Case Study:
Customer Churn Prediction Model
Case Study:
Customer Churn Prediction Model
Client Challenge
The client's Postpaid division experienced a large volume of voluntary churn in the past year and required a propensity model to have the ability to proactively identify customers who are likely to churn.
The ability to proactively identify customers who are likely to churn would provide the client with an opportunity to retain these customers and ultimately decrease voluntary churn rates.
The Solution
The BSC team collated a data set which contains information about each postpaid customer from various data sets.
Attributes sourced, include usage, spend, location, device, credit information, price plan, demographics and tenure. The data set was then ingested into a propensity model developed by the team.
The output of the propensity model was a ranking of customers who are most likely to churn. This enabled the client to target these customers with more aggressive deals to retain them.
Solution Value
- Leverage the model to identify customers with a high likelihood of churning.
- Understand key drivers that increase/decrease the likelihood of churning.
- Identify customers who should be prioritized for churn interventions.
- Continue executing churn reduction initiatives post engagement.
- Measure the success of churn initiatives through continuous reporting.