
The Challenge.
- Acquiring new customers is much more expensive than customer retention.
- Customer churn is an important metric to keep track of in insurance companies.
- Clients exhibit different behavior, making it difficult to accurately predict churn without using advanced analytics.

The Solution.
- Designed and built a Machine Learning model to predict the probability of churning for customers on a monthly basis.
- Created KPIs to capture each client’s service frequency and churn risk over time.
- Divided customers into three churn segments to better customize interactions.

The Results.
- Increased customer retention by identifying high-risk clients and putting retention policies in place.
- Improved customer experience by customizing interactions with different churn segments depending on the risk.
- Improved understanding of optimal times to call clients within their subscription cycle.