Intelligent Enterprise

Better Insight for Business Decisions

RSS
Webcasts
Whitepapers
Subscribe
Home

December 5, 2001

Never the Same

Using value state models can help you classify customers to make better business decisions

By Erik Thomsen

Continued from Page 1

By looking at entities in terms of value states, you can easily see that slightly more than a third of the customers have been and are continuing to provide only medium value. Furthermore, that number is more than three times the number of customers in any other value state. And more than half the customers either were of medium value or have become medium in value.

But knowing the number of customers in each value state isn't enough to make key decisions. You also need to know the dollars sold to each category of customer, average transaction size, and so forth. A small number of customers can account for a large share of the revenue. There are many variables for which it's useful to view their values in terms of how they vary in value state space. Table 5 shows that although only 10 percent of your customers maintained a high value state over the last two time periods, they accounted for about a quarter of your sales.

8. Decide how to handle new customers and defunct customers.

Astute readers may have questioned the value state schema in Table 5, saying that it lacked the concept of the noncustomer, either on the "from" side (a new customer) or on the "to" side (an ex-customer). By creating a no-value bin in the earnings dimension, you can explicitly model these noncustomers. Of course, you must still make business decisions, such as defining the difference between a highly inactive customer and a noncustomer. Regardless of how those definitions are made, you will wind up with a value state model like Table 6.

QUESTIONS TO ASK

Once you have your data in a value state model, you can pose myriad queries to your newly structured information:

What percentage of customers is stable high value, has changed from high to medium value, or is improving or declining in value?
What are the attribute characteristics of these customers?
What percentage of sales accrues to stable medium-value customers?
What percentage of total earnings is from stable high-value customers?

And once you've gained new insights into your customers and their changing value states, you are armed with the appropriate information to take decisive actions, such as:

Sending different mailings to customers in different value states
Offering special discounts to customers who recently transitioned from a high to a medium value state
Offering long-term service relationships to customers who transitioned from low to medium value several periods ago and have stayed medium value since then
Analyzing the attributes of customers in each value category, building a predictive model of customers likely to increase or decrease in value, and then using that model to identify customers for focused marketing activity.

Rate This Article

Comments:

Optional e-mail address:

The pace of change in the business world is only increasing. Using value state models can help you classify customers according to the dynamics of their relationship with your company, make targeted decisions about how to treat different value-classes of customers, and better leverage your CRM investments.

Erik Thomsen [erik@dimsys.com] is cofounder of Power Thinking Tools, which developed the first OLAP engine with integrated statistics, visualization, text processing, and object management. He is a researcher and consultant for Dimensional Systems and focuses on integrated multitechnology analytic solutions. He is the author of OLAP Solutions (John Wiley & Sons, 1997) and coauthor of Microsoft OLAP Solutions (John Wiley & Sons, 1999).