Skip to main content
Predicts which regular givers are at risk of cancelling — so a small save programme can protect a large share of recurring income. What it predicts — For supporters on an active recurring mandate, the probability their subscription ends within the next window (e.g. 3 or 6 months). Who it scores — Only currently-active subscribers as of the scoring date. One-off donors and already-cancelled givers are excluded automatically. The decision it drives — Who to put into a save or thank-and-reassure programme before they lapse: a call, a stewardship email, an impact update. Worked example — You score 12,000 active regular givers each month and flag the top 800 by churn risk. The retention team calls the highest-value 200 and emails the rest; cancellations in that group fall against a holdout. What good looks like — A small, cheap intervention protects a disproportionate share of recurring income.
Pairs well with lifetime value — prioritise saves by what’s at stake.

All models

Back to the full catalogue.

Create & train a model

Set one up on your own data.