Average Inter-Purchase Time Retention & Defection Rate Survival Rate Lifetime Duration P (Active)
Size of Wallet Share of Category Requirement Share of Wallet (Transition Matrix)
V. Kumar and W. Reinartz – Customer Relationship Management
3
Traditional Marketing Metrics
Market Share Sales Growth
V. Kumar and W. Reinartz – Customer Relationship Management
4
Market Share (MS)
Share of a firm’s sales relative to the sales of all firms – across all customers in the given market Measured in percentage Calculated either on a monetary or volumetric basis
Market Share (%) of a firm (j) in a category = 100 × S j S ∑ j j =1
J
Where:
j = firm,
S = sales, ΣSj = sum of sales across all firms in the market Information source Numerator: Sales of the local firm available from internal records Denominator: Category sales from market research reports or competitive intelligence Evaluation Common measure of marketing performance, readily computed Does not provide any information about how sales are distributed across customers
V. Kumar and W. Reinartz – Customer Relationship Management 5
Sales Growth
Compares changes in sales volume or sales value in a given period to sales volume or value in the previous period Measured in percentage Indicates the degree of improvement in sales performance between two or more time periods Sales growth in period t (%) = Where: j = firm, ∆Sjt = change in sales in period t from period t-1, Sjt-1 = sales in period t-1
100 × [∆S jt S jt −1 ]
Information source Numerator and denominator: from internal records Evaluation Quick indicator of current health of a firm Does not provide any information about changes in customer size
V. Kumar and W. Reinartz – Customer Relationship Management
6
Customer Acquisition Metrics
Group of primary customer based metric = customer acquisition metric
Concepts: Acquisition Rate Acquisition Cost
V. Kumar and W. Reinartz – Customer Relationship Management
7
Acquisition Rate
Acquisition = first purchase or purchasing in the first predefined period Acquisition rate (%) = 100*Number of prospects acquired / Number of prospects targeted Denotes average probability of acquiring a customer from a population Always calculated for a group of customers Typically computed on a campaign-by-campaign basis
Information source Numerator: From internal records Denominator: Prospect database and / or market research data
Evaluation Important metric Gives a first indication of the success of a marketing campaign But cannot be considered in isolation
V. Kumar and W. Reinartz – Customer Relationship Management
8
Acquisition Cost
Measured in monetary terms Acquisition cost ($) = Acquisition spending ($) / Number of prospects acquired Precise values for companies targeting prospects through direct mail Less precise for broadcasted communication
Information source Numerator: from internal records Denominator: from internal records
Evaluation Difficult to monitor on a customer by customer basis
V. Kumar and W. Reinartz – Customer Relationship Management
9
Customer Activity Metrics
Average Inter-Purchase Time Retention & Defection Rate Survival Rate Lifetime duration P(Active)
V. Kumar and W. Reinartz – Customer Relationship Management
10
Customer Activity Measurement
Objectives Managing marketing interventions Aligning resource allocation with actual customer behavior Providing key input for customer valuation models such as the net-present value (NPV)
V. Kumar and W. Reinartz – Customer Relationship Management
11
Average Inter-Purchase Time (AIT)
Average inter-purchase time of a customer =
1 / Number of purchase incidences from the first purchase till the current time period
Measured in time periods Important for industries where customers buy on a frequent basis
Information source Sales records
Evaluation Easy to calculate Useful for industries where customers make frequent purchases Firm intervention might be warranted anytime customers fall considerably below their AIT
V. Kumar and W. Reinartz – Customer Relationship Management
12
Retention and Defection Rate
Rrt (%) = 100*Number of customers in cohort buying in (t) | customer in (t-1) / Number of customers in cohort buying in (t-1) Rrt (%) = 100 – Avg. defection rate (%) Avg. lifetime duration = [1 / (1- Rr)] Number of retained customers in period (t+n) = number of acquired customers in cohort at time (t)*Rr Avg. defection rate in t (%) = 100 – Rrt Where: Rrt = Retention rate in period t, n = Number of elapsed periods
Assuming constant retention rates, number of retained customers in any arbitrary period (t+n) = Number of acquired customers in cohort * Retention rate (t+n) Given a retention rate of 75%, variation in defection rate with respect to customer tenure results in an average lifetime duration of four years
V. Kumar and W. Reinartz – Customer Relationship Management
13
Retention and Defection-Example
If the average customer lifetime duration of a group of customers is 4 years, the average retention rate is 1*(1/4) = 0.75 or 75% per year, i.e., on an average, 75% of the customers remain customers in the next period
The effect for a cohort of customers over time – out of 100 customers starting in year 1, about 32 are left at the end of the 4th year Customers starting at the beginning of year 1: 100 Customers remaining at the end of year 1: Customers remaining at the end of year 2: Customers remaining at the end of year 3: Customers remaining at the end of year 4: 75 56.25 42.18 31.64 (0.75*100) (0.75*75) (0.75*56.25) (0.75*42.18)
Assuming constant retention rates, the number of retained customers at the end of year 4 is 100*0.75 4 = 31.64. (Number of acquired customers in cohort * Retention rate (t+n)) The defection rate is 100-75% = 25%
V. Kumar and W. Reinartz – Customer Relationship Management
14
Variation in Defection Rate with Respect to Customer Tenure
30 # of customers defecting 25 20 15 10 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Customer tenure (periods)
Plotting the entire series of customers that defect each period demonstrates variation (or heterogeneity) around the average lifetime duration of 4 years
V. Kumar and W. Reinartz – Customer Relationship Management
15
Projecting Retention Rates
To forecast non-linear retention rates, Rrt = Rc*(1- e-rt) Where: Rrt = predicted retention rate for a given future period, Rc = retention (rate) ceiling, r = coefficient of retention
r = (1/t)*(ln(Rc) – ln(Rc – Rrt ))
V. Kumar and W. Reinartz – Customer Relationship Management
16
Actual and Predicted Retention Rate for a Credit Card Company
Rc = 0.95 means that managers believe the maximum attainable retention rate is 95% The known retention rate in period 9 is 80% while it is 82% in period 10 The parameter r for period 9 is (1/9)*(ln(0.95)-ln(0.95-0.8)) = 0.205. The r for period 10 is (1/10)*(ln(0.95)-ln(0.95-0.82)) = 0.198 for both periods r approximates the value 0.2
17
V. Kumar and W. Reinartz – Customer Relationship Management
Survival Rate
Measured for cohorts of customers Provides a summary measure of how many customers survived from the beginning of the formation of a cohort up to any point in time afterwards SRt (%) = 100 * Rrt * SRt-1 Where: SR = Survival Rate
Number of survivors for period 1 = survival rate for period 1 * number of customers at the beginning
V. Kumar and W. Reinartz – Customer Relationship Management
18
Survival Rate Computation-Example
Number of customers starting at the beginning of year 1: 1,000
Retention rate Period 1 Period 2 Period 3 Period 4 0.55 0.62 0.68 0.73 Survival rate 0.55 0.341 0.231 0.169 Survivors 550 341 231 169
Number of survivors for period 1 = 0.55*1000 = 550 Survival rate for period 2 = retention rate of period 2*survival rate of period 1 Survival rate for period 2 = 0.62*0.55 = 0.341 (=34.1%)
V. Kumar and W. Reinartz – Customer Relationship Management
19
CRM at Work: Amazon
One of the leaders in implementing customer relationship management programs on the Web Unique CRM program increased customer acquisition and retention In 1999 Amazon acquired 11 million new customers, nearly tripling its number of customers from 1998 Greatest success in customer retention: Repeat customers during the year accounted for 71% of all sales Success attributed to the attempt to learn about customers and their needs and then using this information to offer value-added features
V. Kumar and W. Reinartz – Customer Relationship Management
20
Lifetime Duration
Average lifetime duration = Where:
∑(t *Number of retained customers in t) / N
t =1
T
N = cohort size, t = time period, T = time horizon
Limitations: information is not always complete making the calculation more challenging Differentiate between complete and incomplete information on customer Complete information = customer’s first and last purchases are assumed to be known Incomplete information = either the time of first purchase, or the time of the last purchase or both are unknown
V. Kumar and W. Reinartz – Customer Relationship Management
21
Customer Lifetime Duration when the Information is Incomplete
V. Kumar and W. Reinartz – Customer Relationship Management
Lifetime Duration
Customer relationships Contractual (“lost-for-good”) = Lifetime duration spans from the beginning until the end of the relationship (e.g.: mobile phone contract) Noncontractual (“always-a-share”) = Whether a customer is active at a given point in time (e.g.: department store purchase) One-off purchases
V. Kumar and W. Reinartz – Customer Relationship Management
23
P(Active)
Probability of a customer being active in time t P(Active) = τ Where:
n
n = the number of purchases in a given period,
τ = is the time of the last purchase (expressed as a fraction of the
observation period)
Non-contractual case For an advanced application see: Reinartz, Werner and V. Kumar (2000, 2002)
V. Kumar and W. Reinartz – Customer Relationship Management
24
Estimation of P(Active)-Example
Customer 1
Customer 2
Month 1
Month 8
Month 12
An x indicates that a purchase was made by a customer in that month
To compute the P(Active) of each of the two customers in the 12th month of activity Customer 1: T1 = (8/12) = 0.6667 and n1 = 4 P(Active)1 = (0.6667)4 = 0.197 Customer 2: T2 = (8/12) = 0.6667 and n = 2 P(Active)2 = (0.6667)2 = 0.444
V. Kumar and W. Reinartz – Customer Relationship Management
25
Comprehensive Example of Customer Acitivity Maesures
Actual retention pattern of a direct marketing firm
Cohort of 7500 customers at the outset, maximum retention rate is 0.80 and the coefficient of retention r is 0.5; after period 10 the company retains approximately 80% of customer base
V. Kumar and W. Reinartz – Customer Relationship Management
26
Popular Customer-Based Value
Size of Wallet Share of Category Equipment Share of Wallet Transition Matrix
V. Kumar and W. Reinartz – Customer Relationship Management
27
Customer-Based Value Metrics
Customerbased Value Metrics
Popular
Strategic
Size Of Wallet
Share of Category Reqt.
Share of Wallet
Transition Matrix
RFM
Past Customer Value
LTV Metrics
Customer Equity
V. Kumar and W. Reinartz – Customer Relationship Management
28
Size of Wallet
Size of Wallet ($) of customer i in a category = ∑ Sij
j =1
J
Where:
i = a particular customer, j = firm, J = all firms offering products in the considered category, Sj = sales value (in category) to customer i by firm j, j = 1,…,J
Information source Primary market research
Evaluation Critical measure for customer-centric organizations based on the assumption that a large wallet size indicates more revenues and profits
Example A consumer spends on average $400 on groceries in different supermarkets per month. Thus his/her size of wallet is $400.
V. Kumar and W. Reinartz – Customer Relationship Management
29
Share of Category Requirement (SCR)
aSCR (%) of firm (or brand) j0 in a category = Where: j0 = focal firm or brand, i = customer,
∑ Vij0 / ∑ ∑ Vij * 100
i =1
i =1
j =1
I
I
J
I = all customers buying in focal category, J = all firms or brands available in focal category, Vij = purchase volume of customer i from firm (or brand) j
V. Kumar and W. Reinartz – Customer Relationship Management
30
Share of Category Requirement (SCR)
Example Calculation of aSCR – purchases during a 3-month period Brand SAMA has a MS of 33% (i.e., 8 purchases out of a total of 24) and an aSCR of 42.1% (i.e., 8 purchases out of 19, made by its two buyers) This shows that even though SAMA‘s MS is already substantial, its aSCR is even higher
Brand SAMA Customer 1 Customer 2 Customer 3 Total 2 6 0 8 Brand SOMO 8 0 4 12 Brand SUMU 0 3 1 4 Total 10 9 5 24
V. Kumar and W. Reinartz – Customer Relationship Management
31
Share of Category Requirement (SCR)
iSCR (%) of customer i0 that a firm x (or brand) j0 satisfies = Vi0j0 / Where: j0 = focal firm or brand, i0 = focal customer, J = all firms or brands available in focal category,
∑ Vi0j * 100
j =1
J
Vij = purchase volume of customer i from firm (or brand) j
V. Kumar and W. Reinartz – Customer Relationship Management
32
Share of Category Requirement (SCR)
Example: Individual SCR-ratios Customer 3 has the highest iSCR PEAR Computers should identify high iSCR customers such as customer 3, and target more of its marketing efforts (mailers, advertisements etc.) towards such customers and their respective requirements Also, customer 3’s size of wallet (column A), is the largest
A Total requirement of notebook computers per customer in 2010 Customer 1 Customer 2 Customer 3 100 1,000 2,000 B Total number of notebook Computers purchased from PEAR Computers per customer in 2010 20 200 500 B/A Share of category requirement for PEAR computers per customer in 2010 .20 .20 .25
V. Kumar and W. Reinartz – Customer Relationship Management
33
Share of Category Requirement (SCR)
Information source Numerator: volumetric sales of the focal firm from internal records Denominator: total volumetric purchases of the focal firm’s buyer base – through market and distribution panels, or primary market research (surveys) and extrapolated to the entire buyer base
Evaluation Accepted measure of customer loyalty for FMCG categories SCR controls for the total volume of segments / individuals category requirements Does not indicate if a high iSCR customer will generate substantial revenues or profits Can only be achieved by knowing the customer’s size of wallet
V. Kumar and W. Reinartz – Customer Relationship Management
34
Share of Wallet (SW)
Individual Share of Wallet (iSW) iSW (%) of firm j0 to customer i = Sij0 / Where: j = firm, i = customer, Sij = sales of firm j to customer I, J = all firms who offer the category under consideration
∑
j =1
J
Sij * 100
Example If a consumer spends $400 monthly on groceries, $300 thereof are spend at the supermarket “BINGO” Consequently “BINGO”’s iSW for this particular consumer amounts 75%
V. Kumar and W. Reinartz – Customer Relationship Management
35
Share of Wallet (SW)
Aggregate Share of Wallet (aSW) (brand or firm level)
∑ Sij * 100 aSW (%) of firm j0 = ∑ Sij0 / ∑ i =1 i =1 j =1
J
I
I
Where:
j = firm, i = customer, Sij = sales of firm j to customer I, J = all firms who offer the category under consideration, I = all customers
Example The aSW is “BINGO”’s sales (value) in period t ($ 750,000) divided by the total grocery expenditures of “BINGO”’s customers in the same period ($1,250,000) 750,000/1,250,000 = 60%
V. Kumar and W. Reinartz – Customer Relationship Management
36
Share of Wallet (SW)
Information source Numerator: From internal records Denominator: Through market and distribution panels, or primary market research (surveys) and extrapolated to the entire buyer base
Evaluation Important measure of customer loyalty The iSW sheds light on how important the firm is for an individual customer in terms of his expenditures in the category The aSW indicates how important (value wise) a specific firm is for its customer base in terms of their expenditures in the category
V. Kumar and W. Reinartz – Customer Relationship Management
37
Applications of SCR and SW
SCR – for categories where the variance of customer expenditures is relatively small SW – if the variance of consumer expenditures is relatively high Share of wallet and size of wallet simultaneously – with same share of wallet, different attractiveness as customers Example:
Share-of-Wallet Buyer 1 Buyer 2 50% 50% Size-of-Wallet $400 $50 Absolute expenses with firm $200 $25
Absolute attractiveness of Buyer 1 is eight times higher even though the SW is the same as for Buyer 2
V. Kumar and W. Reinartz – Customer Relationship Management
38
Segmenting Customers Along Share of Wallet and Size of Wallet
High
Hold on
Maintain and guard
Share of Wallet
Low
Do nothing
Target for additional selling
Small
Size of Wallet
Large
The matrix shows that the recommended strategies for various segments differ substantively The firm makes optimal resource allocation decisions only by segmenting customers along the two dimensions simultaneously
V. Kumar and W. Reinartz – Customer Relationship Management
39
Share of Wallet and Market Share (MS)
MS of firm j0 (%) = Where:
∑(iSW of customer i to firm j0*Size of Wallet of customer i) / ∑∑Sij
i =1 i =1 j =1
I
I
J
j = firm, i = customer, Sij = sales of firm j to customer i, J = all firms who offer the category under consideration, I = all customers
Difference of share of wallet to market share: MS is calculated across buyers and non-buyers, whereas SW is calculated only among actual buyers
V. Kumar and W. Reinartz – Customer Relationship Management
40
Share of Wallet and Market Share (MS)
Example The supermarket “BINGO” has 5,000 customers with an average expense of $150 at “BINGO” per month (SW*size of wallet) The total grocery sales in “BINGO”’s trade area are $5,000,000 per month “BINGO”’s market share is (5,000 * $150) / $5,000,000 = 15% Implication: although “BINGO” has an overall low MS, it has a high SW for those consumers buying “BINGO” “BINGO” is a niche player with very loyal customers
V. Kumar and W. Reinartz – Customer Relationship Management
41
Transition Matrix
Brand purchased next time A Brand currently purchased A B C 70% 10% 25% B 20% 80% 15% C 10% 10% 60%
Characterizes a customer’s likelihood to buy over time or a brand’s likelihood to be bought Example The probability that a consumer of Brand A will switch to Brand B and then come back to Brand A in the next two purchase occasions is 20%*10% = 2% If, on average a customer purchases twice per period, the two purchases could be composed as: AA, AB, AC, BA, BB, BC, CA, CB, or CC It is possible to compute the probability of each of these outcomes if the brand that the customer bought last is known
V. Kumar and W. Reinartz – Customer Relationship Management 42
Summary
In the absence of individual customer data, companies used to rely on traditional marketing metrics like market share and sales growth Acquisition measurement metrics detect the customer level success of marketing efforts to acquire new customers Customer activity metrics track customer activities after the acquisition stage Lifetime duration is a very important metric in the calculation of the customer lifetime value and is different in contractual and non-contractual situations Firms use different surrogate measures of customer value to prioritize their customers and to differentially invest in them Firms can use information about size of wallet and share of wallet together for the optimal allocation of resources Transition matrix measures the probability for a customer to purchase a particular brand providing the previous purchased brand is known
V. Kumar and W. Reinartz – Customer Relationship Management