3. Analytical Aspects of CRM
Data as Infrastructure
3
Sourced from
Diverse sources
Centralised
Used by the entire
Organisation
Requires Quality
Control
4. Analytical Aspects of CRM
Managing Data Quality- Issues
• Current , Complete, Correct, Unique
8. Database Types
Relational Database Non Relational database
Structured Tables
Uses Structured Query Language
Simple , Accurate
Easy to access
Flexible
Performance issues
Used by Content Management
systems like Youtube etc.
Handling unstructured data
Fast
Open source
Readability
Limited functionality
9. Developing a Customer Profile
Key Characteristics Used
Segment
Transactions
Satisfaction
score
Identity &
Lifestyle
Channel
Pref.
Products
Customer
value
Includes Hard & Soft Elements
19. Data Analysis
Types of Questions & Objectives
19
Segmentation &
Selections
• Create
homogenous
addressable
groups
• Creation of Lists
for Marketing
Campaigns
Acquisition &
Selections
• Channels of
interaction and
onboarding
• Data Quality of
Customer list
Customer
Analyses &
selections
• Retention
• Relationship
Development
• Cross Selling Opp
• Deep Selling Opp.
• Upselling Opp.
Channel &
Comms.
Analyses
• Campaign
effectiveness
• Channels impact
• Lists quality
20. Data Analysis
Preparing for Analysis
20
Selection of Variables
Random Sample Analysis
Training set to develop the model
Validation set to Test Reliability
21. Data Analysis
A/ B Testing
21
Two treatments and one acts as the control for the other
22. Final Data Analysis
Data Mining Techniques
22
Neural
Networks
• Computing systems
with interconnected
nodes like neurons
in the human
brain.
• Using algorithms,
they can recognize
hidden patterns
and correlations in
raw data, cluster
and classify it, and
– over time –
continuously learn
and improve
Evolutionary
Computation
• A Collection of
Algorithms based
on evolutionary
Biology Processes
• Can also be defined
as Optimisation
Program routines
for a defined
Population,
eliminating
members basis
factors till the fittest
is arrived at.
Association
Rules
• Conclusions based
on the relationship
between
characteristics of a
defined group and
aspects of their
behavior.
Case Based
Reasoning
• A current problem
is addressed via
precedents or
historical data.
25. Segments
Traits
25
Size , Purchasing Power & Characteristics of the segment
Measurability
Large enough in size to be Profitable
Substantial
Can be reached and Serviced effectively
Accessibility
Allows effective Marketing Plans to be created for it .
Actionable
Unique and respond to stimuli differently
Differentiated
27. Segmentation Techniques
Cluster Analysis & K Means Cluster Analysis
27
k-means cluster analysis is an algorithm that groups similar objects into
groups called clusters. The endpoint of cluster analysis is a set of clusters,
where each cluster is distinct from each other cluster, and the objects within
each cluster are broadly similar to each other.
33. 33
Segmentation Research
Chaid- Chi Square Automated Detection
• Variable chosen by
Marketer
• Creates Multiple sub-
segments
• More accurate than
RFM
Used to build a predictive model to outline a specific customer group or segment
34. 34
Segmentation Research
CART Analysis
• Two groups/ variable
• Yields Only Two segments
• Works on a Ratio Scale
• The best split point of each
input is obtained.
• Based on the best split points
of each input in Step 1, the
new “best” split point is
identified.
• Split the chosen input
according to the “best” split
point.
• Continue splitting until a
stopping rule is satisfied or no
further desirable splitting is
available.
35. 35
Segmentation Research
Response Gain Chart
• Based on historical
data and test models
• Enables choosing the
most responsive
sections
• Relationship Phase a
key variable
37. 37
Retention & Cross Sell Analyses
The Importance of Customer Retention
• Customer – Definition ?
38. 38
Customer Retention Analyses
Choosing the Variables
Stayers Vs.
Quitters
Which
Variables?
Revenue
Response to Comms.
Time Period
External Factors
Cust. Experience
Cust. Satisfaction
Destination? Internal or External
39. 39
Customer Retention Analyses
Developing the Model
Data set: 50:50 ratio of current & former customers
Training & Validation set
Correct Classification of Observations
Combination of Classification for best results
40. 40
Cross Selling
What is it
Opportunities
Product
Centric
Pdts from
different
cat.
Expanded
Product
Over Time
More of the
same
Two
different
Products
Purchased
later
45. 45
Cross Selling
Analysis in Action- Factors
During Customer
contact
Based on
Probabilities
Context dependent
Outreach
Based on
Probabilities
Segment Rankings
Costs & Margins
Testing the
Probabilities
47. 47
Impact of Marketing Activities
Evaluating The Sales Process
• Helps in Targeting
• Generate Cross Sell
Calendar
• Align with
Organisation’s Objective
• Enable Faster closing
• Reduce Costs
• Contribute to LTV
48. 48
Marketing Activities
Learning with Experiments
Experiment
conditions
Internal
Sanctity of
environment
Cause & Effect
Measurement
External
Mirror Real world
Representation of
sample, stimuli &
Response
Variables
• Experimental Vs Control
Group
• Null measurements
• Application of Stimuli
• No. of Experimental
Groups used
49. 49
The Learning Organisation
Creates Knowledge
from Experience
Team Learning -
One to Many
Helps develop
mental models
Shared vision
Links Learnings to
Operations
Feeds in to Strategy
Cross functional Knowledge
Sharing & Mgmt
52. 52
Evaluating Results
Lifetime Value
• Avg Transactions x # of Transactions x Retention Period- Costs
• Past Data used to Project future expected earnings
• Determines the Profitability of the Customer
• Determines the effectiveness of Strategy & Marketing Plans
• Enables forecasting & Strategy formation
• Applicable for a limited time period in the future
54. 54
Evaluating Results
Non Economic Value
Goes Beyond
immediate
transactions
Dependant on Value
attributed by Buyer
& Supplier
Deduced from
Relationship
Patterns/Lifestyles
etc.
Examples
- Complaints
- Departing Customers
- Receivables