From a distance a chargeback and a return look like the same event: a customer keeps their money and the store does not keep the sale. Underneath they behave so differently that a statement grouping them together will misprice both. A return brings the stock back, usually saleable, and costs a label and some minutes at the bench. A dispute brings nothing back at all, charges a fee whether you win or lose, and consumes real time in evidence. This post is the difference, and how a P&L should book each one so the numbers stay honest.
- A return recovers the goods; a chargeback usually does not, so the stock is a loss on top of the money.
- The dispute fee is charged whichever way the case goes, so even a won dispute costs real money.
- Return processing belongs to returns only: charging a label and a bench to a dispute invents work that never happened.
- On a statement the disputed amount is a return, and the dispute fee sits with transaction fees.
The goods are the difference
This is the whole distinction in one sentence: a return is a reversal, a chargeback is a loss. When a return arrives, most of the value comes back with it and the cost is the margin plus the round trip. When a dispute is lost, the customer keeps the product and the money, so the store loses the goods at cost as well as the revenue, which frequently makes one dispute worth several returns.
The full anatomy of the cheaper case is worked through in what a return really costs, and it is worth reading first, because the surprise there is that a return costs more than the margin it reverses. The surprise here is larger.
The fee is charged either way
Card schemes charge a dispute fee per case, and winning does not refund it. Add the time somebody spends gathering order records, tracking and delivery confirmation before a deadline, and a won dispute still costs real money: on ordinary figures around €38 before anything else happens.
That is the fact that changes behaviour, because it means a store with a healthy win rate is still paying monthly for the privilege of defending itself, and prevention is the only lever that removes the cost rather than recovering part of it.
How each one is booked
On a nouz statement the disputed amount is treated as a return, because the money went back to the customer, and the dispute fee sits on the same line as the other card charges, because that is what it is. What a dispute never triggers is restocking: no parcel came back, so charging a return label and a bench cost against it would invent work nobody did.
| Return | Chargeback lost | |
|---|---|---|
| Money back to the customer | yes | yes |
| Goods recovered | usually, often saleable | no |
| Outbound parcel | spent either way | spent either way |
| Payment fee | kept by the gateway | kept, plus a dispute fee |
| Return label and bench | charged | not charged, nothing came back |
| Your time | minutes | half an hour of evidence |
Pricing one dispute properly
Put the pieces together on an ordinary order and the size becomes obvious. The money goes back gross, the goods stay with the customer at cost, the parcel was already shipped, the scheme charges its fee, and somebody spends half an hour on evidence. A chargeback cost calculator weights that against your own win rate, and on typical figures a single dispute costs well over a hundred euros on an order worth a fraction of it.
Expressed in the unit that changes decisions, that is two or three healthy orders replacing one bad one, every time. Six disputes a month is therefore a five-figure annual number for many stores, which is enough to justify the unglamorous prevention work that never gets scheduled.
Which one to work on first
Returns are usually the bigger number and chargebacks the sharper one. If your return rate is high, the money is in prevention at the product page: sizing, photography, delivery estimates. If disputes are rising, the causes are different and mostly about recognition: an unrecognisable billing descriptor, a late delivery, support that could not be reached.
There is also a threshold worth knowing about. Beyond a certain dispute ratio, card schemes put merchants into monitoring programmes with penalties attached, which turns an ordinary cost line into a business continuity question. That is a good reason to price disputes properly long before they become a problem.
The practical rule
Keep them on separate lines and count them separately, even when both are small. A blended figure for money handed back hides the fact that one kind is recoverable and the other is not, and the two need entirely different fixes. A statement that can tell you a bad month for returns from a bad month for fraud has told you something you can act on the same day.