Amazon can now tell EU buyers to keep a return

Somewhere in Seller Central there is now a box. You type a number into it, anywhere from 1 to 75 euros, or 1 to 75 pounds in the UK store, and that number is the price below which Amazon may tell your customer to keep the thing and have their money back.

Not send it back. Keep it.

This is FBA Returnless Resolutions, live since late October in the UK, German, French, Italian and Spanish stores. The logic is not mad. A cheap phone case can cost more to collect, inspect, repackage and restock than it is worth, so everyone agrees to pretend the transaction never happened. You save the return carriage, the return processing and the storage.

What you do not get back is the phone case. There is no inventory reimbursement, and you may not resell the item even if the customer posts it to you anyway. The unit simply leaves the business.

So the box is really asking you something else. Which of your products would you rather write off than ever see again?

What is actually live

The programme runs in five European stores: the United Kingdom, Germany, France, Italy and Spain. It is configured inside Seller Central, and that is the only place it appears. Nothing about it is visible on a product page or in a buyer’s account, so a seller who has not been into the setting has no way of seeing it exists.

Scope is yours. Apply it to the whole catalogue with named ASINs excluded, or to selected ASINs only. Take the second option. One threshold across a mixed catalogue will be wrong for most of it.

Some goods are out of scope whatever the threshold. Dangerous goods, recalled items, gift cards, and heavy and bulky products are excluded. There is no fee to use the programme, and it applies to returns opened after it is switched on rather than to anything already in flight.

Eligibility depends on account health. Amazon requires the seller to be in good standing, so this is not a lever for an account already in trouble.

From the publisher

Everything we know about running Amazon, in one course.

The commercial strategy, taught by operators who run our own brands.

Check out Amazon Grand Masters

The trade is not symmetric

The saving is easy to name. No return carriage, no return processing fee, and none of the storage the unit would have eaten while it waited to be graded and put back on sale. On a cheap item in a European store, those three routinely cost more than the thing itself.

The cost is also real, and it is larger than it first reads. Returns resolved this way are not eligible for inventory reimbursement, so there is no partial credit for the lost unit. And items a customer has been told to keep cannot be resold. The unit does not come back damaged, or come back at all. It leaves the business.

So the sum is not return cost against nothing. It is return cost against landed unit cost, meaning everything you spent getting that unit into a European warehouse: the goods, the freight, the duty, the inbound handling. Most sellers guess that number low. A 12 euro item with 4 euros behind it is a different decision from a 12 euro item with 9 euros behind it, and the box in Seller Central cannot tell them apart.

One term repays a second reading. These resolutions still count toward the seller’s return rate. What they do not attract is the high return rate fee, which is not charged on returns resolved this way. That combination is easy to misread. The metric still moves, so the category-level consequences of a high return rate still apply. Only the fee is lifted.

Underneath the mechanics, the cost of a cheap return moves off the logistics line and onto the inventory line. The same money leaves either way. It just stops showing up as a fulfilment charge and starts showing up as stock that went missing, and those two lines usually belong to different people. Accountants at TrueBooks, reading Amazon’s seller announcement, set out the mechanics the same way: the refund reverses the sale, the kept unit becomes a separate stock loss at cost charged to cost of sales or a write-off account, and the refund reduces output VAT in the period it is made. Amazon publishes no report isolating these outcomes, so the write-offs have to be traced from return records by hand.

Amazon decides, not you

The threshold is yours. Each individual decision is Amazon’s.

Having set a ceiling, the seller is not choosing which requests get a returnless resolution. Amazon weighs each request on its own, taking in the item price, that customer’s return history, the product category and the seller’s account status. The seller is granting permission up to a value, not issuing an instruction.

This distinction decides how the setting should be modelled. A seller cannot forecast the saving by counting returns under 75 euros and multiplying by a shipping cost, because Amazon will grant the resolution on some of those and not on others, on criteria that are partly about the buyer rather than the product. The honest planning assumption is a reduction in reverse logistics of unknown size, bounded above by the volume below the threshold.

The number nobody has published

The missing figure is the grant rate: how often Amazon actually says keep it. Amazon has not published that for any category or any store, so nobody can size the saving in advance.

That absence is the reason to start low rather than to start at the ceiling. A threshold set at 75 euros on a catalogue a seller has not measured is an open commitment of unknown size, made before peak, in the quarter when return volume is at its highest. A threshold set at the value where landed cost clearly sits below reverse-logistics cost is a bounded one, and it produces the data to raise it later.

The second absence is category behaviour. Amazon names product category as one of the inputs but has not said how it is weighted, and the goods where reverse logistics costs most, the heavy and bulky ones, are excluded from the programme altogether. What is left is a population of smaller items where the saving per unit is lower, and the weighting that decides which of them qualify has not been published.

The third is the one a seller can partly close without Amazon. Customer return history is an input, which means the same ASIN at the same price will be resolved differently for different buyers. A seller cannot see that history, but can see its own return reasons, and an ASIN whose returns are mostly buyers changing their minds is a different risk from one whose returns are mostly faults.

What a seller does differently on Monday morning

This is a configuration decision with a real deadline behind it, because European return volume rises through November and December and a threshold set in January is a threshold set after the expensive quarter.

• Pull the landed unit cost for your twenty highest-return ASINs, including inbound freight, duty and inbound processing. Not the sale price, and not the cost of goods alone. That number is the floor of the decision.
• Compare it against what a return on each of those actually costs you in that store, carriage plus processing plus the storage consumed before the unit is back on sale. Where the landed cost is clearly below that total, the ASIN is a candidate.
• Set the threshold on selected ASINs rather than the whole catalogue, at the value the first two steps produce, and not at the ceiling because the ceiling was offered.
• Check your account standing before planning around any of it, because eligibility depends on it.
• Treat your return rate as unchanged for planning purposes, because the fee relief does not alter the metric.

Where the honest uncertainty sits

Amazon has not published a grant rate, so the size of the saving cannot be estimated from anything currently available. Any projection a seller builds this week rests on an assumption about how often Amazon will say yes, and that assumption has no published basis.

Amazon has published no breakdown of how category weighs against customer history in an individual decision, and no reimbursement treatment beyond the exclusion itself.

And there is no European track record to read yet. Behaviour in other Amazon stores is not a safe guide, because return rates, carriage costs and consumer return rights all differ by market.

The three-line version

• If you run the operation: pull landed unit cost against return cost for your twenty highest-return ASINs this week, and set the threshold per ASIN rather than across the catalogue.
• If you have just taken on returns: the gap to name is that nobody can currently size the saving, because the grant rate is unpublished, and a ceiling set at 75 euros before peak is an open commitment.
• If you sign the budget: approve a bounded threshold on measured ASINs. Refuse any business case that forecasts a saving from the volume below the threshold, because that volume is the maximum, not the expectation.

Sources

• TrueBooks: https://truebooks.co.uk/news/amazon-fba-returnless-resolutions-2026
• Ecommerce News Europe: https://ecommercenews.eu/amazon-introduces-returnless-refunds-in-europe/