What is ACoS? How to work out your break-even ACoS on Amazon

Most Amazon advertising reports answer the question “how efficient were my ads?” A founder needs the answer to a different one: “did advertising leave the business with more money than it started with?” The two can point in opposite directions. This guide shows how to get from the first to the second, using figures you already have in Seller Central.

What ACoS, ROAS and TACoS measure

Three ratios do most of the work, and one more number, your break-even, tells you how to read them.

Amazon defines ACoS as ad spend divided by attributed sales¹ and describes ROAS as its inverse¹². So a 25% ACoS is a 4.0× ROAS, and a 33% ACoS is roughly 3.0×. Neither is good or bad on its own; each only means something next to the break-even for the product being advertised.

“Attributed” is doing a lot of work. For sellers, Amazon credits a Sponsored Products sale to an ad when the shopper buys within seven days of clicking it, on a last-click basis; for vendors the window is fourteen days³. A purchase on day eight, or one that follows an ad the shopper saw but never clicked, does not appear in ACoS at all. That is one reason ACoS on its own understates what advertising does, and why TACoS exists.

TACoS is not a single figure in the Amazon Ads console. Amazon’s own guides describe it as ad spend as a percentage of total sales⁸, and its API documentation builds it by combining advertising data with Selling Partner sales data⁹. In practice you take total ad spend from Amazon Ads and total sales from the Business Reports in Seller Central, where the figure is called Ordered Product Sales⁷.

ACoS vs ROAS

ACoS and ROAS describe the same campaign from opposite ends. ACoS is a cost ratio: ad spend as a percentage of ad-attributed sales¹. ROAS is a return ratio: ad-attributed sales for each £1 of ad spend². One is the reciprocal of the other, so converting between them is a single division, and our ROAS calculator does it alongside your break-even ROAS.

The practical differences are about habit, not information. In our experience, teams that also buy Meta or Google ads tend to find ROAS easier to compare across channels, while Amazon specialists tend to think in ACoS. Lower is better for ACoS and higher is better for ROAS, which is the most common source of confusion in a shared report.

Both share the same blind spots. Both use the same attributed sales, which for sellers’ Sponsored Products means purchases within seven days of a click, last-click³, and both exclude VAT from those sales³. Neither tells you whether the ad made money until you compare it with the product’s break-even.

How to work out your break-even ACoS

Break-even ACoS is the point at which the money an ad-attributed sale brings in exactly covers what it cost to make, sell, ship and advertise.

The hard part is being honest about “every cost”. Here is a worked example for a single product on Amazon.co.uk. The price, fees and VAT treatment follow Amazon’s and HMRC’s published rules; the fulfilment fee and cost of goods are illustrative, so replace them with your own.

The UK standard rate of VAT is 20%⁵. Amazon.co.uk’s referral fee is 15% for everything outside the categories with their own rates⁴, and Amazon calculates it on a fee base that includes “any taxes and customs duties"⁶. The same fee schedule adds a 2% digital services fee to selling and FBA fees on UK sales⁴, which turns a 15% referral fee into 15.3% in practice. Our guide to Amazon selling fees in the UK lists the rest.

That leaves £5.52 before advertising. Amazon Ads reports this sale at £20.83, the price without VAT, so the margin to compare with ACoS is £5.52 ÷ £20.83 = 26.5%. This product’s break-even ACoS is 26.5%, and its break-even ROAS is 3.77×. An ACoS of 25% makes about 1.5p on every £1 of ad sales; an ACoS of 30%, which many sellers would still call acceptable, loses about 3.5p.

Keep the price basis the same on both sides

The same £5.52 is 22.1% of the £24.99 the shopper pays. Measured that way, a 25% ACoS would look like a loss when it is not. Amazon settled which basis applies in its attribution help, updated on 27 August 2026: “Attributed sales don’t include tax. Inclusive tax such as VAT is subtracted from sales totals.“³ Some industry guides had said the opposite for European marketplaces¹¹, so if you set targets from one of them, recheck them now.

Advertising fees are also quoted excluding VAT¹²; if your business is VAT-registered, the ex-VAT spend is your real cost. For TACoS and profit after advertising, the rule is the same: take total sales and margin on one basis, and check whether the sales figure you are using includes VAT before you combine it with anything else.

What is a good ACoS?

A good ACoS is any ACoS below the break-even for the product you are advertising, so the answer moves with every product’s margin. Because break-even ACoS is simply the margin before advertising, the right target moves with every product’s economics:

In a catalogue of any size, margins vary widely from product to product. A single account-wide ACoS target will be too tight for your best-margin products, starving them of growth they could afford, and too loose for your thinnest, funding sales that lose money. Set the break-even per product, or at least per group of products with similar margins.

We need to be entrepreneurs first and look at the P&L.

TACoS: the number that connects advertising to profit

ACoS judges the ads. TACoS judges what they cost the business, and it leads straight to the figure that matters.

Take an account selling £120,000 a month at an average 28% margin before advertising, spending £14,000 a month on ads. TACoS is 11.67%, so profit after advertising is £120,000 × (28% − 11.67%) = £19,600 a month. Each point of TACoS is worth £1,200 a month here.

Now compare two ways to “grow” the same account. The figures are illustrative.

Doubling the ad budget buys an extra £20,000 of sales a month and costs £8,400 a month in profit: turnover up 17%, profit down 43%. Cutting the targets that lose money gives up £3,000 of sales and adds £2,160 a month. On a sales dashboard the first plan looks like success. On the P&L it is the opposite.

This is also why a falling TACoS alongside rising sales is the signal to look for. It usually means advertising is helping products rank and sell organically, so total sales grow faster than spend.

The three levers that move ACoS

ACoS can be broken into three parts, which shows that bids are only one of the levers:

With a cost per click of £0.90, a 10% conversion rate and a £24.99 average order (£20.83 without VAT, the figure Amazon Ads reports), ACoS is 43.2%. Lift conversion to 13% with better images, a stronger offer or more reviews, and ACoS falls to 33.2% without touching a bid. Add a bundle that raises the average order to £29.99 and it falls again, to 27.7%.

Conversion matters more as shoppers use AI assistants. Amazon reported in October 2025 that customers using Rufus, its shopping assistant, were 60% more likely to complete a purchase¹⁴, and in May 2026 it brought Rufus into a new assistant, Alexa for Shopping, in the US¹⁵. Listings that answer shoppers’ questions clearly tend to convert better whichever way the shopper arrives, and every point of conversion flows straight into ACoS.

When it makes sense to run above break-even

Advertising above break-even is not always a mistake. It is a cost, and some costs are worth paying:

• A launch. A new product needs sales and reviews before it can rank organically. Paying for that early momentum can be the cheapest route to a product that later sells without ads.
• Defending a hero product. When a competitor bids aggressively on your brand or your best-selling product’s main search terms, losing the placement can cost more than holding it.
• Clearing slow stock. If the alternative is months more storage or a removal fee, a loss-making ad sale can still be the smaller loss.

The rule we apply is simple: set the budget, the target ACoS and the end date before you start, and review it weekly. A deliberate investment has a finish line. A default does not.

Common mistakes

1. One ACoS target for the whole account. Break-even differs by product, so a single target is wrong for most of them.
2. Mixing price bases. Amazon Ads reports sales without VAT. A margin calculated on the VAT-inclusive price makes profitable ads look like losses; a TACoS built from ex-VAT ad sales and VAT-inclusive total sales is wrong in the other direction.
3. Forgetting the fees that move. The 2% digital services fee⁴, a 1.5% fuel and logistics surcharge on UK fulfilment fees since 17 April 2026¹⁰, lower FBA parcel fees and a wider low-price FBA band from December 2025 and January 2026¹³, and peak fulfilment fees from 15 October 2026 to 14 January 2027¹⁶ all change the break-even.
4. Judging ads on ACoS alone. ACoS ignores sales that ads help to create but don’t get credit for. Watch TACoS and organic sales alongside it.
5. Treating last-click attribution as the whole truth. A seven-day, last-click window³ misses some of the value of advertising, and credits some sales that would have happened anyway.

How to check your own account in ten minutes

1. In Seller Central, open Reports, then Business Reports, and download Detail Page Sales and Traffic by Child Item for the last 30 days. Ordered Product Sales is your total sales for each product⁷.
2. In the Amazon Ads console, export spend and sales by advertised product for the same dates.
3. For each product, work out the margin before advertising using the method above, or enter its price and costs in our FBA calculator for the UK. That is its break-even ACoS.
4. Compare each product’s ACoS with its break-even, and work out TACoS for the account and for your biggest products.
5. Multiply total sales by margin minus TACoS. That is what your advertising leaves you with.

If the answer surprises you, you are not alone, and it is usually fixable. Most of the gains come from a few decisions: which products deserve more budget, which search terms to stop paying for, and which listings need work before they are worth advertising at all.