BRIDGEi2i Whitepaper - The Science of Customer Experience Management

The Science of Customer
Experience Management
White Paper

Today there is a lot of buzz around customer
experience. Many companies have realized that
investments in customer experience improvement
is important not just because it helps to boost the
bottom lines of their businesses but because it
takes at least 4 to 6 times more cost to acquire a
new customer than to retain an existing customer.
Modern day communicating platforms and tools
have given more muscle power to customer's
voice. On one hand, your customers could become
brand ambassadors of your business if they are
delighted about doing business with you. On the
flip side one sad customer would soon turn to
thousands of unhappy customers at the blink of an
eye. Now, businesses have an added advantage
unlike a few decades ago. They can derive long
term success by leveraging technology retain their
customers longer by providing them a better
customer experience. With the help of
technological improvements, businesses can
understand their customers like never before.
A proactive strategist would not only leverage the
information to meet today's customer needs but
also align the business for tomorrow. Though
businesses now have more data about their
customers, it is not always effectively utilised to
manage customer experience.
Analytics per-se should be applied at every
customer touch point to derive insights which
could be used to devise strategic initiatives that
can increase customer retention and enhance
customer life time value by providing better
business experience and value to the customers.
Is Customer Experience Innovation being done right?
At a broad level, businesses have not leveraged technology to understand the customers as much as
customers do to evaluate a product / service. This is also a reason as to why the average tenure of companies
in the S&P 500 has come down from 61 years in 1958 to 18 years in 2012. At the current churn rate, about
75% of S&P 500 will be replaced by 2027.
Some companies have a notion that they pay key attention to customer experience improvement efforts. In a
research study conducted by Forrester, among 100 customer experience professionals, about 73 % of the
Interviewees said they plan to launch innovative customer experience in the upcoming year and about 66%
of them told that they have already launched innovative customer experience program. If so much of
innovation is happening in this space and happening right, we should have probably seen most of the
companies achieve a good Customer Experience Index rating by its customers. But in 2013, only 8% of the
companies received top grade from the customers in the annual Benchmarking Survey. A lot of them
certainly missed the bus here!
The Science of Customer
Experience Management

The scope of this white paper
In this white paper we shall talk about:
 Data sources for customer experience management
 Steps for effectively managing customer experience
 Approach note on quantitative data analysis
 Approach note on unstructured data analysis for mining insights about customer experience
with examples.
 Sample industry use cases that showcase benefits of implementing customer experience
management solution.
 Note on the need for real time customer experience tracking and management by enabling real time
analytics infrastructure
 While in this paper we attempt to cover reasonable depth on basic principles and approaches on
customer experience management using analytics, details about the working of advanced analytics
algorithms are not under the scope of this white paper.
DATA SOURCES FOR CUSTOMER EXPERIENCE MANAGEMENT
We will look at various data sources and categorize them under:
1. Internal – Both structured and unstructured
2. External - Structured and unstructured

1. Feedback Surveys: (Internal data - Structured)
Periodic and trigger based surveys form the core of customer experience management as these surveys give
us the most direct information about customer's perception about the quality of product /service. Periodic
surveys help measure overall customer experience by time, whereas trigger based surveys help capture the
most recent customer experience as soon as the customer interacts with the business. This event could be:
product purchase, product service or product enquiry etc.
2. Transaction information: (Internal data – Structured and Unstructured)
Structured data
Product Information such as product purchase details, purchase frequency, product type, purchase value,
product returns /abandonments, order renewals, renewal dates, etc.
Website Information such website navigation pattern, most visited web pages, inbound / outbound page
visits, product views.
Unstructured data:
Reason for product return, customer service feedback, reason for service request, searched terms in the
website, customer support information – both voice and text data from call centres etc.
3. Customer demographics: (Internal data – Structured)
Personal Information such as - gender, age, job, title, residence information etc.
4. Social Media Information: (External data – Unstructured)
Facebook likes, shares, twitter sentiment and other customer interactions with the business on social media.
Customer profile data can be of incredible use for improving customer experience. Many innovative
companies have started to tap the data that is available in customer's social profile. The data is also highly
accurate and reliable as this information is declared by customers themselves. Demographic and
psychographic details about individuals from social profiles provide a more rounded view. This information
can also be leveraged for customer segmentation. For example, a typical Facebook profile would contain
information such as birth date, marital status, location, interests and a lot of other personal data. A LinkedIn
profile contains professional credentials about the individual. This information can be used to segment and
profile customers to tailor offers and promotions that can tremendously improve customer experience.

5 Simple Steps To Effectively Manage Customer Experience
1. Create a customer centric data repository
Customers form the core of any business and this should be reflected in the way we collect and manage data.
Information captured from all the data sources should be stored in a customer centric way. We must move
away from channel centricity. All the data sources and customer touch points should be mapped to the
customer. These customer touch points could be internal (managed by the company) or external (social
media) touch points. The touch points should in turn mapped to channels of communication.
2. Capture transactional and service data
Capture all the transaction information such as number of products purchased, order/ subscription value,
order/renewal dates, product abandonments, product returns, product purchase channel, product enquiries
etc. All inbound and outbound service requests must be mapped with request details, communication date,
communication channels, resolution details, customer satisfaction rating, and customer feedback.
If there are any interactions with the customer on social media, it is imperative to capture those interactions
as well.
3. Track customer experience scores
Surveys are the most effective tools used by businesses today to capture feedback from their customers. A
well designed survey will give enough information about your customer's comfort level of doing business
with you. You can drill down to the factors that your customers appreciate about doing business with you
and to the factors which the customers want you to improve. You can also get an idea about the reasons /
factors responsible for your customers moving on to your competitor.
In order to derive the insights mentioned above, it is not just enough to analyse the survey data in isolation.
Integrate the survey responses (at the customer level) with your internal data repository and conduct
data analytics.
 Slicing and dicing of responses by various dimensions of your survey data will help you identify the
customer segments of importance.
 Advanced analytics (key driver analysis, statistical significance testing, correlational analysis, banner
tables) done on quantitative responses after integrating survey responses with your data repository
will help you to identify the drivers of engagement/satisfaction.
 Advancement in technology has made it possible to have a structured approach to mining insights
from comments captured from the open ended questions captured in the surveys unstructured data.
(We will get into that later in this white paper).
Companies must have a good understanding of the key drivers of engagement/satisfaction of the stake
holders involved with their businesses in order to provide a better stake holder experience. Key driver analysis
done on the survey data (satisfaction score as a dependent variable) allows us to zero down on the key drivers
of engagement.
This exercise typically is complex, exploratory involving a lot of statistical analysis needed to be performed by
data scientists and generally takes couple of weeks to get the results. BRIDGEi2i's survey analytics
application Surveyi2i hides all the complexities at the backend and provides you with a simple interface to
choose the algorithm of interest to perform the analysis. It has algorithms which combine machine learning
and statistical techniques, which could be used for analysing a variety of surveys.

One should also perform key attribute analysis to identify the features of demographic profiles (of
stakeholders) that influence engagement/satisfaction. Cross tabulations and banner tables with significance
tests should then be performed to validate the parameters that influence overall experience. Such analysis
when done by the customer segments (identified by analysing various sources of data) will help us
accurately identify drivers and sentiments specific to the customer segment
4. Track behavioural changes and communicate them timely
Businesses must allow the customers to communicate in the channel of their comfort. Information about
customer's preferred channel of communication must be made available for every customer. Track
behavioural changes such as changes in: purchase pattern, purchase frequency, service request type, request
frequency, preferred purchase channel, payment mode etc. Establish a real time analytics framework to track
these changes in conjunction with the most recent customer feedback (surveys). Real time analytics
capabilities enable businesses to quickly react to changing environment and facilitate timely interaction
with the customers. Timely interaction with the customers through their most preferred channel of
communication is the key for conversion and provides businesses with the best opportunity to improve
customer experience.
5. Evaluate the impact of your customer experience improvement efforts
It is necessary to quantify the impact of your customer experience improvement efforts. The metrics to
choose for quantifying customer experience are subjective and vary by industry vertical. For example, in the
retail industry, the yardstick to measure the impact of customer improvement efforts would be the
measurement of net change in: sales / revenue, customer attrition rates, average cost to serve a customer,
number of new customer additions. These numbers will certainly help you identify if you are in the right
direction with respect to your customer experience improvement efforts.
A Novel Approach For Mining Insights From Unstructured Data Using Analytics
There is certainly no doubt that there are a lot of insights hidden in comments collected from various
customer touch points. It is very difficult to manually eyeball all the comments given by the customer and
derive insights from it. We can however design a process, where we get an understanding of our customer's
voice, without losing information or putting in labourious manual efforts for reading the comments.
Putting math behind unstructured data
This idea is derived from the concept of “collective Intelligence”. It is like the voting system in a democracy,
where the nation is expected to make the right decision of choosing a political party. Before we go deeper in
to the steps, let us cover the business context with 5 basic components of unstructured data analysis:
1. Word cloud - Provides a view of recurrent words appearing in the comments. Might act as the first
step in building a taxonomy.
2. Theme cloud – Will give a gist of topics mentioned by the respondents, again aiding development of
a customized taxonomy.
3. Word Associations – It gives an idea words / phrases that has strong associations with the
key features.
4. Sentiment Analysis - It takes the tonality and mood of the sentences in to consideration and gives
the distribution for positive and negative sentiments.
5. Category Analysis – Provides the summarised view of respondents’ thoughts after a taxonomy is
built (Find topic clusters by categorizing synonymous/ representative words under each cluster).

4 Useful Steps For Extracting Insights From Unstructured Data
Step 1 – Analysis of key words and phrases
Start applying the word cloud and noun phrase analysis to get an idea of the subjects discussed by the
respondents. Do this analysis across various customer segments such as customer segmentation based on:
age, revenue, average purchase value, purchase frequency, geography, products purchased, family size, last
satisfaction rating, purchase channel etc.
Step 2 – Building a taxonomy of key features
Now that we have an idea about the most frequently repeated words and nouns, the next step is to
categorize synonymous words and noun phrases in to clusters and give a name to each cluster. Each cluster
name must collectively denote the words represented in the cluster.
Create an exhaustive list of the unique subjects discussed by the respondents by iteratively running the word
cloud and noun phrase analysis across various dimensions identified earlier.
Step 3– Analysis of associated sentiments
In order to get an idea about respondents’ mood and tonality, conduct sentiment analysis for each category
by identifying the degree of positivity or negativity for each sentiment (sentences) and sum up the sentiment
scores for each category.
Customization of the sentiment dictionary and continuous enhancement of the business rules applied can
make such analysis more accurate. Sentiments observed at a feature level is far more actionable than a high
level sentiment analysis.
Step 4 – Key driver analysis on verbatim sentences
It is possible to find the drivers of satisfaction from unstructured data by considering the percentage of
people who spoke about specific category. Identify the presence of such categories in each sentence (given
by the customer in surveys, service feedback etc.). Now regress the presence or sentiment of each category
against the customer satisfaction rating (captured in latest customer satisfaction survey).
Following this, we can achieve drivers of satisfaction for each segment. However such insights have to be
supplemented with average satisfaction scores, sentiment scores and themes discussed across segments.
Now, at an aggregate level, we would not only be able to identify the drivers of satisfaction (from
unstructured data), but also find the sentiment for every category. You save precious time without the need
of manually going through individual sentences without missing much of information but at the same time.
Let us see the above steps in action with a few use cases.

Use Cases
Use Case 1 – Unstructured data analysis done for one of the world leading logistics service provider
Snapshot of the steps followed
Sample word cloud and word phrases
Sample Categories

Sample of top issues raised and negativity associated for each category
Key drivers of satisfaction derived from unstructured data
What kind of insights am I expected to get?
Use Case 2 – Telecom Service Provider
A communication service provider wants to improve the customer experience in its business. BRIDGEi2i
designed a customer experience measurement survey and analysed it in conjunction with the internal /
external data sources.
Following are the snapshots of the insights provided:
 Dynamic customer experience tracking, via reports shared at daily frequency. Identification of the
root cause or sources of satisfaction / dissatisfaction.
 Identification of Product / Service gap through unstructured data analysis done on the Customer's
feedback through surveys and social media.
 Identification of non-profitable customer segments and deriving a rate plan change for
achieving profitability
 Customer segmentation based on: Big ticket customers, churn rate, life time value, interaction
channels etc., and identification of drivers of customer experience for each segment.
 Most preferred interaction channel analysis at individual and customer segment level analysis.
 Recommendation of yield optimisation strategies that can boost revenue and customer experience.
 Product bundling, up sell cross sell suggestions that are likely to improve customer experience and
revenue - based on customer usage patterns
 Optimisation of IVR service and recommendation of self-service portals based on call centre data
analysis. This is done by analysing the call drop out trends at IVR paths, identifying repeat calls.
Changes to the IVR are validation of customer satisfaction levels post implementation and number
of repeat incidences.

Use Case 3 – Retail groceries chain in US
One of the leading retail chains in the US wanted to measure customer experience levels and improve
the customer engagement and experience. BRIDGEi2i set up their customer experience management
infrastructure and delivered actionable insights.
 Identified the reasons for customers’ preference for a particular retail store and reasons for
customers preference.
 Customers’ retailing experience measurement across various channels. Channels that met
their expectations vs. channels that did not meet his expectations. Reasons for satisfaction
and dissatisfaction.
 Relevant product bundling, cross sell, up sell opportunities and discount coupons were suggested for
each customer segment based on purchase pattern and purchase frequency.
 Proactive follow ups, reminders to improve customer experience and pave way for customer loyalty.
 Customer rewards program that incentivise loyal customers and boost customer experience.
Use Case 4 – User experience for technology support services
A global technology company wanted to improve customer experience for their IT services. They wanted to
create elevate the user experience by converging user feedbacks across different platforms. Millions of user
feedback entries came in regularly through emails, forums, tickets and surveys. The challenge was to collate
this large amount of structured and unstructured data to understand the problems addressed at different
points in time and cluster them across different services and regions.
BRIDGEi2i performed extensive text mining on the unstructured data of emails and threads between the
initiator and the IT resolver platform to gain insights like top conversations, sentiments etc. on the customers
requiring immediate attention and systems requiring immediate resolution. BRIDGEi2i also analysed the
structured big data of service tickets to identify the common IT related issues and performed survey analysis
to highlight customer rating and sentiments along with key areas of improvement.
The user experience analysis provided a detailed drill down into their IT services & products across different
regions/departments. Insights about challenges for the user experience across different regions along with
areas of improvements and sentiments across top conversation categories with high priority incidences
helped the client make user experience much easier to maintain, reducing their maintenance costs and
improve user experience.
Use Case 5 – Customer service improvement for a group insurance service provider
The client is a large provider of group insurance. Hundreds of queries regarding claims and other related
service are made by the customers at the company's contact centre every day. The objective of the project
was to analyse the huge volume of unstructured text in contact centre agents’ comments to assess the key
drivers of service call volume and other related aspects.
BRIDGEi2i used text mining and advanced semantic analysis tools to understand the typical call initiation
reasons and time varying nature of communication between the agent and the customer. Agent notes were
textual, highly abbreviated and contained spelling mistakes which needed to be removed. Several publicly
available lexicons and grammar dictionaries are used to enable this. The histories of conversations were
broken in to single conversation to enable separate analysis as well as analysis of call resolution through
multiple calls.

Semantics based text mining methods were used to segment discussions in to various topics which allowed
identification of various reasons for calls. Analysis of service instances needing multiple calls provided
insights on key bottle necks in the claim management process. Analysis of time taken for issue resolution by
issue category helped set the right expectations on resolution time with the customers.
Insights obtained from this exercise helped the insurance company identify key bottle necks in the
process and conceptualize measures to significantly decrease service instances as well as improve
customer satisfaction.
Use Case 6– Identifying key improvement areas for a global entertainment company
The client is one of the world's largest golf entertainment companies with assets in multiple cities across US.
As an initiative to improve brand presence and perception, the client intends to leverage social media as a
platform to communicate and interact with its patrons. The client is interested in understanding:
 The reach of its social media promotion activities
 Innovative methods to identify and manage consumer sentiments as soon as a negative event has
been triggered.
BRIDGEi2i identified the appropriate sources of information, the metrics to track and the mechanism to do
this. The team developed an automated interactive social media dashboard that identifies consumer
sentiments and pin points key offerings that require immediate attention along with a host of metrics that
discover patterns in promotion response. BRIDGEi2i created 4 important metrics (share of voice, level of
engagement, feature level sentiment and attention areas) that summarizes the client's brand presence and
perception in social media.
The client was able to use these metrics to identify focus areas in terms of a brand presence or perception
improvement conversation. This enabled separate analysis as well as analysis of call resolution through
multiple calls.
Why is it critical to have a real time customer experience management solution in place?
Earlier, we touched upon the need of real-time customer experience management by using real time
analytics. Now, let us see what we should do to establish a real time analytics framework.
1. Automate data load to your customer centric data repository. Build APIs to connect with upstream
systems. Load the data in to your repository only after data extraction and transformation.
2. Choose metrics that you want to track. Create simple report templates with charts containing the
metrics of your interest. It would be a good idea to create report templates based on the report
recipient list and reporting frequency
3. Establish various triggers for analysis and reporting. Triggers could be time based or event based.
Define the events along with threshold value. (The analysis would be triggered when the threshold
limit is reached)
4. Create platform-neutral front end tools for digital report consumption. Keep the reporting interface
light and visually rich to ensure report compatibility across multiple devices (laptop, tablets, mobiles)
5. Enable the recipients to forward the report and connect with concerned authorities to decide the
next course of action. All the reports should be accessible from your centralized customer experience
management platform and all the actions performed / decisions taken must be recorded. This will
facilitate bottom up analysis at a later date.
6. Have a tracking mechanism in place. For example if a same survey is executed multiple times across
different time periods, it would be useful to save the analysis and reporting settings just once and
track the performance over a period of time.

Office
Bangalore: 389, 2nd Floor, 9th Main, HSR Layout, Sector – 7, Bangalore – 560 102
Phone: +91-80-42102154
US: 1013 Centre Road, ST # 403S, Wilmington, New Castle, DE 19805
Phone: +1 858 312 1075
Web: www.bridgei2i.com | email: enquiries@bridgei2i.com
Facebook | Twitter | Google+ | LinkedIn: BRIDGEi2i
Such an infrastructure will help your organization take decisions around improving customer experience.
The proliferation of data at an individual consumer level across social media and customer touch points,
when complimented by advancement in big data analytics using cloud computing technologies allows
businesses get a singular view of the customer .
Our customer experience management platform ExTrack, enables businesses to effectively manage customer
experience. The insights provided by this platform allow businesses to provide better experience to the
customer and orchestrate a single view of brand across channels. The objective is simple – Increase in brand
loyalty and growth to the bottom line!
Shown below is the high level view of ExTrack solution.
About BRIDGEi2i
BRIDGEi2i provides Business Analytics Solutions to enterprises globally, enabling them to achieve
accelerated business impact harnessing the power of data. These analytics services and technology
solutions enable business managers to consume more meaningful information from big data, generate
actionable insights from complex business problems and make data driven decisions across pan-
enterprise processes to create sustainable business impact. BRIDGEi2i has featured among the top 10
analytics and big data start-ups in several coveted publications.
Research has shown that after every bad experience of doing business with a company, 65 % of customers
share that bad experience with their friends, 34 % send feedback to the company, while about 25 % express
it on social media. Customer experience, if not given enough attention at the right time will lead to
exponential revenue loss. It is time to accept the fact that, customer experience management is no more a
luxury for businesses, it is now the new norm of doing business.
Reference:FORRESTERResearchQ4,2012,Globalcustomerexperiencepeerresearchpanel,onlinesurvey

BRIDGEi2i Whitepaper - The Science of Customer Experience Management

More Related Content

What's hot (19)

Similar to BRIDGEi2i Whitepaper - The Science of Customer Experience Management (20)

More from BRIDGEi2i Analytics Solutions (20)

Recently uploaded (20)

BRIDGEi2i Whitepaper - The Science of Customer Experience Management