CASE STUDY
Repurchase Propensity Score for a Pizza Chain
BUSINESS PROBLEM
- Create a solution to identify the specific customers who are likely to make a purchase in the future, enabling micro-targeting and upsell / cross-sell opportunities .
PERFORMANCE
- The model has a good discriminatory power to classify the customers as ‘likely to repurchase’ or ‘not’.
- Testing Scenarios – the model could identify more than 58% of the target customers who had the highest Repurchase Probability Score.
SOLUTION
- Computed the Repurchase Probability Score by leveraging the Poisson Regression Model.
- Transaction Data was used to analyse Purchase pattern behaviour.
- Performance metric checks on the Validation data were performed to select the final model.
- The final process returns ‘Repurchase Propensity Score’ for all customers, finally segmented into 10 bands.
BENEFITS
- Provides a guide to develop marketing strategies based on the prediction that the customer will be purchasing in the next period based on past data.
- Implementation of advanced analytics solutions to drive business growth and decision making.
- Promotes ‘Customer Retention’ and ‘Customer Experience’.
PERFORMANCE OF THE MODEL - KS PLOT
