Personalization vs.

Customization

User experience issues- A case study

Vinay Mohanty & Devesh Jagatram

CONFIDENTIAL

Tuesday, December 7, 2021

All Rights Reserved
Contents

» Personalization - An Overview
» Personalization vs Customization
» Ideal Personalization Scenario
» Best Practice Case Studies
» Customization (My Yahoo!)
» Personalization (Amazon & Gmail)
» ANNEXURE - I

2
Personalization - An overview

» What is Personalization?

Personalization is the process of tailoring content to individual users' characteristics or

preferences.  It is a means of meeting the customer's needs more effectively and
efficiently, making interactions faster and easier and, consequently, increasing customer
satisfaction and retention.

3
Personalization - An overview

» Web personalization is delivering different and unique content to each

individual customer based on the customer's interests. It's your own version of
the web site, a site just for you.

» The real benefit of personalization occurs when an individual has an "aha!"

experience that occurs when the content adapts itself based on the person's
profile, and provides something new, different, and possibly unexpected.

4
Personalization vs Customization
Customization Personalization

You explicitly specify what you want. It Personalization uses implicit interests
gives explicit user control (you are and learns what you like from your
what you say you are) actions (you are what you click on
and what you buy)

E.g. choosing which information E.g. getting the information related to

elements might be presented on a one’s preferences and regular activities
home page, choosing display colors,
etc.

Customization is at a generic Personalization is at the data/ content

"feature" level, where you see/hide etc served level
what you want

My Yahoo is an example of Amazon.com is an example of

customization where you tell the site personalization where the latter learns
what you want your interests and adapts to your needs

5
 Personalization vs Customization

Customization
my.yahoo.com
www.amazon.com
Personalization

6
Ideal Personalization Scenario

» A sample scenario depicting the ultimate goal of personalization.

The holy grail of e-commerce is always to achieve a real world experience , online.
Something like …

»Sales Person: "Hi, how may I help you?“

»Customer: "I'm looking for a new digital camera.“
»Sales Person: "Great, did you have a model in mind?“
»Customer: "Uh, no, I'm not sure which model I need. I was hoping you could help me.“
»Sales Person: "Great! Well, let me ask you a few questions ...“

Personalization aims at creating such (like the above conversation) a

compelling and strong user experience in on the web by anticipating the user’s
needs i.e. the questions the customer is most likely to ask.

7
Best Practice Case Study for Customization

» My Yahoo!
» http://my.yahoo.com

» Need for Customization

» Prioritize content
» User Controlled content
» Customized Look & Feel for each user
» Promote regular usage of My Yahoo!

8
Customization features in My Yahoo!

Name

Single point access

with latest activities
related to each service

9
Customization features in My Yahoo!

Choice of services to
be listed (users can
show / hide sections)

10
 Customization features in My Yahoo!
Prioritizing services by
specifying their
placement on screen

Choice of colors / layout

and other settings
pertaining to UI

11
Customization features in My Yahoo!

Allowing users to add

non-Yahoo! sections –
providing access to all
frequently used services
/ sites from one point

12
Best Practice Case Study 1 :for Personalization

» Amazon
» http://www.amazon.com

» Need for Personalization

» Promoting sales of various products
» Build a relationship that encourages the customer to return
» Provide customized experience for online shopping to make it as real as in-person
shopping
» Target each individual customer differently by providing one-to-one marketing
» Predict customer needs and purchase patterns
» Identify customer characteristics by initiating 2 way communication
» Make the interaction efficient and satisfying for both parties

13
Personalization features in Amazon

Recording customer preferences by

asking her to rate/tag products
( 2 way communication)

14
Personalization features in Amazon

Advertisements related to
what’s on wishlists

Advertisements related to
what’s on cart

15
Personalization features in Amazon

Name

Recommendations based
on preferences of people
with similar profile

Recommendations based
on browsing history

16
Personalization features in Amazon

Recently Viewed Items

17
Best Practice Case Study - 2 for Personalization

» GMail
» http://www.gmail.com

» Need for Personalization

» Promote sponsor sites
» Predicting sites users might be interested into

18
Personalization features in GMail

Informing users about

sites they may be
interested into based on
their email communication

19
ANNEXURE - I

20
Why personalize the web ?

» In an age of information explosion,

personalization offers a way to focus on
customers / users

» Personalization doesn't waste time showing

you what everyone else sees and then asking Business Applications

you to search or display / hide things

» Better serve the customer by anticipating

needs
Customer Database

» Make the interaction efficient and satisfying

for both parties

» Build a relationship that encourages the

customer to return for purchases

21
Types of Personalization

There are three types of Personalization:

Content 1

» Simple filtering Decision Content 2

A site displays content based on predefined Content 3

groups of site visitors.

» Rules engines Content 1

In a rules based system, the site owner Content 2

defines a set of business rules which
determine what category of content is shown Content 3

when a certain profile type visits the site.

» Collaborative filtering Similar preferences

The preferences of the community of people User

with similar characteristics are used to
predict appropriate content. Users with similar
preferences

Recommended
Content

22
Approaches to gather user data & Best practices

» Fill-in Profile/ forms: The user has to do lot of work filling complex
questionnaires, which repels them.

» Click-stream Analysis/ Web Usage Mining: It does group profiling, but privacy
issues remain.

» Cookies: ‘Tag’ / identify the user and his browser, but security issues remain
with cookies.

» Measure directly by asking the user to rate products.

» Collaborative Filtering: It is fast emerging as best practice for personalization

as it doesn’t have the above drawbacks.

23
Collaborative Filtering – A Best Practice

» In Collaborative Filtering, choice of content is based on the assumption that this

particular user will value that which the like-minded people also enjoyed. The
preferences of the community of like-minded people are used to predict
appropriate content.

» The user’s tastes are either inferred from their previous actions (for example
buying a book, or viewing a product is assumed to show an interest (or taste)
for that product).

» This method has an advantage of speed and efficiency in computation, thus

delivering rapid feedback. The reliance on a ‘critical mass’ of users can be a
problem for collaborative filtering.

» Recommendation systems use collaborative algorithms (they are the programs

which attempt to predict items like movies, music, books, news and web pages
that a user may be interested in, given some information about the user's
profile).

24
How personalization software can help ?

» Personalization software contain a dynamic recommendation system based on

collaborative filtering. E.g. IBM Like Minds is software that could be used with
any e-commerce application. It analyze user interactions that occur on websites
and generate real time predictions and recommendations for users.

» For example, the following types of transactions may be recorded:

» Products a user has purchased

» Items added or removed from a shopping basket

» A history of the user's navigation throughout the application

» Products that go best with a product that the user has already selected

» Any action or series of actions that are meaningful for a site

25
Domains where we can use personalization software

The personalization software could be used in a variety of situations:

» e-Retailer promotion and personalization web sites

» Financial portal content recommendation and personalization web sites

» Help desk or on-line technical support content recommendation web sites

» Gift recommendations for e-Retailer

» Music, movie, book, or other product rating and recommendations

» Travel - trip planners

26
Personalization software products

» Red Dot. Live server

» Plum Tree (Enterprise web Develop Kit, EDK)

» IBM Like minds

» Autonomy's Agentware 1.1

» Response Logic

» Broadvision Personalization tools

27
Is Personalization needed? Where will it not work?

» Computer , Don’t 2nd guess me !

» “Don't stereotype me - just give me the options because I prefer choosing for myself
rather than having the computer tell me what's good for me”

» What users want on one visit may be very different from what they want on the next.

» Don’t ignore basic design & User Experience

» Web personalization is much over-rated and mainly used as a poor excuse for not
designing a navigable website.

» There is no substitute for good design, usability engineering and proper information
structure.

28
Is Personalization needed? Where will it not work?

» Privacy / ethical issues?

» Consumers are clearly expressing an interest in a personalized experience. But at the

same time, they are expressing significant fear about security of their personal
information.

» “Should I share my likes & dislikes with anyone for a more personalized online
experience?”

» Paradox of the active user

» Users never want to waste time teaching/training the system about themselves by filling
complex forms, questionnaires. They want to do, explore and want results, right away.

» Good personalization requires the system to know a lot about the user. This is in direct
conflict with the paradox of the active user.

29
Thank You

30