Customer Relationship Management: Should you really acquire that customer?

I was reading up on Customer Lifetime Value concept and its relevance for taking customer acquisition costs related decisions. A thought struck me as to how in the present market scenario, when budgets are tightening, marketers are faced with the challenge of tradeoff between retaining existing and acquiring new customers. General convention is that new customers are expensive to find and acquire. This has brought about an urgent need to make an objective decision about how much time, money and effort should be spent on acquiring a new customer.

Hence the concept of Customer Lifetime Value (CLV) [that projects the value of a customer over the entire useful relationship period with the organization] looks relevant given the current recessionary trends. We can use CLV to take go/no-go decisions on customer acquisition by doing a cost benefit analysis, in terms of how much value will a prospect/ customer potentially bring in vs. how much the organization should spend to acquire it.

The concept can also be used for enhanced segmentation by complementing the demographic parameters and customer preferences, hitherto used for segmentation, with the potential that prospects/ customers have in terms of future cash flows. We can use any standard OLAP tool to calculate CLV to provide intelligence to business by using the existing application and data.

Some of the pertinent questions that come to mind are:

How is CLV a superior method than traditional marketing and sales budgets used to determine resources that organization should invest to achieve the targeted volumes and revenue?
At what levels (customer/prospect/segment/organization etc.) can CLV be calculated and what can be the possible benefits / applications at each of them?
How can CLV concept be built using a standard OLAP tool instead of a sophisticated data mining tool?

Let us discuss this concept and use the forum to explore the possibilities that it can open in terms of its application and business value that can be derived from it.

Comments

Interesting read. The problem arises when the 'relationship' is blindly substituted with a mere number. Values/prices are for objects and products. Period. Customers are built on relationships. So how would we bring the R into this equation? Churn perhaps? Also would it be right in saying that companies ought to differentiate based on relationship and not the cost?

Posted by: Deepak | May 4, 2009 10:48 PM

A very relevant thought. The biggest problem in the use of CLV is that it is often mistaken as a substitute to relationship. My understanding on the subject says that CLV is an indicator of potential which allows marketers to set their expectations from the customer accordingly and gives a good insight to them on deciding upon the maximum cost of acquisition that the organization should spend such that the customer remains profitable

Posted by: Nitin Rai | May 14, 2009 7:14 AM

While the concept of CLV may primarily be used at an organization level so as to determine the appropriate cost of acquisition per customer, it can also be used to determine a customer’s potential as not only an input for acquisition but also for determining the potential from a sales point of view.

For this, CLV at an organization level may not hold relevance since individual customers may behave differently. However, calculating CLV at customer level also possesses challenges especially in case of new customers since the absence of historical data points acts as a deterrent for prediction.

CLV at best can be calculated at segment level where the entire customer set can be divided into mutually exclusive segments based on parameters that determine/influence buying & purchasing patterns. Once the segments are created, customer lifetime value can be determined for each segment and hence the buying / consumption potential for each individual customer within that segment can be determined.

Accuracy of CLV for a segment depends highly on the ability of the organization to identify those parameters which affect the buying/purchasing patterns and the class intervals within them. While some parameters may be very generic, I think the best approach is to do regression analysis to identify the parameters and determine the intervals.

Talking about bringing R into the equation, I feel relationship as a differentiating factor should be used while calculating Loyalty of a customer. Customer Loyalty and Lifetime Value together provide a good insight as to how a customer should be tapped to derive the best returns for the organization.

Posted by: Mitul | May 18, 2009 10:33 AM