Amazon’s US FBA fees rise $0.19 a unit in January

Nineteen cents.

That is what Amazon says next year costs. On 5 November it told sellers in its US store that Fulfilment by Amazon fees, the per-unit charge for letting Amazon hold your stock and ship it to the buyer, rise by an average of 0.19 dollars from 15 January 2027. Amazon puts that at roughly 0.8% of the average item’s selling price.

Nineteen cents is a number built to be waved through. A rounding error. A thing you mention in the third slide and nobody asks about.

Do not wave it through, for two reasons.

The first is that the average is an average of Amazon’s catalogue, not yours, and underneath it sits a band table that charges the cheapest goods most. The second is that there are two dates in this announcement, not one, and they pull in opposite directions. One of them might even be good news.

There is also a third thing, which is a number Amazon did not publish. We will come back to it, because it is the one that would let you check any of this.

The average is not the number you pay

Here are the bands, as Amazon gave them.

Small standard-size goods take the increase hardest and most consistently. Below 10 dollars the fee rises 20 cents per unit. Between 10 and 50 dollars it rises 40 cents. Above 50 dollars, 70 cents.

Large standard-size goods are treated differently. Between 10 and 50 dollars the rise is 12 cents, a third of what a small standard item in the same price band absorbs. Above 50 dollars it is 47 cents. And below 10 dollars, large standard-size non-apparel does not rise at all.

Read that back and the shape is unmistakable. The cheapest small items pay the most, proportionally, by a wide margin.

Twenty cents on a 9 dollar phone case is a different event from 70 cents on a 60 dollar one, even though 70 is the bigger number. On that 9 dollar case at a 15% net margin, 20 cents takes roughly an eighth of the margin. Gone. And it is taken from exactly the inventory most sellers built their velocity on, the cheap fast movers where the unit economics were always tight and volume did the work.

The rate card bears it out. A small standard item of 2 ounces or less priced under 10 dollars goes from 2.43 dollars to 2.64 dollars. In the 10 to 50 dollar band the same item goes from 3.32 to 3.75. Above 50 dollars, 3.58 to 4.31. Those 2026 figures are non-peak rates, so a peak-season comparison will not match them.

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

This is the US store, and that is not a technicality

Every figure above belongs to Amazon’s US store. The announcement names the 2027 US referral and FBA fees, it went to sellers in the US store, and nothing in it sets out a European schedule.

That cuts two ways, and they are not the same.

Sell into the US store from Europe and this is your January. Fees attach to units, not to the seller’s address, so a European business with American FBA inventory absorbs the same bands as an American one.

Sell only in European stores and nothing here changes your fees in January. There is no published European timetable and no European band table. Amazon’s European fee cycles have historically run on their own schedule and their own numbers, and anyone pricing 2027 European stock off these figures is pricing off the wrong country.

What the announcement is worth to a European-only seller is a shape, not a rate: cheap, small, fast-moving goods are the ones being repriced, and that is the pattern to watch for when a European schedule lands.

The second date runs the other way

Here is the part most sellers will read last and should read first, because for some of them it is a refund.

From 1 February 2027, storage changes, and the direction is opposite. Amazon is removing two storage surcharges. The storage utilisation surcharge goes entirely. The aged inventory surcharge is folded together with monthly storage into a single rate based on how long stock has been sitting. Amazon says the package will mean flat or lower fees for inventory that turns in under six months.

So January punishes cheap goods and February rewards fast ones. A seller shifting low-value stock quickly takes the fulfilment hit and gets some of it back on storage. A seller sitting on slow, expensive inventory takes nothing in January and meets the age-based rate in February with no surcharge structure left to dilute it.

One condition in there is worth reading twice. The low-inventory-level fee applies only when inventory falls below 28 days of supply, measured against actual shipments in both the four weeks before and the four weeks after the assessment date. Two windows, not one, and one of them has not happened yet when the assessment is made. Run deliberately lean and you are exposed to a measurement that partly depends on demand that does not exist yet.

For the business operator

The work is a band-mapping exercise and it is finite.

Split your US catalogue by size tier first, then by the three price bands, and attach the right increase to each line. A catalogue weighted to sub-10 dollar small standard goods faces something close to 20 cents a unit on its highest-volume lines, which is nothing like the headline figure.

Where the increase takes more than a tenth of net margin on a line, that SKU needs a decision before January: reprice, repack into a cheaper size tier where the dimensions genuinely allow it, or cull. Repacking is the only one of the three that changes which band you are in, and the only one needing lead time, which is why this is November work.

From the consumer’s side

The band structure has a shopper inside it, and she explains why the cheap end is where the pressure lands.

Someone buying a 9 dollar accessory is the most price-sensitive customer on the platform, and the one most likely to be comparing two near-identical listings on price alone. That is the segment where passing 20 cents through is most visible, and also where absorbing it hurts most. The seller is squeezed from both ends by the same behaviour.

At the other end, a buyer of a 60 dollar item is weighing reviews, delivery date and brand as much as price, and 70 cents disappears into the noise. The larger absolute increase is the easier one to pass on.

So a flat percentage price rise across the catalogue gets both ends wrong. The increase is regressive and the tolerance for passing it on runs the same way.

The number Amazon did not publish

Which brings us back to the thing we left hanging.

Amazon said the rise is roughly 0.8% of the average item’s selling price. It did not say what that selling price is. It has not published how the average is weighted, or which period of unit sales produced it.

That matters more than it sounds. Without the denominator, 0.8% cannot be checked against any individual catalogue. It is a true statement about a population nobody outside Amazon can see, offered in place of the one number that would let a seller test whether their own mix is better or worse than the average they are being quoted.

The band figures are averages within each band too, rather than a single rate per band, so the increase on a specific ASIN may sit either side of the numbers here. The full rate card is the only place a per-SKU figure can be read.

And the February storage rates have not been published in the same detail as the January bands. So the claim that fees go flat or lower for stock turning in under six months is Amazon’s characterisation of its own change, not something a seller can currently check against a table.

Which is the honest summary of the whole announcement. One precise number, one vague percentage, one missing denominator, and a table that does the real work if you go and find your own SKUs in it.

The three-line version

• If you run the operation: map your US catalogue to the three price bands by size tier this month, and measure the increase against net margin per unit rather than against price. Sub-10 dollar small standard goods are where the damage is.
• If you have just taken on fulfilment: the gap to name is that 0.19 dollars describes Amazon’s catalogue and not yours, and that no European schedule has been published, so nothing here tells you what European fees do in 2027.
• If you sign the budget: there is nothing to buy. Approve the repacking lead time for SKUs that can genuinely change size tier before January, and refuse any 2027 model that applies 0.19 dollars per unit across the catalogue or treats these figures as applying to European stores.

Sources

• PPC Land: https://ppc.land/amazon-sellers-face-average-0-19-fba-fee-rise-from-january-15-2027/
• EcomCrew: https://www.ecomcrew.com/amazons-2027-fee-math-doesnt-stop-at-19-cents/