A leading digital B2B media platform in Western Europe was struggling with customer base management.
Our client lost over one-third of its customer base, a loss that newly acquired customers could not compensate.
Many customers left during the onboarding phase and within the first year of the customer relationship – but a lot of “older” clients also churned.
When the company approached us for help, we recognized the need for a churn reduction program based on a predictive churn model.
That’s where we introduced MyBase, our platform designed to improve customer base management and boost net retention and customer loyalty.
Using MyBase, our team immediately developed a three-step approach to design and implement an effective predictive churn model.
The first step involved building a database with optimal variables for churn modeling based on in-depth analysis of churn drivers for onboarding, first year churners, and older churners. We chose variables with high accuracy for predicting churn in the second step and identified the statistical model that best predicts churn using machine learning. Finally, we assigned a risk score (0-1) from the model to each client ID and created a warning system combined with appropriate churn reduction measures to reduce churn.
Additionally, we developed customer segmentation based on churn risk and value and used it to prioritize customers. A toolbox with rapid response measures to engage high-risk customers was also designed.
Our client successfully predicted churn with over 90 percent accuracy.
Combined with appropriate churn reduction measures, it was able to significantly reduce overall churn. For example, churn rate in the first year customer group dropped by more than 4 percentage points.
お問い合わせ
サイモン・クチャーのエキスパートがいつでもご相談に応じます。お気軽にお問い合わせください。