A prominent insurance company in India, offering a diverse range of products covering various stages of a customer’s life cycle, including children’s future plans, wealth protection plans, retirement, and pension solutions, health plans, traditional term plans, and Unit Linked Insurance Plans (“ULIPs”), aimed to identify customers with a higher propensity for cross-selling and to enhance customer loyalty and lifetime value.
Overview
Approach
To achieve these objectives, we implemented a data-driven approach:
- Variable Transformation: We transformed key variables related to cross-sell rates, including the face premium ratio band, final annual premium band, policy status bands, and more.
- Segmentation: We segmented customers into categories such as urban mass, rural, and urban affluent to tailor our strategies.
- Advanced Modeling: We employed K-means clustering and logistic regression techniques at the client level.
- Two-Step Modeling: Our approach consisted of two steps:
- Step 1: Cross-Sell Propensity Score
- Step 2: Product-Specific Cross-Sell Propensity Score
Outcome
- The model significantly improved the identification of potential cross-sell customers compared to random selection.
- Overall cross-sell rates increased by an impressive 8%.
This case study demonstrates how our data-driven approach can help you identify the right customers and boost your cross-sell success.
