Hey Product Hunt! 👋 SaaS churn tools have a dirty secret: they tell you *who* churned, but not *why* a specific intervention would have prevented it. We built RetentionLens because we kept seeing the same problem — CS teams spending budget on customers who were going to stay anyway, while the ones who actually needed a call slipped through. So we went deeper: → Survival analysis (Kaplan-Meier + Cox) gives you true time-to-churn curves, not just monthly snapshots → Segment-aware ML trains separate churn models for SMB, mid-market and enterprise — because what predicts churn in each segment is genuinely different → Causal uplift modeling (T-Learner HTE) separates correlation from causation — it estimates the *incremental* retention lift of each intervention, not just who looks healthy → Budget-constrained portfolio optimizer allocates your CS capacity to the accounts where the expected value is highest Connect Stripe, get your first predictions in minutes. No SQL, no data team. Happy to answer any questions about the methodology, the math behind the uplift model, or how to read a survival curve. This is the stuff I genuinely love talking about.

RetentionLens

Predict who churns. Prove the ROI of every retention action.

Predict who churns. Prove the ROI of every retention action.