USING CUSTOMER RELATED DATA

Dr. Yaser Al-Abdi

 CHAPTER OBJECTIVES
• How analytical CRM supports strategic and operational CRM.
• How analytics support customer management strategy and tactics,
throughout the customer lifecycle, in the sales, marketing and
customer service functions.
• How standard reports, online analytical processing (OLAP) and data
mining generate insights for CRM users.
• That data mining works in a number of ways: by describing and
visualizing, classification, estimation, prediction, affinity grouping
and clustering.
• The types of analytics that apply to structured, unstructured and
‘big’ data.
• Why it is important to understand the differences between nominal,
ordinal, interval and ratio data before selected analytical procedures.
 Customer Related Data
• Analytical CRM is the process through which
organizations transform customer-related data into
actionable insight for either strategic or tactical
purposes.
• Analytics can provide a deeper insight into the
customer, and the cost to serve each customer.
• Operational CRM is also supported by analysis of
customer-related data, which involves deployment of
automated solutions in the sales, marketing and
service areas.
 Customer Related Data
• CRM software applications generally allow users to
produce simple descriptive reports, but for deeper
insights other forms of analysis are often needed.
• Amazon.com who messaged their online customers
with ‘people who bought this also bought that’
recommendations based on simple correlations.
– They work by identifying and statistically quantifying
relationships in historical transactional and demographic
data – this is called model development.
 ANALYTICS FOR CRM STRATEGY AND TACTICS

• Customer lifecycle as an evolution through

three phases: customer acquisition, customer
retention and customer development
• The goal of customer development is to
increase the value to the business of retained
customers, by cross-selling and up-selling, or
adjusting service levels to improve customer
profitability
 ANALYTICS FOR STRUCTURED AND
UNSTRUCTURED DATA
• The most advanced form of unstructured data
analytics currently is text analytics. Text analytics
extracts relevant information from unstructured text
files, and transforms it into structured information.
• There are a number of text mining tools that can
help. SAS Text Miner®, for example, enables users to
convert text, audio and other files into a format from
which it is possible to extract information by revealing
the themes and concepts that are concealed in them.
 ANALYTICS FOR STRUCTURED AND
UNSTRUCTURED DATA
• Sentiment analysis. An important goal of
analyzing textual data is to get an insight into
what customers feel and think about a
company, brand, product, service, person or
group.
– Typically, if a customer feels good towards an
entity, it is classified as a positive sentiment. If the
perception is bad, it can be considered as a
negative sentiment.
 BIG DATA ANALYTICS
• Big data is characterized by:
• Volume. Whilst some big data assets do
include structured data, much big data are
unstructured.
– The volume of data is set to increase dramatically
with the advent of the ‘Internet of Things’ – the
online linking of traditionally ‘dumb’ products into
an intelligent system.
 BIG DATA ANALYTICS
• Variety. Big data are collected from new
sources that have not been mined for insight
in the past.
– Traditional analytical processes applied to
structured data cannot cope with the
heterogeneity of big data, which includes email,
social media posts, video, images, blogs, location
and sensor data
 BIG DATA ANALYTICS
• Velocity. Big data are not just batched data,
but also streamed and produced in real time.
Streamed data do not reside quietly in back-
office relational databases ready to be
analyzed periodically. Streamed data update
continually
 ANALYTICS FOR STRUCTURED DATA
• There are several different types of data kept in relational
databases.
• A fundamental distinction is made between categorical and
continuous data.
– Categorical data, also known as discrete data, are data about entities
that can be sorted into groups or categories, for example product
types, gender or country. Categorical data that are unordered are
called nominal data (e.g. customer name). Categorical data that are
ordered are called ordinal data (e.g. a list of customers ordered by
sales revenues).
– Continuous data are data that can take on any value within a finite or
infinite range. Continuous data can be either interval data or ratio data.
Interval data are measured along a continuum that has no fixed and
non-arbitrary zero point. Temperature scales are interval scales.
 Nominal, ordinal, interval and ratio scales

• Nominal data are used only to classify, identify or

categorize. Unique Customer Numbers, gender
classification (M/F) and email addresses are examples
of nominal data.
• Ordinal data rank the variable being measured. Ordinal
data tell you that an observed case has more or less of
some characteristic than another observed case. you
would know that the number one ranked customer
generates most sales, but the size of the gap between
first and second could not be known from the rank
order.
 Nominal, ordinal, interval and ratio scales

• Interval data: not only do interval data identify rank

orders, but the distance between the rankings is also
known. Many survey instruments use interval scales
to measure attitudes and opinions.
• Ratio data, unlike interval data, have a fixed and
absolute zero point. Ratio measures also have all the
properties of nominal, ordinal and interval measures.
– Ratio data, for example, would enable you to claim that a
customer paying $150 for a room spent twice as much as
a customer paying $75.
THREE WAYS TO GENERATE ANALYTICAL INSIGHT

• Standard reports
• Reporting can take the form of simple lists of
information such as key accounts and annual
revenues, to more sophisticated reports on certain
performance metrics.
• Most CRM technologies enable the automated
creation of periodic reports. Examples include
monthly reports to sales management about sales
rep activity and performance against quota, and
daily reports of call centre activity
 THREE WAYS TO GENERATE ANALYTICAL
INSIGHT
• Online analytical processing (OLAP)
• data stored in a data mart to be subjected to
analysis and ad hoc enquiry.
• A data mart is typically a subset of data that is
held in a larger data warehouse. For example,
a sales department would own a data mart
containing only sales-related information.
 Data Mining
• Data mining is the application of descriptive and
predictive analytics to large datasets to support the
marketing, sales and service functions.
– Sometimes the purpose of analytics is simply to describe some
phenomenon.
– Once a description has been produced, many analytics
packages offer users an array of visualization tools such as
charts, graphs, plots, maps, dashboards, hierarchies and
networks of many kinds to help users understand the
information.
– you might have developed a hierarchy of existing customers
based on their CLV, and created a word profile of each group.
 Data Mining
• Analysts use affinity grouping procedures to
find out which things go together. Affinity
grouping is based on finding associations
between data.
– Affinity groupings can be used to identify cross-
selling opportunities, or plan store layouts so that
associated items are located close to each other.
 Data Mining
• There are two approaches to data mining:
– Directed data mining (also called supervised,
predictive or targeted data mining) has the goal of
predicting some future event or value.
– Undirected (or unsupervised) data mining is
simply exploration of a dataset to see what can be
learned. It is about discovering new patterns in
the data.