Ecommerce return rate: what is normal?

In short. There is no official European return figure: the two most solid sources point to roughly one purchase in four coming back (24.1% of parcels in Germany, 23.7% in Spain outside food), but those figures are not comparable with yours if you measure with a different denominator. Before you try to bring it down, split it in two — legal withdrawal and commercial return: with both you can prevent the causes before the purchase, but only in the commercial one can you change the rules.

The ecommerce return rate is the most quoted and worst compared metric in online retail. We have all read that "the sector is at 30%", nobody says what that 30% is a percentage of, and expensive decisions get made along the way: switching logistics provider, tightening a policy or — most dangerous of all — adding friction to a right that European law protects. This guide explains what the sources you can actually cite say, why your number is not comparable with any report, what each return really costs you, and which reduction tactics are legal in the European Union and which can earn you a penalty. It is a satellite of our complete guide to Shopify returns.

What is the return rate and how is it calculated?

The return rate is the share of what you sell that ends up coming back, and it takes three different calculations that give three different numbers. None of them is the correct one: there is one that fits each decision. The expensive mistake is mixing them.

The difference is not cosmetic. A shop selling baskets of three or four items can have a rate of 22% by orders and 14% by units, simply because most returns bring one item back and keep the rest. And the revenue rate almost always sits below the other two, because what gets returned most is rarely the most expensive thing in the catalogue (these figures are an illustrative example, not market data).

There is a fourth trap, and it is about timing: returns arrive weeks after the sale. If you divide August's returns by August's sales while you are growing, your rate will look better than it is, because the numerator belongs to a smaller month of sales. The clean way to measure it is by cohorts: take one month's orders and count how many have come back 90 days later. It is slower, and it is the only number that survives a meeting.

What return rate is normal in 2026?

As far as we know, no European institution publishes a return rate that is comparable across countries. Neither Eurostat nor the Commission maintains a series for this indicator, so every figure in circulation comes from academic research or from industry vendor reports. That does not invalidate them — some are excellent — but it does mean citing the source, the date and the denominator every single time. These are the ones we use:

Three practical readings of that table:

1. The European order of magnitude is "one in four", in the two large markets that are actually measured with rigour. Germany and Spain land in almost the same place by different routes.
2. Fashion is not just another category: it is almost the whole of returns. If you sell clothing or footwear, comparing yourself to "the ecommerce average" tells you nothing, because you are the one generating that average.
3. The sector split in the Spanish report is not a set of return rates. Clothing being 31% does not mean 31% of clothing is returned: it means 31% of the value returned in Spain is clothing. That is exactly the kind of confusion that turns a good data point into a bad decision.

Why is your rate not comparable with the report you just read?

Because you almost never share a denominator, a catalogue or a market. Before you conclude that you are doing badly, check these six differences:

• A different denominator. Orders, units or revenue, as we have just seen. It is the first question to ask any figure.
• Catalogue mix. 20% in an accessories shop and 20% in a dress shop are two opposite realities.
• Market mix. Selling into Germany raises the average; selling into markets with less of a returning habit lowers it.
• Cancellations counted as returns. An order cancelled before it left the warehouse is not a return: there is no reverse logistics and nobody touched the product. If they share a bucket, your rate is inflated.
• Time window. Calendar month versus 90-day cohort, again.
• Returns you never record. The ones settled by email, by WhatsApp or over the counter that never enter the system. This is the biggest bias and the least discussed.

And here is why this year deserves a closer look at your series: from 19 June 2026, Directive (EU) 2023/2673 requires shops that direct their activity at EU consumers to offer a clearly identified withdrawal function for the contracts where that right exists. Many requests that used to arrive through informal channels will start coming through a button that does leave a record. If your measured rate jumps in the second half of the year, the first hypothesis should not be that your customers have changed their behaviour: it is that you are finally counting them. With one honest caveat: part of the rise may be real, because an accessible channel also lets people exercise the right who used to give up in the face of friction — friction that, incidentally, was never lawful.

Legal withdrawal and commercial return: two rates, not one

With both you can prevent the causes, but only in one can you change the rules, so measuring them together stops you knowing whether you are improving. We develop the difference in withdrawal vs. return vs. exchange, but in metric terms it comes down to this:

The operational consequence is direct: with withdrawal you can work on its causes before the purchase, never on its exercise; the goal is to prevent it and, when it does come, to make it cheap and frictionless; the goal with a commercial return can indeed be to reduce it or redirect it into an exchange. If you mix them in the same indicator, any improvement in the second gets buried under the noise of the first. The detail of which products fall outside the right is in exceptions to the right of withdrawal, and who bears the transport, in who pays return shipping.

What does a return really cost you?

Almost always more than the spreadsheet says, because half the costs never come with an invoice. Rather than giving you an average figure that will look nothing like yours, here are the components to add up to work out your own:

Multiply the result by your number of returns and compare it with your gross margin: that ratio is the metric that really decides whether you have a problem. A 30% rate at three euros per return is a healthy business; a 12% rate at twenty euros a time is not.

One point almost no guide mentions and worth keeping on the radar: Regulation (EU) 2024/1781 (ESPR) bans the destruction of certain unsold consumer products — clothing, accessories and footwear listed in Annex VII — and its definition of an unsold product includes goods returned by the consumer within the withdrawal period. That said, the ban applies to large companies from 19 July 2026 and to medium-sized companies from 19 July 2030, and micro and small companies are exempt. If your shop is small, it does not bind you today; it does show which way the wind is blowing, and it already affects the big brands you compete with.

How to reduce your return rate without breaking the law

This is where most "tactics to reduce returns" lists become dangerous in Europe, because they are written with the United States in mind, where the right of withdrawal does not exist as such. This table separates what you can do from what you cannot:

Let us pause on the second-to-last row, because it is the most tempting one. The German Federal Court of Justice (BGH) held on 16 March 2016 in case VIII ZR 146/15 that the law does not make the exercise of withdrawal conditional on the consumer having a legitimate interest: they may withdraw for whatever reason they like, including having found the product cheaper somewhere else, and excluding the right is only possible in exceptional situations, such as fraud. Translated: within the legal period there is no such thing as "a customer who returns too much". You may decide who you sell to in the future — that is freedom of contract, with its own limits — but you cannot deny withdrawal from a contract already concluded. If you are thinking of systematising anything along those lines, talk to a lawyer first.

Seven levers that do work (and are legal)

1. Product pages with no ambiguity. The real measurements of the item, not of the size label; fabric and stretch; photos on several bodies and in neutral light. Most avoidable returns are decided here, before checkout.
2. Your own size guide, not the supplier's. If your customers write "runs small" in the reviews, that information is worth more than the manufacturer's chart.
3. Reviews with context. Height, usual size and size bought. It is fit data your customers generate for free.
4. A structured return reason, compulsory to analyse (even if optional to fill in). Without a taxonomy of reasons there is no diagnosis: "did not like it" and "arrived damaged" call for opposite solutions.
5. Chase the outlier products, not the average. Sort your catalogue by return rate per unit and work on the worst 5%. It usually concentrates a disproportionate share of the cost.
6. Offer the exchange before the refund, with the same accessibility. An exchange keeps the revenue without touching any right, as long as the refund stays one click away and carries the same visual weight.
7. Shorten the return cycle. The sooner the item is back in sellable stock, the less you lose. There is more money here than in arguing about the percentage.

Notice that none of the seven adds friction for a customer who has already decided to return. That is the acid test: if a tactic only works because someone gives up halfway through, it is not return reduction, it is a dark pattern under another name.

How to measure your return rate in Shopify

Shopify does not show a "return rate" as a headline metric: you have to build it. With these elements you have enough for a decent series:

1. Analytics → Reports, the "Orders and reversals by product" report: units sold and units removed from the order, by product. Mind the word "reversal", because it mixes things up: it covers returns, refunds, cancellations and order edits alike. It is your starting point, not your numerator — for the rate by units you have to keep only the physical returns, cross-checking it against the returns created on each order, and drop whatever never left the warehouse.
2. Return reasons. Shopify records a reason per order line when you create the return from the admin. Always use it, with a short taxonomy, and review it every month.
3. Refund reports for the rate by revenue, with the same filter: keep only merchandise refunds tied to a physical return — leave out pre-fulfilment cancellations, price adjustments, shipping and goodwill — and attribute them to the month of the original order, not the month you paid them out.
4. Cohorts by hand. Export one month's orders and the returns associated with them at 90 days. It is tedious once a quarter, and it is the number that really counts.

The availability and detail of the reports depend on your Shopify plan, so check what you can see in your own admin before designing the dashboard. If you would rather have the whole flow produce clean data at source, how to manage a return in Shopify step by step explains the full process.

At returnEasier the separation lives where it actually matters: in the record. The withdrawal button is on every plan, including the free one, and every request is sealed with its date, its time and its type — legal withdrawal or commercial return — plus the reason whenever the customer chooses to give one. The returns analytics dashboard on the Scale plan, with its configurable alerts for when a product spikes, gives you the aggregate series, not a rate per type; the split by type comes today from the auditable export, which carries the type of every request and is on every plan.

Common mistakes

• Comparing your number with a headline without checking the denominator. This is the mistake that causes almost all the others.
• Putting cancellations in the same bucket. An order that never shipped has no reverse logistics.
• Measuring by calendar month while you grow. It flatters the result and hides the problem.
• Setting a rate target without looking at the unit cost. A cheap 30% is a better business than an expensive 12%.
• Treating legal withdrawal as an indicator to minimise. It is a right: you handle it efficiently, you do not fight it.
• Copying reduction tactics written for the United States. Restocking fees and short periods do not transfer to the EU.
• Not recording informal returns. If they get settled on WhatsApp and never enter the system, your series measures nothing.

Frequently asked questions

What return rate is normal? Roughly one purchase in four, according to the two most solid European sources: 24.1% of B2C parcels in Germany (University of Bamberg) and 23.7% of non-food purchases in Spain (ZigZag and Retail Economics). In fashion it is considerably higher; in electronics and groceries, considerably lower.

How is it calculated? What comes back divided by what you sold, choosing a denominator: orders, units or revenue. All three are valid and give different numbers for the same shop.

Can I refuse a customer who returns too much? Not within the legal period: the BGH held in case VIII ZR 146/15 (16 March 2016) that the motives are irrelevant and that exclusion is only possible in exceptional cases, such as fraud.

Will my rate go up with the withdrawal button? Mostly the one you measure, because from 19 June 2026 requests come through a channel that leaves a record. Part of the rise may be real — a button with no login lets people exercise the right who used to give up — and that friction was never lawful. What you can reduce are the causes, before the purchase.

Can I charge a handling fee? Not in a withdrawal: Article 14(1) of Directive 2011/83/EU only covers the direct cost of the return, and only if you informed beforehand. The diminished value under Art. 14(2) is a different route.

And in fashion, what target should I set? None taken from the sector. Measure your own series by cohorts, sort the catalogue by rate per unit and work on the worst 5%: that moves the needle, a round number does not.

Conclusion

If you sell on Shopify to European consumers, the most profitable thing you can do with your return rate this week is not to lower it: it is to stop measuring it badly. Pick a denominator, split it into legal withdrawal and commercial return, measure by 90-day cohorts and work out your real cost per return. With those four decisions you will already know whether you have a problem and where it is.

And when you move on to reducing it, do it ahead of the purchase — photos, measurements, sizes, expectations — never behind it. In Europe, every bit of friction you add after "I want to return this" is a legal risk dressed up as a margin improvement.

💡 Ready to comply effortlessly? returnEasier installs the compliant withdrawal button on your Shopify shop in minutes, with every request recorded, sealed and kept separate from your commercial returns. Try it free — 3 trial returns, no card.

---

Official sources

• Directive 2011/83/EU on consumer rights (arts. 13 and 14) — EUR-Lex
• Directive (EU) 2023/2673 — EUR-Lex
• Regulation (EU) 2024/1781 (ESPR), art. 25 and Annex VII — EUR-Lex
• BGH, judgment of 16 March 2016, VIII ZR 146/15 — Bundesgerichtshof
• Forschungsgruppe Retourenmanagement, University of Bamberg — record of 550 million return parcels in 2025
• Informe Benchmark Anual de Devoluciones España 2025 (ZigZag and Retail Economics) — summary in Marketing4eCommerce

Informational content; not legal advice. For specific cases, consult a lawyer specialised in consumer law.