Exchanges instead of refunds: retain revenue, no dark patterns

In brief. An exchange instead of a refund retains the full returned amount, sometimes more, but you only win it when the customer genuinely prefers it: the stock exists, shipping costs them nothing and the replacement arrives soon. Pushing it through design — preselecting it, hiding the refund, charging a fee only if they ask for money — is not optimisation: on a withdrawal it is unlawful, and outside one it costs you the customer.

Almost every piece written about exchanges instead of returns comes from the United States and repeats the same recipe: make the portal exchange-first, put the exchange first and the refund behind it. In a European shop that recipe has a structural problem, because the refund is not an item on your menu: it is a right when the customer withdraws. This article explains how to lift your exchange rate without crossing that line, which kinds of exchange exist and which ones convert, and exactly where a legitimate incentive turns into a dark pattern. It is a satellite of our pillar guide on turning returns into retained revenue.

Why does an exchange retain more than any other alternative?

Because it is the only resolution where the money neither leaves nor sits pending: it turns straight into another product. An exchange for another variant at the same price retains 100% of the returned amount. An exchange for a more expensive item beats that, because the customer pays the difference. And unlike store credit, it creates no liability waiting to be redeemed.

That is the comparison that settles the decision:

That is why the exchange comes first in the order of levers: it is the most profitable and also the cheapest to set up. Its limit is not strategic, it is inventory. If you do not have the variant, the exchange becomes a refund with extra steps, which is worse than a plain refund.

Mind the denominator, which is the part most often inflated: what you retain is measured against the returned amount, not the order total. On a partial return from a multi-line order, swapping one size retains the value of that line; the rest of the order was never at stake.

And one measurement point is worth fixing before we go on: issued store credit is not retained revenue, it is a liability. Only what the customer actually spends counts. That calculation, with the formula, is in the pillar guide for this cluster.

The legal nuance almost nobody mentions: an exchange does not settle a withdrawal

A withdrawal is not «settled» with an exchange, because it extinguishes the obligations to perform the contract. Article 12 of Directive 2011/83/EU governs the effects of withdrawal: the parties are released from their obligations to perform the distance contract. Whatever comes next — another size, another colour, another product — is legally a new purchase, not an amendment of the old one.

Three practical consequences follow, and no American guide mentions them:

1. The refund clock does not stop because you shipped the replacement. Article 13(1) requires you to return the money without undue delay and, at the latest, within 14 days of their withdrawal notice, using the same means of payment they used. Mind that order, because it gets read backwards all the time: the 14 days are the backstop, not a grace period; if the goods are already back on day 2, the refund is due then. And it still holds unless they expressly agree to another means of payment and incur no fees. Each Member State has transposed this into its own consumer law, so check the national text for the markets you sell to. An exchange you set in motion is not on its own that express agreement: if the customer has not accepted the exchange in place of the money, you still owe the refund even after shipping the new size. If they did expressly accept it, on the record and at no cost to them, the amount of the goods is settled by the exchange and you do not owe that cash on top — but the burden of evidencing the agreement is yours, and silence, a preticked box or a clause in your terms do not count as one. Mind what that agreement covers: Article 13(1) requires you to reimburse all payments received, including delivery costs of the original order (the least expensive standard type; not the premium the customer chose on top). If they withdrew from the whole order and accept a replacement for the goods, that does not by itself settle the shipping they paid: either the agreement expressly covers it, or you refund it. We develop it in who pays return shipping in the EU. With one qualification that does protect you: you may withhold the reimbursement until you receive the goods back or the customer supplies evidence of having sent them, whichever comes first, unless you offered to collect the goods yourself (Article 13(3)). That lever is there to wait for the goods, not to wait for the exchange to complete, and withholding does not restart the clock: the 14 days run from the moment the withdrawal is communicated to you. We develop it in when the 14-day period starts.
2. «Exchange» and «withdrawal» are not alternative buckets, and the test is not what the customer asks for. Article 11 only requires an unequivocal statement of the decision to withdraw: they need not invoke the law or use the word «withdrawal». So look at the statement, not at the replacement. «Could you swap it for an L?» declares nothing: it is a commercial request. «I am cancelling the purchase, refund me» does declare it. And there is a third case, the most common and the worst handled: «I am returning the order and I want an L», which is unequivocal in neither direction — it can be a withdrawal followed by a new purchase, or an exchange in which the return is merely the vehicle. There the right move is not to presume but to ask: one written line («would you rather we cancelled the purchase and refunded you, or the exchange?»), and you keep the answer. Presuming withdrawal cancels a purchase they may have wanted to keep; presuming exchange applies your policy's conditions to a statutory right. And settle it quickly, because their withdrawal period keeps running meanwhile.
3. You can offer both at once, and that is the sensible move. Nothing stops you saying «we are refunding you, and if you like we will hold the size M for you right now». What you cannot do is make one conditional on the other.

In your commercial flow — your policy, your deadlines, your conditions — none of this applies: there you decide which options you offer and in what order. The full separation between the two worlds is in withdrawal vs. return vs. exchange.

The three kinds of exchange (and which suits your catalogue)

Not every «exchange» is the same thing, and the kind you offer sets your retention ceiling for you. Worth knowing which one you are running before you read the metrics.

The variant exchange is the floor: cheap, obvious and capped by your inventory. Shop for an exchange has the highest ceiling, because it removes that cap and because a share of customers pick something dearer — there the return ends up being a sale. The advanced exchange converts hardest and pushes the most financial risk onto you: if the original parcel never arrives, you gave a product away.

⚠️ Careful with advanced exchanges in the legal flow. If you build it around a card hold, you cannot require that from someone who is withdrawing: it would attach a condition to a right. Offer it inside your commercial flow and keep the withdrawal clean.

Exchange conversion is not won on the choice screen

When a customer picks the refund with the exchange right in front of them, it is almost never the button design: it is what sits behind the buttons. Four levers decide most of it, and all of them are touched before you redesign anything.

1. Real stock, visible at the moment of choosing. Offering a size you cannot ship is not a conversion: it is a return that comes back tomorrow with an angry customer attached. If the portal does not read live inventory, the exchange is a promise.

2. Who pays for return shipping. This is the cost that most often turns an exchange into a refund, and it is usually smaller than the margin you retain. One asymmetry works well and is defensible: exchange shipping on you, refund shipping on the customer within your commercial policy and having told them beforehand. On a withdrawal the split is set by law and not by your policy: you may only pass on the direct costs of return, and only if you informed them first. We unpack it in who pays return shipping in the EU.

3. How long the replacement takes. The exchange competes with the alternative of taking the money and buying again, which arrives in 24 or 48 hours. If your exchange takes two weeks because you wait to receive the parcel, inspect it and only then ship, the customer does the maths and picks the refund. Shortening that cycle — auto-approving low-risk exchanges, shipping on drop-off confirmation — moves more than any incentive.

4. What happens to the price difference. If the customer does not know whether you will charge them, how much and how, they do not choose. Say it on the same screen, with the amount already calculated.

That last row is there on purpose: it is where nearly everyone starts and the one that pays least.

Exchanging down: what happens to the difference

When the chosen item is worth less than the returned one, there is customer money left over, and who decides where it goes depends on the flow. It is the case that raises the most questions and the one most easily got wrong.

In your commercial flow you decide, with one condition: announce it before the customer chooses. Returning the difference to the means of payment is the cleanest option; giving it as store credit is legitimate if you explain it on the same screen and not as a surprise at the end.

On a withdrawal, the difference is money the customer paid that no longer corresponds to anything: it goes back the way it came, unless they expressly agree to another route and incur no fees (Article 13(1)). The same applies if they paid in several ways: each part goes back by its own route. That detail, with the German case law behind it, is in refund or store credit: what customers can demand.

There is also a Shopify technical limit worth knowing before you promise credit: to issue store credit you need new customer accounts enabled and an order with a customer attached. A guest order cannot receive credit, only a refund. And since the legal flow under Article 11a has to work without registration, the guest who withdraws is precisely the case where credit is not even technically possible. The setup is covered in store credit in Shopify.

Eight dark patterns of the exchange screen

The difference between incentivising and manipulating is whether the customer could have chosen the opposite with the same effort. These eight show up again and again in returns portals; the table separates what you may do in your policy from what you may not when the customer exercises their right.

The two columns are not a double standard: they are two different contracts. In the commercial one you offer a voluntary service and have plenty of design room; in the legal one the customer is exercising a right and the design cannot weigh on it.

And none of this waits for a new law. Dark patterns — designs that manipulate users into a decision they did not want — are already punishable today under Directive 2005/29/EC on unfair commercial practices, which penalises practices that materially distort the decision of the average consumer. Mind the nuance, because it decides how your screen gets judged: no rule on its own prohibits giving prominence to a voluntary alternative; what is assessed is the design as a whole and its effect on the decision. That is why the ⚠️ rows in the commercial column are not a permission slip: they are cells that depend on how the whole screen ends up. And the penalty regime introduced by Directive (EU) 2019/2161 sets, for widespread cross-border infringements, a maximum fine of at least 4% of the trader's annual turnover in the member state concerned. The three clocks running at once — button, dark patterns and the future Digital Fairness Act — are laid out in the three legal timelines for your shop.

What do the available data say about exchanges?

That the exchange is already the majority resolution among brands that run it well, although the best public data are not European. Worth saying up front: there is no comparable EU study, so these numbers are an order of magnitude, not a benchmark for your market.

Both sources are competitors of ours and we will say so: Loop Returns is an American returns app and Reversso is Chilean. Their data beat any estimate we could improvise, and their bias is obvious — they measure shops that already use an exchange tool. The useful reading is the direction, not the decimal: when the exchange is built properly, it stops being the exception.

How to build it in Shopify without breaking anything

Shopify already calculates the difference on an exchange from the admin; what it does not ship is the separation between the legal and the commercial flow. When you create the return, return fees and exchange items are applied against the returned items to determine whether a refund is due or payment needs to be collected: if the new item is worth less, you issue the refund; if it is worth more, you collect the balance with an invoice or by capturing payment. The full walkthrough is in how to manage a return in Shopify.

What returnEasier adds on top is that separation, built into the product rather than left to your judgement:

• The commercial screen always offers the refund, and always first, for every combination of plan and settings. Exchange and store credit are added behind it when you enable them.
• Nothing is preselected and every option renders through the same markup and the same visual weight. The exchange then asks for the variant and shop-for-an-exchange opens the catalogue — data intrinsic to each resolution — but not one step is added to the refund path.
• The withdrawal button lives in its own circuit, without login and with a fixed label, separate from all of the above.
• The commercial window is configurable up to 60 days — the cap comes from the Shopify API, which without the historical-orders scope only returns the last 60 days of orders — and is independent of the legal 14 days.

The legal flow is on every plan, including the free one. Exchanges, shop for an exchange, store credit and automatic resolution rules start on the Pro plan (USD 24.99/month).

Common mistakes

• Treating the exchange as the default resolution of a withdrawal. It is not: a withdrawal extinguishes the obligations to perform the contract and its normal resolution is the refund.
• Measuring your exchange rate over all returns. Put legal withdrawals in the denominator and your metric will always look bad, pushing you to squeeze where you must not.
• Offering exchanges without live stock. You convert today and take it back tomorrow, with a customer worse off than at the start.
• Charging for exchange shipping to «protect margin». That is usually the one euro that turns the exchange into a refund.
• Changing the buttons before the process. Redesigning the screen is the cheap part; stock, lead time and shipping are what decide.

Frequently asked questions

Is it legal to offer an exchange instead of a refund? Offering it, yes; imposing it, no. On a withdrawal, Article 13(1) of Directive 2011/83/EU requires reimbursement using the same means of payment, unless the customer expressly agrees otherwise and incurs no fees.

Can I preselect the exchange in my portal? In the commercial flow, yes, with the refund visible and accessible. On a withdrawal, no: a preselected box is not express agreement.

Does an exchange close a withdrawal? No. Article 12 of Directive 2011/83/EU releases the parties from performing the contract; whatever ships afterwards is a new purchase, and the 14-day refund clock keeps running unless the customer expressly accepted the exchange in place of the money, at no cost to them. You may withhold payment until the goods come back or the customer proves dispatch, unless you offered to collect them yourself (Article 13(3)), but withholding does not restart the clock.

What if the customer exchanges for something cheaper? In your policy you decide, announcing it beforehand. On a withdrawal the difference goes back to the original means of payment, unless expressly agreed otherwise and at no cost.

Why do they pick the refund even when I offer an exchange? Because of stock, shipping, lead time, or because the price difference is unclear. Rarely because of the button design.

Conclusion

If you sell on Shopify to European consumers and want to lift your exchange rate, the cheap order is this: first show real stock at the moment of choosing, then absorb the exchange shipping, then shorten the lead time until the replacement goes out, and finally make the price difference unmistakable. All four are operational, none of them touches the legal flow, and every one pays better than redesigning a screen.

And hold the line on the other side: on a withdrawal the exchange is an offer, never a default exit. Not because an inspector is coming tomorrow, but because the most expensive leak in a return is not the amount: it is the customer who does not come back.

💡 Ready to retain without risking compliance? returnEasier keeps a compliant withdrawal button — no login, refund to the original means of payment, stamped with date, time and hash — separate from your commercial flow of exchanges, shop for an exchange and store credit (Pro plans and above). Try it free — 3 trial returns, no card.

---

Official sources

• Directive 2011/83/EU on consumer rights (arts. 11, 12, 13 and 14) — EUR-Lex
• Directive (EU) 2023/2673 — EUR-Lex
• Directive 2005/29/EC on unfair commercial practices — EUR-Lex
• Directive (EU) 2019/2161 (Omnibus Directive) — EUR-Lex
• Real Decreto Legislativo 1/2007 (TRLGDCU), art. 107 — BOE
• Shopify Help Center — Creating and managing returns and exchanges
• Shopify Help Center — Store credit
• Retention benchmarks report 2026 — Loop Returns
• Figures from Reversso's 2025 annual report (Chile, Mexico and Colombia) — Marketing4eCommerce

Shopify is a registered trademark of Shopify Inc.; Loop and Loop Returns are trademarks of Loop Returns, Inc.; Reversso is a trademark of Reversso SpA. Third-party data captured on 21 September 2026.

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