Supporting sales activities with “Approach list” created based on learning technologies Qorus-NTT DATA Innovation in Insurance Awards 2025 - Winner

Aflac is a life insurance company with a focus on third-sector cancer insurance and medical insurance, with product sales through sales agencies. However, third-sector competition is becoming more intense, making closings more difficult only through conventional sales methods. To resolve this issue, we have considered a data-driven approach to enhance sales efficiency by identifying customers with high insurance needs and proposing appropriate products.

In this project, we designed a solution for increasing sales efficiency through a machine learning model, enabling more effective agency sales activities. With methods based on conventional experience, it is difficult to effectively use vast amounts of customer data due to limits on identifying optimum customers and products. First, we built a machine learning model using customer attributes and touchpoint data to forecast product conversion rates, making it possible to identify priority customers and products for proposal. Then we optimized approach timing by projecting days and times that would be more difficult for phone contacts, enabling agencies to identify priority-contact customers from client lists for more efficient sales activities. Then we increased forecasting accuracy by using transfer learning in product fields having little correct data. This series of initiatives allowed us to increase sales productivity through increased conversion rates.