Spot at-risk customers before they lapse

Customers signal they're leaving before they go: past the reorder window, engagement drops, an unresolved complaint. A simple at-risk score catches them while a save is still cheap.

Churn has warning signs: past the expected reorder window, opens and clicks fading, an unresolved support issue, shrinking order value. A small store can run a simple RFM-style at-risk score without any fancy tooling, and AI can build it from an order export. This note is the strategy layer; the winback flow itself and the email mechanics live elsewhere.

Almost no customer vanishes without warning. They drift past their usual reorder window, stop opening emails, or go quiet after a complaint.

Scoring those signals weekly costs nothing and catches people while a save is still cheap. AI can build the score straight from your order export.

What you'll learn

Anchor everything to your reorder cycle
Layer the softer signals
Score simply, act by bucket
Route each bucket to its play