Subscription · EU
Churn prediction that saved millions in revenue
A European subscription business was losing customers without warning. The brief: predict who is about to leave, explain why, and target retention spend where it actually pays back.
Monthly churn
−34%
after retention plays
Model accuracy
0.91
AUC
Revenue retained
€2.4M
annualized
Campaign ROI
5.8×
on retention spend
Churn risk by customer segment
average probability of leaving
Customer base by segment
share of all customers
◆ Findings
- Two segments — at-risk and dormant — carried almost all of the churn.
- The biggest warning sign was a drop in usage, weeks before anyone cancelled.
- Retention budget was sprayed evenly instead of where it could change the outcome.
✓ What we did
- Scored every customer weekly with a clear, explainable risk reason.
- Triggered targeted retention offers only for high-risk, high-value accounts.
- Gave the team a dashboard to watch risk move and measure each campaign.
Have a project like this?
Tell us what you are building. We reply within one business day with a clear plan and a fixed quote.
Start your project