Customer portfolio
management
(CPM)
How to find customer?
Instructor: MSc. Ngo Hien Dan
ngohiendan.cs2@[Link]
Business Portfolio: This refers to the entire collection of assets (products, services,
business resources, investment and business units,..) that a company holds. Each
asset has a different profile (i.e., set of attributes) and poses different potential risks
and rewards for the company.
Customer portfolio: A customer portfolio is the collection of mutually exclusive
customers or customer groups that make up a company’s entire customer base,
where each customer or customer group varies in terms of its risk/reward profile.
Customer portfolio management (CPM): Customer portfolio management (CPM)
prefers to activities which aims to optimize business performance across the entire
customer base by delivering superior value propositions with different cost for
different segments of customers.
Basic principles of CPM
1. Market segmentation: “Market segmentation is the process of dividing up a
market into more-or-less homogenous subsets for which it is possible to create
different value propositions.”
1.1. Market segmentation goal
For customer acquisition: Segment potential customer/market to determine which
customers to attract.
For customer retention and development: Segment existing customer/market to
create different value for each customer group.
1.2. Identifying market: Consumer markets; Reseller markets; Business markets;
Government markets and International markets.
Business market Illustration
segmentation criteria
International Standard An internationally agreed standard for classifying goods and service producers
Industrial Classification
Dispersion Geographically concentrated or dispersed
Size Large, medium, small businesses: classified by number of employees, number of
customers, profit or turnover
Account status Global account, national account, regional account, A or B or C class accounts
Account value <$50,000, <$100,000, <$200,000, <$500,000
Buying processes Open tender, sealed bid, Internet auction, centralized, decentralized
Buying criteria Continuity of supply (reliability), product quality, price, customization, just-in-time,
service support before after or sale
Propensity to switch Satisfied with current suppliers, dissatisfied
Current share of customer Sole supplier, majority supplier, minority supplier, non-supplier
spend in the category
Geography City, region, country, trading bloc (ASEAN, EU)
Buying style Risk averse, innovator
1.3. Customer portfolio segmentation
Grouping customers based on demographics, purchasing behavior, profitability, or
other criteria to better understand and target each segment.
1.3.1. Customer portfolio segmentation can help company to: Identifying customer
segments to attract; Identifying customer segments to retain and Identifying customer
segments to withdraw.
1.3.2. Customer portfolio segmentation approaches
1/Intuitive segmentation approaches: Rely on personal perceptions, knowledge,
experiences and observation of decision-makers in the company.
2/ Analytical (Data-based) segmentation approaches: Involves collecting customer
portfolio-related data (purchase history, demographics, engagement, etc.) from
multiple sources across the company to create more precise customer segments.
Pros and Cons
1/Intuitive segmentation approaches:
Pros: Does not require large datasets or complex analytics; Useful when
launching a new business with little historical data; Useful when making quick
strategic decisions based on market intuition.
Cons: Relies on assumptions about customer behavior; Time limitations and
information constraints; Can be subjective and inconsistent.
2/ Analytical (Data-based) segmentation approaches:
Pros: Useful when a company has large amounts of structured data; More
‘scientific’ techniques which involves high levels of accuracy; Company can
develop a higher level of confidence to gain customer’s insights and
segmentation.
Cons: Requires time and effort to collect, clean, analyze and interpret
customer-related data.
Intuitive and data-based segmentation approaches
Intuitive Data-based
Brain-storm segmentation variables Obtain customer data
• gender, life-style age, • internal and external
• SIC, size, location
Produce word-profiles Analyze customer data
Compute sizes of segments Identify high/medium/low value customer segments
Assess company/segment fit Profile customers within segments
• age, gender, life-style
Make targeting decision • location SIC, size,
• one/several/all segments?
Assess company/segment fit
Make targeting decision
• one/several/all segments?
Define customer segments: In B2B market contexts
Customer Focus
segmentation base
International Standard An internationally agreed standard for classifying goods and service producers
Industrial according to standard industry descriptors
Classification
Size The number of employees, number of customers, profit and/or turnover of the
organization compared to other organizations
Buying approach The ways in which the organization approaches suppliers to solicit offers common
approaches can include open tender, sealed bid, and Internet auction
Buying process control The degree to which the organization uses a central procurement and/or contract
management approach as opposed to a delegated one
Buying process formality The degree to which the organization uses a strict, formal buying process involving
multiple stakeholders as opposed to an informal one
Market position The position where the organization sits with respect to its competitors in their
customer markets
Organizational health The likelihood that the organization will continue to survive, noting their anticipated
financial performance, the status of their target markets, and the degree of
competition they face (among other factors)
Define customer segments: In B2C market contexts
Customer
segmentation Focus Variables
base
Behavioral The major patterns of activity that members of Benefits sought, and purchase occasion,
the customer segment exhibit as these relate to usage consumed, volume loyalty status
purchases (recency-frequency-monetary spend,
share of category spend)
Geographic The locations where members of the customer Country, region, TV region, city, city
segment are most likely to make purchase size, postcode, residential neighborhood
decisions
Demographic The structural attributes of members of the Age, gender, occupational status,
customer segment likely to impact purchases household size, marital status, terminal
educational age, household income,
stage of family life cycle, religion, ethnic
origin, nationality, language group
Psychographic The psychological and lifestyle characteristics of Attitudes, beliefs, views, lifestyle,
members of the customer segment likely to impact personality, values
purchases
Customer portfolio segmentation: SWOT Matrix
Define customer segments: BCG Matrix - The growth-share matrix
From a CRM perspective
A customer with a
balanced portfolio of
products has greater
Customer Lifetime Value
(CLV) potential for a
supplier than a customer
with an unbalanced
portfolio.
Define customer segments: McKinsey/GE customer portfolio matrix
Company's competitive position
High
Market
Medium
Attractiveness
Low
Strong Average Weak
Attractive markets
Medium-priority markets
Unattractive markets
Basic principles of CPM
2. Sale forecasting: Sales forecasting involves estimating the revenue that a
customer segment will produce for the company over a specified period (such as
quarterly or annually). One major issue with sales forecasting is that the data
available takes a historical or, at best, a present-day view.
2.1. Some sale forecasting methods
2.1.1. Qualitative methods: Qualitative methods are the most widely used
forecasting methods.
1/Customer surveys: ask consumers or purchasing staff to give their opinion on
what they are likely to buy during the forecast period.
2/ Sales team estimates: can be useful when salespeople have built close
relationships with their customers. A key account management team might be
well placed to generate several individual forecasts from the team membership.
Basic principles of CPM
2.1.2. Some sale forecasting methods: Quantitative
1/ Time series: Time series approaches take historical data and extrapolate them
forward in a linear or curvilinear trend. This approach makes sense when there
are historical sales data, and the assumption can be safely made that the future
will reflect the past.
2/ Causal methods
Leading indicators: The current activity or event indicates that another activity or
event will occur in the future.
Regression models: Use the data for several anticipated variables to estimate
future demand. The variable being predicted is called the dependent variable; the
variables used as predictors are called independent variables.
Basic principles of CPM
3. Activity-based costing (ABC): Activity-based costing (ABC) can help
companies understand the relative costs to serve each customer segment.
Activity-based costing (ABC) can be applied to 2 different cost types
Volume-Based Costs (Product-related): costs can vary with the size of the order
but the cost per unit of product is fixed for any order and any customer (E.g.
Material and direct-labor costs)
Order-related costs (customer-related): vary depending on the product and
process requirements of each specific customer, cohort, or segment. (E.g.
Custom packaging, order processing, unique delivery requirements, special
handling,...)
Basic principles of CPM
4. Customer lifetime Value (CLV) estimation
Historical CLV computation
Total Revenue
CLV = ARPU = Number of Customers
𝐴𝑅𝑃𝑈 = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑅𝑒𝑣𝑒𝑛𝑢𝑒 𝑃𝑒𝑟 𝑈𝑠𝑒𝑟
Predictive CLV computation
T × AOV × AGM × ALT
CLV = TR − CS TR = Total Revenue
Number of customers at the end of period GM = (
TR
) × 100
CS = Cost of sales
T = Average number of transaction (per month)
AOV = Average order value (per month)
AGM = Average Gross Margin (% or number) GM month 1 + ⋯ + GM (month n)
AGM =
ALT = Average Customer Lifetime (months) n
1 CB − CE
ALT = CB = Customer at the beginning of a month
Churn Rate (%) Churn Rate = CE = Customer at the end of a month
CB
Traditional CLV computation (Sumil Gupta and Donald Lehmann)
CLV = customer lifetime value
𝑟 m = margin or profit from a customer per period (e.g. per year)
CLV= m
1+𝑖 −𝑟 i = the discount rate (e.g. 0.8 or 80%)
r = the retention rate (e.g. 0.12 or 12%)
Discount rate: The rate of future cash flow are discounted to the present value
Discounted cash flow (DCF): The total present value of expected future cash flow,
discount rate is used to reduce the value of the future cash flow
A higher discount rate reduces the present value of future cash flows, suggesting
higher perceived risk or cost of capital. A lower discount rate increases the present
value of future cash flows, indicating a lower perceived risk or cost of capital.
Discounted cash flow (DCF) formula t = time period
𝐶𝐹t
DCF= t CF = the cash flow in period t
1+𝑟
This formula sums all future cash flows,
each discounted back to its present value. r = the discount rate
Traditional CLV computation (Sumil Gupta and Donald Lehmann)
Discount rate (i)
Retention rate (r)
10% 12% 14% 16%
60% 1.20 1.15 1.11 1.07
70% 1.75 1.67 1.59 1.52
80% 2.67 2.50 2.35 2.22
90% 4.50 4.09 3.75 3.46
Margin Multiple (m)
To calculate CLV, the company need to know:
For an existing customer:
1 What is the probability that the customer will buy products and services from the
company in the future, period by period?
2 What will be the gross margins on those purchases period-by-period?
3 What will be the cost of serving the customer, period-by-period?
For new customers, ask one more question:
4 What is the cost of acquiring the customer?
Finally, to bring future margins back to today’s value, another question needs to be
answered for both existing and new customers:
5 What discount rate should be applied to future net margins?
Notes: CLV calculations rely on a set of assumptions that the company makes about the
future purchase behaviors of current or potential customers. This means that any CLV
calculation is a ‘best guess’.
Basic principles of CPM
5. Data Mining
5.1 Data mining helps CPM to answer the following questions:
How can businesses segment the market to identify potential customers?
How can businesses match their existing customers into customer groups?
Which customers offer the greatest future potential?
5.2. Data mining includes:
Directed data mining: uses decision tree to target variables and relevant forecast
variables
Undirected data mining: uses cluster analysis to find relationships between
underlying data or structures.
CPM Models: Shapiro et al.’s customer portfolio matrix
High
Carriage
Passive
trade
Perceived
price
Bargain
Agressive
basement
Low
Low High
Cost-to-serve
CPM Models: Ritter and Andersen’s six -pack model
Notice
• Do not recommend customer loyalty
programs for skeptics;
• Reduce discount for cherry
pickers;
• Uninvite time bandits to free
seminars;
• Do not use satisfaction measures to
evaluate cowboys;
• Use share of wallet to evaluate
potentials.