Definition
Repeat purchase rate is the share of a store's customers who have placed more than one order, measured over a stated window. It can be calculated across everyone who bought in a period, or followed properly by cohort, tracking the customers who first bought in one month and asking how many of them came back within ninety days, six months or a year. The cohort version is slower to read and far more useful.
Formula
Repeat purchase rate = Customers with two or more orders ÷ Total customers in the window × 100
- Customers with two or more orders
- Distinct customers whose order count in the window is at least two, matched on the customer record rather than on the email typed at checkout
- Total customers in the window
- Every distinct customer who ordered in the window, which for a cohort view means only the customers whose first order fell in the starting month
One acquisition cohort followed for a year
- Customers whose first order was in March
- 1,200
- Ordered again within 90 days
- 216, or 18 percent
- Ordered again within 180 days
- 312, or 26 percent
- Ordered again within 365 days
- 420, or 35 percent
- Average orders per repeat customer in the year
- 2.6
- Share of the cohort's revenue from second and later orders
- 44 percent
Illustrative figures. The same cohort produces three different repeat rates depending on where the window closes, which is why the window has to be part of the metric's name whenever it is reported.
Why it matters
Repeat purchase rate matters because it decides what a store can afford to pay for a customer. A first order that barely covers its acquisition cost is a fine trade when a third of those buyers come back without further media spend, and a poor one when almost nobody does. It is also the metric that separates a brand from a series of transactions. Improving it works through product quality, the unboxing and the reorder reminder rather than through more traffic, so the gains do not compete with rising media costs. For stores in consumable categories it is the single number that predicts whether growth compounds or has to be bought again every month.
Where it goes wrong
- Measuring across all customers in a period rather than by cohort, which makes the rate rise simply because the store is older and the customer base has had more time to buy again
- Matching customers on the email typed at checkout, so the same person buying as a guest twice with two addresses appears as two first-time buyers and the rate is understated
- Ignoring the natural reorder cycle: a category people restock twice a year cannot be judged on a ninety day window, and a store that measures it that way will conclude retention is broken
- Counting subscription renewals alongside one-off reorders without labelling them, which flatters the figure and hides whether non-subscribers ever return
- Reading the rate without cancellations and refunds removed, since an order that was returned in full is not a repeat purchase in any sense that helps
Questions about repeat purchase rate
What is a reasonable repeat purchase rate for an online store?
It depends almost entirely on what you sell and how often people need it. A coffee roaster, a supplement brand and a pet food store live on reorders, while a mattress or a wedding band is bought once and recommended afterwards. Rather than borrowing a target, measure your own cohorts, then set the goal against the reorder cycle the product actually has.
How is repeat purchase rate different from customer retention rate?
Repeat purchase rate counts whether a customer ever came back at all. Retention rate usually measures whether customers active in one period are still active in the next, which suits subscriptions and memberships where a relationship is either continuing or has ended. For a store selling one-off orders with irregular timing, the repeat measure is the more honest of the two.
Which levers move repeat purchase rate fastest?
The ones closest to the first order. A post-purchase sequence that sets expectations and helps the customer use the product, a reorder prompt timed to when the item actually runs out, replenishment or subscription options at the right moment, and a delivery experience that arrives when promised. Discounting later orders raises the count but takes the margin the repeat business was supposed to provide.