Break-even ACoS is your gross profit margin before advertising, expressed as a percentage of the selling price. Spend below it and the order makes money. Spend above it and you lose money on every unit. Amazon puts it plainly in its own advertising guide: “Break-even ACOS is directly linked to your profit margin. In order to maintain a profit, your Amazon ACOS needs to be lower than your profit margin.”
So it isn’t a target you pick. It’s a number your unit economics already decided. Your only job is to calculate it honestly, and most sellers don’t, because the standard worked examples floating around leave out four real costs.
The formula
Break-even ACoS % = ((Selling price − COGS − all Amazon fees − other per-unit costs) ÷ Selling price) × 100
Which is the same as saying: break-even ACoS equals pre-ad gross margin. Every point of ACoS below it is a point of margin you keep.
What most calculations leave out
Look at the worked examples in circulation and you’ll see the same four lines: selling price, landed COGS, referral fee, FBA fulfilment fee. Subtract, divide, done.
That’s the calculation that produces a break-even ACoS several points higher than reality, and it’s why campaigns that look profitable in a spreadsheet lose money in the P&L.
The authoritative list of what Amazon actually charges you per unit is Amazon’s own SKU Economics report, and it runs to fourteen fee types: FBA fulfilment fee, monthly storage, low-inventory-level fee, referral fee, aged inventory surcharge, inbound placement service fee, Sponsored Products charges, inbound transportation, per-item selling fee, closing fees, removal fee, disposal fee, refund administration fee, and FBA prep.
Four of those get missed almost universally:
Storage. US standard-size storage runs $0.78 per cubic foot from January to September and $2.40 from October to December, before a utilisation surcharge that can add $0.44–$1.88. That’s a threefold seasonal swing. Cubic feet = (L × W × H in inches) ÷ 1,728.
Returns processing. Apparel and shoes are charged on every returned unit with no threshold. Every other category is charged only on returns above a category rate — Grocery 2.9%, Home & Kitchen 8.1%, Consumer Electronics 11.2%, Backpacks and Luggage 12.8%. If you’re above your threshold, this is a per-unit cost on the whole overage.
Refund administration. The lesser of $5.00 or 20% of the referral fee, on every refunded order.
The digital services fee. UK sellers pay 2% on top of Selling on Amazon fees in the UK, US, Canada, Germany and several other stores. From 20 March 2026, a UK-established seller pays 3% on both Selling on Amazon fees and FBA fees for sales in the French, Italian and Spanish stores. If you sell into the EU from a UK entity, this belongs in your margin build.
Get your referral fee right — most people assume 15%
The US referral schedule runs from 3% to 45%, with a $0.30 minimum per unit in most categories. Fifteen per cent covers Home and Kitchen, Sports and Outdoors, Toys and Games, Tools and Home Improvement, Office Products, Footwear, Lawn and Garden and a handful of others. Plenty of categories sit elsewhere:
| Category | US referral fee |
|---|---|
| Amazon Device Accessories | 45% |
| Pet Products (veterinary diets) | 22% |
| Automotive, Base Equipment Power Tools, Business/Industrial/Scientific | 12% |
| Tires | 10% |
| Computers, Consumer Electronics, Full-Size Appliances, Video Game Consoles | 8% |
| Baby Products, Beauty/Health/Personal Care | 8% at or under $10, 15% above |
| Grocery | 8% at or under $15, 15% above |
| Clothing | 5% to $15, 10% to $20, 17% above |
And watch the mechanics, because two different structures are in play. Clothing, Baby, Beauty, Grocery and Lawn Mowers use a whole-price threshold — cross it and the higher rate applies to the entire price. Compact Appliances, Electronics Accessories, Furniture, Jewelry, Watches and Fine Art use a marginal structure, where each band applies only to its portion. Amazon’s own example: a $7,000 watch pays 16% on the first $1,500 plus 3% on the remaining $5,500, so $405 rather than $1,120.
Practitioners get this wrong regularly, and a misread referral fee moves break-even ACoS by several points.
UK sellers: your fees went down in 2026
Worth knowing if you’re modelling both stores, because the US and UK moved in opposite directions this year.
Amazon cut European referral fees. From 15 December 2025, Clothing under £15 fell from 8% to 5% and £15–£20 fell from 15% to 10%. From 5 January 2026, a new Home Products category cut items under £20 from 15% to 8%; Pet Clothing and Food under £10 fell from 15% to 5%; Grocery, Gourmet and Vitamins under £10 fell from 8% to 5%. Low-Price FBA extended to items up to £20, worth about £0.40 a unit. FBA parcel fees fell by roughly £0.26 on average.
Amazon’s stated net effect across Europe is a reduction of about £0.15 per unit, partly offset by higher storage, return-to-seller and liquidation fees. UK rates also differ from the US: Computers and Consumer Electronics are 7% rather than 8%, Tools and Home Improvement is 13% rather than 15%, Tyres 7% rather than 10%. The UK minimum referral fee is £0.25.
For the US, Amazon announced no referral fee changes for 2026. FBA fees did change — that’s covered separately.
A worked example, done properly
A £34 kitchen product, UK, sold FBA.
| Line | Amount |
|---|---|
| Selling price (ex VAT) | £34.00 |
| Landed COGS (manufacture + freight + duty) | −£9.60 |
| Referral fee (Home & Kitchen, 15%) | −£5.10 |
| FBA fulfilment fee | −£3.05 |
| Monthly storage, amortised per unit | −£0.18 |
| Returns processing, amortised over units sold | −£0.41 |
| Refund administration, amortised | −£0.12 |
| Digital services fee (2% of Amazon fees) | −£0.17 |
| Pre-ad gross profit | £15.37 |
| Break-even ACoS | 45.2% |
Run the same product through the four-line version everyone publishes — price, COGS, referral, FBA, and you get £16.25, or a 47.8% break-even ACoS. Two and a half points of difference, which on £200,000 of ad-driven revenue is £5,000 of margin you thought you had.
The four lines people skip are individually small. That’s exactly why they get skipped, and exactly why they add up.
About the “30/30/30 rule”
You’ll see it everywhere: 30% landed COGS, 30% Amazon fees, 30% gross margin, sometimes called the one-third pricing rule.
It’s a rule of thumb with no source. There’s no Amazon publication behind it, no academic work, no survey. It circulates in FBA course material and agency posts, and nobody traces it to an origin. As a rough sanity check when you’re pricing a new product it’s fine. As a substitute for calculating your own numbers it’s a way of being confidently wrong.
If you want a sourced reference point instead, Jungle Scout’s 2024 seller survey: 1,064 active sellers, published methodology — found 13% of sellers not currently profitable, 57% above 10% net margin and 28% above 20%. That’s a real distribution rather than a tidy third.
Better still, use Amazon’s own per-SKU tools. The SKU Economics report and the Fee and Economics Preview report give you actual charged fees at MSKU level, and the preview estimates up to 120 days forward — including announced fee changes, if your date range covers the effective date. That’s your data, not an industry average.
Break-even is not your target
Break-even tells you where you stop making money. It doesn’t tell you where to aim, and it says nothing about whether the business is growing — that’s what TACoS is for. Amazon declines to give a target number, “There isn’t a definitive number for a good Amazon ACOS. It’s dependent on your industry, company size, and campaign frequency.”
Practitioner frameworks set the gap by intent rather than by benchmark. On a 45% break-even:
| Goal | Target ACoS | On 45% break-even |
|---|---|---|
| Maximum profitability | 10–15 points below break-even | 30–35% |
| Growth or launch | At or slightly above | 45–50% |
| Defending a listing | At break-even | 45% |
| Clearing stock | Above break-even, below total unit cost | 55–65% |
Treat that as a framework, not a law. It’s agency practice, not research.
Recalculate quarterly
Break-even ACoS is not a number you work out once and pin to the wall. Storage triples in Q4. Freight moves. Your return rate crosses a category threshold. Amazon reprices FBA in January and the digital services fee changed in March.
A break-even ACoS calculated in January and still in use in November is describing a product that no longer exists. Recalculate every quarter, and always after a fee change or a price move.
FAQ
What is break-even ACoS? The ACoS at which an advertised order makes exactly zero profit. It equals your gross margin before ad spend, as a percentage of selling price. Below it you profit; above it you lose money per unit.
How do I calculate break-even ACoS? Subtract landed COGS, all Amazon fees and per-unit costs from the selling price, divide the result by the selling price and multiply by 100. Include storage, returns processing, refund administration and: for UK sellers — the digital services fee, or you’ll overstate it.
Is 15% the Amazon referral fee? For many categories, yes, but the US schedule spans 3% to 45%. Consumer Electronics and Computers are 8%, Automotive 12%, Amazon Device Accessories 45%. Clothing, Baby, Beauty and Grocery use price thresholds; Jewelry, Watches and Furniture use marginal bands.
Should my target ACoS equal my break-even ACoS? Only when defending a listing or launching. For profit, practitioners typically target 10–15 points below break-even. Amazon publishes no recommended figure.
Is the 30/30/30 rule accurate? It’s an unsourced heuristic, not a benchmark. Useful for a rough gut-check when pricing; not a substitute for the SKU Economics report.
Sami Sultan runs Ecom Enable, an Amazon growth partner for DTC brands. Fee figures verified against Amazon Seller Central on 27 August 2026. Referral rates, FBA fees and the digital services fee all changed in 2026 — check the live schedule before pricing.
Sources: Amazon Seller Central referral fee schedule, SKU Economics fields, storage fees, digital services fee, returns processing fee; Amazon Ads on ACoS; 2026 European fee update; Jungle Scout seller survey.