Predict policy level lapses for insurance provider


Leading South East Asia-based insurance provider catering to small and medium businesses.


The client wanted to develop a model to predict lapses at the policy level. The aim of the model was creating cost-effective interventions to target high-value customers with high likelihood of lapse.


SG Analytics decided to design a classification model to predict the probability of lapse, using the following approach:
  • We created a data mart that combined demographic data with policy characteristics.
  • Our data scientists developed a logistic model to predict the probability of a customer lapse on premium payments using variables such as age, gender, marital status, income, household size, mode of payment, payment frequency, timeliness of payment, etc.
  • The team also designed other models based on Kernel-based SVM Regression and Tree-based Regression techniques.
  • Finally, we calculated the lifetime value index for all the customers. Based on the lifetime value, we identified the customers that should be targeted for business growth.


Achieved 85% accuracy on unseen data, with an MCC score of 71.
Reduced probability of lapsing from 16% to 7%.