Underwriting beyond
data enrichment.

Working with a high-net-worth insurance company, we applied machine learning methods to support underwriting analysis.

The work focused on prediction, feature engineering, customer segmentation, interpretation and price elasticity.

The analysis examined customer characteristics relevant to retention and acquisition and provided information for model validation.

Churn analysis identified a group of customers who were most profitable for the company. A range of machine learning models was used to examine features associated with customers leaving. Gain charts were used to identify prospective customers more likely to convert and existing customers less likely to leave.