer obsession” is reflected in the metrics the company
reports to shareholders each year, which emphasize customer growth,
repeat business, and brand—rather than short-term financial results.47
The company’s entrepreneurial culture is enabled by a single sign-off
process for innovation ideas: rather than needing to work her way up a
line of command, a manager needs approval from only one senior
executive to begin the first steps on a new project.48 Perhaps most
famously, Amazon banned PowerPoint presentations from meetings
throughout the company. Instead, every presenter must prepare a
written six-page narrative laying out their proposal and offering data
arguing both for and against. The first half-hour of every meeting is set
aside for everyone to read the memo in silence, take notes, and write
their follow-up questions. The six-page memo was instituted precisely
because it forces a more thoughtful and data-driven approach to
decision making.

Tool: The Culture-Process Map

Our final tool is the Culture-Process Map (see figure 7.3). The purpose
of this tool is to help any organization grow a culture that supports its
strategy for the future and reflects its unique character. The Culture-
Process Map guides leaders through three essential tasks: defining the
culture they need, communicating that culture to others, and enabling
that culture by aligning business processes.
Figure 7.3.
The Culture-Process Map

1. Define Your Culture

Step 1 of the Culture-Process Map is to define your desired culture in

terms of a set of clear principles and the behaviors they require. (Think
of Ford Motor Company’s principle, “Solve the Problem,” and the
required behavior, “don’t pass the problem to somebody else; it stops
with you.”) Focus on the most important principles that will have the
greatest impact. Some of these may be long-standing, but others
should be new principles that define important shifts in culture for the
future.

FROM . . . TO
Start by defining a few shifts in culture that are most needed for your
organization to thrive in the future. In every legacy organization that I
have worked with, the pre-digital-era culture has needed to evolve for
the digital era. Try writing these culture shifts as “from . . . to”
statements. Think of Nadella’s push to change Microsoft from a know-
it-all culture to a learn-it-all culture. Review table 7.2 for the list of
common culture shifts in DX. Are any of these shifts relevant to your
organization? In most legacy businesses, you’ll find more than one.

THIS, NOT THAT

Next, identify aspects of your current culture that you want to keep and
reinforce. What makes your culture distinctive and powerful? Try writing
contrasting statements in the form of “We are this, not that.” One
example is, “We have a bias for action, not for deliberation.” Think of
these as hiring criteria that would help ensure a culture fit for new
employees: if you are “this” you can thrive here, but if you are “that,”
you won’t do well, even if you are skilled and talented. Avoid generic
statements that would be universal to any company, for example, “we
are hard-working, not lazy.” Your “that” should be a quality you will
avoid but that a different organization might choose to prioritize in its
own culture and hiring.

PRINCIPLES
Combine all of the positive ideas from what you wrote (your “to” from
each “from . . . to” statement and your “this” from each “this, not that”)
to create one list of principles you aspire to. Give a name to each
principle on your list. Pick a short, memorable phrase (two to five
words) that uses simple language, for example, “Customer Obsession”
(Amazon), “Learn It All” (Microsoft), “Solve the Problem” (Ford Motor),
or “Freedom and Responsibility” (Netflix). Then, for each principle, add
a longer description of 50 to 500 words about what it looks like in the
context of your business. What behaviors do you expect people to do
or not do in their work? What should they prioritize? Here you can
include the negatives that you seek to avoid (e.g., the “from” in your
“from . . . to” statements and the “that” in your “this, not that”).
 Next, test your principles against the eight qualities of cultural
principles that we saw earlier in the chapter. Are your principles
actionable? Are they relatable to anyone in your organization? Do they
have intrinsic value and motivation? Revise or eliminate any principles
that don’t measure up well to these qualities. Pick the principles that
are most important to your culture. If two seem closely related, try
combining them. Try for a short list (e.g., five to ten principles). Your
goal is to focus on a few critical levers of culture, not a laundry list of
everything about your organization.

2. Communicate Your Culture

Step 2 of the Culture-Process Map is to communicate your principles in

ways that bring them to life as more than just slogans. By linking them
to memorable stories, symbols, and actions, your cultural principles
become something employees will recognize and strive for in their
actions. Think of Ford Motor Company’s Viking funerals or Microsoft’s
story of its first product, which exemplifies the principle of empowering
people.

STORIES
Look for specific stories you can tell to bring your cultural principles to
life. Some stories may relate to a single principle, but others may
illustrate more than one. Look for different types of stories, including
these classics:

• Origin stories—Relate your aspirations today to the early

beginnings of your organization.
• Mission stories—Express your culture through the impact of your
work on others.
• Failure stories—Capture when you fell short of the culture you
aspire to and remind everyone of the work to be done.
• Spotlighting stories—Highlight the everyday actions of people
throughout your organization and how these uphold your culture
and make an impact.
 SYMBOLS AND SYMBOLIC ACTIONS
Look for symbols to illustrate your cultural principles and make them
visible every day at work. Look for examples from each of these types
of symbols:

• Rituals—Actions you regularly perform together that symbolize

some aspect of your culture.
• Language—Words and names that you consciously use or omit to
reinforce a point about your culture and mindset.
• Sights and sounds—Objects, images, or media seen regularly by
employees that remind them of a cultural principle.
• Symbolic actions—Actions you can take as a leader to surprise
and draw the notice of others and that clearly illustrate one of your
cultural principles.

3. Enable Your Culture

Step 3 of the Culture-Process Map is where you align your

organization’s processes with the culture that you seek. This starts with
assessing your current processes. Where are they enabling your
cultural principles? Where are they blocking or hindering them?
Remember that process is critical to scaling culture change. Think of
Netflix’s open document policy or Amazon banning PowerPoint from its
meetings.

BIG THREE PROCESSES

Start with the big three processes and consider how each one currently
shapes your culture:

• Incentives—What are you rewarding employees for doing? What

are you recognizing and praising them for? What do you punish
them for? Where do these incentives match your cultural
principles, and where do they differ?
 • Metrics—What are the metrics you use to manage your business?
What metrics do you use to evaluate individual and team
performance? Where do these metrics align with your intended
culture, and where do they diverge from it?
• Resource allocation—How, when, by whom, and on what basis is
funding disbursed? What about head count or other scarce internal
resources? Do these processes enable the culture you are
seeking, or do they pose obstacles?

For each process, write a stop/keep/start analysis:

• Stop—What current process is impeding the culture you want?

How will you change it?
• Keep—What current process is reinforcing the culture you want?
How will you spread it more broadly?
• Start—What new process could you add to enable the culture you
want? How will you implement it?

OTHER PROCESSES
Next, look more broadly at any other rules or processes that shape
behavior in your organization. Again, consider how each one promotes
or contradicts your cultural principles. These should include the
following:

• Hiring and promoting—How do you hire and acquire new talent?

How do you choose whom to promote into leadership? How can
these processes strengthen your culture rather than dilute it?
• Accountability and approvals—Who signs off on important
decisions? Where does decision making reside? Does this support
the culture you want among employees and teams, or does it
undermine it? Where do bottlenecks occur, and when do the costs
outweigh the benefits?
• Communications and meetings—What processes do you use for
daily communication? How are meetings typically run? Who
attends these meetings and why? How does this match or not
match the culture you want?
 • Tools and data—What tools do employees have access to for their
work? How is data shared across teams, silos, and divisions? Do
employees have what they need to do their best work and live up
to your cultural principles?
• Workspace—What is the design of your workspace, both physical
and virtual? What work do employees do virtually versus in
person? What written and unwritten norms are set for this? How
does that enable or hinder your culture?

Again, try writing a stop/keep/start analysis for each process. What

processes do you need to stop, keep, or start in your organization to
support your culture fully?

How to Use the Culture-Process Map

Unlike other tools, the Culture-Process Map should not be applied in

the same way at every level of the organization (company, business
unit, team) because your cultural principles should be shared by the
entire organization. You do not want a different culture in each part of
the business! Therefore, in step 1 (defining your principles), top
leadership must guide the process. They should involve as many
stakeholders as possible from all levels, bringing their input and
perspectives to what the firm’s culture should be. But the final principles
should be approved at the top and shared throughout the firm.
By contrast, step 2 (communicating the culture) should be
everyone’s job. Each business unit, function, and team should
continually look for stories and symbols that illustrate the cultural
principles. And each should share them so different parts of the
organization can learn from each other.
Step 3 (enabling the culture through process) should also be
everyone’s job. At every level, leaders should be looking at processes
and asking how they align with the desired culture. I recommend a
formal review once a year, with input from stakeholders inside and
outside the company. Ask your customers and your partners, How is it
to do business with us? Where are we upholding our professed culture
in our interactions with others? Where could we do better? Culture is a
journey that is never complete!
 Capabilities for a Bottom-Up Organization

Any digital strategy for the future requires investing in and growing new
digital capabilities. But for real transformation to happen, businesses
must focus on capabilities that will enable them to operate in a more
bottom-up fashion, with change driven by employees at every level of
the organization. Rather than building capabilities in service of a top-
down master plan, this calls for building capabilities that foster a more
empowered organization.
A bottom-up organization calls for investing in specific kinds of
technology capabilities. This means modular IT architecture with APIs
and microservices that are not just scalable and resilient but that offer
greater flexibility for every team to work faster and connect with others
outside the firm. It means data assets that are not just rich in data from
diverse sources but synchronized and accessible to people across the
organization. It means IT governance that provides oversight from the
center but allows application and innovation by every team at the edge
of the organization.
A bottom-up organization calls for growing specific talent
capabilities as well. This means embedding new digital skills
throughout the organization rather than hoarding them in central
functions. It means staffing teams with the diverse mix of skills that they
need to be self-sufficient and training teams in iterative experimentation
with methods like agile, lean start-up, and design thinking. And it
means attracting and retaining the best digital talent by giving them
autonomy, mastery, and purpose in their work.
A bottom-up organization also calls for cultivating the right culture.
This means defining cultural principles and behaviors for bottom-up
action—such as risk taking, autonomy, and responsibility. It means
engaging everyone in discussing and debating those principles,
applying them in every project and meeting, and telling their own
stories about them. And it means shaping every business process to
enable employees to show initiative and act independently in support of
your shared vision and strategy.
No digital transformation can succeed in the future if it relies on the
capabilities of the past. Without capabilities that match the ambition of
your digital strategy, the best vision for transformation will forever
remain a distant goal, just out of reach. In Step 5 of the DX Roadmap,
we have seen how any organization can grow the technology, talent,
and culture it needs to deliver lasting growth. We have seen how the
right IT infrastructure, data, and tech governance can transform work
and unlock the potential of teams at scale. We have learned why digital
skills require managing the entire talent life cycle. And we have seen
the power of building a culture that is aligned with your digital vision,
and that is embraced by all and enabled by every business process.
At the start of this book, I identified five steps to successful digital
transformation—vision, priorities, experimentation, governance, and
capabilities. Over the last five chapters, we have seen how the DX
Roadmap can enable any organization to master each of these critical
elements and find its unique path to growth. Through the five steps of
the DX Roadmap, your organization can define a shared vision of your
digital future, pick your most important strategic priorities, rapidly
validate new ventures, manage growth at scale within and beyond your
core, and grow the capabilities essential to your long-term success.
As I have emphasized throughout this book, the DX Roadmap is not
a linear process, with centralized planning leading to completion at an
ordained end date. DX has no finish line because its goal is to create a
more adaptive organization able to respond to waves of change that
will not abate. The Roadmap process itself is therefore iterative: you
adapt your vision, priorities, ventures, governance, and capabilities as
you go, based on what you learn.
Real DX must engage every level of the organization and every
business function. Each step of the Roadmap, properly applied, should
work in service of building a more bottom-up organization—where
leaders push decision making down, where insights flow from the
bottom up, and where strategy and innovation happen at every level. In
the book’s conclusion, we will return to this theme of the bottom-up
organization and reflect on the dramatically different kind of leadership
it demands in our dynamic digital era.
 Conclusion

Three truths appear in all the stories we have seen about companies
on the path to digital transformation (DX).
First, DX is iterative. It cannot start with a long planning process
followed by dutiful execution of predetermined steps. Instead,
successful transformation starts small, learns from what works, and
adapts as it goes.
Second, DX has no end date. It is not a two- or three-year project to
complete. DX is an ongoing evolution to become a more adaptive
organization, one that can thrive in an era of accelerating change.
Third, DX must happen from the bottom up. It cannot be centrally
planned and directed by a CEO or CDO acting alone. It must be driven
at the same time by the vice president of supply chain, the head of a
country unit, the email marketing manager, and the lead of a single
product team. Change must come from every part of the organization
and every job title.
This last point is particularly critical. Throughout the book, we have
seen how each step of the Digital Transformation (DX) Roadmap is
based on the idea of a bottom-up organization: a shared vision
“cascades up” the organization, not just down. Strategic problems and
opportunities are defined at every level. Digital ventures start in every
business function and are given the resources and governance to
scale. Every team is empowered by having access to data and
technology, the skills to move fast, and a shared culture of ownership
and accountability.

From Top-Down to Bottom-Up

The bottom-up model is a radical shift from the top-down model of

management that defined large organizations in the twentieth century.
Top-down management theory was built on an inherent trade-off. It
sacrifices agility and speed for command and control. It is inflexible by
design, with fixed processes and employees slotted into siloed roles.
Work is designed to ensure predictability, consistency, and efficiency at
scale. Numerous cross-departmental meetings are required to approve
any change in standard practice. In many large organizations, the
biggest impediment to speed is not resources or external constraints
but the decision-making process itself. That trade-off of speed for
consistency may have made sense in a more predictable, slow-
changing business environment. But in the digital era, organizations
need a different model that enables rapid change.
Today, we see leading organizations switching away from the top-
down style of management. The U.S. military was famous for
management by command and control; indeed, it was where many
twentieth-century management theories were first tested. But that same
top-down decision making is untenable in our twenty-first-century world
of volatility, uncertainty, complexity, and ambiguity (VUCA). As
described in Stanley McChrystal’s book Team of Teams, unpredictable
threats and competitors forced the U.S. Army to shift to a new model
based on bottom-up decision making from small teams that are highly
aligned but loosely coupled.1
Digital-native businesses like Amazon, Netflix, and Alibaba have all
embraced the bottom-up model. As they grew from start-ups, these
companies achieved speed and flexibility at scale by embracing three
key principles:
First, decision making is pushed down to the lowest level. One of
the bedrock principles of agile software methods is the use of self-
organizing teams that, once given clearly defined goals, are allowed to
figure out on their own how to best achieve them. Netflix’s culture
document states: “We pride ourselves on how few, not how many,
decisions senior management makes.” It goes on to explain, “We
believe we are most effective and innovative when employees
throughout the company make and own decisions.”2
Second, information flows up and down the organization and from
the outside in. The most critical market knowledge sits at the periphery
and is usually sensed first in the lower levels of the organization chart.
Intel’s former CEO Andy Grove told the story of how his frontline
marketers and plant supervisors detected a shift in customer demand
away from memory chips—and even started responding to this shift—
fully two years before his top management team recognized what was
happening. “Our most significant strategic decision was made . . . by
the marketing and investment decisions of frontline managers who
really knew what was going on.”3
Third, innovation and change originate at every level. Digital-native
businesses embrace a model where leadership sets high-level strategy,
but it is up to individual teams to set the mission, vision, and strategy
for individual products. Amazon’s most profitable business, Amazon
Web Services (AWS), started with an idea from a junior-level engineer,
Benjamin Black, sketched in a six-page memo. The New York Times’s
approach to data-driven journalism was kicked off by an intern named
Josh Katz when he took an assignment to write an article on a scientific
paper and turned it into an interactive quiz that became the most-read
article of the year. As Jeff Bezos of Amazon explains, “Distribution of
invention throughout the company—not limited to the company’s senior
leaders—is the only way to get robust, high-throughput innovation.”4

Rethinking Leadership for the Bottom-Up Organization

If the digital era requires a shift from prioritizing control to prioritizing

speed and autonomy, then our old model of leadership must change as
well. In the past, the leader was decider in chief. But in the digital era,
the goal of a leader should be to make as few decisions as possible.
Every leader will inevitably have to make some decisions—but these
should be only the toughest and most important ones, when that level
of leadership is truly required.
But what does that leave for the leader to do? In my own work and
research, I have identified three essential jobs of leaders in successful
bottom-up organizations: defining, communicating, and enabling. We
have seen examples of each of these leadership roles throughout the
DX Roadmap, from vision to priorities, experimentation, governance,
and capabilities and culture. Together, these jobs define a new,
emerging model of leadership in the digital era, as shown in table C.1.
Table C.1.
The Three Jobs and Three Roles of Leaders in the Digital Era
Job of Leaders Role of
Leaders
Define a vision of where we are going and why Leader
as
author
Communicate that vision in words, stories, Leader
symbols, and actions as
teacher
Enable others to bring that vision to life Leader
as
servant

Define

The first job of a leader is to define a vision of where the organization is

going and why. This could be a statement of purpose or North Star
impact that the business wants to achieve, a statement of strategy with
key opportunities or problems to be solved, or a statement of the
culture that the organization is striving for. The aim is to provide guiding
direction to the efforts of others and to inspire them with a clear sense
of purpose for their work.
A great leader will articulate where you are going and why—but not
how you should get there. They set the context for others to act—but
do not craft their operating plans. They will vigorously debate questions
of strategy but trust others to execute and choose the right tactics. “If I
say ‘alignment,’ you could think that means ‘Do as the leadership tells
you,’ ” Espen Egil Hansen says. “But that is not it at all. What you don’t
want is leaders that go into details that hinder innovation and learning
and speed.”
In defining the organization’s vision, the leader takes on a role much
like an author. They start by studying and learning from many inputs,
inside and outside the organization. This requires getting in the
trenches and listening to the perspectives of customers, partners, and
employees at every level. Steve Jobs was famous for reading and
answering email sent to steve@apple.com so that he could learn from
customers directly. The leader must then work to synthesize all these
perspectives and insights. Then they must simplify them—looking for
common themes and a central thread—to define a future vision that will
guide the actions of others.

Communicate

The second job of a leader is to communicate their vision to every

stakeholder inside and outside the organization. Effective leaders
communicate both what their vision is and why it matters. Or, as Lucy
Kueng summarizes it, “This is our problem, where we are going, and
why it is necessary.”5
Leaders communicate these ideas in carefully chosen words,
stories, and symbols to make sure the ideas are clear and
unforgettable. In his very first letter to shareholders in 1997, Bezos
introduced the phrase “Day 1,” which became a rallying cry within
Amazon for long-term thinking and a mindset that Amazon is always at
the start of its journey. As Bezos wrote two decades later, “Day 2 is
stasis. Followed by irrelevance. Followed by excruciating, painful
decline. Followed by death. And that is why it is always Day 1.”6
Leaders not only communicate; they overcommunicate. They don’t
just tell their story to employees on the biggest stage at their largest
annual meeting. Leaders communicate through every tool and means
possible—in public forums and private conversations. They repeat the
same ideas and themes over and over. They are relentless. At Acuity
Insurance, CEO Ben Salzmann is in a perpetual state of
communication with employees and partners alike—from social media
posts to company meetings in a theater built in-the-round in their
headquarters, to his weekly “gossip” audio messages sent by voicemail
to every employee.
Great communicators don’t just speak; they ask questions, and they
listen. Rather than declare “We need to do X!,” leaders ask “How might
we achieve Y?” They are approachable and regularly step out of their
office to seek the perspectives of customers, partners, and employees
at every level. By listening, they find out if others have heard them—
and if not, why not? What barriers remain? Leaders remember the
adage “The single biggest problem in communication is the illusion that
it has taken place.”7
Through it all, leaders recognize that every interaction they have—
with an employee, a customer, or a shareholder—is a chance to teach
what they believe about their organization. As social entrepreneur
Wendy Kopp says, “Leadership is teaching.”8 Or, in the words of
Antoine de Saint-Exupéry: “If you want to build a ship, don’t drum up
the people to gather wood, divide the work, and give orders. Instead,
teach them to yearn for the vast and endless sea.”9

Enable

The third job of a leader is to enable others to bring their vision to life.
This, too, is a reversal of command-and-control leadership. Bottom-up
leaders lead less by their own actions than by enabling the actions of
others.
The first way that leaders enable others is by removing obstacles or
roadblocks to their work. As Citibank’s Valla Vakili, puts it, “Legacy
processes, legacy ways of working. These are often just very simple
things that get in the way of the creativity and productivity of the people
you already have.” It is a leader’s job to align every process they
approve—from compensation to organizational structure, to metrics
and KPIs—with the vision they have defined and communicated.
Processes carelessly chosen or unthinkingly inherited can undermine
culture, strategy, and the most purposeful efforts of your employees.
Leaders enable the work of others through their choices of who to
hire and who to promote to key positions. Leaders give those people
the autonomy to act and provide them with the tools and technologies
that will make a difference in their work. Leaders help their people grow
their own abilities, developing the talents they need to realize the goals
they set for themselves. And leaders advocate internally for those who
need support to make change happen.
Leaders empower others by getting them the resources they need.
This starts with the allocation of financial capital in support of strategic
priorities—that is, putting their money where their mouth is. They also
empower by allocating human capital—the most vital resource for any
new initiative. The last resource a leader must allocate is their mental
bandwidth. When the New York Times Company committed to a digital-
first future, one of the key changes was to reallocate the attention of
top leaders. The paper’s long-cherished print edition, would continue to
be produced (because its subscribers were among the most loyal, and
its advertising business was profitable though declining). But the
company was reorganized so that news would be developed in a
digital-first approach, with a separate team pulling from that digital
content to assemble each day’s print edition. A few experienced
managers were charged with overseeing the declining print operation.
This allowed the rest of the Times’s leadership to focus all their
attention on future growth areas such as audio, video, paid apps, and
the all-important digital news subscriptions.
By enabling others—through attention, resources, advocacy, and
the removal of obstacles—leaders act in service of others in the
organization. This third job of leaders resembles Robert K. Greenleaf’s
concept of “servant leadership”—which defines leadership as service to
the needs and growth of others.10
Rethinking leadership around these three roles—author, teach,
servant—may be hard for longtime leaders. Those who have risen
under an older model may feel they are giving up power and influence
as defined by the old model of control. But a deeper understanding of
bottom-up leadership reveals that power and influence are simply
reconfigured in the three roles.
A leader’s instinct to act as decider in chief can be especially hard
to resist during a digital transformation, as countless new decisions
must be made—related to investments, hiring, workflows, technologies,
products, and more. But instead of lapsing into old habits, leaders
should focus on pushing these responsibilities down in the
organization. Find your most talented people to do the highest-level
planning. And push all individual decisions down to the level of the
people who are actually doing the work. By defining a vision,
communicating it powerfully, and enabling others to bring it to life, the
leader plays an essential role in driving change.
In the digital era, the need for transformation is inescapable. But while
DX is hard, it is possible for any organization. By following the five
steps of the DX Roadmap—defining a shared vision; picking the
problems that matter most; validating new ventures; scaling change
with new governance; and growing tech, talent, and culture—any
organization can grow and transform for the digital era.
I wish you well in your own efforts as you find your path forward.
Remember, as you pursue new digital ventures, always focus on
impact and value creation, never on technology for its own sake. And
as you work to transform your people, processes, and culture for the
future, remember that the journey never ends. Change is always
ahead.
 MORE TOOLS FOR YOUR BUSINESS

You can find additional resources for applying the Digital

Transformation (DX) Roadmap by visiting www.davidrogers.digital
and subscribing to my newsletter. These resources include the
following:

• Downloadable PDFs of the strategic tools in this book and my

prior books
• Video tutorials on applying key concepts from the DX Roadmap
• New case studies, interviews, and research
• My latest advice on leading digital transformation

I hope you will subscribe, send me your questions, and keep in

touch!
 SELF-ASSESSMENT:
Is Your Organization Ready for DX?

This assessment tool is designed to uncover areas of strength

versus weakness in your organization as it seeks to adapt and grow
in the constantly changing digital era.

Who should complete the assessment?

I recommend having a wide range of managers and executives
complete the assessment tool while keeping their individual identities
and responses confidential.

How to complete the assessment?

Each question is presented as a pair of contrasting statements.
Read each pair and reflect on the current state of your own
business. Choose the number on the scale from 1 to 7 that you think
reflects where your organization stands in relation to the two
statements: 1 indicates fully aligned with the statement on the left, 7
with the statement on the right.

How to score the assessment?

Questions with a low score (1–3) indicate current weakness in the
organization that will pose challenges for any DX effort. Questions
with a high score (4–7) indicate an area of greater strength in the
organization.
As you combine responses from multiple employees, be sure to
capture not just the average score on each question but also the
range of scores and frequency of each score. Some of the greatest
insights may come from a question that produces divergent answers
(e.g., many responses of 1 or 2, and many of 6 or 7).

Discussion and insights:

After creating a report on the numerical answers, convene a forum
with those who completed the assessment tool. Use an independent
moderator to review the scores and lead a discussion about why
respondents answered as they did. This guided discussion should
reveal important challenges facing the organization that may not be
apparent to senior leadership. The insights from this assessment tool
can then be used to focus your own efforts to implement the Digital
Transformation (DX) Roadmap.
Vision (Step 1)
Our employees fear 1 Employees at every level
change and lack a clear 2 of our business
sense of where the firm 3 understand our digital
is going. 4 agenda and push it
5 forward.
6
7
Backing for our digital 1 Support for our digital
investments is weak 2 investments is strong
from investors, CFO, 3 from investors, CFO, and
and P&L heads. 4 P&L heads.
5
6
7
Our digital initiatives are 1 We only invest in digital
generic, following the 2 initiatives where we have
examples of peers. 3 a competitive advantage.
4
5
6
7
We use generic digital 1 The business impact of
maturity metrics to 2 our digital efforts is
guide efforts. 3 clearly defined, with
4 metrics to measure and
5 track results.
 6
7
Our firm follows the 1 Our firm leads the
market, reacts to others, 2 market, alert to critical
and is surprised by new 3 trends while there is time
entrants. 4 to choose a course.
5
6
7
Priorities (Step 2)
Our digital 1 Clear priorities provide
transformation is a 2 direction to digital
series of scattered 3 transformation across our
projects with no clear 4 organization.
direction. 5
6
7
Our digital efforts are 1 Our digital efforts are
defined by the 2 defined by the problems
technologies they use. 3 they solve and
4 opportunities they
5 pursue.
6
7
Our digital efforts are 1 Our digital efforts are
focused solely on 2 focused on future growth
operations, cost cutting, 3 as well as improving our
and optimizing our 4 current business.
current business. 5
6
7
A few people in our 1 Each of our departments
organization drive digital 2 is pursuing its own digital
while the rest stick to 3 ventures, with a backlog
old ways of working. 4 of ideas to try next.
5
 6
7
Our transformation is 1 Our transformation is
disconnected from 2 linked to the needs of the
business needs and 3 business and gaining
losing support over 4 support over time.
time. 5
6
7
Experimentation (Step 3)
Our approach to 1 Our approach to
innovation is focused on 2 innovation is focused on
coming up with a few 3 testing many ideas to
great ideas. 4 learn which work best.
5
6
7
Important decisions are 1 Important decisions are
made based on 2 made based on
business cases, third- 3 experimentation and
party data, and expert 4 learning from the
opinion. 5 customer.
6
7
Once our teams start a 1 In any project, our teams
project, they are 2 stay focused on the
committed to building 3 problem but flexible on
the solution in full. 4 the solution.
5
6
7
We view failures as 1 We keep our failures
costly, so our fear of 2 cheap and maintain a
taking risks is high. 3 bias toward risk taking.
4
5
 6
7
Even our good ideas 1 Our good ideas grow fast
move slowly and don’t 2 and deliver business
seem to move the 3 value at scale.
needle on our business. 4
5
6
7
Governance (Step 4)
A top executive must 1 We have established
personally approve any 2 structures that provide
new innovation for it to 3 resources and
happen. 4 governance for
5 innovation.
6
7
Our new ventures move 1 Our new ventures move
slowly and are led by 2 fast and are led by highly
traditional teams in 3 independent,
functional silos. 4 multifunctional teams.
5
6
7
Allocating resources to 1 Our resource allocation
new ventures is slowed 2 happens quickly through
by our annual budgeting 3 iterative funding.
cycle. 4
5
6
7
Our innovation is limited 1 We have a steady
to a few big projects, 2 pipeline of innovations,
and they are hard to 3 and it is managed with
shut down once they 4 smart shutdowns to free
are started. 5 up resources.
 6
7
The only new ventures 1 Our governance model
that gain support are 2 supports ventures with
low-risk innovations in 3 low and high uncertainty,
our core business. 4 both in our core and
5 beyond it.
6
7
Capabilities (Step 5)
Our inflexible IT 1 Our modular IT systems
systems reinforce our 2 integrate across our
silos and limit our 3 organization and connect
collaboration. 4 easily with outside
5 partners.
6
7
Our data is 1 Our data provides a
contradictory, 2 single source of truth to
incomplete, and 3 managers across the
inaccessible to 4 company.
managers in real time. 5
6
7
Our centralized IT 1 Our IT governance
governance causes 2 provides oversight while
bottlenecks for new 3 keeping innovation in the
projects. 4 hands of the business.
5
6
7
Our employees lack 1 Our own employees can
digital skills, so digital 2 build and iterate digital
projects must be 3 solutions.
outsourced. 4
5
 6
7
Our top-down culture 1 Our culture and
and bureaucracy stifle 2 processes empower
employees and breed 3 employees to drive
cynicism and inertia. 4 bottom-up change.
5
6
7
CASES AND EXAMPLES BY INDUSTRY

Agribusiness
Deere & Company

Airlines
Southwest Airlines

Automotive
Ford Motor Company
Tesla
Volkswagen

Banking and Financial Services

Citi
Intuit
Itaú Unibanco
Mastercard Inc.
Nasdaq
National Commercial Bank (NCB)

Consumer Durables
BSH Home Appliances
Haier

Consumer Electronics
Apple
Nokia
Samsung Group

Consumer Internet
Facebook/Meta
Google/Alphabet
Zoom

Consumer Packaged Goods

PepsiCo
Procter & Gamble

Design
IDEO

Energy
Schlumberger
Fashion and Apparel
Nike

Government
U.S. Army

Health Care
Bristol Myers Squibb
Johnson & Johnson
Merck Animal Health
Pfizer
Zoetis

Hospitality
Airbnb

Industrial Manufacturing
Air Liquide
BASF
General Electric
United Technology Co./Raytheon Technologies

Insurance
Acuity Insurance

Journalism
Aftenposten
Huffington Post
New York Times
Wall Street Journal
Washington Post

Logistics
CMA CGM

Marketing Technology
Adobe
Optimizely

Media and Entertainment

Axel Springer
CNN/Warner Bros. Discovery
Netflix
Schibsted
Sony Group
The Walt Disney Co.
YouTube

Membership Services
AARP
Canadian Automobile Association
FIA

Nonprofit
The Gates Foundation

Restaurants
Domino’s Pizza, Inc.
McDonald’s
Panera

Retail
Alibaba
Amazon
Walmart

Start-Ups
Blinds.com
Colombier Group
CupClub
Diapers.com
Muuse
Qikfox
VG Nett

Technology Hardware
Intel
Thomas A. Edison, Inc.
Xerox

Technology Services
Amazon Web Services
Ant Financial
Cisco
HCL Technologies
IBM
Microsoft
SAP

Telecom
Safaricom
Virgin Media 02
Vodafone

Transportation
Uber
Waze
 VISUAL OVERVIEW OF THE DX
PLAYBOOK AND DX ROADMAP

Figure A.1.
The DX Playbook overview
Figure A.2.
The DX Roadmap overview
 NOTES

1. The DX Roadmap
1. Rachel McAthy, “Pulitzer Goes to New York Times
‘Snow Fall’ Journalist,” April 16, 2013,
https://www.journalism.co.uk/news/new-york-times-digital-
snowfall-feature-wins-pulitzer/s2/a552683/.
2. Kyle Massey, “The Old Page 1 Meeting, R.I.P.:
Updating a Times Tradition for the Digital Age,” New York
Times, May 12, 2015, https://www.nytimes.com/times-
insider/2015/05/12/the-old-page-1-meeting-r-i-p-updating-a-
times-tradition-for-the-digital-age/#more-10891.
3. Gabriel Snyder, “The New York Times Claws Its Way
into the Future,” Wired, February 12, 2017,
https://www.wired.com/2017/02/new-york-times-digital-
journalism/.
4. Joshua Benton, “The Leaked New York Times
Innovation Report Is One of the Key Documents of This
Media Age,” Nieman Lab, May 15, 2014,
https://www.niemanlab.org/2014/05/the-leaked-new-york-
times-innovation-report-is-one-of-the-key-documents-of-this-
media-age/, 44.
5. Amy Watson, “New York Times Company’s Revenue
2021,” Statista, March 21, 2022,
https://www.statista.com/statistics/192848/revenue-of-the-
new-york-times-company-since-2006/.
6. “NYT Innovation Report 2014,” Scribd, March 24,
2014, https://www.scribd.com/doc/224332847/NYT-
Innovation-Report-2014, p. 72.
 7. BCG Global found that 70 percent of digital
transformations fall short of their objectives, often with
profound consequences. See Patrick Forth et al., “Flipping the
Odds of Digital Transformation Success,” BCG Global,
October 29, 2020,
https://www.bcg.com/publications/2020/increasing-odds-of-
success-in-digital-transformation. McKinsey’s research found
that more than 70 percent fail, and only 14 percent succeed in
a sustainable way. See Hortense de la Boutetière, Alberto
Montagner, and Angelika Reich, “Unlocking Success in Digital
Transformations,” McKinsey & Company, October 29, 2019,
https://www.mckinsey.com/business-
functions/organization/our-insights/unlocking-success-in-
digital-transformations.
8. Steve Lohr, “G.E. to Spin off Its Digital Business,” New
York Times, December 13, 2018,
https://www.nytimes.com/2018/12/13/business/ge-digital-
spinoff.html.
9. Anand Birje and David Rogers, “Digital Acceleration for
Business Resilience,” HCL Technologies, 2021,
https://www.hcltech.com/digital-analytics-
services/campaign/digital-acceleration-report-2021.
10. Lauren Forristal, “Disney+ Reaches 164.2m
Subscribers as It Prepares for Ad-Supported Tier Launch,”
TechCrunch, November 8, 2022,
https://techcrunch.com/2022/11/08/disney-reports-fourth-
quarter-results-2022/.
11. CEO slide deck:
https://businessleadersformichigan.com/wp-
content/uploads/2016/11/Patrick-Doyle-Presentation-
FINAL.pdf highlights several of these innovations, plus: Share
price: $4.97 (2009) to $155.01 (2016). See Patrick Doyle,
“Failure Is an Option—Business Leaders for Michigan,”
Business Leaders for Michigan, November 2016,
https://businessleadersformichigan.com/wp-
content/uploads/2016/11/Patrick-Doyle-Presentation-
FINAL.pdf.
12. Gabriel Snyder, “The New York Times Is Clawing Its
Way into the Future,” Wired, February 12, 2017,
https://www.wired.com/2017/02/new-york-times-digital-
journalism/.
13. Sara Fischer, “New York Times Surpasses 10 Million
Subscriptions,” Axios, February 2, 2022,
https://www.axios.com/new-york-times-10-million-
subscriptions-eb401cfb-2135-4845-b873-8b3b5f7fd10d.html.
14. New York Times Company, “The New York Times
Company 2021 Annual Report,” March 11, 2022,
https://nytco-assets.nytimes.com/2022/03/The-New-York-
Times-Company-2021-Annual-Report.pdf.
15. This is a paraphrase of a saying that is famously
misattributed to Darwin: “It is not the strongest of the species
that survives, nor the most intelligent. It is the one that is most
adaptable to change.” This epigram does encapsulate one of
the key insights of Darwin’s theory of natural selection, but it
was never written by him! Rather, it began as a portion of a
1963 speech by Leon C. Megginson that discussed Darwin’s
ideas; Megginson’s words were then repeated and
condensed by various management writers before becoming
the popular saying misattributed to Darwin. A full explanation
can be found at
https://quoteinvestigator.com/2014/05/04/adapt/.

2. DX and the Challenge of Innovation

1. Vijay Govindarajan and Anup Srivastava, “Strategy
When Creative Destruction Accelerates,” Working Paper No.
2836135, Tuck School of Business, 2016,
https://ssrn.com/abstract=2836135 or
http://dx.doi.org/10.2139/ssrn.2836135.
2. Todd Spangler, “Netflix Aims to Launch Cheaper, Ad-
Supported Plan in Early 2023,” Yahoo! Finance, July 19,
2022, https://finance.yahoo.com/news/netflix-aims-launch-
cheaper-ad-203926425.html.
3. Sara Fischer, “Big Cuts Coming for CNN+ After Slow
Start.” Axios, April 12, 2022.
https://www.axios.com/2022/04/12/cnn-plus-cuts-warner-
brothers-discovery.
4. Jason Kilar, Twitter, March 29, 2022, 3:35 p.m.,
https://twitter.com/jasonkilar/status/1508890566276362241.
5. Lucia Moses et al., “ ‘Hubris. Nothing More.’ Insiders
Blame Jeff Zucker and Jason Kilar for the Rapid Demise of
CNN+ as Warner Bros. Discovery Leadership Looks
Forward.,” Business Insider, April 12, 2022,
https://www.businessinsider.com/cnn-plus-failure-blame-
zucker-kilar-hubris-warner-bros-discovery-2022-4.
6. Michjael M. Grynbaum, John Koblin, and Benjamin
Mullin, “CNN+ Streaming Service Will Shut Down Weeks After
Its Start,” New York Times, April 21, 2022,
https://www.nytimes.com/2022/04/21/business/cnn-plus-
shutting-down.html.
7. Ted Johnson and Dade Hayes, “CNN+ Debuts: Is It the
Next News Innovation or Too Late to the Streaming Wars?,”
Deadline, March 28, 2022, https://deadline.com/2022/03/cnn-
plus-launch-streaming-service-preview-1234987770/.
8. Alex Sherman, “CNN+ Struggles to Lure Viewers in Its
Early Days, Drawing Fewer Than 10,000 Daily Users,” CNBC,
April 12, 2022, https://www.cnbc.com/2022/04/12/cnn-plus-
low-viewership-numbers-warner-bros-discovery.html.
9. Austin Carr, “The Inside Story of Jeff Bezos’s Fire
Phone Debacle,” Fast Company, January 6, 2015,
https://www.fastcompany.com/3039887/under-fire. Sources in
the article said that sales were in the tens of thousands of
units before the company’s radical price cut. The first such cut
happened forty-five days after launch.
10. Benjamin Black, “EC2 Origins,” Benjamin Black
Causes Trouble Here, January 25, 2009,
https://blog.b3k.us/2009/01/25/ec2-origins.html.
 11. Rachel King, “Amazon Breaks out Cloud Results for
First Time on Q1 Earnings Report,” ZDNET, April 23, 2015,
https://www.zdnet.com/article/amazon-breaks-out-cloud-
results-for-first-time-on-q1-earnings-report/.
12. Todd Bishop, “Amazon Web Services Posts Record
$13.5B in *Profits* for 2020 in Andy Jassy’s Aws Swan Song,”
GeekWire, February 2, 2021,
https://www.geekwire.com/2021/amazon-web-services-posts-
record-13-5b-profits-2020-andy-jassys-aws-swan-song/.
13. Tom Huddleston, “Zoom’s Founder Left a 6-Figure
Job Because He Wasn’t Happy—and Following His Heart
Made Him a Billionaire,” CNBC, August 21, 2019,
https://www.cnbc.com/2019/08/21/zoom-founder-left-job-
because-he-wasnt-happy-became-billionaire.html. In addition
to this published account of Eric Yuan’s departure, a former
Cisco executive told me that Yuan had pitched his idea for
Zoom to Cisco leaders, and the idea was shot down as too far
from Cisco’s enterprise focus.
14. Mansoor Iqbal, “Zoom Revenue and Usage Statistics
(2022),” Business of Apps, June 30, 2022,
https://www.businessofapps.com/data/zoom-statistics/,
accessed December 14, 2022. March 2019 was 10 million
daily meeting participants; March 2020 was 200 million.
15. Charles O’Reilly, Michael Tushman, and J. Bruce
Herrald, “Organizational Ambidexterity: IBM and Emerging
Business Opportunities,” California Management Review
(May 1, 2009), https://ssrn.com/abstract=1418194.
16. Malcolm Gladwell, “Creation Myth,” The New Yorker,
May 9, 2011,
https://www.newyorker.com/magazine/2011/05/16/creation-
myth.
17. Clayton M. Christensen, The Innovator’s Dilemma:
When New Technologies Cause Great Firms to Fail (Boston,
MA: Harvard Business Review Press, 2016).
18. Theodore Levitt, “Marketing Myopia,” Harvard
Business Review 38, no. 4 (1960): 24–47. Still a classic more
than fifty years later, the article is republished online at
https://hbr.org/2004/07/marketing-myopia.
19. David L. Rogers, in The Digital Transformation
Playbook: Rethink Your Business for the Digital Age(New
York: Columbia University Press, 2016), 127.
20. “Harvard i-Lab: Fireside Chat with Michael Skok and
Andy Jassy: The History of Amazon Web Services,” YouTube,
2013, https://www.youtube.com/watch?v=d2dy GDqrXLo.
21. Julie Bort, “Amazon’s Game-Changing Cloud Was
Built by Some Guys in South Africa,” Business Insider, March
28, 2012,
https://archive.ph/20130119102209/http:/www.businessinsider
.com/amazons-game-changing-cloud-was-built-by-some-
guys-in-south-africa-2012-3. Longtime Amazon executive
David Glick tells me that locating the EC2 team in Cape Town
had a dual purpose: Pinkham, a South African national, was
facing trouble renewing his U.S. work visa.

3. Step 1: Define a Shared Vision

1. Bill Ford, “A Future Beyond Traffic Gridlock,” TED Talk,
accessed December 14, 2022,
https://www.ted.com/talks/bill_ford_a_future_beyond_traffic_g
ridlock.
2. Lucy Kueng, “Transformation Manifesto: 9 Priorities for
Now,” November 2, 2020, http://www.lucykung.com/latest-
news/transformation-manifesto-9-priorities-for-now/.
3. In the memo, Stephen Elop relates a story about a
worker on an oil platform in the North Sea awaking to find the
entire platform on fire after an explosion. The memo is worth
a read at; see Chris Ziegler, “Nokia CEO Stephen Elop Rallies
Troops in Brutally Honest ‘Burning Platform’ Memo? (Update:
It’s Real!),” Engadget, February 8, 2011,
https://www.engadget.com/2011-02-08-nokia-ceo-stephen-
elop-rallies-troops-in-brutally-honest-burnin.html.
 4. The theory of extrinsic motivation (based on external
rewards) versus intrinsic motivation (based on the rewards of
the work itself) comes from the self-determination theory
(SDT) of motivation. A good overview of the work of Richard
M. Ryan and Edward L. Deci can be found at Delia O’Hara,
“The Intrinsic Motivation of Richard Ryan and Edward Deci,”
American Psychological Association, December 18, 2017,
https://www.apa.org/members/content/intrinsic-motivation.
5. Ford, “A Future Beyond Traffic Gridlock.”
6. Daniel Goleman defined six different leadership styles
or roles that leaders could adopt and advocated developing
the ability to combine them depending on circumstance. But
his empirical research found that the authoritative style—
associated with using a narrative to align others with a vision
—had the strongest positive impact of the six styles. See
Daniel Goleman, “Leadership That Gets Results,” Harvard
Business Review, March-April 2000,
https://hbr.org/2000/03/leadership-that-gets-results.
7. In McKinsey’s study, the factor of a “clear change
story” showed a 3.1 times spread between the 30 percent of
companies that were successfully transforming and the 70
percent in its study that were not—the greatest difference
among all the factors reported. See Hortense de la
Boutetière, Alberto Montagner, and Angelika Reich,
“Unlocking Success in Digital Transformations,” McKinsey &
Company, October 29, 2018,
https://www.mckinsey.com/business-
functions/organization/our-insights/unlocking-success-in-
digital-transformations.
8. Satya Nadella et al., “Learning to Lead,” in Hit Refresh:
The Quest to Rediscover Microsoft’s Soul and Imagine a
Better Future for Everyone (New York: Harper Business,
2018), 62.
9. As Philip Bobbitt wrote, Parmenides’ fallacy “occurs
when one tries to assess a future state of affairs by
measuring it against the present, as opposed to comparing it
to other possible futures.” Bobbitt first used the term in a 2003
op-ed for the New York Times; see Philip Bobbitt, “Today’s
War Is Against Tomorrow’s Iraq,” New York Times, March 10,
2003, https://www.nytimes.com/2003/03/10/opinion/today-s-
war-is-against-tomorrow-s-iraq.html. He developed the
concept in his later books, including Terror and Consent: The
Wars for the Twenty-First Century (New York: Alfred A. Knopf,
2018) and The Garments of Court and Palace: Machiavelli
and the World That He Made (New York: Grove Press, 2013).
The fallacy has since been discussed in a business context
by Clayton Christensen, Margie Warrell, and others.
10. This quote is widely attributed to President John
Fitzgerald Kennedy, but I am unable to find the source or
speech where he said it, including searching the online
archive of the JFK Presidential Library at “Home: JFK
Library,” accessed December 14, 2022,
https://www.jfklibrary.org/.
11. Dee-Ann Durbin and Tom Krisher, “Fields out at Ford;
New CEO Hackett Known for Turnarounds.” Chicago Tribune,
June 4, 2018, https://www.chicagotribune.com/business/ct-
ford-ceo-20170521-story.html.
12. Mark W. Johnson and Josh Suskewicz. Lead from the
Future: How to Turn Visionary Thinking into Breakthrough
Growth (Boston: Harvard Business Review Press, 2020), 210.
13. In a 1985 Playboy magazine interview, Steve Jobs
describes the type of person who worked at Apple as
“[s]omeone who really wants to get in a little over his head
and make a dent in the universe.” See David Scheff, “Steven
Jobs Playboy Interview,” Playboy (February 1985): 58.
14. “The CEO Test: Master the Challenges That Make or
Break All Leaders,” YouTube, 2021,
https://www.youtube.com/watch?
v=WXyFu53wMV8&list=PL38520A76CC5A4EE6&i
ndex=3.
15. John E. Doerr’s book Measure What Matters
provides an excellent deep dive into the practice of OKRs.
See John E. Doerr, Measure What Matters: OKRs, the Simple
Idea That Drives 10x Growth (London: Portfolio, 2018).
16. Doerr, Measure What Matters, 154–71.
17. D. E. Hunt, Beginning with Ourselves: In Practice,
Theory and Human Affairs (Cambridge, MA: Brookline Books,
1987), 4, 30.
18. The term “business theory” is inspired by Todd
Zenger’s writing on the importance of each company having a
“corporate theory”—although a business theory can be
applied not just to a company but to any new strategy or
change in resource allocation. See Todd Zenger, “What Is the
Theory of Your Firm?,” Harvard Business Review (June
2013): 126.
19. Mitchell Gordon, “Disney’s Land: Walt’s Profit
Formula: Dream, Diversify—and Never Miss an Angle; Here’s
How His Divisions Complement Each Other,” Wall Street
Journal, February 4, 1958, p. 1
20. This was explained numerous times by Jeff Bezos,
including in this interview: Jeff Bezos, “Interview with Adi
Ignatius,” Harvard Business Review, podcast audio, January
4, 2013, https://hbr.org/podcast/2013/01/jeff-bezos-on-
leading-for-the. Bezos also discussed it in his 2008 letter to
shareholders; see Jeff Bezos, “Letter to Amazon
Shareholders,” 2009,
https://ir.aboutamazon.com/files/doc_financials/annual/Amazo
n _SH_Letter_2008.pdf.
21. Zenger, “What Is the Theory of Your Firm?,” 126.
22. In 2021, National Commercial Bank merged with
Samba Financial Group to become Saudi National Bank.
23. Bob Iger, interview with Kara Swisher, Sway, podcast
audio, January 27, 2022,
https://www.nytimes.com/2022/01/27/opinion/sway-kara-
swisher-bob-iger.html?show Transcript=1.
24. Disney+ launched on November 12, 2019. Disney
stock closed November 11 at 136.74. It closed November 13
at 148.72.
 25. Bezos, “Interview with Adi Ignatius.”
26. “Hackett CEO News Conference.mp4,” Dropbox,
accessed January 5, 2023,
https://www.dropbox.com/s/k84legr519o0xpl/Hackett%20CEO
%20News%20conference.mp4?dl=0&mod=article_inline.
This quote occurs in the video at 17:10.

4. Step 2: Pick the Problems That Matter Most

1. Michael Porter, “What Is Strategy?,” Harvard Business
Review (November–December 1996): 60,
https://hbr.org/1996/11/what-is-strategy.
2. Thomas Wedell-Wedellsborg, What’s Your Problem?
To Solve Your Toughest Problems, Change the Problems You
Solve (Boston: Harvard Business Review Press, 2020).
3. Minda Zetlin, “This Video Is How Microsoft CEO Satya
Nadella Introduced Himself to an Audience of 17,000 and It
Was Perfect,” Inc., accessed December 16, 2022,
https://www.inc.com/minda-zetlin/satya-nadella-microsoft-
xbox-adaptive-controller-super-bowl-video-disabled-gamers-
owen-sirmons.html.
4. Kyle Evans, “Product Thinking vs. Project Thinking,”
Medium (Product Coalition, October 21, 2018),
https://productcoalition.com/product-thinking-vs-project-
thinking-380692a2d4e.
5. Among other places, this idea was prominently
articulated in Amazon’s 2008 shareholder letter; see Jeff
Bezos, “2008 Letter to Amazon Shareholders,” 2009,
https://ir.aboutamazon.com/files/doc_financials/annual/Amazo
n_SH_Letter_2008.pdf.
6. Jeff Bezos, “Interview with Adi Ignatius,” Harvard
Business Review, podcast audio, January 4, 2013,
https://hbr.org/podcast/2013/01/jeff-bezos-on-leading-for-the.
7. Another example comes from Mastercard Labs, which
has a similar process for nurturing early ideas via its five-day
“launchpad workshops,” in which customers are brought in to
help confirm their needs, pain points, and whether the
proposed innovation offers real value.
8. Bezos, “2008 Letter to Amazon Shareholders.”
9. This categorization is often credited to Kevin Fong,
longtime managing partner at Mayfield Fund in Silicon Valley.
One such source is Omer Khan, “Candy, Vitamin or Painkiller:
Which One Is Your Product?,” SaaS Club, accessed
December 16, 2022, https://saasclub.io/candy-vitamin-
painkiller-which-one-is-your-product/. Fong also described a
third category, which he called “candy,” for ideas that provided
fleeting benefits; he did not like to invest in ideas from this
third category.
10. Alexander Osterwalder et al., Value Proposition
Design: How to Create Products and Services Customers
Want (Hoboken, NJ: John Wiley, 2014).
11. Cited by early Uber investor Chris Sacca in 2015; see
Chris Sacca, “Why I’d Never Want to Compete with Uber’s
Travis Kalanick,” Fortune, February 4, 2015,
https://fortune.com/2015/02/04/why-id-never-want-to-
compete-with-ubers-travis-kalanick/.
12. Noriaki Kano et al., “Attractive Quality and Must-Be
Quality,” Journal of the Japanese Society for Quality Control
14, no. 2 (1984): 147–56,
https://web.archive.org/web/20110813145926/http:/ci.nii.ac.jp/
Detail/detail.do?LOCALID=ART0003570680&lang=en. For a
good summary of how the Kano model evolved, see “Kano
Model: What Is the Kano Model? Definition and Overview of
Kano,” September 2, 2021,
https://www.productplan.com/glossary/kano-model/.
13. Liz Tay, “Google Has Updated Its 9 Principles Of
Innovation: Here They Are and the Products They Have
Enabled,” Business Insider Australia, November 19, 2013,
https://www.businessinsider.com.au/google-has-updated-its-
9-principles-of-innovation-here-they-are-and-the-products-
they-have-enabled-2013-11.
 14. John E. Doerr, Measure What Matters: OKRs, the
Simple Idea That Drives 10× Growth (London: Portfolio,
2018), 127.
15. “From AT&T to Xerox: 90+ Corporate Innovation
Labs: CB Insights,” CB Insights Research, August 28, 2021,
https://www.cbinsights.com/research/corporate-innovation-
labs/.
16. Warren Berger, “The Secret Phrase Top Innovators
Use,” Harvard Business Review (September 12, 201),
https://hbr.org/2012/09/the-secret-phrase-top-innovato.
17. Emily Chasan, “Don’t Toss That Cup: McDonald’s
and Starbucks Are Developing Reusables,” Bloomberg.com,
February 18, 2020,
https://www.bloomberg.com/news/articles/2020-02-
18/reusable-coffee-cups-being-tested-for-mcdonald-s-and-
starbucks.
18. Porter, “What Is Strategy?,” 60.
19. “Steve Blank: The Key to Startup Success? ‘Get Out
of the Building.’ ” Inc., accessed January 6, 2023,
https://www.inc.com/steve-blank/key-to-success-getting-out-
of-building.html.
20. Colin Bryar and Bill Carr, Working Backwards (New
York: St. Martin’s Press, 2021), 98–120.
21. David Leonhardt et al., “Journalism That Stands
Apart” (New York: New York Times, 2017), Section: “The way
we work: Every department should have a clear vision that is
well understood by its staff.”
22. Donald N. Sull, “Closing the Gap Between Strategy
and Execution,” MIT Sloan Management Review (July 1,
2007), https://sloanreview.mit.edu/article/closing-the-gap-
between-strategy-and-execution/.

5. Step 3: Validate New Ventures

 1. Sarah Nassauer, “WSJ News Exclusive: Walmart
Scraps Plan to Have Robots Scan Shelves,” Wall Street
Journal, November 2, 2020, https://on.wsj.com/3c04VQF.
2. Tom Ward, “From Ground-Breaking to Breaking
Ground: Walmart Begins to Scale Market Fulfillment Centers,”
Walmart Corporate, January 27, 2021,
https://corporate.walmart.com/newsroom/2021/01/27/from-
ground-breaking-to-breaking-ground-walmart-begins-to-scale-
local-fulfillment-centers.
3. Melissa Repko, “Walmart Drew One in Four Dollars
Spent on Click and Collect—with Room to Grow in 2022,”
CNBC, December 30, 2021,
https://www.cnbc.com/2021/12/30/walmart-drew-one-in-four-
dollars-on-click-and-collect-market-researcher.html.
4. Sarah Nassauer, “Walmart Pushes New Delivery
Services for a Post-Pandemic World,” Wall Street Journal,
February 28, 2022, https://www.wsj.com/articles/walmart-
pushes-new-delivery-services-for-a-post-pandemic-world-
11645971260.
5. Dean Baquet, “The New York Times and Journalism’s
Future,” presentation at the INMA World Conference of
Media, New York, May 17, 2019.
6. Steve Blank, “No Plan Survives First Contact with
Customers—Business Plans Versus Business Models,”
SteveBlank.com, April 8, 2010,
https://steveblank.com/2010/04/08/no-plan-survives-first-
contact-with-customers-%E2%80%93-business-plans-versus-
business-models/.
7. Jonathan Becher, “RIP ROI: Time-to-Market Is the
New Indicator of Success,” LinkedIn, August 8, 2016,
https://www.linkedin.com/pulse/rip-roi-time-to-market-new-
indicator-success-jonathan-becher/.
8. Steven G. Blank and Bob Dorf, The Startup Owner’s
Manual: The Step-by-Step Guide for Building a Great
Company (Pescadero, CA: K & S Ranch, 2012), 551.
 9. Eric Ries, The Lean Startup: How Constant Innovation
Creates Radically Successful Businesses (London: Penguin
Business, 2019).
10. I am indebted to Bob Dorf’s lectures in my classes at
Columbia Business School for sharing this story of the origins
of Diapers.com.
11. Marc Randolph, “Please Mr. Postman,” in That Will
Never Work: The Birth of Netflix and the Amazing Life of an
Idea (New York: Back Bay Books, 2022), 24–37.
12. Eric Von Hippel, “Lead Users: A Source of Novel
Product Concepts,” Management Science 32 (1986): 791–
806, doi:10.1287/mnsc.32.7.791.
13. The concept of product-market fit was developed and
named by Andy Rachleff, who is currently the CEO and
cofounder of Wealthfront and is a cofounder of Benchmark
Capital. Steven G. Blank and Bob Dorf also use the term
“problem/solution fit” in The Startup Owner’s Manual.
14. Alberto Savoia, The Right It: Why So Many Ideas Fail
and How to Make Sure Yours Succeed (New York: Harper
One, 2019).
15. Tesla unveiled the new car on April 1, 2016, and had
200,000 orders in a little over twenty-four hours from
customers who would not receive a car until late 2017 or
2018. See Chris Isidore, “Tesla Got 200,000 Orders for the
Model 3 in about One Day,” CNNMoney, April 1, 2016,
https://money.cnn.com/2016/04/01/news/companies/tesla-
model-3-stock-price/index.html.
16. Tim Harford, Fifty Things That Made the Modern
Economy (London: Abacus, 2018).
17. Julie Jargon, “How Panera Solved Its ‘Mosh Pit’
Problem,” Wall Street Journal, June 2, 2017,
https://www.wsj.com/articles/how-panera-solved-its-mosh-pit-
problem-1496395801.
18. Eric Ries, “Test,” in The Lean Startup: How Today’s
Entrepreneurs Use Continuous Innovation to Create Radically
Successful Businesses (New York Currency, 2017), 99–102.
 19. In 2018, Reid Hoffman claimed that he coined this
aphorism “more than a decade ago.” See Reid Hoffman, “If
There Aren’t Any Typos in This Essay, We Launched Too
Late!,” LinkedIn, March 29, 2017,
https://www.linkedin.com/pulse/arent-any-typos-essay-we-
launched-too-late-reid-hoffman/.
20. Mark W. Johnson and Josh Suskewicz, “How to
Jump-Start the Clean-Tech Economy,” Harvard Business
Review (November 2009): 87.
21. Clayton M. Christensen and Michael E. Raynor, The
Innovator’s Solution: Creating and Sustaining Successful
Growth (Boston: Harvard Business School Press, 2003), 74–
80, 96. Christensen and Raynor credit Richard Pedi with
coining the phrase “jobs to be done,” Anthony Ulwick with
developing closely related concepts, and David Sundahl with
assisting in their own formulation. The job-to-be-done concept
has been further explored in various articles by Christensen
with other coauthors.
22. David L. Rogers, in The Digital Transformation
Playbook: Rethink Your Business for the Digital Age (New
York: Columbia University Press, 2016), 56.
23. Geoffrey A. Moore, Crossing the Chasm: Marketing
and Selling Disruptive Products to Mainstream Customers
(New York: HarperBusiness, 2014).
24. Mark W. Johnson and Josh Suskewicz, “How to
Jump-Start the Clean-Tech Economy,” Harvard Business
Review (November 2009): 87.

6. Step 4: Manage Growth at Scale

1. Penny Crosman, “Welcome to Open Mic Night at a Citi
Fintech Unit,” American Banker, November 22, 2017,
https://www.americanbanker.com/news/welcome-to-open-mic-
night-at-a-citi-fintech-unit.
2. An original source by Margaret Mead has not been
found; the quote was first attributed to her by Donald Keys
shortly after her death. See “Never Doubt That a Small Group
of Thoughtful, Committed Citizens Can Change the World;
Indeed, It’s the Only Thing That Ever Has,” Quote
Investigator, November 12, 2017,
https://quoteinvestigator.com/2017/11/12/change-world/.
3. J. Richard Hackman and Neil Vidmar, “Effects of Size
and Task Type on Group Performance and Member
Reactions,” Sociometry 33, no. 1 (1970): 37–54,
https://doi.org/10.2307/2786271.
4. Eric Ries, The Startup Way: How Modern Companies
Use Entrepreneurial Management to Transform Culture and
Drive Long-Term Growth (New York: Currency, 2017).
5. Mark Wilson, “Adobe’s Kickbox: The Kit to Launch
Your Next Big Idea,” Fast Company, February 9, 2015,
https://www.fastcompany.com/3042128/adobes-kickbox-the-
kit-to-launch-your-next-big-idea.
6. Crosman, “Welcome to Open Mic Night at a Citi
Fintech Unit.”
7. Crosman, “Welcome to Open Mic Night at a Citi
Fintech Unit.”
8. Lucy Kueng, “Going Digital: A Roadmap for
Organisational Transformation,” Reuters Institute for the
Study of Journalism and University of Oxford, November
2017, p. 16,
https://reutersinstitute.politics.ox.ac.uk/sites/default/files/2017-
11/Going%20Digital.pdf.
9. Mark W. Johnson and Josh Suskewicz. Lead from the
Future: How to Turn Visionary Thinking into Breakthrough
Growth (Boston: Harvard Business Review Press, 2020), 115.
10. Eric Ries, The Startup Way: How Modern Companies
Use Entrepreneurial Management to Transform Culture and
Drive Long-Term Growth (New York: Currency, 2017), 294.
11. Lucy Kueng, “Going Digital,” 16.
12. Steven Levy, “Google Glass 2.0 Is a Startling Second
Act,” Wired, July 18, 2017,
https://www.wired.com/story/google-glass-2-is-here/.
13. “NYT Innovation Report 2014,” Scribd, March 24,
2014, https://www.scribd.com/doc/224332847/NYT-
Innovation-Report-2014, p. 75.
14. Susan Wojcicki, “The Eight Pillars of Innovation,”
Google, July 2011, https://www.thinkwithgoogle.com/future-of-
marketing/creativity/8-pillars-of-innovation/.
15. Alex Morrell, “We Spoke with Citi’s Innovation Chief
About Which Fintechs It Wants to Invest in, How Its Internal
‘Shark Tank’ Judges Know When to Kill an Idea, and Why
Red Tape Helps Some Startups Flourish,” Business Insider,
February 8, 2019, https://www.businessinsider.com/vanessa-
colella-citi-ventures-innovation-interview-2019-2.
16. Robert D. Hof, “Amazon’s Risky Bet,” BusinessWeek,
November 13, 2006. The magazine’s cover image can be
seen at Jeff Bezos, Twitter, May 18, 2022, 3:11 p.m.,
https://twitter.com/jeffbezos/status/1527003895393812480.
17. For a good postmortem on the failed GE digital effort,
see Alex Moazed, “Why GE Digital Failed,” Inc., accessed
January 4, 2023, https://www.inc.com/alex-moazed/why-ge-
digital-didnt-make-it-big.html. Another good analysis is Ted
Mann and Thomas Gryta, “The Dimming of GE’s Bold Digital
Dreams,” Wall Street Journal, July 18, 2020,
https://www.wsj.com/articles/the-dimming-of-ges-bold-digital-
dreams-11595044802?mod=djemalertNEWS.
18. Tenets are a set of principles that each team creates
to guide its everyday decision making. The practice of tenets
was first introduced at Amazon on a team led by my friend
David Glick. During nearly twenty years at Amazon, Glick led
many two-pizza teams in areas such as warehousing,
logistics, pricing, and merchant fulfillment. When he was
leading a team working on pricing, Glick met with Jeff Bezos
to hash out the guiding strategy for their work. Bezos declared
in the meeting, “We keep our prices very, very low because
we think that earns customer trust, and over the long term, we
take it as an article of faith that customer trust will drive long-
term free cash flow.” Glick wrote that down and captured four
other ideas that seemed most central to the direction they had
agreed on. He refined the wording, dubbed them his team’s
five tenets, and started putting them at the very top of every
memo produced by his team as a reminder of what they were
aiming for. After several monthly meetings, Bezos remarked,
“I really like that this team has their tenets right at the top.”
Then he turned to his technical adviser and said “Ahmed, go
make sure everybody does this!” And henceforth, 100,000
people at Amazon were told they had to define their own
guiding tenets for decision making by their teams. “I got a lot
of hate email,” Glick said, “although they also asked me if I
could send them my list of tenets.”

7. Step 5: Grow Tech, Talent, and Culture

1. William Boston, “How Volkswagen’s $50 Billion Plan to
Beat Tesla Short-Circuited,” Wall Street Journal, January 19,
2021, https://www.wsj.com/articles/how-volkswagens-50-
billion-plan-to-beat-tesla-short-circuited-11611073974.
2. Boston, “How Volkswagen’s $50 Billion Plan to Beat
Tesla Short-Circuited.”
3. Henry Man, “Volkswagen to Develop In-House
Software for Next-Gen Cars,” CarExpert, June 22, 2020,
https://www.carexpert.com.au/car-news/volkswagen-to-
develop-in-house-infotainment-software
4. Boston, “How Volkswagen’s $50 Billion Plan to Beat
Tesla Short-Circuited.”
5. Herbert Diess, LinkedIn, February 2022,
https://www.linkedin.com/posts/herbertdiess_i-am-happy-that-
lynn-longo-as-our-new-cariad-activity-689893566
0502487040-h_zv.
6. Boston, “How Volkswagen’s $50 Billion Plan to Beat
Tesla Short-Circuited.”
7. Jeff Lawson, Ask Your Developer: How to Harness the
Power of Software Developers and Win in the 21st Century
(New York: Harper Business, 2021), 3–4. Lawson cited Bezos
as saying this at the first all-hands meeting he attended, after
joining Amazon in September 2004.
8. Mark J. Greeven, Howard Yu, and Jialu Shan, “Why
Companies Must Embrace Microservices and Modular
Thinking,” MIT Sloan Management Review (June 28, 2021),
https://sloanreview.mit.edu/article/why-companies-must-
embrace-microservices-and-modular-thinking.
9. Greeven, Yu, and Shan, “Why Companies Must
Embrace Microservices.”
10. This is my definition of technical debt, meant to be
broad enough to capture the full range of the application of
the idea today, which has grown far beyond “sloppy code.”
The idea dates back at least to 1980—when Meir Manny
Lehman wrote, “As an evolving program is continually
changed, its complexity . . . increases unless work is done to
maintain or reduce it.” See Meir Manny Lehman, “Laws of
Software Evolution Revisited,” in Software Process
Technology: 5th European Workshop, EWSPT ʼ96, Nancy,
France, October 9–11, 1996: Proceedings (Berlin: Springer,
1996), 108–124. The metaphor of “debt” was later coined by
Ward Cunningham, “The WyCash Portfolio Management
System,” March 26, 1992, http://c2.com/doc/oopsla92.html.
Smart thinking includes Martin Fowler, “TechnicalDebt,” May
21, 2019, https://martinfowler.com/bliki/TechnicalDebt.html.
11. Werner Vogels, “Modern Applications at AWS,” All
Things Distributed, August 28, 2019,
https://www.allthingsdistributed.com/2019/08/modern-
applications-at-aws.html.
12. The 2002 memo declaring Bezo’s application
programming interface (API) mandate is cherished legend,
but no contemporary copies or accounts of it exist. A 2011
social media post from Amazon insider Steve Yegge
attempted to paraphrase the original memo, saying that “[it]
went something along these lines,” but Yegge’s own wording
has since been repeated by others as the original memo
itself. Yegge’s post was on the now-defunct Google Plus but
has been archived here: Steve Yegge, Google Plus, October
11, 2011, https://gist.github.com/chitchcock/1281611.
13. Vogels, “Modern Applications at AWS.”
14. This account times the Netflix transition as happening
in 2009–2011. shriramvenugopal, “The Story of Netflix and
Microservices,” Geeks for Geeks, May 17, 2020,
https://www.geeksforgeeks.org/the-story-of-netflix-and-
microservices/.
15. Joshua Benton, “The Leaked New York Times
Innovation Report Is One of the Key Documents of This
Media Age,” Nieman Lab, May 15, 2014,
https://www.niemanlab.org/2014/05/the-leaked-new-york-
times-innovation-report-is-one-of-the-key-documents-of-this-
media-age/.
16. For larger companies, a third option is to acquire a
business with the capability you need. Thus, you will
sometimes hear the choice framed as “build versus buy
versus partner”—where “buy” means to acquire a firm, and
“partner” means to buy services, components, or technology
from an outside partner.
17. Lawson, Ask Your Developer, 4.
18. Angus Loten, “PepsiCo Bottles Tech Collaboration
Effort into New Digital Hubs,” Wall Street Journal, October 28,
2021, https://www.wsj.com/articles/pepsico-bottles-tech-
collaboration-effort-into-new-digital-hubs-11635457546.
19. Boston, “How Volkswagen’s $50 Billion Plan to Beat
Tesla Short-Circuited.”
20. Franklin Foer, “Jeff Bezos’s Master Plan,” The
Atlantic, November 2019,
https://www.theatlantic.com/magazine/archive/2019/11/what-
jeff-bezos-wants/598363/.
21. The idea of the “T-shaped” person was espoused for
consultants by McKinsey in the 1980s and later embraced in
both agile software development and by design thinking firms
like IDEO.
 22. Benton, “The Leaked New York Times Innovation
Report Is One of the Key Documents.”
23. Aaron Aboagye, Ani Mukkavilli, and Jeremy
Schneider, “Four Myths About Building a Software Business,”
McKinsey & Company, April 30, 2021,
https://www.mckinsey.com/capabilities/mckinsey-digital/our-
insights/four-myths-about-building-a-software-business.
24. Liad Agmon, “Dynamic Yield Joins the McDonald’s
Family,” Dynamic Yield, accessed February 7, 2023,
https://www.dynamicyield.com/blog/dynamic-yield-joins-
mcdonalds/.
25. A study by McKinsey of 2,000 M&A transactions
found that nondigital firms that began with an “anchor
acquisitions” of a digital company worth $1 billion or more saw
total returns to shareholders five times higher than those seen
in nondigital companies that started with multiple smaller
digital acquisitions. See Aboagye, Mukkavilli, and Schneider,
“Four Myths About Building a Software Business.”
26. Daniel Pink, Drive: The Surprising Truth about What
Motivates Us (New York: Riverhead Books, 2013).
27. Satya Nadella, 2015 Microsoft shareholder meeting,
https://www.youtube.com/watch?v=TDYAGKHFIjM.
28. Ann Rhoades, in Built on Values: Creating an
Enviable Culture That Outperforms the Competition (San
Francisco: Jossey-Bass, 2011), 19. Kelleher is not alone in
thinking of culture in terms of norms of behavior. Leadership
scholar John Kotter has defined company culture as a “group
norms of behavior and the underlying shared values that help
keep those norms in place.” See John Kotter, “The Key to
Changing Organizational Culture,” Forbes, September 27,
2012, https://www.forbes.com/sites/johnkotter/2012/09/27/the-
key-to-changing-organizational-culture/.
29. Ben Horowitz, What You Do Is Who You Are: How to
Create Your Business Culture (New York: HarperBusiness,
2019), 2–3.
 30. “Creating a Culture of Innovation,” Google, accessed
April 15, 2020, https://gsuite.google.co.in/intl/en_in/learn-
more/creating_a_culture_of_innovation.html.
31. “Ten Things We Know to Be True,” Google, accessed
January 5, 2023, https://about.google/philosophy/.
32. Pauline Meyer, “Tesla Inc.’s Organizational Culture &
Its Characteristics (Analysis),” Panmore Institute, February
22, 2019, http://panmore.com/tesla-motors-inc-organizational-
culture-characteristics-analysis.
33. “Southwest Careers,” Southwest Airline, accessed
April 10, 2020, https://careers.southwestair.com/culture.
34. “Leadership Principles,” Amazon Jobs, accessed
January 5, 2023, https://www.amazon.jobs/content/en/our-
workplace/leadership-principles.
35 Samir Lakhani, “Things I Liked About Amazon,”
Medium, August 28, 2017,
https://medium.com/@samirlakhani/things-i-liked-about-
amazon-4495ef06fbda.
36. Reed Hastings, “Freedom & Responsibility Culture
(Version 1),” Slideshare, June 30, 2011,
https://www.slideshare.net/reed2001/culture-2009.
37. “Netflix Culture—Seeking Excellence,” Netflix,
accessed January 5, 2023, https://jobs.netflix.com/culture.
38. Michael Lewis, “How Two Trailblazing Psychologists
Turned the World of Decision Science Upside Down,” Vanity
Fair, November 14, 2016,
https://www.vanityfair.com/news/2016/11/decision-science-
daniel-kahneman-amos-tversky.
39. Krzysztof Majdan and Michael Wasowski, “We Sat
Down with Microsoft’s CEO to Discuss the Past, Present and
Future of the Company,” Business Insider, April 20, 2017,
https://www.businessinsider.com/satya-nadella-microsoft-ceo-
qa-2017-4. Satya Nadella called the same story Microsoft’s
‘creation myth’ in Tim O’Reilly, “We Must Find a Grand
Purpose for AI,” LinkedIn, September 11, 2018,
https://www.linkedin.com/pulse/conversation-satya-nadella-
his-new-book-hit-refresh-tim-o-reilly/.
40. Minda Zetlin, “This Video Is How Microsoft CEO
Satya Nadella Introduced Himself to an Audience of 17,000
and It Was Perfect,” Inc., March 30, 2019,
https://www.inc.com/minda-zetlin/satya-nadella-microsoft-
xbox-adaptive-controller-super-bowl-video-disabled-gamers-
owen-sirmons.html.
41. Lucy Kueng. “Why Media Companies Need to Stop
Focusing on Content,” presentation at the INMA World
Conference of Media, Washington, DC, May 17, 2018.
42. Robin D. Schatz, “How Blinds.com Searched Its Soul
—and Found Home Depot,” Inc., May 2014,
https://www.inc.com/magazine/201405/robin-schatz/how-
blinds-com-acquired-by-home-depot.html.
43. Greylock, “Culture Is How You Act When No One Is
Looking,” Medium, June 1, 2017,
https://news.greylock.com/culture-is-how-you-act-when-no-
one-is-looking-f29d5dd16ecb.
44. Steven Kerr, “On the Folly of Rewarding A, While
Hoping for B,” Academy of Management Journal 18, no. 4
(1975): 769–83. Updated by the author in 1995: Steven Kerr,
“On the Folly of Rewarding A, While Hoping for B,” Academy
of Management Executive 9, no. 1 (1995): 7–14,
https://www.ou.edu/russell/UGcomp/Kerr.pdf
45. Ryan Felton, “Tesla Switching to 24/7 Shifts to Push
for 6,000 Model 3s per Week by June, Elon Musk Says,”
Jalopnik, April 17, 2018, https://jalopnik.com/tesla-switching-
to-24-7-shifts-to-push-for-6-000-model-1825335216.
46. “Netflix Culture—Seeking Excellence.”
47. Jeffrey P. Bezos, “1997 Letter to Shareholders,”
1998,
https://s2.q4cdn.com/299287126/files/doc_financials/annual/S
hareholderletter97.pdf.
48. This process of single-executive approval was
described by Doug Herrington, senior vice president, North
America consumer, Amazon while speaking at Princeton
University’s Keller Center. See: “Ten Rules of Innovating at
Amazon,” Keller Center at Princeton University, January 18,
2018, https://kellercenter.princeton.edu/stories/ten-rules-
innovating-amazon.

Conclusion
1. Stanley A. McChrystal, David Silverman, Tantum
Collins, and Chris Fussell, Team of Teams (London: Portfolio
Penguin, 2015).
2. “Netflix Culture—Seeking Excellence,” Netflix,
accessed January 5, 2023, https://jobs.netflix.com/culture.
3. C. A. Bartlett and S. Ghoshal, “Changing the Role of
Top Management: Beyond Strategy to Purpose,” Harvard
Business Review 72, no. 6 (November–December 1994): 79–
88, https://hbr.org/1994/11/beyond-strategy-to-purpose
4. Jeff Bezos, “2013 Letter to Shareholders,” 2014,
https://ir.aboutamazon.com/files/doc_financials/annual/2013-
Letter-to-Shareholders.pdf.
5. Lucy Kueng, “Why Media Companies Need to Stop
Focusing on Content,” presentation at the INMA World
Conference of Media, Washington, DC, May 17, 2018.
6. Jeff Bezos, “2016 Letter to Amazon Shareholders,”
2017,
https://ir.aboutamazon.com/files/doc_financials/annual/2016-
Letter-to-Shareholders.pdf.
7. This quote is popularly attributed to George Bernard
Shaw—but there is no evidence. The earliest known source
was in business writing: “The Biggest Problem in
Communication Is the Illusion That It Has Taken Place,”
Quote Investigator, November 3, 2018,
https://quoteinvestigator.com/2014/08/31/illusion/.
8. Alan Deutschman, Walk the Walk: The #1 Rule for
Real Leaders (London: Portfolio, 2011), 158. He may be
paraphrasing Wendy Kopp.
9. “Netflix Culture—Seeking Excellence,” Netflix,
accessed June 17, 2023, https://jobs.netflix.com/culture.
https://jobs.netflix.com/culture. The origin of this quote is not
completely clear and may be a paraphrase. See “Teach Them
to Yearn for the Vast and Endless Sea,” Quote Investigator,
August 25, 2015,
https://quoteinvestigator.com/2015/08/25/sea/.
10. First formulated in his 1970 essay, Robert K.
Greenleaf, “The Servant as Leader,” Greenleaf Organization
(Cambridge, MA: Center for Applied Studies, 1970), which
credits inspiration to Hermann Hesse’s 1932 novel Journey to
the East. Greenleaf expanded his thinking in 1977; see
Robert Greenleaf, Servant Leadership (Mahwah, NJ: Paulist
Press, 1977).
 INDEX

Page numbers refer to the print edition but are hyperlinked to the appropriate
location in the e-book.

Page numbers in italics represent figures or tables.

AARP, 49

accountability: in Culture-Process Map, 261; in teams, 175

actionable principles, 258
Acuity Insurance, 62, 269; future landscape, 53–54; P/O statements for, 97
AdSense, 193
Adobe, 180
advertising, on Netflix, 24
advocacy, 177
Aftenposten, 244–246, 253, 255
agile software development, 9, 20, 34–36, 88, 104, 106–107, 120–122, 173–174, 233–234,
242, 247, 262, 266, 295n21
AI. See artificial intelligence
Airbnb, 29
Airgas, 188
Air Liquide, 11, 65, 92, 132, 188; innovation structures, 211; P/O statements for, 97
Akamai, 32
Alibaba, 24, 25, 39, 62
Alphabet, 41
Alto, 32
Amazon, 5, 25, 29, 34, 46, 69, 173, 236, 269; culture at, 249–250, 256; Leadership
Principles, 249; shareholders of, 73; virtuous cycle of, 69, 70
Amazon Alexa, 10
Amazon Marketplace, 132, 192
Amazon Prime, 30
Amazon Web Services (AWS), 30–31, 94, 196, 227
ambidextrous organization, 203
analysis paralysis, 26–27
anchor acquisitions, 295n25
Android, 24, 29–30
Ant Financial, 4
APIs. See application programming interfaces
Apple, 26, 63
Apple Watch, 10
application programming interfaces (APIs), 225, 294n12
Aristotle, 252
ARPU. See average revenue per user
artificial intelligence (AI), 65, 90
Audi, 221
Audm, 17
automotive industry, 44–45, 198
autonomy, 270; of teams, 174
average revenue per user (ARPU), 147
AWS. See Amazon Web Services
Axel Springer, 190
Azure, 26

backlogs, 191–192
Banga, Ajay, 66
Baquet, Dean, 119–120
Barnes & Noble, 25
Baron, Marty, 253
Basadur, Min, 100, 111
BASF, 170
BAU. See business-as-usual
BCG, 4
Becher, Jonathan, 121
behavior, culture as, 247–251
behavioral data, 127
Bezos, Jeff, 29, 69, 89, 191, 200, 227, 269, 293n18
Bharara, Vinit, 123
Birje, Anand, 232
Black, Benjamin, 30–31
Blank, Steve, 35, 106, 124, 125
Blinds.com, 253
Block, 4
blockchain, 65, 90
boards, 172; decision rights, 179; growth, 175–179; innovation, 212, 214–216
Bobbitt, Philip, 54–55, 286n9
BOPS. See buy-online-pickup-at-store
bottom up strategy: decision-making, 266–267; DX, 15–16; experimentation, 164–165;
governance, 217–218; information flows in, 267; innovation in, 267; leadership, 267–
271; organizational structure, 262–263; P/O statements, 113–115; shared vision, 81–82
bravery, 255
Bristol Myers Squibb, 230, 232
Brown, Joe, 123
Bryant, Adam, 64
BSH Home Appliances: future landscape of, 53; innovation structures, 211
Buffett, Warren, 73
burning platform 46, 286n3
business-as-usual (BAU), 9; funding, 186, 187; teams, 175, 176
business plans, writing, 27
business theory, 13, 48, 68–73, 80–83, 106, 288n18; benefits of, 72–75; causal theory in,
80–81; shareholder communications and, 73–75; success in, 80; value drivers in, 80
business validation, 129, 143–148, 157–159; CLV in, 158; cost structure in, 158; learning in,
147–148; metrics in, 147–148; MVPs in, 146–147; smart shutdowns in, 189; value
capture in, 157–158
Business Week, 200
buy-online-pickup-at-store (BOPS), 101
Buzzfeed, 2–3

CAA. See Canadian Automobile Association

Campisi, Vince, 72, 89, 248
Canadian Automobile Association (CAA), 56
cannibalization, 33
capabilities: failure symptoms, 224; importance of, 222–225; stakes of, 224; success
symptoms, 224. See also talent capabilities; technology capabilities
capability change, 94; in DX, 9–10, 15; at New York Times Company, 17–18
CARIAD, 221
causal theory, in Shared Vision Map, 80–81
CDO. See chief digital officers
CEOs, 47
challenge of proximity, 30, 40, 203
challenge of uncertainty, 29, 35, 37, 116
channel level, 98
Chase, 62
chief digital officers (CDO), 5, 47
Christensen, Clayton, 152
Cisco, 32
Citibank, 62, 101, 167, 168, 173, 177, 186; Discovery 10X, 131, 168, 185, 193, 218;
innovation structures, 211; P/O statements, 103
CitiConnect for Blockchain, 168
Citi Ventures, 167
Citi Ventures Studio, 169
Cline, Patsy, 126
cloud computing, 26
CLV. See customer lifetime value
CMA CGM, 235
CNN, 62
CNN+, 28–29, 39
Colella, Vanessa, 131, 167, 169, 180, 186, 255
combinations, talent capabilities, 235–239
Comcast, 192
communication: of culture, 251–256, 258–259; in Culture-Process Map, 261; leadership
and, 269
competition, 50; in Four Stages of Validation, 148–150
competitive differentiation, 153
Competitive Value Train, 52, 108
compliance, 171
composable technology, 232
consumable technology, 232
Cooper-Hewitt National Design Museum, 100
copper, 144
copycat products, 143
Corporate Innovation Stack, 212–217; innovation boards, 214–216; innovation structure,
212–214; innovation teams, 216–217; layers of, 212
corporate VC, 206
cost/risk savings, 157
cost structure: in business validation, 158; of innovation, 145
COVID-19, 5, 119
Crossing the Chasm (Moore), 161
cultural principles; actionable, 258; in Culture-Process Map, 258; relatable, 258, tenets at
Amazon, 293n18
cultural symbols, 251–256; in Culture-Process Map, 259
culture, 244, 262; at Amazon, 249–250, 256; as behavior, 247–251; communication of,
251–256, 258–259; defining, 249–251; at digital-native businesses, 255–256; in DX,
246–247, 251; enabling, 254–256, 259–260; leadership and, 249; at Netflix, 250; at
New York Times Company, 246–247; resource allocation and, 254; scaling, 254–255; at
Tesla, 255–256
Culture-Process Map, 256–263, 257; accountability in, 261; big three processes, 260;
communication in, 261; communication of culture in, 258–259; culture enabling in, 259–
260; data in, 261; incentives in, 260; metrics in, 260; principles in, 258; resource
allocation in, 260; step 1, 257, 261; step 2, 257–258, 261; symbols in, 259; use of, 261–
262; workspace in, 261
CUPID, 169
curiosity, 255
customer delight statement, 94
customer demand, 136
customer experience (CX), 71, 72
customer identification, in problem validation, 132–134
customer insight tools, 106–108
customer journey mapping, 107–108
customer lifetime value (CLV), 145, 147; in business validation, 158
customer network behaviors, 108
customer opportunities, 93–94
customers, 49–50; focus, 33; in Shared Vision Map, 74
CX. See customer experience
cybersecurity, 58, 167

Darwin, Charles, 284n15

data, in Culture-Process Map, 261
data assets, 228–229, 241
data governance, 229–231, 241
data value templates, 108
Davy, Humphry, 144
decision-making: bottom up, 266–267; HIP-PO, 165; in planning, 27
decision rights, 178; boards, 179; in innovation team, 217; team, 179
Deere & Company, 11
definition of success, 66–68, 80, 101, 175, 177, 217
Delabroy, Olivier, 65, 136, 188, 201
design, governance and, 206–210
design thinking, 9, 20, 35–36, 88, 93, 106–107, 120–122, 133, 169, 173, 224, 234, 243, 262
Diapers.com, 146
Diconium, 222
Diess, Herbert, 220
Digital Accelerators, 205, 208–209
digital-first companies, 47
digital-native businesses, 247; culture at, 255–256
digital revolution, 6
digital strategy tools, 108
digital transformation (DX), 1; barriers to, 7–11; bottom up, 15–16; capability change in, 9–
10, 15; culture in, 246–247, 251; defining, 5–6; difficulty of, 6–7; digital strategy in, 11–
12; experimentation in, 8–9, 14–15; failure of, 4–5, 19, 20; future landscape, 48–55;
governance flexibility in, 9; growth priorities in, 8; as innovation, 22–24; mistakes in, 6–
7; organization change in, 11–12; P/O statements for, 97; priorities in, 14; roadmap, 12–
22, 87, 120, 218–219, 281–282; self-assessment, 275–277; shared vision in, 7–8, 13–
14; steps, 13; successful, 10–11, 19, 20; visual overview of, 281–282
Digital Transformation Playbook (Rogers), ix, 11, 34, 51–52, 104, 108, 152, 160, 165, 228,
281
digitization, 8; of New York Times Company, 1–3
Discovery 10X, Citibank, 131, 168, 185, 193, 218
Disney, Walt, 69, 253
Disneyland, 69
disruption, 52
diversity, openness to, 255
Doerr, John, 67
Domino’s Pizza, 10, 65
DoorDash, 101
Dorf, Bob, 35, 124, 125; on MVPs, 126
dual transformation, 203–204
DX. See digital transformation
Dynamic Yield, 236
e-books, 25
Edison, Thomas, 144
Elop, Stephen 46, 286n3
emerging technologies, 242
empathy, 255
empowerment, 270–271
enterprise level, 98
environmental changes, 161
existing solutions, 139
exits, 244; talent capabilities and, 238–239
experimentation: bottom up, 164–165; in DX, 8–9, 14–15; importance of, 119–128; in
innovation team, 217; innovation through, 121–122; of start-ups, 34–37
extrinsic motivation, 46, 286n4

Facebook, 39, 100

failure: capabilities and symptoms of, 224; of DX, 4–5, 19, 20
fashion brands, 197
feature roadmap, 153
feature rollout, 160–161
Fédération Internationale de l’Automobile (FIA), 61, 193
Fire Phone, 29
Fire tablets, 29
fixed costs, 158
flywheel, 69
Fong, Kevin, 93–94, 289n9
Ford, Bill, 44, 81, 83
Ford Motor, 46, 253, 255, 257; Smart Mobility unit, 208
Four Religions of Iterative Innovation, 35–36, 120
Four Stages of Validation, 14, 123, 127–150; business validation, 129, 143–148, 157–159,
189; competition in, 148–150; elements tested in, 128–129; metrics in, 148–150;
overlapping stages, 129–131, 130; problem validation, 129, 131–135, 151–152, 189;
product validation, 129, 139–143, 189; sequencing, 129–131; solution validation, 129,
135–139, 152–155, 189; stakeholders in, 163–164; validation in, 128–129
functional MVPs, 124–125, 146
funding: BAU, 186, 187; governance and, 171; for innovation, 205; innovation structures,
207; iterative, 37–39, 172, 177, 181–188, 186
future landscape, DX, 48–55; acuity Insurance, 53–54; appearance of, 52; BSH Home
Appliances, 53; Merck Animal Health, 52–53; scanning, 49–51; Shared Vision Map, 74–
77

Gates Foundation, 67, 147

GE Digital, 206
General Electric (GE), 4–5, 6, 22–23, 99, 177
Gerstner, Lou, 32
Glick, David, 217, 293n18
Global Consumer Bank, 168
golden handcuffs, 238
Goleman, Daniel, 47; on leadership, 286n6
Google, 5, 24, 26, 34, 63, 67, 90–91, 173, 236; reciprocal advantage, 60
Google Answers, 193
Google Glass, 192
Google Maps, 56; reciprocal advantage, 59
go-to-market skills, 243
governance, 171, 175; bottom up, 217–218; critical functions of, 230; data, 229–231, 241;
design and, 206–210; funding and, 171; ideation contrasted with, 204–205; importance
of, 169–172; of innovation boards, 217–218; rules, 171; in Three Paths to Growth, 200–
205, 201
governance flexibility: in DX, 9; at New York Times Company, 17
Greenleaf, Robert K., 270
greenlighting, 172, 179–180; ventures, 177
Greeven, Mark J., 223
Grove, Andy, 267
growth. See Three Paths to Growth
growth boards, 175–179. See also innovation boards
growth priorities, in DX, 8

hackathons, 205, 252

Hackett, Jim, 55, 253
Hackman, Richard, 173
Haier, 223
Hansen, Espen Egil, 244–245, 249, 268
Haque, Imran, 84, 235
Hashem, Omar, 70
Hassanyeh, Sami, 49
Hastings, Reed, 23, 24, 126, 250
HCL Technologies, 232
highest paid person’s opinion (HIP-PO) decision-making, 165
high-fidelity MVPs, 125
Hilgenberg, Dirk, 221
HIP-PO. See highest paid person’s opinion
hiring, 243, 260; talent capabilities and, 236–237
Hoffman, Reid, 141
Horowitz, Ben, 248
HR. See human resources
Huffington Post, 2–3, 228
Hughes, Thomas, 144
human resources (HR), 20

IBM, 22–23, 32, 234

ID.3, 220
ideal customer, 162
ideation, governance contrasted with, 204–205
IDEO, 100
Iger, Bob, 73
illustrative MVPs, 124, 138–139
Immelt, Jeffrey, 5
incentives, in Culture-Process Map, 260
incubators, start-up, 206
industry cases and examples, 279–280
information flows, 267
information technology (IT), 7, 9–10; infrastructure, 225–228; modular, 227–228; monolithic,
227–228
InHome, 154
initial public offering (IPO), 56
innovation, 199; in bottom up strategy, 267; challenges, 205; beyond core, 30–43; cost
structure of, 145; divide, 24–25; DX a, 22–24; experimentation and, 121–122; Four
Religions of Iterative Innovation, 35–36, 120; funding for, 205; independence of, 39;
island of, 165–166; labs, 205; Shotts on, 118; skills, 243; smart shutdowns and, 188–
193; structures, 172, 205–212; teams, 173–175, 176; uncertainty and, 25–26, 183;
upsides and downsides, 39–40; at Walmart, 117–118
innovation boards, 212; charter, 214–216; Corporate Innovation Stack, 214–216;
governance of, 217–218; handoffs, 215; iterative funding, 215; members of, 215;
pipeline, 215; portfolio, 215; process, 215
Innovation Report, 3–4, 16, 246
innovation structures: Air Liquide, 211; BSH Home Appliances, 211; charter, 213–214;
Citibank, 211; Corporate Innovation Stack, 212–214; funding, 207; independence of,
214; key elements of, 207; leadership in, 214; learning in, 214; mandates, 207, 213;
metrics, 207, 214; multiple, 210–212; process, 214; recruiting in, 214; resources, 214;
talent pools in, 214
innovation team: decision rights in, 217; experimentation in, 217; metrics, 216–217; P/O
statements, 216; process, 216–217
innovation teams, 212; charter, 216–217; Corporate Innovation Stack, 216–217
Institutional Clients Group, 168
insurance company, 197
Intel, 67, 267
internal rate of return (IRR), 148
Internet of Things (IoT), 84–85, 229
intrinsic motivation, 46
Intuit, 67; Succar on, 125; unique advantages, 58
involuntary exits, 238
IoT. See Internet of Things
iPhone, 25, 29–30
IPO. See initial public offering
iPod, 93
IRR. See internal rate of return
island of innovation, 165–166
IT. See information technology
Itaú Unibanco, 161
iterative funding, 172, 177, 186; innovation boards, 215; learning and, 184–186; net present
value in, 181–184; option value in, 181–184; in practice, 186–188; uncertainty in, 38,
181–184; by VCs, 37–39
iterative MVPs, 124
iterative processes, 242

Jassy, Andy, 31, 41

Jetblack, 117, 119–120, 165
Jobs, Steve, 32, 63, 93, 287n13
job-to-be-done, 152
Johnson & Johnson, 190, 234, 235

Kahneman, Daniel, 252

Kanioura, Athina, 231
Kano model, 93
Keeper Tests, 250
Kelleher, Herb, 248
Kennedy, John F., 55
Kenya, 139–140
Kerr, Stephen, 254
Kindle, 25, 93, 97–98
Klevorn, Marcy, 83, 255
know your customer (KYC), 62
Kodak, 4
Kopp, Wendy, 269
Kotter, John, 295n28
Kueng, Lucy, 45, 269
KYC. See know your customer

Lacerda, Antonio, 170, 208

lagging metrics, 127
Lamborghini, 221
Larson, Jay, 162–163
Latin America, 46, 99
Lawson, Jeff, 231
leadership, 200; bottom up strategy, 267–271; communication and, 269; culture and, 249;
defining, 268; enabling, 270; Goleman on, 286n6; in innovation structures, 214
Leadership Principles, Amazon, 249
leading metrics, 127
lead users, 134
Leahy, Terry, 41
lean start-up, 20, 34–35, 89, 93, 106, 120, 123, 126, 133, 169, 173, 217, 224, 234, 243, 262
learning, 121, 180; acceleration of, 123–124; in business validation, 147–148; in innovation
structures, 214; iterative funding and, 184–186; MVPs and, 123–124; in problem
validation, 134–135; in product validation, 142–143; sharing of, 192–193; in smart
shutdowns, 192–193; in solution validation, 137–138; uncertainty and, 184
legal regulation, 62
Lehman, Meir Manny, 294n10
Levitt, Ted, 33, 89
liaising, 177
light bulb, 144
Liguori, Stephen, 99, 177
LinkedIn, 222
local infrastructure, 62
Lonie, Susie, 139
Lore, Marc, 123

mandates, innovation structures, 207

March, James G., 203
marginal costs, 158
Mastercard, 10, 39, 52; P/O statements for, 97; Start Path, 209–210, 218; unique
advantages, 56, 58; zooming out, 66
Mastercard Labs, 289n7
M&A teams. See mergers and acquisitions teams
McChrystal, Stanley, 266
McDonald’s, 101, 236; P/O statements, 103
McGrath, Rita, 181, 182
McKinsey, 4, 47, 287n7, 295n25
Mead, Margaret, 173
Megginson, Leon C., 284n15
Merck Animal Health, 52–53
mergers and acquisitions (M&A) teams, 206
Merkel, Angela, 220
MEs. See microenterprises
Meta, 173, 236
metaverse, 65
metrics, 33, 155; in business validation, 147–148; in Culture-Process Map, 260; focus on,
127; in Four Stages of Validation, 148–150; innovation structures, 207, 214; innovation
team, 216–217; lagging, 127; leading, 127; MVPs, 126–127; of P/O matrix success,
112–113; in problem validation, 134–135; in product validation, 142–143; in solution
validation, 137–138
microenterprises (MEs), 223
microservices, 227
Microsoft, 26, 63, 247, 251
Microsoft Windows, 32
milestones, 186
minimum viable products (MVPs), 209; in business validation, 146–147; defining, 123; Dorf
on, 126; functional, 124–125, 146; high-fidelity, 125; illustrative, 124, 138–139; learning
 and, 123–124; metrics, 126–127; multiple, 125–127; in problem validation, 133–134; in
product validation, 141–142; in solution validation, 136–137; Succar on, 125–126
modular development, 242
modular IT, 227–228
monolithic IT, 227–228
motivation: extrinsic, 46; intrinsic, 46
M-Pesa, 139–140
multifunctional teams, 174, 234–235
multisided market. See platform business model
MVPs. See minimum viable products

Nadella, Satya, 88, 247, 251, 252

Nasdaq, 168
National Commerce Bank (NCB), 62, 70, 225
NBMs. See new business models
NCB. See National Commerce Bank
negative urgency, 46
Nest, 24
Netflix, 29, 255–256, 266–267; advertising on, 24; culture at, 250; IPO, 56; reinvention of,
23–24; unique advantages, 56
net present value, 181–184
Net Promoter Score, 142
new business models (NBMs), 71
new markets, 161
newspapers, 197
New York Times Company, 1–4, 16–18, 31, 40–41, 62, 67, 102–103, 113, 119, 192; 228,
235–236, 246, 267, 270–271; capability change at, 17–18; culture at, 246; data assets,
228; definition of success, 67; digital revenue, 18; digitization of, 1–3; governance
flexibility at, 17; Innovation Report on, 3–4, 16, 246; journalists who write code, 235;
New York Times Electronic Media Company, 1; North Star impact, 67; P/O statements,
102, 103; revenue, 3; share learning widely, 192; strategy goals, 114; success in
following the DX Roadmap, 16–18; talent lifecycle, 236;
NextGen Cup Challenge, 101–102
NFTs. See non-fungible tokens
Nike, 163, 229
nimbleocity, 231
Nisenholtz, Martin, 1–2
Nokia, 25, 46
non-fungible tokens (NFTs), 65
nonmonetary value, 157
nonprofits, 147
nontechnical skills, 234–235
North Star impact, 63–65, 106–107; company level, 64; Shared Vision Map, 79–80;
success of, 67
Nőthig, Dániel, 247
Nuance, 32
Nunes, Juliana, 234
Nvidia, 222

objectives and key results (OKRs), 66

Onono lab, 170, 208
“On the Folly of Rewarding A While Hoping for B” (Kerr), 254
operational excellence (OpEx), 71, 72
opportunity lens, 93–96; defining, 94; problem lens and, 95
Optimizely, 162
option value: explanation of, 181; in iterative funding, 181–184
O’Reilly, Charles A., 203
organizational structure, 33; bottom up, 262–263; technology in, 262
organization change, in DX, 11–12
Osterwalder, Alex, 93
other investment opportunities, 148
Otus, 29
oversight, 171
ownership, 61

Page One meetings, 2

Panera, 140
parallel tracks, 242
Parmenides’ fallacy, 54–55, 287n9
partner agreements, 61
partnership, 244; talent capabilities and, 239
PayPal, 4, 25
Peabody Awards, 2
PepsiCo, 231
Pfizer Animal Health, 84–85
physical retailers, 198
Pieper, Mario, 71–72, 89–90, 201
Pinkham, Chris, 30–31, 41, 286n
pipelines: sample, 191; smart shutdowns, 191
pivoting, 126
planning: decision-making in, 27; execution of, 27; perils of, 27; studying, 27; traditional,
26–30
platform business model (or multisided market) 25, 36, 132, 157, 160
P&Ls. See profit and loss statements
POC. See proof of concept
P/O matrix. See problem/opportunity matrix
Porsche, 221
Porter, Michael, 86
positive urgency, 46
P/O statements. See problem/opportunity statements
Power Point, 256, 260
press release/frequently asked questions (PR/FAQ), 107
priorities, 86–88; growth, 8; stakes of, 87; of Zoetis, 93
problem lens, 88–93; focus in, 89–91; limits of, 92–93; opportunity lens and, 95;
stakeholders in, 91–92
problem/opportunity (P/O) matrix, 109–113; brainstorm test, 112; business opportunities in,
111; business problems in, 110; combination in, 111–112; customer opportunities in,
110; customer problems in, 110; customers and level in, 110; quadrants in, 110–111;
refinement in, 112; success metrics, 112–113
problem/opportunity statements (P/O statements), 96–101, 167; for Acuity, 97; for Air
Liquide, 97; bottom up strategy, 113–115; Citibank, 103; customer insight tools, 106–
108; at different levels, 98–99; for DX, 97; ideas sparked by, 99–100; identification of,
104–115; innovation team, 216; for Mastercard, 97; McDonald’s, 103; New York Times
Company, 103; qualities of great, 100–101; shared vision, 104–106; Starbucks, 103;
strategy as ongoing conversation, 114–115; teams and leaders, 115–116; tools, 104–
115; ventures from, 101–104, 103; Walmart, 98, 103; for Zoetis, 97
problem validation, 129, 131–135; customer identification in, 132–134; existing alternatives
in, 132; learning in, 134–135; metrics in, 134–135; MVPs in, 133–134; Rogers Growth
Navigator and, 151–152; smart shutdowns in, 189; threats in, 135
Procter & Gamble, 100, 174, 177
product management, 20, 35–36, 42, 88, 106, 120, 124, 173, 233–234, 243
product validation, 129, 139–143; delivery, 155–156; learning in, 142–143; metrics in, 142–
143; MVPs in, 141–142; right to win in, 156–157; smart shutdowns in, 189; threats in,
143; usage, 155
profit, paths to, 145, 159
profit and loss statements (P&Ls), 8
project management 36, 188, 204
promotion to leadership, 260
proof of concept (POC), 124
proximity, in Three Paths to Growth, 194
Proxymity, 168
psychological data, 127

Qikfox, 65
QuickBooks, 58

RABBITs. See real actual businesses building interesting tech

Randolph, Marc, 23, 126
Random House, 25
Raynor, Michael, 152
real actual businesses building interesting tech (RABBITs), 210
reciprocal advantage, 58–65; Google, 60; Google Maps, 59; Walmart, 59–60, 60
recruiting, in innovation structures, 214
Reid, Chris, 41, 52, 131, 140–141
relatable principles, 258
request for proposal (RFP), 101
resources, 33; culture change and allocation of, 254; Culture-Process Map, 260
retail banks, 197
retaining, 244
return on investment (ROI), 121, 148
revenue models, 157
RFP. See request for proposal
Ries, Eric, 35, 123, 126, 141, 177
right to win, 55–59, 94, 105–106; in product validation, 156–157; Shared Vision Map, 77–
80; unique advantages, 55–57
rituals, 259
Robinson, Frank, 123
Rogers Growth Navigator, 150–151, 154; problem validation and, 151–152; solution
validation and, 152–155; tips, 159–160; top-line summary, 153–154
ROI. See return on investment

SaaS. See software-as-a-service

Safaricom, 139
Saint-Exupéry, Antoine de, 269
Salzmann, Ben, 53, 231, 269
Samsung Group, 59
Savoia, Alberto, 124, 137
scale economies, 158
scaling operations, 161
Schibsted, 190, 244
Schlumberger, 113
scientists, 120–121
Scrum, 35, 121, 140, 173, 217, 234, 243, 247
self-assessment, 275–277
Serra, Steve, 114, 208–209
shadow accounting, 41
Shan, Jialu, 223
Shapiro, Danny, 222
shared vision: from bottom up, 81–82; in DX transformation, 7–8, 13–14; importance of, 45–
47; P/O statements, 104–106; stakes of, 47
Shared Vision Map: customers in, 74; future landscape, 74–77; inaction in, 76; North Star
impact, 79–80; right to win, 77–80; statement of impact in, 79–80; structural trends in,
75; technology in, 74–75; unique advantages in, 77–79
shareholders: of Amazon, 73; business theory and communications of, 73–75; of Disney, 73
Shearman, Tim, 56
Shotts, Jeff, 56, 117; on innovation, 118
single-threaded teams, 174
skepticism, 127
skin in the game, 199
Skoda, 221
Smart Bow, 85
Smart Mobility unit, 208
smart shutdowns, 172; barriers to, 190–194; in business validation, 189; extraction of value
from, 192; innovation and, 188–193; learning in, 192–193; people distinguished from
projects in, 193; pipelines, 191; in problem validation, 189; in product validation, 189; as
routine, 190; in solution validation, 189
“Snow Fall,” 1
software-as-a-service (SaaS), 92, 132, 162, 229
software engineering, 242
solution validation, 129, 135–139; learning in, 137–138; metrics in, 137–138; MVPs in, 136–
137; Rogers Growth Navigator and, 152–155; smart shutdowns in, 189; threats in, 138–
139; value proposition in, 152–153
“Solve the Problem,” 249, 255, 258
Sony Group, 230
Southwest Airlines, 248
Squid Game, 24
stakeholders: in problem lens, 91–92; validation for, 163–164
Standalone Products and Ventures Group, 41
Starbucks, 101; P/O statements, 103
Start Path, Mastercard, 209–210, 218
start-ups: experimentation of, 34–37; incubators, 206; lean, 35
statement of impact, 79–80
Steinfeld, Jay, 253
stories, 251–256
strategic constraints, 61–63
strategic guidance, 177
stretch goal,10x, 94
Stripe, 4
structural trends, 50–51; in Shared Vision Map, 75
Succar, Rania, 58, 121; on Intuit, 125; on MVPs, 125–126
Sull, Donald, 114
Sulzbreger, Arthur Ochs, 1–2, 18
Super Bowl ads, 252
survival rates, 191

talent capabilities, 233–239; assessment of, 242–243; bridge plan, 243–244; combinations,
235–239; exit phase, 238–239; hiring phase, 236–237; levels, 235–239; life cycle, 236–
239; nontechnical skills, 234–235; partnership phase, 239; in Tech and Talent Map,
242–244; technical skills, 234–235; training phase, 237–238
talent pools, in innovation structures, 214
TAM. See total addressable market
Tan, Fiona, 226
Team of Teams (McChrystal), 266
teams, 172; accountability in, 175; autonomous, 174; BAU, 175, 176; decision rights, 179;
innovation, 173–175, 176, 212; M&A, 206; multifunctional, 174, 234–235; single-
threaded, 174; small, 173
Tech and Talent Map, 239–244; talent capabilities in, 242–244; technology capabilities in,
240–242
tech infrastructure, 240–241
technical debt, 226, 294n10
technical skills, 234–235
technology, 50; composable, 232; consumable, 232; distraction of, 90–91; emerging, 242; in
organizational structure, 262; in Shared Vision Map, 74–75. See also specific topics
technology capabilities: assessment of, 240–241; bridge plan, 241–242; building v. buying,
231–232; data governance, 229–231; IT infrastructure, 225–228; staging, 232–233; in
Tech and Talent Map, 240–242
TED conference, 44
telco, 198
10x stretch goal, 94
Tesco, 41
Tesla, 25, 291n15; culture at, 255–256
Thiel, Peter, 39
threats: in problem validation, 135; in product validation, 143; in solution validation, 138–
139
Three Paths to Growth, 41, 194–217, 195; challenges in, 196–200; defined, 194–196; in
different industries, 197; dual transformation, 203–204; governance in, 200–205, 201;
necessity of all three paths in, 203–205; Path 1 ventures, 195; Path 2 ventures, 195–
196; Path 3 ventures, 196; proximity in, 194; uncertainty in, 194
time frame, 159
Times Select, 192
top-line summary, 153–154
total addressable market (TAM), 152
traditional planning, 26–30
training, 244; talent capabilities and, 237–238
transparency, 175
Tushman, Michael L., 203
Twilio, 231
Tyto, 29

Uber, 56, 132

Ulbrich, Thomas, 220
uncertainty: innovation and, 25–26, 183; in iterative funding, 38, 181–184; learning and,
184; in Three Paths to Growth, 194
Under Armour, 67
unique advantages, 55–57, 57; CAA, 56; Intuit, 58; Mastercard, 56, 58; Netflix, 56; in
Shared Vision Map, 77–79; utilizing, 57–58; Walmart, 56
United Technologies Corporation (UTC), 72, 89, 114, 208, 248
Upshot, The, 17
U.S. Army, 266
UTC. See United Technologies Corporation

Vakili, Valla, 169, 173, 270

value capture, in business validation, 157–158
value drivers, 71–72; in Shared Vision Map, 80
value elements, 152
value proposition, 135; roadmap, 51–52, 108; in solution validation, 152–153
venture capital (VCs), 93, 168, 186; corporate, 206; investment stages, 37; iterative funding
by, 37–39
ventures: backlog, 104, 115; green-lighting, 177
VG Nett, 244
Virgin Media 02, 238
virtual reality (VR), 93
virtuous cycle, of Amazon, 69, 70
vision, narrow, 33
visual overview of DX, 281–282
Vodafone, 139
Vogels, Werner, 227–228
Volkswagen, 220–221
voluntary exits, 238
Vox, 2–3
VR. See virtual reality
Vudu, 192
VW.os 2.0, 221, 232

Walker, Yana, 230

Wall Street Journal, 162–163
Walmart, 101, 173, 226; InHome, 154; innovation at, 117–118; P/O statements, 98, 103;
reciprocal advantage, 59–60, 60; unique advantages, 56
Walmart Labs, 117
Walt Disney Co., 10, 69, 73; shareholders of, 73
Washington Post, 191, 253
Waze, 56
Wedell-Wedellsborg, Thomas, 88
What’s Your Problem? (Wedell-Wedellsborg), 88
What You Do Is Who You Are (Horowitz), 248
Wohland, Philipp, 238
Wojcicki, Susan, 193
workspace, in Culture-Process Map, 261
World Wide Web, 1

X (company), 41
Xbox Adaptive Controller, 252
Xerox, 32–33

YouTube, 24, 26, 82, 94

Yu, Howard, 223
Yuan, Eric, 32
Zelle, 25
Zenger, Todd, 288n18
Zoetis, 84, 85, 88, 95, 106; P/O statements for, 97; priorities of, 93
zombie projects, 190
Zoom, 32
zooming out, 65–68, 79; Mastercard, 66
 ABOUT THE AUTHOR

David Rogers is the world’s leading expert on digital transformation,

a member of the faculty at Columbia Business School, and the
author of four previous books. His landmark bestseller, The Digital
Transformation Playbook, was the first book on digital transformation
and put the topic on the map. Rogers defined the discipline by
arguing that digital transformation (DX) is not about technology; it is
about strategy, leadership, and new ways of thinking. With this
sequel, The Digital Transformation Roadmap, Rogers tackles the
biggest barriers to DX success and offers a blueprint to rebuild any
organization for continuous digital change.
Rogers has helped companies around the world transform their
business for the digital age, working with senior leaders at
corporations including Google, Microsoft, Citigroup, Visa, HSBC,
Unilever, Procter & Gamble, Merck, GE, Toyota, Cartier, Pernod
Ricard, China Eastern Airlines, NC Bank Saudi, and Acuity
Insurance, among others.
Rogers regularly delivers keynote addresses at conferences on
all six continents and has appeared on CNN, ABC News, CNBC, and
Channel News Asia, and in the New York Times, the Financial
Times, the Wall Street Journal, and The Economist.
At Columbia Business School, Rogers is faculty director of
executive education programs on digital business strategy and on
leading digital transformation. He has taught over 25,000 executives
through his programs in New York City, in Silicon Valley, and online.
His recent research has focused on new business models,
innovating through experimentation, governance for growth, and
barriers to change in digital transformation.
 For new tools and content from David, visit
www.davidrogers.digital.