What is a good repeat purchase rate for a DTC brand?

Why this one number decides profitability

Acquisition costs on Meta and Google have roughly doubled since 2021, and for most 7-8 figure brands the first order barely breaks even after CAC. The second order is where the margin lives: no acquisition cost, higher average order value, and a customer who already trusts you.

The economics compound from there. A brand whose repeat customers generate enough contribution margin to cover overheads and fund next month’s acquisition grows on its own cash; a brand that cannot is buying every month’s revenue from scratch, however impressive the top line looks. That is why Everboost treats repeat purchase rate as the number client programmes are judged on, and why the groundwork matters early: most of what determines whether a second order ever happens is set in the first 30 days after the first one, whether or not the order itself lands that soon.

The formula, and the three choices that change the answer

Repeat purchase rate = (customers with a 2nd order within the window ÷ customers acquired in the cohort) × 100

Three definitional choices swing the result by multiples, and most published benchmarks state none of them:

1. The window. 30, 60, 90 or 365 days from first order. The same brand can read 9% at 30 days, 18% at 90 and 28% at 12 months.
2. The population. New-customer cohorts only, or every customer who ordered in the period (which double-counts loyalists and flatters the number).
3. Maturity. Only count cohorts old enough to have finished the window. A customer acquired 20 days ago cannot fail a 90-day test they have not completed.

This is why the internet disagrees with itself. Industry reports built on 12-month data, such as Bluecore’s benchmark study, put the ecommerce average at 25 to 30%. Subscription-oriented frameworks measure month-one-to-month-two rebills and treat 60%+ as the healthy floor, numbers that would classify nearly every one-time-purchase brand as failing if applied literally. A 90-day new-customer cohort measure sits between the two. All three are internally consistent; quoting any of them without its definition is how founders end up panicking over a healthy number or celebrating a broken one.

The benchmarks, with their definitions attached

12-month window, all-customer basis (published benchmark data, primarily Bluecore’s customer growth report and Shopify’s averages):

Category	12-month repeat rate
Grocery and food delivery	40%+
Health and supplements	~29%
Beauty and cosmetics	21-26%
Fashion and apparel	20-26%
Home and furniture	~15%
Luxury goods	~10%
Ecommerce average	25-30% (Shopify stores ~27%)

90-day window, matured new-customer cohorts (Everboost’s working benchmarks for DTC health and wellness, the definition we run client programmes on):

Category	Poor	Typical	Strong
Supplements and consumables	Under 12%	15-22%	25%+
Skincare and haircare	Under 10%	12-18%	20%+
Food and beverage	Under 15%	18-25%	30%+
Devices and tools	Under 5%	6-10%	12%+

For calibration across both tables: most DTC brands sit between 0.3 and 1.5 repeat orders per customer per year, and the difference between the ends of that range is rarely the product. It is whether anyone built a deliberate path to the second order.

Check your own number

The calculator assumes a matured, new-customer, 90-day cohort. If your inputs came from a lifetime dashboard number, the verdict will mislead you, which is rather the article’s point.

How do you know if yours is actually bad?

PhycoHealth, an Australian marine-science supplement brand, came to Everboost with a 15.4% repeat rate and a 90-day median time to reorder. On paper, typical. In practice the product itself ran out in 30 to 45 days, so roughly 1 in 7 customers came back in a month while the rest quietly lapsed between empty tub and reorder prompt.

That is the diagnostic that matters more than any benchmark: compare your median time to second order against your product’s natural usage cycle. If customers finish the product in 45 days but reorder at day 90, the problem is not demand. Nobody asked for the order at the right moment. And even where the reorder itself sits months out, the groundwork for it, onboarding, education, the first well-timed nudge, happens in the first month whether or not the order does.

What actually moves repeat purchase rate?

Across Everboost client programmes, these are the levers that move the number reliably, starting with the three that earn their place in every engagement:

1. Time the ask to the usage cycle. A replenishment flow that lands as the tub runs low beats any discount. Discount Supplements cut median time to second order from 63 to 43 days this way, lifting the matured 60-day repeat rate by 12.2%.
2. Sell the routine, not the refill. In one client’s data, 62% of second orders were a different product: customers were building a routine, not replenishing. Cross-sell journeys built on that insight outperform buy-it-again emails.
3. Fix deliverability before touching creative. If a fifth of email lands in spam, no copy improvement matters. One grooming brand moved sender score from 59 to 80 and spam placement from 20.8% to 8.9% before campaign volume did anything.
4. Design the front-end offer to engineer the second order. A 50%-off first order that reverts to 10% from order two anchors customers to a price you will never repeat, and the dropoff shows it. A bundle built from complementary, routine-enhancing products starts the habit the second order depends on.
5. Use free gifts to shorten time to value. A first-order gift chosen to speed up results or embed the routine, a dosing aid, a companion mini, a habit-forming accessory, moves the second order in a way a discount of the same cost never does.
6. Pair fast and slow consumption in the basket. One fixed reality first: usage speed is one of the most fixed behaviours in ecommerce, and you will not email someone into emptying a jar faster. If the hero product lasts 100 days, no replenishment flow produces 60-day reorders. Putting a 30-day consumable alongside it in the first order changes what the customer owns, so the fast product sets the reorder rhythm and the slow one rides along.
7. Match subscription frequency to real consumption. Shipping every 30 days to someone who consumes in 45 does not accelerate reorders; it manufactures “too much product” cancellations, among the most common and most avoidable in subscription data. Smaller first-order sizes can also shorten the reorder clock, at an honest cost to first-order AOV worth pricing deliberately.
8. Collect zero-party data you actually use. Asking preferences at sign-up is common; feeding the answers into genuinely tailored post-purchase education and recommendations is rare, and it is the difference between personalisation as a survey and personalisation as a reason to come back.
9. Anchor subscription retention on loss aversion. A gifting ladder across the early months, with milestone rewards and a mystery gift sitting in next month’s order, has subscribers second-guessing the cancellation because leaving means visibly giving something up.

The measurement trap almost everyone falls into

Growing brands routinely conclude retention is worsening when it is not. Acquisition spikes flood the base with people who have not had time to reorder, so the blended rate falls even while every cohort improves. When acquisition slows, the same arithmetic flatters a programme that did nothing. Judge the metric on matured, like-for-like cohorts, treat any movement during a big acquisition swing as suspect until cohorts confirm it, and remember that the attributed revenue number in your ESP is a different measurement answering a different question.

A good repeat purchase rate, in the end, is one measured honestly, benchmarked against its own definition, and improving cohort over cohort. The brands that get this right stop asking whether their number beats an average and start asking whether this month’s cohort came back faster than last month’s, which is the version of the question that compounds.