ANZ Onepath

We were able to model what factors led to a customer choosing not to renew a product using machine learning techniques with a strong emphasis on explainable and interpretable results. After modelling the probability of churn we then assessed customer lifetime value to decide which customers should be targeted for different retention strategies.

This product was embedded into the business and the modelling was set up to run as part of the day-to-day reporting, giving stakeholders information to act on as part of their role. There were also other benefits of the analysis which allowed for a more in-depth view of the metrics that were being reported on.