Serial returners: the 11% who drive a quarter of your returns

Most merchants track their return rate as a single number. It hides something important: your returns are not spread evenly across your customer base. A small, identifiable group of serial returners is generating a wildly disproportionate share of them, and unless you look at returns per customer rather than per order, you will never see it.

The numbers are stark. Research by ZigZag and Retail Economics, reported by The Guardian, found that serial returners make up just 11% of online shoppers who return anything, yet they generate roughly a quarter of all online non-food returns. Each one sends back around £1,400 of merchandise a year, which Blue Yonder puts at close to $1,500 per person, per year. Widen the lens slightly and the concentration gets worse: serial and slow returners together are only 22% of returners but produce 45.5% of returns.

If that ratio holds even loosely in your store, then roughly one customer in ten is setting your reverse logistics costs, your restocking labour, and a meaningful chunk of your margin. This post is about finding those customers in your own data and deciding what, if anything, to do about them.

Serial returners are an abuse problem, not a fraud problem

This distinction gets blurred constantly in returns coverage, and getting it wrong leads merchants to treat ordinary customers like criminals. It is worth being precise.

Return fraud is illegal. It covers empty-box returns, fake receipts, returning stolen merchandise, and counterfeit swaps. It has criminal intent behind it and it is the subject of our guide to the most common return fraud tactics.

Return abuse is legal. It is excessive but genuine returning: real orders, real products going back, real refunds owed. Nobody is lying. The customer is simply using your policy far more heavily than the pricing of your products assumes. Serial returning sits squarely here.

The Appriss Retail 2026 Total Retail Loss Benchmark Report puts hard dollars on the split. Of the $706 billion in US returns during 2025, $100 billion (14.2%) was preventable loss. Within that, abuse accounted for 12% of all returns, worth $86 billion, while outright fraud accounted for 2% of returns, worth $14 billion. Abuse is roughly six times larger than fraud, and Appriss notes it is harder to catch precisely because "it looks like normal customer behavior."

The report also makes a point most merchants underrate: most people doing this have no idea they are doing anything wrong. They are operating inside what they perceive as acceptable boundaries. That should shape your response. You are not building a case against a fraudster; you are correcting an expectation you accidentally set.

A serial returner is not just a bracketer

Bracketing, buying three sizes to try at home and sending two back, is a single-order behaviour. It is widespread, it is largely rational given inconsistent apparel sizing, and it is best solved with better sizing information and an exchange-first flow. We covered that separately in our post on turning bracketed multi-item returns into exchanges.

Serial returning is a relationship-level pattern. It is the same customer, quarter after quarter, returning the majority of what they buy across many separate orders. A bracketer who orders three sizes and keeps one has a 67% return rate on that order and may still be a perfectly profitable customer over a year. A serial returner has a sustained return rate across their entire history, and the cumulative processing cost outruns the margin they leave behind.

The overlap is real (habitual bracketing is one route into serial returning) but they are not the same population, and a policy built to catch one will misfire on the other. Treat single-order bracketing as a merchandising problem. Treat sustained multi-order return rates as a customer-economics problem.

One more caution before you go hunting. Appriss found that the top 1% of customers can generate up to 50% of total sales while returning 8% less than the average shopper, and that blanket policies designed to catch bad behaviour disproportionately trip up high-volume good customers, simply because they transact more often. Volume of returns is not the signal. Rate of returns relative to spend is.

How to spot serial returners in your Shopify data

Shopify does not give you a per-customer return-rate metric out of the box, and neither does Exchange It. There is no automated serial-returner score, no AI risk model, and no "flag this customer" button in the app. What you do have is enough raw material to do this manually, once a quarter, in about an hour.

Here is a practical method:

Export your return history. On Exchange It's Advanced plan, data export gives you the underlying return records rather than just the dashboard summary.
Group by customer email, not by order. This is the whole trick. Sum lifetime order value and lifetime returned value for each customer, then compute returned value divided by ordered value.
Set a floor of at least three orders. A customer with one order and one return is a data point, not a pattern. ASOS uses three orders as its own threshold, which is a reasonable place to start.
Sort by return rate, then look at the top 5%. Compare their rate against your store average. In apparel, a store average around 25% to 30% is unremarkable; a customer sitting at 70% or higher across many orders is a different animal.
Cross-reference the reasons. Exchange It's Advanced analytics include a Return Reasons breakdown, return volume over time, and a Top 5 Return Variant view. If your heaviest returners are all citing fit on the same three SKUs, you have a product problem, not a customer problem. Fix the size chart before you touch the policy.

That last step matters more than anything else on the list. A high individual return rate can be the honest consequence of a bad size guide, a misleading product photo, or a colour that renders differently on every screen. Genuinely faulty and incorrect items should never count toward anyone's return rate, and every retailer that has published a fair-use policy has carved them out explicitly.

What ASOS and PrettyLittleThing learned the hard way

Two UK fashion retailers have run this experiment in public, and the contrast between them is instructive.

ASOS took the transparent, graduated route. After introducing a £3.95 returns fee in September 2024 and closing the accounts of a small group of persistently high returners in June 2025, ASOS rolled out a personal return-rate tool in January 2026. According to Drapers, customers can now see their own return rate in the app alongside tips for reducing it. The mechanics, as summarised by law firm Fox Williams: customers with a return rate of 70% or more across three or more orders in a rolling 12 months pay the £3.95 fee unless they keep more than £40 of the order; above 80% across five or more orders, a second £3.95 handling fee applies. Under 70%, returns stay free. Faulty and incorrect items remain free to return for everyone.

Three design choices make that work. The rule is published, the customer can see their own number, and improving the number removes the fee. It is a nudge with an exit, not a punishment.

PrettyLittleThing took the abrupt route and reversed course. PLT scrapped free returns in June 2024, introducing a £1.99 fee, and began deactivating accounts over high return rates, including for members of its paid loyalty programme. The backlash was severe enough that the company reinstated free returns for loyalty customers within months.

Research from Iowa State University suggests why the framing mattered so much. In their study, when a return policy change was presented as generalised, 64% of participants expressed negative emotions and 42% said they would shop elsewhere. When the same change was presented as targeted at people abusing the policy, only one participant said they would switch retailers. Same restriction, radically different reception, decided entirely by how it was explained.

One legal note if you sell into the UK or EU: Fox Williams flags open questions about whether behaviour-based return fees sit comfortably with the Consumer Contracts Regulations, particularly where statutory cancellation rights are involved. Get advice before you build a fee schedule around customer scoring.

Five levers you can pull without banning anyone

You almost certainly do not need account bans. Appriss found that a graduated "warn and approve" tier, warning high-risk customers before ever declining them, reduced abusive returns by 90% without measurable damage to loyalty. Friction and information do most of the work. Here is what is actually available to a Shopify merchant today.

1. Tighten the window, selectively

Long return windows are generous to everyone and expensive because of a few. Exchange It's return rules support configurable time-window eligibility, so requests outside your window are filtered automatically rather than landing in your inbox for a judgement call. Shorter windows also mean items come back while they are still sellable at full price, which is where most of the recovered margin lives.

2. Exclude the categories that get abused

Non-returnable item filtering and market-based product filtering let you carve out specific products or regions from your standard return terms. Occasion wear, final-sale lines, and heavily discounted stock are the usual candidates. Wardrobing concentrates in exactly these categories, and excluding them is a rule about products, which nobody experiences as a personal accusation.

3. Charge a restocking fee instead of banning

A percentage-based restocking fee, configurable per return line item, recovers processing cost on the returns that actually happen without denying anyone service. Appriss estimates returns processing costs around 30% of an item's retail value, so even a modest fee closes a real gap. We wrote a full breakdown of how to set restocking fees without damaging conversion.

4. Make exchanges and store credit the easiest path

This is the highest-leverage move and the least confrontational. A return that becomes an exchange keeps the revenue and often costs less to process than a refund. Serial returners are unusually policy-aware: the ZigZag Annual Returns Benchmark found they are the segment most open to paying an upfront fee for better returns terms, which suggests they respond to incentives rather than simply resenting them. Put the exchange option first in the portal and make store credit visibly more attractive than a refund.

5. Publish the rule before you enforce it

Every retailer that has handled this well has done the same thing: written it down, made it visible, and given the customer a way to improve. Your returns and exchange policy is where a fair-use clause belongs. A single sentence noting that you reserve the right to apply fees or restrictions to accounts with persistently unusual return activity does most of the deterrent work on its own, and it protects you if you ever need to act.

What you should not do is act on a hunch. The NRF and Happy Returns 2025 Retail Returns Landscape found that 82% of consumers treat free returns as an important factor when choosing where to shop. The cost of wrongly restricting a good customer is a lost relationship. The cost of tolerating a genuine serial returner for one more quarter is a few hundred dollars. Measure first.

Exchange It gives you the practical pieces for that measurement and response: configurable return rules on the Standard plan at $9.99 a month, and return analytics plus data export on Advanced at $19.99 a month, both with a 7-day free trial. There is no automated serial-returner detection in the app, and we would rather say so plainly than sell you a scoring model you would need to second-guess anyway. The work of identifying chronic returners is manual, quarterly, and worth doing. Install Exchange It on your Shopify store and start with the exports.

Serial Returners on Shopify: Spot Them and Respond