Business Analytics for Sales and Marketing Managers: How to Compete in the Information Age

PROGRAM-LEVEL LEAD INFORMATION

The team was finally formed in September 2004. It identified as its first task the collection of lead information about the different kinds of churn that the company experienced. This was chosen as the first focus because it became apparent that customers left for various reasons, that is, there were many kinds of churn.

First of all, the telecom provider itself churned customers who did not pay their bills. Second, customers churned themselves. Customers could churn if they had received a better offer from a competitor or if they were dissatisfied with the treatment they had received from the company.

Analysis revealed that there were even more kinds of churn. Exhibit 9.1 presents the ways customers could leave the company in a mutually exclusive and covering manner. The different churn types will be presented in more detail in the paragraphs to come. As mentioned, there were two kinds of churn: company-initiated churn and customer-initiated churn. Based on data warehouse information, this split was found to be 30–70:

• Focusing on the customer-imitated churn, based on interviews with customers who had churned, it was estimated, that 70% of these customers overall were satisfied with the treatment they had received from the company; they simply had received a better offer from a competitor. To stop this sort of churn, we had to develop a way to give the right customers the right offer at the right time. Hence churn prediction would become what generated ...