CRM Unit 5 - Customer Analytics Part I
CRM Unit 5 - Customer Analytics Part I
Management
University of Economics & Finance
Chapter 5. Customer
Analytics Part I
Lecture 5
1
Overview
Topics discussed:
▪ Traditional Marketing Metrics
▪ Customer Acquisition Metrics
▪ Customer Activity Metrics
▪ Popular Customer-based Value Metrics
Traditional and
Table 5.1 Metrics used in customer analytics part
Customer Based 1
Marketing Metrics
2
Traditional
Marketing Metrics
• Market Share
• Sales Growth
3
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:
∆𝑆𝑗𝑡
𝑆𝐺𝑡 = × 100%
𝑆𝑗𝑡−1
Where:
j = firm,
∆Sjt = change in sales in period t from period t-1,
Sjt-1 = sales in period t-1
• Evaluation
• Quick indicator of current health of a firm
• Does not provide any information about changes in customer size
• Concepts:
• Acquisition Rate
• Acquisition Cost
4
Acquisition Rate
• Acquisition = first purchase or purchasing in the first predefined period
# 𝑜𝑓 𝑃𝑟𝑜𝑠𝑝𝑒𝑐𝑡𝑠 𝑎𝑐𝑞𝑢𝑖𝑟𝑒𝑑
𝐴𝑐𝑞𝑢𝑖𝑠𝑖𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑒 = × 100%
# 𝑜𝑓 𝑃𝑟𝑜𝑠𝑝𝑒𝑐𝑡𝑠 𝑡𝑎𝑟𝑔𝑒𝑡
• 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
• Information source
• Numerator: from internal records
• Denominator: from internal records
• Evaluation
• Difficult to monitor on a customer by customer basis
10
10
5
Customer Activity Metrics
Objective of Customer Activity Measurement:
• Managing marketing interventions
11
11
3. Survival Rate
4. Lifetime duration
5. P(Active)
12
12
6
1. Average Inter-Purchase Time (AIT)
Average Inter-Purchase Time (AIT) is the average time elapsing between
purchases.
1
𝐴𝐼𝑇 𝑜𝑓 𝑎𝑐𝑢𝑠𝑡𝑜𝑚𝑒𝑟 =
# 𝑜𝑓 𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑠 𝑑𝑢𝑟𝑖𝑛𝑔 𝑎 𝑠𝑝𝑒𝑐𝑖𝑓𝑖𝑒𝑑 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
13
13
14
14
7
Retention and Defection Rate
Assuming constant retention rates, number of retained customers in any
arbitrary period (t+n) = Number of acquired customers in cohort *
Retention rate (t+n):
𝑅𝑐𝑡+𝑛 = 𝐴𝑐𝑡 × 𝑅𝑟
Where: Rrt = Retention rate in period t,
n = Number of elapsed periods
E.g.: Given a retention rate of 75%, variation in defection rate with respect
to customer tenure results in an average lifetime duration of 4 years
As already mentioned, the concepts of defection and retention are closely
related. Defection rate is calculated as follows:
𝐴𝑣𝑔. 𝑑𝑒𝑓𝑒𝑐𝑡𝑖𝑜𝑛 𝑟𝑎𝑡𝑒 𝑖𝑛 𝑡 = 100% − 𝑅𝑟𝑡
15
15
16
16
8
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
17
17
18
18
9
Actual and Predicted Retention Rate for a Credit
Card Company
100%
90% Actual
Retention rate (%)
80% retention
rate
70%
60% Predicted
retention
50% rate
40%
1 2 3 4 5 6 7 8 9 1011121314151617181920
Period
• 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
19
19
20
20
10
Survival Rate Computation-Example
• Number of customers starting at the beginning of year 1: 1,000
• Survival rate for period 2 = retention rate of period 2 survival rate of period 1
21
21
22
22
11
4. Lifetime Duration
σ𝑇𝑡=1 𝑡 × #𝑜𝑓 𝑟𝑒𝑡𝑎𝑖𝑛𝑒𝑑 𝑐𝑢𝑠𝑡𝑜𝑚𝑒𝑟𝑠 𝑖𝑛 𝑡
𝐴𝑣𝑔. 𝑙𝑖𝑓𝑒𝑡𝑖𝑚𝑒 𝑑𝑢𝑟𝑎𝑡𝑖𝑜𝑛 =
𝑁
• Where:
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
23
23
Observation Window
• Buyer 1: complete information
• Buyer 2 : left-censored
• Buyer 3: right-censored
• Buyer 4: left-and-right-censored
24
24
12
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)
25
25
5. P(Active)
• Probability of a customer being active in time t
P(Active) = 𝜏 𝑛
• Where:
• n = the number of purchases in a given period,
= is the time of the last purchase (expressed as a fraction of the
observation period)
• 𝜏 = the time of the last purchase (expressed as a fraction of the
observation period)
26
26
13
Estimation of P(Active)-Example
Customer 1
Customer 2
▪ 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
27
27
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
28
28
14
Popular Customer-Based Value
• Size of Wallet
• Share of Wallet
• Transition Matrix
29
29
Customer-
based
Value
Metrics
Popular Strategic
30
30
15
Size of Wallet
𝐽
31
31
• Where:
• j0 = focal firm or brand,
• i = customer,
• 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
32
32
16
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
33
33
𝑉𝑖0𝑗0
iSCR of customer i0 that a firm × (or brand) j0 satisfies = 𝐽 × 100%
σ𝑗=1 𝑉𝑖0𝑗
• Where:
• j0 = focal firm or brand,
• i0 = focal customer,
• J = all firms or brands available in focal category,
• Vij = purchase volume of customer i from firm (or brand) j
34
34
17
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 B B/A
Total requirement of Total number of notebook Share of category
notebook computers Computers purchased from requirement for PEAR
per customer in PEAR Computers per computers per customer in
2010 customer in 2010 2010
Customer 1 100 20 .20
Customer 2 1,000 200 .20
Customer 3 2,000 500 .25
35
35
• 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
36
36
18
Share of Wallet (SW)
• Individual Share of Wallet (iSW)
𝑆𝑖𝑗0
iSW of firm j0 to customer i = 𝐽 × 100%
σ𝑗=1 𝑆𝑖𝑗
• Where: j = firm,
i = customer,
Sij = sales of firm j to customer I,
J = all firms who offer the category under consideration
• 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%
37
37
→ 750,000/1,250,000 = 60%
38
38
19
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
39
39
40
20
Segmenting Customers Along Share of
Wallet and Size of Wallet
Share of
Wallet
Target for
Low Do nothing additional selling
▪ 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
41
41
42
42
21
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
43
43
Transition Matrix
Brand purchased next time
Brand A B C
currently A 70% 20% 10%
purchased B 10% 80% 10%
C 25% 15% 60%
• Characterizes a customer’s likelihood to buy over time or a brand’s likelihood
to be bought
• It is possible to compute the probability of each of these outcomes if the brand that the
customer bought last is known
44
44
22
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
45
45
23