Repeat Purchase Rate Benchmarks: How to Set Targets That Mean Something for Your Store

You looked up a repeat purchase rate benchmark, found a number like 28% or 35%, and now you are trying to work out whether your store's 19% is a problem or just a different way of counting. Most of the time it is the counting. This article shows why published benchmarks rarely compare to your number, gives broad typical ranges by category as a sanity check, and walks through building the internal benchmark from your order export that you can actually manage against.

Key takeaways

• Repeat purchase rate is the share of customers who place a second order. Published benchmarks hide different time windows, denominators and product mixes, so direct comparison is usually meaningless.

• The honest benchmark is internal: compare product families inside your own store on the same definition, then use external ranges only as a rough check.

• Three things make numbers incomparable across brands: the window (90 days, 12 months, lifetime), the denominator (all customers or a cohort of first-time buyers), and the product mix.

• What moves the number is concrete: the product that opened the account, the gap to the second order, the acquisition channel, and whether a reorder reminder exists.

What repeat purchase rate measures, and why the benchmark feels off

You see "30% repeat purchase rate over 12 months" and your store sits at 18%. The first instinct is that something is broken. But you may be comparing a 90-day window to a 12-month one, or a supplement brand to your mixed catalogue of apparel and skincare. The benchmark is not wrong. It just does not describe your situation.

Repeat purchase rate is the percentage of customers who buy more than once. It answers one question: of everyone who bought from you, how many came back? For a store selling replenishable products it is the retention number that matters most, because the transition from first order to second is where lifetime value either starts to compound or does not.

It is different from customer retention rate, which tracks whether customers stay active across periods, and from customer lifetime value, which cares about total spend. It is also different from purchase frequency, which measures how often the returning customers come back. The purchase frequency guide covers that distinction.

Why repeat purchase rate benchmarks are hard to compare

External numbers rarely match yours because three assumptions differ.

• Time window. Some benchmarks quote a 90-day rate, some 365 days, some lifetime. A store can show 20% at 90 days and 34% at 12 months, and both are true. A benchmark without a stated window tells you nothing.

• Denominator. "All customers with at least one order in the last 12 months" is a different pool from "customers whose first order was in January". The first is inflated by returning customers from earlier years. The second isolates first-time buyers and gives a clean read on whether acquisition and the post-purchase experience are working.

• Product mix. A brand where most revenue is 30-day supplements will show a far higher blended rate than one where most revenue is seasonal apparel. A portfolio-level number hides that.

Do not compare your rate to an industry average unless the window, the denominator and the category match. Internal benchmarks by first product family, on consistent cohorts, are more honest and more useful.

Pin down the definitions before you calculate

• First purchase: the earliest completed order for a customer id or email, after merging duplicate records and dropping test orders.

• Repeat purchase: any later order on a separate date. Two orders in one checkout session count as one.

• Repeat customers: customers with at least two qualifying orders.

• Repeat purchase rate: repeat customers divided by total customers in the chosen window, times 100. The cohort version below uses first-time buyers as the denominator instead.

Typical ranges by category, as a sanity check only

These are broad bands for a 12-month window, of the kind you will see quoted in public discussion of e-commerce retention. They are not measurements of any specific dataset, definitions differ from source to source, and the honest use for them is a rough check on your own internal numbers, not a target.

• Supplements and vitamins: roughly 30 to 45%, higher with an active subscription or reorder flow.

• Coffee and tea: roughly 25 to 40%, pushed up by flavour loyalty and subscriptions.

• Pet food and treats: roughly 30 to 45%, depending on price point and auto-replenishment.

• Skincare and beauty: roughly 25 to 40% for routine-based lines. One-off colour cosmetics pull the range down.

• Cleaning and household: roughly 20 to 35%.

• Baby (nappies, formula, wipes): roughly 30 to 50% while the child is in a given stage.

• Apparel and fashion: roughly 20 to 30%, compressed by seasonality and gifting.

• Gifts, décor and durables: roughly 10 to 20%, because customers do not need the same item again soon.

Blended across e-commerce, the figure most often quoted sits somewhere in the 20 to 30% range. Compare each product family inside your own catalogue against the band for that family before reacting to any single blended number.

How to calculate your own benchmark from an order export

Columns you need: customer email or id, order id, order date, product SKU, product type or category, and order value.

Steps:

1. For each customer, find the earliest order date. That is the first purchase.

2. Group customers by the month of their first purchase. Every customer whose first order fell in March forms the March cohort. Count each cohort.

3. For each cohort, count how many placed at least one more order within 90 days of the first. Divide by the cohort size. That is the 90-day repeat purchase rate.

4. Repeat with a 365-day window. Now you have both figures side by side for each month.

5. Split each cohort by the product family of the first order and calculate the rate separately for each.

Merchants can stop at store-wide cohorts. Agencies will usually want the same split by acquisition channel if the export carries it. There is a longer walkthrough of the cohort method in the repeat purchase rate guide.

Affinsy computes these cohort rates, and the second-order window behind them, from an uploaded order history. The steps above work in any spreadsheet.

Worked example: benchmarks inside one store

An illustration with round numbers, not real data.

1,000 new customers placed their first order in January. Within 90 days, 260 placed a second order: 26%. Within 12 months, 340 had: 34%.

Split by first product family:

Inside this store, supplements are the retention engine. Apparel brings in customers who rarely return. Whether 18% is bad depends on the rest of the catalogue and on the margin, not on an external 30%. The useful question is not "is 18% bad?" but "should acquisition shift toward the products that actually create repeat buyers?"

What moves the rate between the first order and the second

• The opener product. The first product sets expectations on quality and value, and some products bring people in cheaply but rarely lead anywhere. Others anchor long customer lifetimes. Market basket analysis across orders shows which products play which role in your own data.

• The gap to the second order. For consumables the natural reorder cycle, 30, 45 or 60 days, should set when the reminder goes out. Too early and it is ignored. Too late and the customer has restocked elsewhere. The replenishment email guide covers how to find the window per product.

• Acquisition channel. Customers won with a deep discount on paid social often repeat less than customers from search or referral. Benchmark within each channel as well.

• Post-purchase communication. A delivery check-in, usage tips for the product they bought, then one single-product reorder email timed to when it runs out. That beats a generic "come back soon".

• The feedback loop. A short review request 10 to 20 days after delivery surfaces the problems that, once fixed, lift the rate more than any campaign.

• Loyalty mechanics. Early access to restocks or a small reward for the second order can help, but they work on top of the timing and the opener product, not instead of them.

How agencies should present benchmarks to clients

• State the exact definition on the slide. "90-day repeat purchase rate among customers whose first order was in the month shown."

• Pair external with internal. One broad typical band for the client's main category next to the client's own last six to twelve cohorts.

• Break out by product family. A small table by first product category often shows that the brand is fine and the problem is customers who start with a clearance SKU or a deep-discount bundle.

• End with actions tied to the data. "Shift paid spend toward product A, which produces twice the 90-day rate of product B." "Extend the coffee reorder flow to day 60."

• Avoid vanity comparisons. Never claim "above industry benchmark" without matching definitions. "Within the expected band for this product type, with room to improve the opener and the timing" is honest and still useful.

From benchmark to roadmap

• Prioritise days 15 to 90 after the first order, aligned to the product's consumption time.

• Re-merchandise the first purchase. Find the first products with the lowest 90-day rate and test different sizes, bundles or price points so more customers start on items that historically create returning customers.

• Tune reminder timing per product family. A 30-day vitamin, a two-week coffee bag and a 60-day serum each get their own window and their own email.

• Layer segmentation afterwards. Once the basic flows work, RFM-style groups (recent, at risk, hibernating) let the message match the stage, not just the date of the first purchase.

• Recalculate quarterly. Compare new cohorts with earlier ones and with the category band, and bring it to the review as a trend rather than a single number.

If you would rather have the cohort table and the product-family split worked out for you, the 48-hour analysis returns both for your store.

FAQ

What is a good repeat purchase rate if I sell several categories?

There is no single number. Build a separate benchmark per major product family and weight by revenue share. If the consumable families sit inside their bands and outperform the durables and gifts, you are on the right track. The blended rate will always be a mix.

How often should I review it?

Refresh monthly, review properly each quarter. A 90-day rate is only meaningful on cohorts at least three months old, and a 12-month rate needs a full year. Agencies can show early 30 and 60-day signals monthly while the longer numbers mature.

Should retention rate replace repeat purchase rate as the main KPI?

They answer different questions. For replenishable brands, "second order within 90 days" works well as the primary KPI for first-purchase cohorts, with retention rate and lifetime value for the long view.

How do subscriptions affect the benchmark?

They compress the repeat decision into the first checkout, so they lift the rate. Calculate it both including and excluding subscription orders. Strong subscription uptake with weak one-time repeat behaviour means the offer is good and the wider experience needs work.

Can I benchmark if I mostly sell durables?

Yes, but expect the low bands and read them differently. Look at repeats on accessories, refills and adjacent categories over a longer window, and put more weight on acquisition efficiency and first-order profit.