Predictive modeling helps identify customers who are likely to cancel their policies or switch insurers based on historical customer behavior data. Once an insurer identifies behavior that may lead to lapse, the insurer can implement retention strategies by offering more bespoke policies to suite their customers’ needs.
Before
Long-life Insurance Company offers a range of life and funeral insurance policies. It uses qualitative methods to market policies and has no defined strategy to retain existing customers that are likely to lapse.
After
Long-life Insurance Company now has a predictive model that identifies customers likely to lapse. This gives the insurer the opportunity to implement retention strategies to prevent lapse.
Approach
A predictive model is built using the company’s historical data on both its existing and lapsed customers. The input data included in the model contains demographic (age, gender, region, income, etc.), behavioral data (payment behavior, customer contact frequency, etc.) and psychographic data (customer’s response to the insurer’s marketing and operational campaigns). A logistic regression model is built to determine the relationship between the variables in the input data and the customer’s likelihood to lapse.
Model Results
The results are profiles of customers likely to lapse on their policies or switch insurers as well as those likely to stay with the insurer.
Results
- Increased customer retention.
- Opportunities to cross-sell and up-sell to existing customers and offer more bespoke insurance policies.
