4 Observing renewal and churn

The essence of fighting churn with data is learning from the natural experiments that occur every time a customer chooses to stay with or churn from the service. A natural experiment in this context means a situation that tests an outcome you are interested in, but you didn’t set it up like a formal experiment. These experiments are the churns and renewals that have already occurred, and the results are waiting for you in your data warehouse. Why aren’t you learning from the results already? Actually, observing these experiments and reading the results can be a little tricky if you’ve never done it before. This chapter teaches you the right way to observe the customer experiments that have already taken place in your own data.

The scenario in this chapter assumes you have already produced behavioral metrics (as described in chapter 3) and calculated some kind of churn rate measurement (chapter 2). This chapter is a preparation step for the churn analysis. You are going to collect observations of customer metrics at known times when customers churned or continued with the service. In relation to the overall book scenario introduced in chapter 1, this chapter focuses on the processes highlighted in figure 4.1.

4.1 Introduction to datasets

4.2 How to observe customers

4.2.1 Observation lead time

4.2.2 Observing sequences of renewals and a churn

4.2.3 Overview of creating a dataset from subscriptions

4.3 Identifying active periods from subscriptions

4.3.1 Active periods

4.3.2 Schema for storing active periods

4.3.3 Finding active periods that are ongoing

4.3.4 Finding active periods ending in churn

4.4 Identifying active periods for nonsubscription products

4.4.1 Active period definition

4.4.2 Process for forming datasets from events

4.4.3 SQL for calculating active weeks