Separating financial from commercial customer churn: A modeling step towards resolving the conflict between the sales and credit department

In subscription services, customers who leave the company can be divided into two groups: customers who do not renew their fixed-term contract at the end of that contract, and others who just stop paying during their contract to which they are legally bound. Those two separate processes are often modeled together in a so-called churnprediction model, but are actually two different processes. The first type of churn can be considered commercial churn, i.e., customers making a studied choice not to renew their subscriptions. The second phenomenon is defined as financial churn, people who stop paying because they can no longer afford the service. The so-called marketing dilemma arises, as conflicting interests exist between the sales and marketing department on the one hand, and the legal and credit department on the other hand. This paper shows that the two different processes mentioned can be separated by using information from the internal database of the company and that previous bad-payment behavior is more important as a driver for financial than for commercial churn. Finally, it is shown on real-life data that one can more accurately predict financial churn than commercial churn (increasing within period as well as out-of-period prediction performance). Conversely, when trying to persuade customers to stay with the company, the impact of ‘loyalty’ actions is far greater with potential commercial churners as compared to financial churners. Evidence comes from a real-life field experiment.