Case Study: Leading Voluntary Insurance Provider achieves 2x uplift in churn prediction with Quantiphi

Developing a one-stop platform to identify customer groups having a high possibility of churning out

Quantiphi worked with Leading Voluntary Insurance Provider, a top voluntary insurance company in the U.S. and a leading individual insurance provider in Japan, to tackle customer churn prediction for its Group business. The client needed a more effective, quantitative way to identify group customers at high risk of churning, since the existing process relied on qualitative analysis of servicing data and was complicated by sparse and missing attributes.

Quantiphi implemented a churn prediction platform using Google Cloud Storage, Google Compute Engine, and BigQuery, along with Random Forest and Gradient Boosted Decision Trees, and delivered the results through Tableau dashboards. The solution helped Quantiphi achieve more than a 2x uplift over the existing baseline Risk Score Model and made model insights accessible to both analysts and business users by highlighting the key features influencing churn.