Tesco, the largest supermarket chain in the UK, has developed a CRM strategy that is the envy of many of its competitors. Principally a food retailer, in a mature market that has grown little in the last 20 years, Tesco realized that the only route to domestic growth was taking market share from competitors. Consequently, the development of a CRM strategy was seen as being imperative.
In developing its CRM strategy, Tesco firstly undertook customer portfolio analysis (CPa) to examine its customer base. It found that the top 100 customers were worth the same as the bottom 4000. It also found that the bottom 25% of customers represented only 2% of sales, and that the top 5% of customers were responsible for 20% of sales.
The results of this analysis were used to segment Tesco’s customers and to develop its loyalty programs.
C A S E I L L U S T R AT I O N 5 . 4
CUsToMER PoRTFolIo ManagEMEnT
managing the account might require the services of a large number of people – sales man- ager, customer service executive, and applications engineer among others. The customer might demand and receive customized product, delivery in less- than- container loads, just- in- time, extended due dates for payment, and deep discounts on price. These costs- to- serve erode margin and might ultimately make the big customers unprofitable. Very often it is the mid- range sales volume customers that are the most profitable. Figure 5.7 shows the profitability of customers who have been previously clustered according to volume. The chart shows that the top 20% of customers by volume are unprofitable, just like the bottom 20% by volume.
When Kanthal, a Swedish manufacturer of electrical resistance heating elements, intro- duced activity- based costing, they found that only 40% of their customers were profitable.
Two of their top three customers by sales volume were among the most unprofitable. The most profitable 5% of customers generated 150% of profits. The least profitable 10% lost 120%
of profit. A major challenge for Kanthal was deciding what to do with the unprofitable cus- tomers.16 Their options included implementation of open- book accounting so their custom- ers could see how much it cost to serve them, negotiation of service levels with customers, introducing transparent rules for migrating customers up and down the service level ladder, simplifying and standardizing the order process, introducing a self- service portal, negotiating
Sales
Top 20%
by volume
Bottom 20%
by volume Quintile
Figure 5.6 The Pareto principle, or 80:20 rule
Quintile Profitability
Volume Quintile Figure 5.7 Customer profitability by sales volume quintile
price increases, sorting product lines into those that could be delivered ex- stock and oth- ers for which advance orders were required, and rewarding account managers for customer profitability – both per cent margin and total krona (crown) value.
Since the early 1980s a number of alternative models have been proposed for assess- ing B2B companies’ customer portfolios.17 They generally classify existing customers using a matrix and measurement approach. Many of these contributions have their origins in the work of the IMP (Industrial Marketing and Purchasing) group that we introduced in Chap- ter 2.18 Many of the models are conceptual, but some models have been applied or subjected to empirical testing.19
Bi- variate CPM model
Benson Shapiro and his colleagues developed a customer portfolio model that importantly incorporated the idea of cost- to- serve into the evaluation of customer value.20 Figure 5.8 presents the matrix they developed.
In this model, customers are classified according to the price they pay and the costs incurred by the company to acquire and serve them. Four classes of customer are identified:
carriage trade (often newly acquired customers who are costly to serve but pay a relatively high price), passive customers, aggressive customers and bargain basement customers.
The important contribution of this customer portfolio model is that it highlights the
CUsToMER PoRTFolIo ManagEMEnT
issue of customer profitability, by recognizing that costs are not evenly distributed across the customer base. Some customers are more costly to win and serve than others, and, if this is accompanied by a relatively low received price, the customer may be unprofitable.
Table 5.10 shows how costs- to- serve can vary before the sale, in production, in distribution and after the sale.
Although this bi- variate customer portfolio model uses cost- to- serve and received price to estimate customer profitability, it fails to recognize that customers may create other forms of value for suppliers. A number of tri- partite CPM models have tried to address this deficiency.
Received price
High
low
Passive Carriage
trade
bargain
basement aggressive
low High Cost- to- serve
Figure 5.8 shapiro et al.’s customer portfolio matrix
Table 5.10 How costs vary between customers
Pre- sale costs Production costs Distribution costs Post- sale costs geographic location:
close v. distant order size shipment
consolidation Training Prospecting set- up time Preferred
transportation mode
Installation
sampling scrap rate back- haul opportunity Technical support Human resource:
management v. reps Customization location: close v.
distant Repairs and
maintenance service: design support,
applications engineering order timing logistics support, e.g.
field inventory
Tri- variate CPM models
Peter Turnbull and Judy Zolkiewski have developed the three- dimensional CPM frame- work shown in Figure 5.9.21 The dimensions they propose are cost- to- serve, net price and relationship value. The first two variables are adopted from the Shapiro model. Rela- tionship value, the third dimension, allows for other strategic issues to be taken into account. Relationship value is “softer” or more judgmental than the other two dimensions.
Among the questions considered when forming a judgement on relationship value are the following:
• Are the goods or services critical to the customer?
• Is the customer a major generator of volume for the supplier?
• Would the customer be hard to replace if they switched to another supplier?
• Does the customer generate cost savings for the supplier?
Thomas Ritter and Henrik Andersen’s more recent modeling also takes a broader view of customer value, by incorporating not only the profitability of the customer, but their growth potential and level of commitment, using both objective data and judgments from account managers.22 This model produces a portfolio of six different clusters of customer, two of which produce neither profit nor growth potential as shown in Figure 5.10. The authors have subjected the model to empirical testing and conclude that it “is a useful tool for managers to gain a better understanding of their customer portfolio”. They find that each
High
Net price
Low Low
Cost to serve
High Low
Relationship value
High
Figure 5.9 Turnbull and Zolkiewski’s 3D Customer Classification Matrix
CUsToMER PoRTFolIo ManagEMEnT
GROWTH POTENTIAL
N O
N O
Y E S Y
E S
NO
Cherry Pickers:
Customers who actually prefer another supplier but are forced to buy from you due to a lack of alternatives
Skeptics:
Customers who used to prefer you as a supplier due to product excellence or relational bonds but who are skeptical at the moment
Cowboys:
Customers who press prices, law and ethics beyond the limits
Time Bandits:
Customers who prefer you as their supplier but who do not adequately pay for the resources they use
PROFITABILITY
COMMITMENT
YES
Potentials:
Customers who prefer you as their supplier but who still buy other places
True Loyalists:
Customers who prefer you as their supplier and buy therefore everything from you
Figure 5.10 Ritter and andersen’s 6- pack model