The Enabler of Customer Equity: Predictive Analytics

1
The Enabler of Customer Equity Predictive
Analytics

• Robert C. Blattberg
• Polk Bros Distinguished Professor of Retailing
• Kellogg School of Management
• Northwestern University

2
Purpose of the Presentation

• Show how scientific tools such as predictive
  analyticals are critical to the management of
  customer equity
• Highlight the importance of the quality of
  predictive analytics
• Describe how predictive analytics has led to
  successful business strategies

3
Outline of Presentation

• Definition of Customer Equity?
• Customer Equity drives strategy?
• Science is important in Customer Equity
  management?
• Customer Equity managerial myths?
• Cases studies of firms using analytical methods
  to increase Customer Equity?
• Concluding comments

4
CUSTOMER EQUITY
Basic Concept

• Customer equity is the long-run value of a
  customer to the firm.
• It focuses on the asset value of the customer,
  not the transaction value of the customer

5
CUSTOMER EQUITY
Definition of Customer Equity
The sum of all future customer revenue
streams minus product and servicing costs,
acquisition costs and remarketing costs
6
CUSTOMER EQUITY AS A TOTAL MARKETING SYSTEM
Database Marketing
Customer Equity
7
What is Customer Equity and Why is Important?
8
Brief History of US Credit Card Industry Pure
Play Entry

• In the 80s non-banks began to recognize that
  banks were not aggressively marketing credit
  cards
• Banks relied on current customer base and viewed
  it as add-on sales
• Non-banks aggressively acquired customers through
  mechanisms such as balance transfers and low
  introductory rates

9
LESSONS LEARNED

• Each stage of the credit card industry saw
  competitors enter through a Customer Equity
  strategy
• Banks failed to recognize the need to acquire
  aggressively to preempt competition and failed to
  understand the trade-off between acquisition and
  retention
• The pure play competitors did not have any
  add-on products and as competition entered,
  low-introductory rates caused low retention rates
  and with no add-on selling, they lost their
  preeminent position
• The players did not recognize the need to use of
  Customer Equity strategies and hence failed to
  capitalize on the future strategic moves

10
Lessons Learned

• The current successful players all have Customer
  Equity strategies
• Citi acquisition through branding
• Capitol One acquisition through database
  marketing
• MBNA retention through affinity cards
• Amex add-on selling

11
The Importance of Science in Customer Equity
Management
12
What is Science?

• The state of knowing knowledge as distinguished
  from ignorance or misunderstanding
• A knowledge or system of knowledge covering
  general truths or the operation of general laws
  especially as obtained and tested through
  scientific method

13
Why is Science Important?

• Science, if focused on real-world problems, can
  provide general laws and principles managers can
  use to run their businesses
• Without science the world is left to managers who
  may or may not have a reasonable gut feel for
  which tactics and strategies work
• Example Money Ball and the Oakland As
• Science also forces managers and consultants to
  use the appropriate measures and methods

14
Marketers Have Relied on Myths, Not Science, to
Manage Customer Equity

• Customer Equity management is rife with
  panaceas and recommendations, some of which are
  clearly false and others not based on sound
  empirical evidence

15
Marketers Have Relied on Myths, Not Science, to
Manage Customer Equity

• The result
• Strategies and tactics which either do not
  increase long-term profitability and / or
  customer equity
• Recommendations that are no better than flipping
  coins.

16
Examples of Customer Equity Myths
17
Examples of Myths About Customer Relationship
Management Strategies

• Zero customer defects is the goal of CRM and the
  focus of the firm should be higher retention
  rates

18
The General Problem in Strategizing About
Retention Rates

• No economic or financial model of retention rates
  
• Failure to recognition there are costs and
  benefits of retention
• What is the cost of increasing retention by 1?
  What is the economic benefit in terms of
  increased life-time value?
• What is the cost of increasing retention rates
  versus allowing some customers to defect?
• Raising prices may cause lower retention rates
  but also higher profits

19
Examples of Customer Equity Myths Acquisition
Pricing

• Low acquisition prices bring in low retention
  rate customers (cherry pickers who do not want a
  long-term relationship with the firm)
• Little empirical evidence this is true but strong
  opinion among managers

20
Examples of Customer Equity Myths Acquisition
Pricing

• Recent research Anderson Simester
• Anderson Simester used control experiments to
  test whether cherry pickers are attracted by
  deep discounts used to convert prospects to
  customers
• Key finding Deep discounts (v. shallow
  discounts) increase the future purchases of new
  customers
• Results are counter to the generally believed
  view that deep discount customers are less
  valuable

21
Examples of CRM Myths Reward Best Customers

• Reward your best historical customers
• Question Can a firm determine who their best
  future customers are based on historical data?

22
Examples of CRM Myths Reward Best Customer

• Recent research - Malthouse and Blattberg
• Use historical data to predict who best future
  customers are
• Compared predicted best customers to actual best
  customers
• Create buckets of customers
• Top 20 percent and bottom 80 percent
• Platinum, gold, iron, lead (Zeithaml et. al.
  2001)
• Platinum are top 5 of customers
• Gold are next 15
• Iron are next 30
• Lead are bottom 50

23
Examples of CRM Myths Reward Best Customer

• Results
• Best historically valued customers are not a good
  predictor of best future customers
• 55-20 rule of the top 20 of customers based
  on long-term value, 55 were predicted to be
  average customersOf the platinum customers (top
  5 percent), 81 percent of misclassified
• Conclusion Rewarding Best historical
  customers is highly risky because future best
  customers may defect

24
Why Should a Business Become More Scientific
25
Answer

• Without science, firm operates on myths, many of
  which aint so.

26
How Does a Business Enter the Realm of Science?

• Science requires the systematic analysis of data
  with holdout samples and testing
• Predictive analytics and statistical modeling are
  the cornerstones of scientific research
• Provides the systematic analysis of data and
  testing of hypotheses

27
Why Does The Quality of the Methods Matter?
28
Potential Myth Number of Relationships
Increases Retention Rate

• In banking, large consulting firm touts the
  number of relationship a firm has with customer,
  influences the retention rate
• To prove the hypothesis, they plot number of
  customers relationships versus the retention rate
  
• The plot shows higher retention rates are
  correlated with the number of relationships

29
Potential Myth Number of Relationships
Increases Retention Rate

• Implication focus on add-on selling to
  increase retention rates
• Economic payout can be lower for additional
  product sales because of the payout from higher
  retention rates

30
Potential Myth Number of Relationships
Increases Retention Rate

• Alternative explanation Satisfied customers buy
  additional products and services and hence have
  higher retention rates (because of satisfaction)
• Evaluation of add-on selling economic analysis is
  different because now add-on selling products /
  services must pay for themselves

31
Potential Myth Number of Relationships
Increases Retention Rate

• Which is correct?
• Need systematic analyses and models to find the
  answer

32
Application of Analytical Methods to Customer
Equity Management
33
Case Studies Capital One (Acquisition Marketing)

• Problem
• Acquisition cost was critical in the credit card
  industry as retention rates and spending per card
  declined because of competition
• How could Capital One generate customers at a low
  acquisition cost

34
Case Studies Capital One (Acquisition Marketing)

• Capital Ones strategy
• Use datamining to create customer segments
• Design credit products to meet segment needs

35
Case Studies Capital One (Acquisition Marketing)

• Capital One has the lowest acquisition cost of
  any major credit card issuer

36
Case Studies Capital One (Acquisition Marketing)

• Lessons learned
• Datamining combined with product management can
  significantly lower acquisition costs

37
Case Studies Harrahs Casinos (Retention
Marketing)

• Problem
• Gaming industry had been an acquisition oriented
  business with entertainment as the critical draw
• However, it was very expensive to constantly
  attract new customers without focusing on
  retention

38
Case Studies Harrahs Casinos (Retention
Marketing)

• Harrahs strategy
• Invest in data acquisition through a card used at
  the Casino to offer rewards programs based on
  usage
• Mine the data using statistical methods to
  determine which customers are likely to migrate
  to higher levels of activity

39
Case Studies Harrahs Casinos (Retention
Marketing)

• Results
• 15 growth rates over the last three years
• Lessons learned
• A strategy associated with retention is necessary
  but not sufficient
• To be successful at retention marketing requires
  data and analytical tools

40
Case Studies Tesco (Add-on Selling)

• Problem
• Growth of discounters (Wal-Mart) in UK (just as
  in every other major market) has put traditional
  grocery retailers under tremendous pressure
• Retailers were losing transactions size as well
  as frequency of customer visits but not total
  customers

41
Case Studies Tesco (Add-on Selling)

• Tescos strategy
• Tescos solution was to use a frequent shopper
  card to collect data which could then be mined to
  determine shopping styles of customers
• Data mining focused on learning how to profile
  customers buying behavior so that appropriate
  related products could be offered as well as
  setting spending goals for customers to incent
  them to migrate to a higher spending levels

42
Case Studies Tesco (Add-on Selling)

• Lessons Learned
• Collection of frequent shopper data is not enough
  to generate incremental sales
• Data mining is the critical requirement
• Without datamining frequent shopper cards have
  little value

43
Concluding Comments

• As is obvious, the critical factor to be
  effective at Customer Equity management requires
  good databases and good models
• Most firms have some type of customer database
  but the weakness is in the analysis

44
CONCLUDING COMMENTS

• To develop excellent Customer Equity management
  requires
• A financial business case to demonstrate the
  potential payout of developing these databases
• Analytical tools and software to transform data
  into information and decisions
• Software that makes the results accessible to the
  users

45
Final Observation

• Will Rogers said
• It aint what you dont know that hurts you,
  its what you know that aint so.

46
Thank You