The published data puts the average store's repeat purchase rate at about 25%, with most of the field between 10% and 40%. Here is where those numbers come from, why the median matters more than the average, and the three levers with receipts.
Repeat purchase rate is the retention metric I trust most for stores, because it contains no philosophy. No churn window to argue about, no definition of “active.” Just: of everyone who ever bought, how many bought twice?
The reason to care is concentration. In the Gorgias analysis that Omnisend cites, repeat customers made up 21% of a typical store’s customer base but produced 44% of its revenue and 46% of its orders. Roughly a fifth of the names on the list carrying half the business. Whatever moves that fifth moves everything.
The most granular public measurement of Shopify-store repeat behavior comes from Little Stream Software, the developer behind the Repeat Customer Insights app, which published a data study of stores on the platform. Three numbers from it are worth pinning above your dashboard:
Average repeat purchase rate: 25%. Median: 22.5%. The gap between them is the usual story: a minority of high-repeat stores pulls the average up, so the median is the honest midpoint. The study makes the same observation across its metrics, with medians routinely about a third below averages.
Standard deviation: 14.6%. Which means about two-thirds of measured stores land between roughly 10% and 40%. That band is the field. A supplement brand at 38% and a furniture store at 12% are both unremarkable once you account for what they sell.
Median orders per customer: 1.45. The bluntest number in the study. The typical store’s typical customer does not even reliably reach a second order, and one-time customers outnumber repeaters about 3 to 1. For historical context, the study notes Adobe’s older index used 27% as the share of one-time buyers who return, close to the same neighborhood.
Reference points from the Little Stream data study: median 22.5%, average 25%, typical band 10% to 40%. In Shopify admin the inputs come from Customers filtered by number of orders.
One measurement discipline before the levers: the lifetime rate is a lagging indicator, because every one-time buyer you have ever had stays in the denominator forever. When you change something, judge it on cohorts, the repeat rate of this month’s first-time buyers measured 90 days later, or the all-time number will hide a working experiment for quarters.
Lever one: the post-purchase flow. The window between order and delivery is the highest-attention moment a store gets: order follow-up emails open at 47.70% in Omnisend’s 2025 dataset. A buyer who just paid is deciding, mostly unconsciously, whether this store becomes a habit. The post-purchase flow is the cheapest tool aimed exactly at the 1.45-orders problem, because its entire purpose is manufacturing order number two.
Lever two: a standing reason to return. Rewards work on the second purchase specifically: in the survey data Omnisend cites, 57% of consumers say loyalty or referral rewards encourage them to buy again, and referred customers spend 11% more on their first order and 8% more on subsequent ones. Smile.io’s 2025 report adds the trend line: purchase frequency grew across every major ecommerce category in 2024, led by CPG at +13.95% year over year. Program design and its costs are covered in our loyalty ideas piece and the loyalty app pricing teardown.
Lever three: recover the drifters. Some share of your one-time buyers intended to return and simply did not. That is what win-back automation is for, and its published economics ($0.51 per email in Omnisend’s data) only need to beat zero, because these are customers you already paid to acquire. It is the weakest of the three levers per send and still free money at the margin.
Notice what is not on the list: acquisition-style discounting to past buyers. A blanket 20%-off blast to your whole file reaches the 44%-of-revenue crowd with a price cut most of them did not need. The levers above are targeted at the moment of decision instead; that is the entire difference between retention marketing and margin donation.
Core benchmarks (25% average, 22.5% median, 14.6% standard deviation, 1.45 median orders per customer, 3:1 one-time to repeat ratio, and the Adobe 27% comparison) are from Little Stream Software's published data study of Shopify stores; Little Stream is the vendor of the Repeat Customer Insights app and measured stores using its platform, so its sample skews toward stores that care about repeat metrics, if anything flattering the benchmark. Revenue concentration figures (21% of customers, 44% of revenue, 46% of orders) are from the Gorgias analysis as cited in Omnisend's repeat customers guide of January 9, 2024, which is also the source for the 57% rewards figure and referral spending lifts. Category frequency trends are from Smile.io's report of February 26, 2025 (100,000+ merchants). Automation economics in the levers section are from Omnisend's benchmarks article of May 12, 2026. Definitions vary between sources (repeat purchase rate vs. repeat customer rate vs. retention rate), so no figure was averaged across datasets.
The honest summary: the median store converts fewer than one buyer in four into a second order, and the published field mostly lives between 10% and 40%. Find where your product’s reorder physics says you should sit, measure by cohort, and put the three levers in place in order of attention: post-purchase first, a return reason second, win-back third. Then let the metric be slow, because it is supposed to be.
Customers who have placed two or more orders, divided by all customers, over a chosen window, times 100. It differs from retention rate, which needs a churn boundary; repeat purchase rate just asks whether a second order ever happened, which is why it is the cleaner metric for stores whose customers buy on irregular schedules.
Above the published median of 22.5% you are in the top half of the measured field; above 40% you are more than one standard deviation out and likely selling consumables or running subscriptions. Below 10% is normal only if your product is genuinely once-a-decade, like mattresses.
Because a minority of high-repeat stores drags the average up. Little Stream measured a 25% average against a 22.5% median and notes the same pattern across its other metrics, with medians often about a third lower than averages. Compare yourself to the median; the average includes businesses that do not resemble yours.
Slowly, because the metric is cumulative: every past one-time buyer stays in the denominator. Expect flow and loyalty changes to show up first in cohort views (repeat rate of this month's new customers at 90 days) long before the all-time number moves. Judge experiments on cohorts, not the lifetime rate.