Customer retention is central to any successful business and machine learning is frequently proposed as a way of addressing churn. It is tempting to dive right into a churn dataset, but improving outcomes requires correctly framing the problem. Doing so at the start will determine whether the business can take action based on the trained model and whether your hard work is valuable or not.
This accelerator blog will teach the problem framing and data management steps required before modelling begins. We will use two examples to illustrate concepts: a B2C retail example, and a B2B example based on DataRobot’s internal churn model.
Look out for Part 2 in this 3 part series on churn modelling for discussion of model training and consumption.
supporting code can be found here