Back to industries
Telecom, Media & Tech
Predictive churn reduction
Client · European telecom operator · 12M mobile lines
End-to-end churn prediction and prevention combining ML scoring, uplift modeling and test-and-learn marketing — reducing annual churn by 1.3 points.
01
Challenge
- Mature, saturated market with 18% annual B2C mobile churn.
- Costly, poorly personalized historical retention actions (push offers, outbound calls).
- Limited ability to identify weak signals of churn.
- Weak measurement of retention efficiency, hard to separate causal from windfall effects.
02
Solution
- Churn prediction model at 30/60/90 days with 400+ features (usage, billing, support, network quality, app behavior).
- Uplift modeling to target customers where retention generates positive uplift vs. blanket campaigns.
- Test-and-learn platform comparing offers and reallocating budget to top performers.
- Channel integration (call center, app, email, SMS) with channel and message personalization.
03
Business impact
- −1.3 pt annual churn on the targeted segment (~110K customers retained).
- −22% retention cost per retained customer thanks to uplift modeling.
- +7 pt engagement on proactive retention actions.
- Approach now extended to B2B SMB segment.
Other industries
Banking
Real-time transaction fraud detection
Insurance
Health underwriting automation with GenAI
Retail
Local assortment personalization & dynamic pricing
Energy
Wind production forecasting & market arbitrage
Manufacturing
Predictive maintenance on F&B production lines
Logistics
Route optimization & demand forecasting
Let's identify where AI can create value in your business.
Book a 30-minute introductory call with a founder.