See when customers come back, and what each reward brings in

Retention used to be four lifetime numbers. They told you whether customers came back, but not when, and not which months of new buyers kept buying.

Time to second purchase. A new card on the Customers tab shows the median number of days to a second order, the share of buyers who return within 30, 90 and 180 days, and a weekly chart of when those second orders land. It's the number to set points expiry, win-back emails and reminders against.

Retention by first purchase month. A cohort table lines up each month's new buyers against the twelve months that follow, so you can see whether the customers you won in spring are still buying in autumn. Hover a cell to see how many of them bought again. Both cards wait for enough data to say something true: at least 10 recorded buyers, and for the second-purchase card at least 5 who bought twice.

What each reward brings in. Top rewards redeemed now puts revenue next to cost: the total of the orders each reward's codes were used on. A new advisor tip points out a reward that customers keep claiming but never use, so you can fix it or swap it for one that gets them back to checkout.

Members against guests. Member impact on the Overview tab compares the average order of members and guests, and says it plainly when the gap is 5% or more: "Members spend 18% more per order than guests".

Charts that follow your filter. The revenue and reward cost chart plots one point per day for any period up to 92 days, and a compact copy now sits on the Overview tab. The sparklines under points earned, points redeemed and rewards redeemed follow the date filter too, and All time shows the last 12 months instead of a single one.

Unredeemed points when rewards have no price. A program whose rewards are all gifts or free shipping has no cash value to put on unspent points, so the liability card shows Unredeemed points for it, and the API returns the points without a money figure.