platform · predictions

Every customer has a rhythm.

Retrics learns it from her own orders — and flags the week she'll need to buy again.

01cadence

Her rhythm, not a segment average.

The window comes from the intervals between her own orders. Lapse risk is drift past her cadence — not days past a store-wide mean.

02health & return probability

Scored against the floor, not hope.

Every return probability is read against the ~2.7%of lapsed customers who come back with no outreach at all. If she'd return anyway, Retrics says so.

03what it feeds

Predictions rank. They never send.

Every score becomes a board, a brief, or a draft waiting in violetfor your approval — nothing moves on a model's say-so.

before you ask

Straight answers.

What about customers with only one order?
One order isn't a rhythm yet, so Retrics falls back to how that product is typically repurchased across your store — and watches the second-order window closely, because that is where the strongest measured effect lives. Once she orders again, her personal cadence takes over.
How accurate are the predictions?
Honestly: they are windows, not appointments — the week a customer is likely to need to reorder, not the day. The method is validated with a holdout-scored backtest, and the ~2.7% natural-return floor is always subtracted before anything is claimed. The full method is public on our methodology page.
Do predictions ever send anything on their own?
No. Predictions only rank and flag. Anything Retrics drafts from a prediction waits in violet — the mark that the AI wrote it — until you approve it. Nothing sends without you.

See your own customers' rhythms.

Connect your store and Retrics scores your real order history the same day.

No credit card · read-only Shopify scopes · live in ~10 minutes