MODULE 2

ARTICULATING BUINESS VALUE AND ANALYTICAL

PROBLEM SOLVING
INTRODUCTION
There are some very common misconceptions and myths about analytical problem solving.
Most candidates simply skim over this phrase on consulting profiles without thinking about the
meaning. This post will tell you what management consulting firms
like McKinsey, Bain and BCG mean by analytical problem solving.

This is important advice so it is worth reading carefully.

What is analytical problem solving

Good analyses are grounded in hypotheses. Can you develop hypotheses? It always
surprises us how many people do not know what is a hypothesis. A hypothesis is not
the problem. It is not a fact. It is not an opinion. It is a statement which captures the
observed phenomenon as well as the likely cause of the phenomenon. Both must be
present for it to be a hypothesis. A surprising number of candidates do not understand
this.

Are you able to reason using only the facts provided? Analytical thinkers are not
unemotional. No one is unemotional. However, analytical thinkers are able to separate
their emotion from the situation and use the data provided to arrive at a conclusion.
Analytical problem solving means reasoning using facts and logic. Past experience or
opinions which cannot be substantiated are ignored.

Can you assemble data and facts to develop an argument or line of reasoning?
Analytical thinkers can take pieces of information, compare them and decide what the
information is saying. They can assemble the information to produce new insight into
the problem rather than simply restating the information.

Business analysis model

Simply put, a business analysis model outlines the steps a business takes to
complete a specific process, such as ordering a product or on boarding a
new hire. Process modeling (or mapping) is key to improving process
efficiency, training, and even complying with industry regulations.

Because there are many different kinds of processes, organizations, and

functions within a business, BAs employ a variety of visual models to map
and analyze data.

For private circulation only Page 1

Take a look at these nine essential business analysis models to include in
your toolbox.

1. Activity diagrams

An activity diagram is a type of UML(The unified modeling language is a

general-purpose visual modeling language that is intended to provide a
standard way to visualize the design of a system.) behavioral diagram that
describes what needs to happen in a system. They are particularly useful for
communicating process and procedure to stakeholders from both the
business and development teams.

A BA might use a UML diagram tool like Lucidchart to create an activity

diagram to map the process of logging in to a website or completing a
transaction like withdrawing or depositing money.

https://www.lucidchart.com/pages/what-is-UML-unified-modeling-
language

For private circulation only Page 2

2. Feature mind maps

Business diagrams aren’t just for late-stage analysis or documentation. They

are also useful during a project’s initial brainstorming phase. Feature mind
maps help BAs organize the sometimes messy brainstorm process so that
ideas, concerns, and requests are clearly captured and categorized.

This visual ensures initial details and ideas don’t fall through the cracks so
you can make informed decisions about project direction, goals, and scope
down the line.

3. Product roadmaps

A product (or feature) roadmap outlines the development and launches of a

product and its features. They are a focused analysis of a product’s
evolution, which helps developers and other stakeholders focus on
initiatives that add direct value to the user.

The beauty of product roadmaps lies in their flexibility and range of

applications. BAs can create different product roadmaps to illustrate
different information, including:

• Maintenance and bug fixes

• Feature releases
• High-level strategic product goals

While product roadmaps are commonly used internally by development

teams, they are also useful resources for other groups like sales.

A defined product outline and schedule helps sales stay on the same page as
the developers so they can deliver accurate, updated information to their
prospects and clients. Because of their versatility and broad applications

For private circulation only Page 3

across teams and organizations, product roadmaps are a core part of an
analyst’s toolbox.

In Lucidchart, you can link data to and set conditional formatting within
your product roadmap to quickly monitor your progress. Check out the
template below!

https://www.youtube.com/watch?v=o8Zi8yvgD9k

4. Organizational charts

An organizational chart outlines the hierarchy of a business or one of its

departments or teams. They are especially helpful reference charts for
employees to quickly understand how the company is organized and
identify key stakeholders and points of contact for projects or queries.

Additionally, organizational charts prove useful for stakeholder analysis and

modeling new groupings and teams following organizational shifts.

5. SWOT analysis

The SWOT analysis is a fundamental tool in a BA’s arsenal. SWOT stands

for strengths, weaknesses, opportunities, and threats. A SWOT analysis
evaluates a business’s strengths and weaknesses and identifies any
opportunities or threats to that business.

SWOT analysis helps stakeholders make strategic decisions regarding their

business. The goal is to capitalize on strengths and opportunities while
reducing the impact of internal or external threats and weaknesses.

From a visual modeling perspective, SWOT analysis is fairly

straightforward. A typical model will have four boxes or quadrants—one
for each category—with bulleted lists outlining the respective results.

https://www.youtube.com/watch?v=mR9eICQJLXA

6. User interface wireframe

Another essential business diagram is the UI wireframe. Software

development teams use wireframes (also called mockups or prototypes) to
visually outline and design a layout for a specific screen. In other words,
wireframes are the blueprints for a website or software program. They help

For private circulation only Page 4

stakeholders assess navigational needs and experience for a successful
practical application.

The level of detail in wireframes range from low-fidelity to high-fidelity

prototypes. Low-fidelity wireframes are the most basic outlines, showing
only the bare-bones layout of the screen. High-fidelity wireframes are
typically rendered in the later planning stages and will include specific UI
elements (e.g., buttons, drop-down bars, text fields, etc.) and represent how
the final implementation should look on the screen.

https://www.youtube.com/watch?v=aqdn7vVKygA

7. Process flow diagram

A process flow diagram (PFD) is typically used in chemical and process

engineering to identify the basic flow of plant processes, but it can also be
used in other fields to help stakeholders understand how their organization
operates.

• A Process Flow Diagram has multiple purposes:

•

• To document a process for better understanding, quality control

and training of employees.
• To standardize a process for optimal efficiency and repeatability.
• To study a process for efficiency and improvement. It helps to
show unnecessary steps, bottlenecks and other inefficiencies.
• To model a better process or create a brand-new process.
• To communicate and collaborate with diagrams that speak to
various roles in the organization or outside of it.

For private circulation only Page 5

 •

8. PESTLE analysis

A PESTLE analysis often goes hand-in-hand with a SWOT analysis.

PESTLE evaluates external factors that could impact business performance.

For private circulation only Page 6

This acronym stands for six elements affecting business: political,
economic, technological, environmental, legal, and sociological.

PESTLE analysis assesses the possible factors within each category, as well
as their potential impact, duration of effect, type of impact (i.e., negative or
positive), and level of importance.

This type of business analysis helps stakeholders manage risk, strategically

plan and review business goals and performance, and potentially gain an
advantage over competitors.

9. Entity-relationship diagram

An entity-relationship diagram (ER diagram) illustrates how entities (e.g.,

people, objects, or concepts) relate to one another in a system. For example,
a logical ER diagram visually shows how the terms in an organization’s
business glossary relate to one another.

ER diagrams comprise three main parts:

• Entities
• Relationships
• Attributes

Attributes apply to the entities, describing further details about the concept.
Relationships are where the key insights from ER diagrams arise. In a visual
model, the relationships between entities are illustrated either numerically
or via crow’s foot notation.

https://www.youtube.com/watch?v=xsg9BDiwiJE

ANALYTICAL VALUE CHAIN

https://www.youtube.com/watch?v=g8p2H7EvoGM

For private circulation only Page 7

https://www.youtube.com/watch?v=g8p2H7EvoGM

The HR Value Chain model is an important tool that shows how the HR department adds
value to business goals.
It is a known fact that HR struggles to demonstrate value in most organizations. One reason is that every
organization is unique. What works for one organization may not work for another. Another reason is
that it is difficult to show value in a practical way. The HR value chain model aims to solve this problem.

The HR value chain is a process that depicts how outcomes associated with HR activities and practices
lead to organizational goals. It shows that the department has several processes and activities that result
in HR outcomes which in turn contribute to organizational objectives.
HR Activities & Processes – Some of the most important examples of HR activities include
Employee Relations, Manpower Forecasts, Training & Development, Compensation, and
Organizational Development.
These activities are measured with the help of efficiency metrics like
Cost of Hire, Learning budget, Number of candidates per post, Time to hire, Number of
vacancies, etc.
While these metrics are great at gauging how efficient HR activities are, they don’t express the
outcomes in terms of quality.

HR Outcomes – These activities lead to HR outcomes which are measured as KPIs. Some of the
examples include Employee engagement, turnover, retention, Individual and Team performance,
Absenteeism rate, Quality of hire, etc. Such metrics give useful insight into how the workforce is
doing.
Organizational Outcomes – HR outcomes are not sufficient; a business wants to make an impact
on the organizational goals in the end. The strategic goals are measured in terms of Productivity,

For private circulation only Page 8

Profit, Customer satisfaction, Market capitalization, Customer loyalty, and more. Such outcomes
add value to an organization and make the business profitable in the long run.
Key Performance Indicators – KPIs are monitored as three different types –
1. customer-focused,
2. financial, and
3. process KPIs.

HR outcomes serve the process of monitoring and the data get fed into the outcomes.

As an example, consider increasing a company’s budget for training as an HR process.

If it results in better workforce performance, it is an HR outcome.
But when the performance boost leads to increased sales, it proves the connection between
HR outcome and organizational objective.

Another example is fair compensation to employees as an HR process. It can result in better

retention which is an HR outcome. This can, in turn, result in cost reduction which is an
organizational objective. Positive relationships between HR practices, outcomes, and
organizational goals eventually lead to recognition of HR value in the organization.

HR Value Chain Templates

https://www.visier.com/blog/eight-step-model-hr-analytics/
https://www.youtube.com/watch?v=es0K_Ahvm6c

Understanding the HR Analytics Value Chain!

This value chain comprises eight distinct steps, each playing a crucial role in aligning Human Resources
(HR) strategies with corporate objectives. The objective is to elucidate the significance of each step and
emphasize the sequential nature of this analytical framework.

For private circulation only Page 9

If we focus on measuring just HRM activities, we will automatically prioritize maximizing efficiency
to reduce costs. However, this may not produce the best long-term results. Instead, we should focus on
measuring HRM outcomes, as this helps to align our processes with our goals.

For example, we would rather spend a few days longer on hiring a new employee (time to hire,
an efficiency metric) if this person will be a better fit in the company (quality of hire, an outcome
metric). The goal should be to get the best person in the right position, not to cut corners and hire
someone as cheaply and quickly as we can.

HRM activities and processes: Efficiency metrics

On the left of the chain, we find the HRM activities. These are measured using the so-called
efficiency metrics. Examples include:

Cost of hire
Time to hire/time to fill
Learning and development budget
Training time in days
Time since last promotion
All these metrics measure HR processes and give information about how efficient the HR function
is. It doesn’t say anything about how well HR is hitting its marks, a.k.a. HR effectiveness.

For private circulation only Page 10

I like to refer to organizations who solely focus on HRM processes level 1 HR organizations. Their
main focus is cost savings, reached by optimizing these efficiency metrics.

For example, if they can lower the cost of hire while keeping the time to hire metric stable, they
are more efficient. This immediately shows the big weakness of these level 1 HR organizations:
they focus on reducing HR cost – and thus approach HR as a cost-center instead of focusing on
the value that HR adds.

In other words, HR efficiency says nothing about how HR contributes to the business.

In the second category, we observe the HRM outcomes. These are the outcomes that are traditionally
seen as important HR KPIs. Examples include:

Engagement
Retention/employee turnover
Absenteeism rate
Individual performance
Team performance
Quality of hire
All these metrics provide information about how well the workforce is doing. This involves both HR
and line management.

For example, when engagement is high, HR is more effective than when engagement is low. The same
holds true for retention and (inversely) for employee absence.

Part of HR effectiveness is how well the intended HR practices are executed by managers. HR can do
a stellar job but with bad managers, employees will be more absent and much more likely to leave!

It is important to realize that most of our HR activities are aimed at achieving positive HR outcomes.
For example:

We don’t want to spend too much time on bringing in new people, otherwise we will lose the best
candidates, bringing our quality of hire metric down
We are training our people to make them perform better and retain them
We engage in wellness promotion in order to lower absence
And so on

For private circulation only Page 11

Level 2 HR organizations focus on HRM outcomes. They don’t focus on cost savings but on how they
can reach their HR outcomes in a cost-efficient way.

Organizational objectives
The last category is organizational objectives. These are the strategic goals that the organization is trying
to reach. Examples of metrics include:

Market share
Profit margins
Market capitalization
Customer satisfaction
Customer loyalty
These are the kind of outcomes that add value to the business and make the business more viable in the
long term.

Level 3 HR organizations focus on the business contribution they make with all of their people policies.
These are truly strategic HR functions.

The HR value chain in practice

A level 1 HR organization will allocate more L&D budget to employees, believing that better-trained
employees will benefit the organization.
A level 2 organization will allocate more L&D budget to employees and follow up by checking if these
investments pay off. They test knowledge retention and check if the investments lead to better individual
performance. If not, they will test and change training programs and/or training providers in order to
optimize return.
A level 3 organization does it the other way around. They know that the L&D spending was increased
because the organization wanted to become more innovative and profitable. This organization will do
all of the above and test how it impacts these two key performance indicators. They will only be satisfied
when there’s a positive relationship between the L&D spending and the key performance indicators.

For private circulation only Page 12

 The HR value chain and analytics

Analytics is a great tool to measure the effectiveness of the HR interventions

aimed at reaching these business outcomes and increasing the overall
organizational effectiveness. This relates to the two models I included at the
beginning of the article, which show the value that is added through HR
practices. In this case by hiring the right people and training them on the job.

This kind of tangible analytics evidence connects what we do in HR to tangible

financial business outcomes, proving once again the added value of HR.

https://www.youtube.com/watch?v=yIa7EL4hx0U
HR CASE STUDY

TRAINING VALUE MEASUREMENT MODELS

Employee Training Metrics

Training metrics are data points used to quantify and validate the effectiveness of
a training program. The goal of any legitimate training solution is to solve
problems. While key performance indicators (KPIs) evaluate progress toward
business goals, training metrics evaluate a training solution’s success.

When rooted in sound theory, training solutions are measurable. Examples of

training metrics include (but aren’t limited to):

• The number of employees who successfully complete training.

• The pass/fail rate of knowledge assessments.
• How well training solutions map to job functions.
• The rate of behavior change as a result of training.
• The impact of training solutions on KPIs.
• The ratio of financial return.

Significance of Training Metrics in Assessing Learning Effectiveness

Analyzing training metrics drives improvement. Often, organizations don’t
factor how to measure the impact of training until after they implement solutions.
Instead, they should identify training metrics in the training needs analysis
(TNA) phase.

https://www.youtube.com/watch?v=jR3enpv0xu0

For private circulation only Page 13

 Coca Cola Case Study | Issues with Workforce Management

Effective training metrics:

• Can help you determine whether the training improves employee and business
performance.
• Can help you identify where employees struggle, feeding into the training design
process.
• Can help you drive business results as employee behavior improves.

Mapping Employee Training Metrics to Training KPIs

Higher-level training metrics should map to training KPIs; otherwise, they won’t
represent accurate results.

For example, in Kirkpatrick’s model, level 3 determines the behavior change that
occurred as a result of a training intervention. The training metrics you gather for
level 3 should map to KPIs that include the desired behavior.

Level 4 maps to the impact training has on business outcomes. Those metrics
should map to the KPIs used to determine how the company is performing.

Strategies To Identify the Right Employee Training Metrics for Training

Program

https://www.youtube.com/watch?v=D_bKaSqN8m0

https://www.aihr.com/blog/training-metrics/

For private circulation only Page 14

 The list below outlines some useful strategies to identify the right training
metrics:

• Consult with key stakeholders before development to identify the metrics they
care about. Make sure to use your learning and development (L&D) expertise to
inform your collaboration.
• Avoid using L&D jargon when collaborating with stakeholders. Modify your
language to suit the audience.
• Focus on objective, quantifiable metrics. Subjective ratings can be misleading
and are difficult to validate.
• Determine the value of measuring the effectiveness of a training program. It takes
effort to evaluate training effectiveness, and training programs that support key
strategic outcomes should be the focus of your measurement efforts.
• Report honestly, even when faced with poor results. Accept and learn from
failure.
• Avoid highlighting low-level metrics, such as enrollment and completion rates.
• https://www.youtube.com/watch?v=D_bKaSqN8m0&t=106s

Kirkpatrick's model

For private circulation only Page 15

A large technical support call center rolled out new screen sharing software for agents to use with the
customers. They're providing training to teach the agents how to use the new software.

THE DEEP AND WIDE APPROACH

Over the past year we’ve seen a significant shift in market as organizations deal with two
conflicting realities:
1. The underlying technology of intelligent automation (IA) is a proven commodity
precluding the need for pilots and proofs-of-concept (POC)
2. The vast majority of IA programs have been predicated on pilots and POCs that have failed
to deliver any meaningful value.
This has led to the next logical question – If small projects deliver limited value, can scaled
deployments deliver scaled value?
Scaling an intelligent automation program continues to be a topic of intense focus in the market.
The analyst community frequently cites problems with scaling as being one of the biggest
concerns for enterprises wishing to increase the impact of their program. But as with many
things involved in setting up an automation program, the absence of a strategy can make scale
hard to come by.
The most common approach we see (and it’s only an approach not a strategy) is to look to
define scale in terms of the number of bots deployed and the number of technologies employed.
This addresses only one aspect of scale.
A less common, but far more impactful approach is to define scale in terms of the value
delivered to the organization from automation. While the metric is more complex (“value” vs
number of bots), this approach forces a rigor around truly transforming processes, addresses
the impact on jobs and workforce, and brings a discipline to the deployment of technology.
If your strategy includes the goal of “transforming” the work being done, you must decide
whether you plan to do the same work with a different mix of labor, (the basis of most
outsourcing arrangements) or if you are open to finding new ways to perform the work to
effectuate the desired outcomes. The implications of what kind of “transformation” you are
seeking are clearly different.
In the same way, your strategy should be clear on what you mean by achieving success
through scale. One mode, typically driven by focusing on the number of bots, might lead to an

For private circulation only Page 16

approach that scales by rapidly spreading across the entirety of the organization. This ‘wide’
approach to scale has some benefits – but can be quite limited and will not help realize the
more profound benefits that can be derived from reimagining processes and functions to take
full advantage of automated capabilities, and ignores that scale can be both horizontal (wide)
and vertical (deep).
During early planning phases, you must consider whether you would rather have automation
at a basic level across the entire set of workstreams, (wide) or if you would prefer to capture
all of the available opportunity in a given business area (deep)? You would likely expect that
the answer is both, but that ignores that the approach to achieving each type of scale can differ,
and is driven by your definition of transformation.
“Transformation” means more than just changing the mix of labor. Following the 3 D’s of
Intelligent Automation as a guide, we can see where the approach to each type of scale can
be different.

The 3 D’s…

1. Decompose – breaking down the processes to be automated and building them back up
with an eye toward efficiency and an “automation-first” design ethic.
2. Digitize – identifying the right tool to move the work in question from delivery by a
human-based workforce to a virtual-based (or blended) workforce that requires digital
input.
3. Deliver – working with all the stakeholders to design and deliver an orchestration of
service between the virtual and digital workforces to improve performance across the
targeted measures.
Here’s a quick look at wide vs. deep scaling:

Wide scale
• Moves rapidly across the business
• Evaluates current processes against defined automation criteria
• Releases fractional capacity back to the business
• Supports an agile model and incremental functionality

Deep scale
• Focuses on specific processes within designated areas of the business
• Takes a deep dive into data requirements
• Consolidates how automation can be deployed to achieve the desired outcome while
further enhancing process performance
• Requires significant investment in design to enable automation development
Further, as design of future state processes will require corresponding changes to associated
jobs and organizational constructs, a deep approach will more closely resemble a BPO or
Shared Service transition where careful attention needs to be paid to the organizational
transformation and how work transitions from the current state to the future state. As a result,
the benefits associated with the new automated process often go well beyond capacity released
to realized savings through tangible cost reduction and measurable impacts in process
performance (compliance, service levels, etc.). The implication of affecting change at this level
is that success cannot be measured at “go-live” and significant effort should be invested in the
workforce transformation and the measurement and realization of benefits.
Overall there is value to be derived from both approaches and trade-offs associated with each.
The wide approach allows you to move quickly, deliver quick wins, and in many cases
minimize initial investment, but the overall value may not be at the level anticipated. Further,
there is a risk that the automation impairs future process transformations.

For private circulation only Page 17

The deep approach will take longer and require capabilities beyond automation
implementation, but the value delivered can be orders of magnitude greater.
As you consider your own automation programs it’s crucial to understand your overall
objectives and the business challenges you are looking to solve, so you can develop a strategy
that is appropriately aligned. In our next blog, we’ll discuss how a focus on “quick wins” can
lead to an approach that runs counter to a larger organizational objective to deliver fundamental
change.

IMPACT OF ANALYTICS IN BUSINESS OUTCOME

Nowadays, businesses can collect a lot of information about customers at every step
they take.

This includes things like using apps, clicking online, and being active on social
media, creating a unique set of data for each person.

But, not too long ago, it would have been weird to think that customers would share
personal details like when they wake up or what they had for breakfast.

Shift in Customer Behavior:

For private circulation only Page 18

 • In the past, people were more private about personal stuff, but now, things
have changed.
• People are more okay with sharing details about their daily lives, like routines
and preferences.

Rising Customer Expectations and Business Benefits:

Because of this change, customers now expect businesses to use data and analytics
to make things better for everyone.

Navigating the Changed Scene:

• Businesses need to be careful when using data. It's like walking on a new path.
• It's important to find a balance between using data for good things and
keeping customer info safe.

Adapting and Doing Analytics Right:

• Businesses have to be flexible and adapt to what customers expect and how
they share data.
• Using data responsibly is a big deal. It means using data in a way that respects
what people are okay with sharing.

1. Proactivity & Anticipating Needs:

Organisations are increasingly under competitive pressure to not only acquire
customers but also understand their customers’ needs to be able to optimise customer
experience and develop longstanding relationships.

By sharing their data and allowing relaxed privacy in its use, customers expect
companies to know them, form relevant interactions, and provide a seamless
experience across all touch points.

For private circulation only Page 19

Thus, companies need to capture and reconcile multiple customer identifiers such as
cell phone, email and address, to one single customer ID.

Customers are increasingly using multiple channels in their interactions with

companies, hence both traditional and digital data sources must be brought together
to understand customers’ behaviours.

Additionally, customers expect and companies need to deliver contextually relevant,

real-time experiences.

2. Mitigating Risk & Fraud:

Security and fraud analytics aims to protect all physical, financial and intellectual
assets from misuse by internal and external threats.

Efficient data and analytics capabilities will deliver optimum levels of fraud
prevention and overall organisational security: deterrence requires mechanisms that
allow companies to quickly detect potentially fraudulent activity and anticipate
future activity, as well as identifying and tracking perpetrators.

Use of statistical, network, path, and big data methodologies for predictive fraud
propensity models leading to alerts will ensure timely responses triggered by real-
time threat detection processes and automated alerts and mitigation.

Data management alongside efficient and transparent reporting of fraud incidents

will result in improved fraud risk management processes.

Furthermore, integration and correlation of data across the enterprise can offer for a
unified view of the fraud across various lines of business, products, and transactions.

Multi-genre analytics and data foundation provide more accurate fraud trend
analyses, forecasts, and anticipation of potential future modus operandi and
identification of vulnerabilities in fraud audits and investigations.

https://www.youtube.com/watch?v=g8UPjC3-H2k

3. Delivering Relevant Products:

Products are the life-blood of any organisation and often the largest investment
companies make.

The product management team’s role is to recognise trends that drive strategic
roadmap for innovation, new features, and services.

For private circulation only Page 20

Effective data collation from 3 party sources where individuals publicise their
rd

thoughts and opinions, combined with analytics will help companies stay
competitive when demand changes or new technology is developed as well as
facilitate anticipation of what the market demands to provide the product before it is
requested.

4. Personalisation & Service:

Companies are still struggling with structured data, and need to be extremely
responsive to cope with the volatility created by customers engaging via digital
technologies today.

Being able to react in real time and make the customer feel personally valued is only
possible through advanced analytics.

Big data offers the opportunity for interactions to be based on the personality of the
customer, by understanding their attitudes and considering factors such as real-time
location to help deliver personalisation in a multi-channel service environment.

https://www.youtube.com/watch?v=i35xfEGAu-I

5. Optimizing & Improving the Customer Experience

Poor management of operations can and will lead to a myriad of costly issues,
including a significant risk of damaging the customer experience, and ultimately
brand loyalty.

Applying analytics for designing, controlling the process and optimizing business
operations in the production of goods or services ensures efficiency and
effectiveness to fulfil customer expectations and achieve operational excellence.

Advanced analytical techniques can be deployed to improve field operations

productivity and efficiency as well as optimize an organisational workforce
according to business needs and customer demand.

Optimum utilisation of data and analytics will also ensure that continuous
improvements are instigated on an on-going basis as a result of end-to-end view and
measurement of key operational metrics.

For example, many organisations, inventory is the largest item in the current assets
category - too much or not enough inventory can directly affect a company’s direct
costs and profitability.

Data and analytics can support inventory management by providing uninterrupted

production, sales, and/or customer-service levels at minimum cost.

For private circulation only Page 21

The use of data and analytics can provide transparency into current and planned
inventory positions as well as deliver insight into drivers of height, composition and
location of stock and aid the determination of inventory strategy and decision
making.

Customers expect a relevant, seamless experience and for companies to know them
wherever they engage.

https://www.youtube.com/results?search_query=how+orders+are+processed+on+amazon
https://www.youtube.com/watch?v=IMPbKVb8y8s

Structure and Team Building:

For private circulation only Page 22

In the initial stages of organizational development, individuals or small groups often handle data
gathering and interpretation. However, as the organization grows, the necessity for a dedicated data
team becomes apparent. This team, alternatively known as the data team or analytics team, assumes the
responsibility of managing and interpreting data. This academic exploration delves into the key players
on a data analytics team and the crucial factors to consider when building such a team.
Key Players on a Data Analytics Team:
1. Data Scientist:
• Utilizes advanced mathematics, programming, and tools for large-scale analysis.
• Informs and shapes data projects through identifying challenges and designing
algorithms.
2. Data Engineer:
• Designs, builds, and maintains datasets for data projects.
• Collaborates with data scientists and analysts, focusing on infrastructure preparation.
3. Data Analyst:
• Performs reporting and direct analysis using cleaned and transformed data.
• Responsible for maintaining dashboards, generating reports, and using data for
forecasting.
4. Advanced Positions:
• Management or leadership roles, such as data manager, data director, and chief data
officer.
• Essential for overseeing and directing the overall data strategy of the organization.
Factors to Consider When Building a Data Team:
1. Team Size:
• Consider the volume of data and the number of projects the team will handle.
• Assess whether the team will serve a specific stakeholder or department or support the
entire organization.
2. Centralization:
• Decide on the level of centralization based on organizational needs—centralized,
decentralized, or a hybrid model.
• Weigh the pros and cons of each approach, understanding its impact on data team
structure and governance processes.
3. Data Strategy:
• Align the data team structure with the overarching data strategy of the organization.
• Consider the level of data-driven decision-making and the required resources for
significant analysis.
The Value of the Data Team:

For private circulation only Page 23

 • Emphasizes the critical role of a highly skilled data team in organizations pursuing data-driven
decision-making.
• Underscores the significance of key roles—data scientists, data engineers, data analysts, and
managerial positions.
• Acknowledges the importance of understanding the diverse professional roles and
responsibilities within a data team.

Building the Cube:

In the realm of Online Analytical Processing, terms like "cubes," "dimensions," and "measures" can
seem perplexing, particularly for those new to modeling. This discussion aims to demystify these terms
and provide clarity on the concept of cubes, their structure, and their utility in OLAP modeling.
Defining Cubes:
• A cube is essentially a multi-dimensional dataset used in OLAP modeling.
• Its primary functions include holding and collecting data from users and performing
calculations as needed.
• Conceptually, a cube's data structure resembles that of a multi-dimensional spreadsheet.
Understanding Dimensions:
• Dimensions play a crucial role in forming cubes, although a detailed exploration of dimensions
will be covered in a subsequent video.
• One-dimensional spreadsheets are familiar, such as creating a list of book genres, quarters,
regions, or products.
Two-Dimensional Analysis:
• Visualizing a two-dimensional spreadsheet is relatively straightforward. For instance, tracking
monthly sales in columns and genre sales in rows constitutes a two-dimensional analysis.
• In this context, "sales" is considered a measure, representing a quantifiable aspect.
Moving to Three Dimensions:
• Expanding to three dimensions can be challenging to grasp initially. Consider the example of a
bookstore owner tracking monthly sales (columns), genre sales (rows), and region (additional
dimension).
• This scenario illustrates a three-dimensional analysis.
Beyond Three Dimensions:
• Despite the term "cube," these structures can encompass more than three dimensions.
• Dimensions, when combined into a cube, become potent tools for comprehensive data analyses.

• Cubes, with their multi-dimensional nature, offer powerful insights when analyzing complex
datasets.

For private circulation only Page 24

 • While dimensions individually may not provide significant insight, their combination within a
cube unleashes their analytical potential.
• The visual representation of a cube facilitates the understanding of how multiple dimensions
contribute to a holistic data analysis.

https://www.youtube.com/watch?v=pN85IFe-VK8

For private circulation only Page 25