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

IDENTIFYING CUSTOMERS FOR SINGLE CAMPAIGNING

If you used the decision tree in the beginning of this chapter (Exhibit 6.1), you were directed to this section because you indicated that you want to start a retention campaign using churn prediction. The outcome of this way of doing campaigning is a percentage figure that indicates the risk of every one of your customers leaving within the next month. Based on this figure and the customers value, you can start contacting all the high-risk customers through mail or by phone. Since you will be starting only one campaign, it is assumed that you are interested only in which customers might leave you when, but not so much why, because they are all going to get the same treatment anyway. An example could be if the project is owned by the call center and the essence of the campaign is to call all customers with a high churn risk. It could also be because you work in an organization with low marketing skills that wants to test the value of this customer-centric process. Finally, this way of campaigning can be a very good way of handling customer segments that are too small to be considered for a separate campaign, as described in the previous section. Therefore, it serves as a safety net that ensures that all high-value customers with a high churn risk also get included in a retention program. You can read more about this way of using churn prediction in Chapter 9.

How to Prioritize Customers

The typical way of running such a project is to ...