Reduced the number of responses from high-mortality applicants
Case Studies / Insurance / Improve Marketing ROI
A large insurance carrier wanted to focus their online marketing efforts on lower risk life insurance prospects.
Our goal was to infer the mortality risk of each applicant from the limited information we could collect online.
We developed an algorithm to select and target lower risk prospects for life insurance based on their online behavior, as well as some offline characteristics. Our model used various techniques like supervised classification, Box-Con transformation, and bootstrapping.
The model was used to filter through external and internal data to identify less risky prospects earlier in the marketing stream. The lowest risk 10% were one-fourth as risky compared to the overall pool. This reshaped their marketing to target more profitable customers.