How ML Can Improve Churn Prediction to Retain More Revenue for Insurers

The insurance company found that customers actively or passively churn due to various reasons. To note, active churn is when someone cancels their policy before its expiration date while passive churn exists when someone simply decides to not renew their policy. There was also a distinct lack of customer demographic data to develop a churn model.

Tesseract Academy developed a churn technique using the Survival Model - an ML model which is more effective in solving and predicting churn. A Survival Model like this can directly model relative risk - that is the risk of one customer vs another customer. It also models this risk over time. The model included alternative data such as the device type, technical specs, when the customer joined, and their country of residence.

Outcome

Due to this solution, the insurance company was able to develop a good understanding of factors that contribute to churn. They now have a predictive model that predicts which customers are at higher risk of churn and when they are about to churn. On top of that, they are able to predict about 89% of the customers that churn with a precision of up to 90%, which means they can target about 4 out of 5 customers that will churn.