Chapter 16:

Applications of CRM in
B2B and B2C Scenarios (Part 1)
Overview

Topics discussed:
§ Measuring customer Profitability
§ The Life-time-Profitability relationship in a non-contractual setting
§ A model for incorporating customers’ projected profitability into lifetime duration
computation
§ A model for identifying the true value of a lost customer

V. Kumar and W. Reinartz – Customer Relationship Management 2

Measuring Customer Profitability

§ Managers are challenged to understand how their marketing efforts affect customers
§ Customers are dynamic and involve several marketing variables:
§ Tenure of the customers with the firm
§ Profitability of the customers
§ Purchase behavior
§ Adoption of multiple channels to purchase
§ Demographic factors governing purchase behavior
§ Customer Lifetime Value (CLV)
§ Multi-period evaluation of a customer’s value to the firm
§ Assists managers to allocate resources optimally and develop customer-level
marketing strategies

V. Kumar and W. Reinartz – Customer Relationship Management 3

Computing CLV

Contractual Non-Contractual
Definition Customers are bounded by a Customers are not bounded by a
contract contract

Examples Cable TV subscription Grocer store purchase

Business Firms get stream revenue or No stream revenue

Implication fixed monthly revenue

Base GC T p̂(Buy it = 1) * ĜC it

T
M̂C it
CLVi = å +å -
t =1 (1 + d) t
t =1 (1 + d) t
(1 + d) t

Baseline Augmented CLV

NPV of future streamed NPV of future cash flows

revenue from the customer from a customer based on
products/service purchased

V. Kumar and W. Reinartz – Customer Relationship Management 4

Drivers of CLV

§ In order to successfully implement a marketing initiative based on CLV, management

must understand the drivers of CLV and how to measure those drivers
§ CLV drivers:
§ Exchange characteristics: variables which affect the customer-firm relationship

§ Spending level § Product returns

§ Cross-buying behavior § Relationship benefits
§ Multi-channel shopping § Loyalty instrument
§ Focused buying § Frequency of marketing contacts
§ Average inter-purchase time § Bi-directional communication
§ Customer characteristics: demographic variables
§ B2B setting: industry, annual revenue, location of the business
§ B2C setting: age, gender, spatial income, physical location of the customers

V. Kumar and W. Reinartz – Customer Relationship Management 5

The Life-time-Profitability Relationship
in a Non-contractual Setting

Background and objective

§ The firm has to ensure that the relationship stays alive since the customer typically splits
his/her category expenses with several firms
§ Main Objective: Rigorous and differentiated empirical analysis of the lifetime-profitability
relationship in a ‘non-contractual‘ context
§ Tests to achieve the objective:
§ The strength of the lifetime duration – profitability relationship
§ Whether profits increase over time (lifetime profitability pattern)
§ Whether the costs of serving long-life customers are actually less
§ Whether long-life customers pay higher prices

V. Kumar and W. Reinartz – Customer Relationship Management 6

The Life-time-Profitability Relationship
in a Non-contractual Setting (2)

Conceptual model
§ To investigate the consequences of customer retention, namely, profitability:
§ Individual customer lifetime profits are modeled as a function of a customer’s
lifetime duration
§ Revenue flows over the course of a customer’s lifetime
§ Firm cost is associated with the marketing exchange

V. Kumar and W. Reinartz – Customer Relationship Management 7

Customer Lifetime and Firm Profitability

Proposition 1: The nature of the lifetime-profitability relationship is positive

Lifetime-Profitability Association Segmentation Scheme

Segment 2 Segment 1
High Long

Lifetime Lifetime
Profit Duration
Segment 4 Segment 3
Low Short

Short Long Low High

Lifetime duration Lifetime Revenue

V. Kumar and W. Reinartz – Customer Relationship Management 8

Customer Lifetime and Firm Profitability (2)

Proposition 2: Profits increase over time

§ Analysis of the dynamic aspects of the lifetime-profitability relationship
§ Non-contractual setting:
§ cost of serving customer can easily exceed the profit margin brought in by the customer
§ Profits may not increase over time
§ Example: catalog shopping or direct mail offerings
§ The customer may end up
buying once a year and
spend a smaller amount
§ Cost of serving the customer
can easily exceed the profit margin
brought in by the customer

V. Kumar and W. Reinartz – Customer Relationship Management 9

Customer Lifetime and Firm Profitability (3)

Proposition 3:The costs of serving longer-life customers are lower

Lower transaction costs The costs associated with

for longer-life versus promotional expenditures
shorter-life customers directed at longer - and
(e.g. retail sector)
vs shorter - life customers
actually differ

Proposition 4: Longer-life customers pay higher prices

Existing customers pay Higher value

consciousness (lower
effectively higher prices vs average prices paid) of
than new ones
long- term customers

V. Kumar and W. Reinartz – Customer Relationship Management 10

Research Methodology

Model for measuring customer lifetime for non-contractual relationships

§ Data: 3 year window of daily sales of an established U.S. catalog retailer
§ Negative Binomial Distribution (NBD)/Pareto model
§ Based on the customer-specific probability of being alive, the model can be used to
determine which customers should be deleted from active status
§ Continuous P(Alive) estimate is transformed into a dichotomous “alive/dead” measure
§ Knowing a person’s “time of birth” and given a specified probability level (threshold), can
approximate when a customer is deemed to have left the relationship
§ The time from birth, t0, until the date associated with the cut-off threshold, tcut-off, constitutes
the lifetime of the customer
§ Calculate finite lifetime for each customer, for profitability analysis

V. Kumar and W. Reinartz – Customer Relationship Management 11

Illustrative Lifetime Determination
of Individual Household

Source: Reinartz W. And Kumar V., “On the Profitability of Long –life Customers in a Non-Contractual Setting: An
empirical Investigation and Implications for Marketing

V. Kumar and W. Reinartz – Customer Relationship Management 12

Measuring Customer Lifetime for
Non-Contractual Relationships

Establishment of cut-off threshold

§ The Length of the lifetime estimate for each customer
§ Establishment of cut-off threshold that produces the highest percentage of correct
classifications

§ Use 0.5 as the cut-off threshold

Lifetime estimation

§ Final lifetime estimates

Mean Lifetime Standard % Right- Minimum Maximum

(months) deviation Censored
Cohort 1 28.7 7.8 41.1 11 36
Cohort 2 27.9 7.9 41.7 12 35

V. Kumar and W. Reinartz – Customer Relationship Management 13

Measuring Customer Lifetime for
Non-Contractual Relationships (contd.)

Profit calculation
Net-present value of profit is calculated on an individual customer basis for the period of
36 months as:
t
æ 1 ö
LT p i = åt =1 (GCti - Cti )ç
36
÷
è 1 + .0125 ø
where LTp i = individual net-present lifetime profit for 36 months
GCti = gross contribution in month t for customer i,
Cti = mailing cost in month t for customer i, and
0.0125 = monthly discount rate (based on 0.15 rate per year)

V. Kumar and W. Reinartz – Customer Relationship Management 14

Test of the Propositions

Proposition 1: The nature of the lifetime-profitability relationship

is positive
§ Could customers with shorter tenure might actually be more profitable than long-term
customers, a claim that runs counter to the theoretical expectations of a relationship
perspective?
§ Which group of customers is of more interest to the firm, the one that buys heavily for a
short period (segment 3) or the one with small spending but with long-term commitment
(segment 2)?

V. Kumar and W. Reinartz – Customer Relationship Management 15

Test of the Propositions (2)

Proposition 2: Profits increase over time

§ Examine the profitability evolution visually
§ Analyze the sign of the slope coefficient
§ The exact specification of the regression is:
Profits = as + b1s * Dummy + b2s * ts+ error
where t = month, bis = regression coefficient, s = segment, and
Dummy = 1 if first purchase month, else 0
Proposition 3: The costs of serving long-life customers are lower
§ Compute the ratio of promotional costs in a given period over the revenues in the same
period
Proposition 4: Longer-life customers pay higher prices
§ Whether longer-life customers do pay higher prices as compared to shorter-life customers

V. Kumar and W. Reinartz – Customer Relationship Management 16

Tests of Propositions - Results
(Cohort 2 results in parentheses)

Segment 2 Segment 1

Lifetime Mailing Lifetime Relativ Mailing

Relative Avg. Average
# of Profit per Cost/ # of Profit per e Profit Cost/
Profit Item Item
customers Customer Sales customers Customer ($/ Sales
Long ($/month) Price Price
($) Ratio ($) month) Ratio
Life
Time
889 50.85 1.43 0.128 47.74 1322 289.83 8.18 0.063* 58.43**
(973) (55.26) (1.56) (0.124) (48.72) (1546) (322.03) (9.31) (0.062)* (58.25)**

Segment 4 Segment 3

Lifetime Mailing Lifetime Relativ Mailing

Relative Avg.
# of Profit per Cost/ # of Profit per e Profit Cost/ Average
Profit Item
Short customers Customer Sales customers Customer ($/ Sales Item Price
($/month) Price
Life ($) Ratio ($) month) Ratio
time
1208 50.49 2.41 0.141 47.97 783 257.96 11.67 0.065 63.54
(1504) (53.67) (2.67) (0.143) (46.80) (942) (284.20) (12.57) (0.064) (64.47)

Low Lifetime Revenue High Lifetime Revenue

* Difference between Segment 1 and Segment 3 is not significant

** Difference between Segment 1 and Segment 3 is significant at a = 0.05

V. Kumar and W. Reinartz – Customer Relationship Management 17

Interpretation of Results

§ Both long-term customers (segment 1) and short-term customers

(segment 3) constitute the core of the firm’s business
§ Relationship between lifetime and profits can be far from being positive and monotonic
§ In terms of profit per month, customers in segment 3 are the most attractive of all
§ Segment 3 customers purchase with high-intensity, thus generating higher profits in a
relatively shorter period of time

V. Kumar and W. Reinartz – Customer Relationship Management 18

Interpretation of Results (2)

§ The mean relative profit for each segment is significantly different from the other
segment at least at a = 0.05 (using the multiple comparison test)
§ Segment 1 customers are the most desirable set for the firm –
represents the loyalty effect at its best
§ For segment 3 customers (high revenue but short lifetime), there appears to be a good
match between offerings and desires but their relationship duration may be
complicated by moderating factors
§ Dissatisfaction might occur for segment 4 whose customers spend the lowest amount

V. Kumar and W. Reinartz – Customer Relationship Management 19

Aggregate Profits ($) for Short-life Segments

20000.00 Short life, low

revenue
18000.00
(Segment 4)
16000.00 Short life, high
14000.00
revenue
($) Segment Profit

(Segment 3)
12000.00

10000.00

8000.00

6000.00

4000.00

2000.00

0.00

-2000.00
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35
Month

V. Kumar and W. Reinartz – Customer Relationship Management 20

Aggregate Profits ($) for Long-life Segments

20000.00
Long life, low
revenue
18000.00 Segment 2)

16000.00 Long life, high

revenue
14000.00 ('Segment 1)
($) Segment Profit

12000.00

10000.00

8000.00

6000.00

4000.00

2000.00

0.00

-2000.00
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35
Month

V. Kumar and W. Reinartz – Customer Relationship Management 21

Regression Analysis of Profits as a Function of Time

Proposition 2: Do profits increase over time?

§ Regression results for t = 1 to 36 months (Cohort 1)

§ Validation results in parentheses (Cohort 2)

Dummy Coefficient Coefficient

Segment Intercept (a) for t =1 (b1) for t (b2) R2
1 12.11 (12.73) 45.77 (46.38) -0.13 (-0.14) .85 (.85)
2 n.s. (n.s.) 30.24 (30.91) 0.07 (0.071) .92 (.91)
3 19.40 (20.9) 57.85 (58.29) -0.70 (-0.75) .95 (.94)
4 3.25 (3.69) 29.53 (31.45) -0.14 (-0.15) .94 (.95)

§ With the exception of Segment 2, the coefficient for the linear effect has a
negative sign – highlights the negative profit trend over time for the 3 segments

§ All the coefficients for time are significant at a = 0.01

V. Kumar and W. Reinartz – Customer Relationship Management 22

Proposition 3: Are the Costs of Serving Long-Life Customers
Lower?

§ The ratio of mailing cost per dollar sales in the longer-life segment (segment 1) is
statistically not different from the mailing cost per dollar sales in the shorter-life
segment (segment 3)
§ In terms of cost efficiency, segments 1 and 3 are the most attractive to the firm,
although they have very different lifetime properties
§ The ratio of mailing cost and revenues – which is one measure of efficiency – need not
necessarily be lower for long-life-customers

V. Kumar and W. Reinartz – Customer Relationship Management 23

Proposition 4: Do Long-Life Customers Pay
Higher Prices?

§ The average price per item for segment 3 is significantly (a = 0.05) different from (and
greater than) that of segment 1
§ The highest average price paid for a single product item is in segment 3, the short-life
segment
§ The highly profitable short-term customers seem to be less sensitive to the
product’s price
§ The higher spending (average prices paid) by segment 3 customers may be due to
some other benefit sought by them

V. Kumar and W. Reinartz – Customer Relationship Management 24

Summary of Findings

§ A strong linear positive association between lifetime and profits does not necessarily
exist
§ A static and a dynamic lifetime-profit analysis can exhibit a much differentiated picture:
profitability can occur for the firm from high and low lifetime customers
§ Profits do not increase with increasing customer tenure: the cost of serving long-life
customers is not lower
§ Long-life customers do not pay higher prices

V. Kumar and W. Reinartz – Customer Relationship Management 25

Model for Incorporating Customers’ Projected Profitability into
Lifetime Duration Computation

§ Background and objective:

§ Critical for managers to understand customer’s profitability and decide when to let
go of an unprofitable customer
§ Key research tasks:
§ Empirically measure lifetime duration for non-contractual customer-firm
relationships, incorporating projected profits
§ Understand the structure of profitable relationships and test the factors that
impact a customer’s profitable lifetime duration
§ Develop managerial implications for building and managing profitable
relationship exchanges

V. Kumar and W. Reinartz – Customer Relationship Management 26

Conceptual Model of Profitable Customer Lifetime

Exchange
characteristics

Profitable Lifetime Customer

Observed heterogeneity Lifetime Duration Profitability
Duration

Non-contractual setting
Revenues Cost

Present Study Earlier Study

B-to-C and B-to-B contexts B-to-C context only

V. Kumar and W. Reinartz – Customer Relationship Management 27

Estimating Profitable Lifetime Duration

§ Determine the contribution margin expected from each customer

in future periods based on the average of the contribution margins in the past
§ Determine for each future period, the probability that the customer will be alive and will
transact with the firm
§ Combine these two components
§ Discount the expected contribution margin in each future period to its
Net Present Value (NPV) using the cost of capital applicable to the firm
§ If in a given month, cost of additional marketing efforts is greater than NPV, determine
that the Profitable Lifetime Duration of the customer has ended

V. Kumar and W. Reinartz – Customer Relationship Management 28

Antecedents of Profitable Lifetime Duration

§ Predictor variables to incorporate into the model:

§ Exchange variables:
§ e.g. amount purchased, degree of cross-buying, degree of focused-buying,
inter-purchase time, number of product returns, ownership of loyalty
instruments, and mailing efforts undertaken by the firm
§ Customer heterogeneity variables:
§ e.g. location and income of the customer
§ Conceptually:

Profitable Lifetime Duration = f (Exchange characteristics, Customer heterogeneity)

V. Kumar and W. Reinartz – Customer Relationship Management 29

Research Methodology

Database structure for B-to-C setting

Mar
‘95 Cohort 3: 2825 observations
Feb Cohort 2: 4965 observations Dec
‘95 ‘97
Jan
‘95 Cohort 1: 4202 observations

Start Finish

§ The customer-firm interaction of Cohort 1 households is tracked for a 36-month

time period, Cohort 2 for 35 and Cohort 3 for a 34-month time period
§ The households are sampled randomly from all households that started in
January, February, and March 1995 respectively
§ The number of purchases ranges from 1 to 46 across the sample with a median
number of 5 purchases, the median interpurchase time is 117 days, and the
median transaction amount is $91 for each purchase
V. Kumar and W. Reinartz – Customer Relationship Management 30
Determining Profitable
Customer Lifetime Duration

§ Calculation of Net Present Value (NPV) of Expected Contribution Margin (ECMit):

t +18 n
æ 1 ö
NPV of ECMit = å
n =t +1
P( Alive)in * AMCM it ç ÷
è1+ r ø

ECMit = estimated expected contribution margin for a given month t

AMCMit = average contribution margin in month t based on all prior
purchases since birth (updated dynamically)
r = discount rate (15% on a yearly basis)
i = the customer
t = month for which NPV is estimated
n = number of months beyond t
P( Alive)in= probability that customer i is alive in month n

V. Kumar and W. Reinartz – Customer Relationship Management 31

Determining Profitable
Customer Lifetime Duration (contd.)

§ Decision of relationship termination

§ Formally, if NPV of ECMit < Cost of Mailing, the firm would decide to
terminate the relationship
§ Decision rule incorporates the cost of mailings and an average flat contribution
margin before mailings of 25 percent
§ Discount rate is assumed to be 15% per year
§ Calculation of finite lifetime estimate
§ Based on decision of relationship termination, average lifetime across Cohort 1 is 29.3
months, Cohort 2 is 28.6 months, and Cohort 3 is 27.8 months
§ Variability in lifetime duration evidenced through
§ a wide range between lowest and highest lifetime estimate
§ the standard deviation of the lifetime estimate
§ the relatively small value of s in the NBD/Pareto model

V. Kumar and W. Reinartz – Customer Relationship Management 32

Analysis

§ In the proportional hazard model, the hazard rate hi(t) for

individual i is assumed to take the form:

hi(t) = h0(t) exit b

where
h0 (t) is the baseline hazard rate
(xitb) is the impact of the independent variables
§ The hazard of a lifetime event of a household i at time t is given as:

hi(t) = h0(t) EXP (β1 Purchase Amountit + β2 Cross Buyingit + β3 Focus of Buyingi + β4
Average Interpurchase Timeit + β5 (Average Interpurchase Timeit)2 + γ1 Returnsit + γ2
Loyalty Instrumenti + γ3 Mailingsit + γ4 Product Categoryi + δ1 Population Densityi + δ2
Incomei + δ3 Agei)

V. Kumar and W. Reinartz – Customer Relationship Management 33

Summary of Results

Hypothesis Description B-to-C Setting B-to-B Setting

Result Result
Profitable Customer Lifetime
Duration is:
1 Positively related to Supported Supported
the customer’s spending level

2a Positively related to degree of Supported Supported

cross-buying behavior exhibited
by customers
2b Related to the Focused Supported Supported
Buying behavior exhibited by
customers However, the relationship is However, the relationship is
negative, indicating that negative, indicating that
buying in only a single buying in only a single
department results in shorter department results in shorter
lifetime duration lifetime duration
3 Related to Average Supported Partial Support
Interpurchase Time in an inverse
U-shaped manner whereby Only the linear term is
intermediate AIT is associated significant
with the longest profitable
lifetime
4 Inversely related to the Not Supported Not Supported
proportion of merchandise However, the interaction of However, the interaction of
returned by the customers Returns with Purchase Returns with Purchase
amount variable is significant amount variable is significant
V. Kumar and W. Reinartz – Customer Relationship Management 34
Summary of Results (2)

Hypothesis Description B-to-C Setting B-to-B Setting

Result Result
Profitable Customer Lifetime
Duration is:
5 Positively related to the Supported Supported
customer’s ownership of the
company’s loyalty instrument (B-
to-C) or the availability of line of
credit (B-to-B)

6 Positively related to the number of Supported Supported

mailing efforts of the company (B-
to-C) or the number of contacts (B-
to-B)

7 Profitable Customer Lifetime Supported Not supported

Duration is higher for customers
living in areas with lower
population density (B-to-C) or
businesses existing in lower
population density (B-to-B)
8 Profitable Customer Lifetime Supported Supported
Duration is positively related to the
Income of the customer (B-to-C) or
Income of the firm (B-to-B)

V. Kumar and W. Reinartz – Customer Relationship Management 35

Estimation of P(Alive)

§ Method-of-moment approach
§ Estimate the four parameters of the NBD/Pareto model (r, a, s, b) with a Fortran routine
with likelihood:

M
L(r , a , s, b ) = Õ P[ X i =xi , ti , Ti r , a , s, b ]
i =1

where M is a random sample of customers and customer i made Xi = xi purchases in (0,Ti)

with the last transaction time at ti
§ The larger the value of the shape parameter r the more homogeneous the population of
customers in terms of purchase rate
§ The larger the value of the shape parameter s the more homogeneous the population of
customers in terms of dropout rate
§ The concentration in dropout rates, b, depends on the parameter s only

V. Kumar and W. Reinartz – Customer Relationship Management 36

Estimation of P(Alive) (2)

§ Probability that a customer with a particular observed transaction history is still alive at
time T since trial:
P [Alive | r, a, s, b, x, t, T] =

-1
ì
ï s éæ a + T
r+x
ö æ b + T ö
s
æ b + T ö
s
ùü ï
í1 + êç ÷ ç ÷ F (a1 , b1 ; c1 ; z1 (t )) - ç ÷ F (a1 , b1 ; c1 ; z1 (T ))ú ý
î r + x + s êëè a + t ø è a + t ø
ï èa +T ø úû ï
þ

where a1 = r+x+s, b1 = s+1, c1 = r+x+s+1, z1(y) = (a-b) / (a+y),

F(a1, b1; c1; z1) is the Gauss hyper geometric function,
r, a, s, b = model parameters,
x = number of purchases,
t = time since trial at which the most recent transaction occurred,
T = time since trial

V. Kumar and W. Reinartz – Customer Relationship Management 37

A Model for Identifying the
True Value of a Lost Customer

Conceptual background
§ The value of a lost customer depends upon whether:
§ the customer defects to a competing firm or disadopts the product category

§ Defection: The firm loses direct sales that customer would have brought in
§ Disadoption: Customer stops purchasing from that product category
§ Affects the long term profitability by:
§ The loss of direct sales
§ Indirect effects of word of mouth, imitation, and other social effects

Customer

V. Kumar and W. Reinartz – Customer Relationship Management 38

Modeling the Effects of Disadoption on
the Value of a Lost Customer

§ Value of an average lost customer (VLC) is calculated as:

VLC = a VLCdisadopter + (1- a) VLCdefectors

a is the proportion of disadopters in a firm’s lost customers

§ Profit impact of a lost customer =

(sales estimate from new product growth model without
disadoption) – (sales estimate when the customer disadopts
after certain time)

V. Kumar and W. Reinartz – Customer Relationship Management 39

Modeling the Effects of Disadoption on
the Value of a Lost Customer (contd.)

Key determinants of the value of a lost customer:

§ The time when customer disadopts has the largest impact on the value of the lost
customer; earlier disadoption causes more loss of money
§ The external influence, p has a negative impact
§ The internal influence, q has a positive impact on penetration because higher q
signifies stronger word of mouth
§ Discount rate has positive impact on the value of lost customer

V. Kumar and W. Reinartz – Customer Relationship Management 40

Summary

§ A study of the Life-time-Profitability relationship in a non-contractual setting highlights

concern with the widespread assumption of a clear-cut positive lifetime-profitability
relationship and underlines the importance of a differentiated analysis
§ The empirical evidence showed that (1) a strong linear positive association between
lifetime and profits does not necessarily exist; (2) profits do not necessarily increase
with increasing customer tenure, (3) the cost of serving long-life customers is not lower,
and (4) long-life customers do not pay higher prices
§ The main drivers of customer’s profitable lifetime duration are classified as exchange
characteristics and customer heterogeneity
§ The key determinants of the value of a lost customer are identified as disadoption time,
external and internal influences and the discount rate
§ The disadoption time is found to have the maximum negative impact on the value
i.e., the earlier a customer disadopts, the higher is the value of the lost customer

V. Kumar and W. Reinartz – Customer Relationship Management 41