The challenge
Understanding and preventing customer churn is critical to the sustainability of any customer-oriented business. An online gaming platform wanted to predict which players were likely to leave and respond with the right retention strategy.
One of the most critical and demanding requirements was a stable data collection and processing pipeline that would not disrupt business operations through downtime.
Our solution
We created a machine learning model that learns player behaviour over the course of the game and predicts the probability of churn for a particular time horizon.
Our data engineering team built a high-quality data pipeline with almost no downtime of the client’s online platform. We also put in place an A/B testing framework to experiment with retention strategies and test newly introduced game features.
- Churn probability prediction for a chosen time horizon
- Data collection and processing pipeline built with almost no platform downtime
- A/B testing framework for retention strategies and new game features
The outcome
The solution helped the client understand why players churned and develop retention strategies that reduced churn to 11.3%.
