Muze AIShort answer
There is no universal good ACoS, and any page quoting one number has not asked what you sell. Break-even ACoS equals your contribution margin: at a 30% margin you break even at 30% ACoS, and at a 45% margin you break even at 45%. Below your break-even the ad is profitable on its own. Above it you are buying something else, usually rank or a launch, and that should be a decision rather than an accident.
The formula, and the only benchmark that means anything
ACoS is ad spend divided by ad-attributed sales, expressed as a percentage. Spend $250 to make $1,000 and your ACoS is 25%. Break-even ACoS is simply your contribution margin, because at that point the ad has consumed exactly the money the sale generated.
ACoS and ROAS are the same fact stated upside down, which matters the moment you run Amazon next to Meta or Google and try to compare them.
- 10% ACoS is a 10.0x ROAS
- 20% ACoS is 5.0x
- 25% ACoS is 4.0x
- 33% ACoS is 3.0x
- 50% ACoS is 2.0x
- 100% ACoS is 1.0x, which is spending a dollar to make a dollar
Contribution margin here means the price after Amazon referral fees, fulfilment, cost of goods, inbound shipping and expected returns. Most sellers compute it on cost of goods alone, which makes break-even look far lower than it is, which makes marginal ASINs look profitable. The definition is in the ACoS glossary entry, and there is a break-even ROAS calculator if you want the arithmetic done for you.
Why published ACoS benchmarks are close to worthless
You will find pages claiming a good ACoS is 15% to 25%. Ask where the number came from and there is no answer, because margin structure varies more inside a category than between categories. A supplement seller at a 60% contribution margin and an electronics reseller at 12% are both Amazon sellers, and one average describes neither.
The second problem is structural. ACoS only counts ad-attributed sales, so a campaign that lifts your organic rank produces revenue the calculation never sees and the ad looks worse than it was. On a product you are actively launching, that is not a rounding error. It is most of the effect.
We are not going to invent a benchmark. Treat pages that do as entertainment.
TACoS tells you whether it is actually working
Total advertising cost of sales divides ad spend by total revenue, organic included. It answers the question ACoS cannot: are ads growing the whole business, or just moving sales from organic into paid?
- TACoS falling while revenue grows: ads are pulling organic up behind them. This is the outcome you want.
- TACoS flat while revenue grows: you are buying growth at a constant rate. Sustainable, not compounding.
- TACoS rising while revenue is flat: you are paying for sales you were already getting.
When a high ACoS is the correct answer
- Launch. A new ASIN has no rank and no reviews. A 60% ACoS for six weeks to build velocity can be the cheapest rank you will ever buy, as long as you decided it in advance and set an end date.
- Defensive branded. Bidding on your own brand looks like paying for sales you would have had. Sometimes it is. Sometimes it is the only thing keeping a competitor's Sponsored Product off your own listing.
- Category research. Auto campaigns exist to find search terms, not to be efficient. Judge them on the terms they surface, then graduate the winners into manual.
- Seasonality. Q4 auction pressure raises everyone's ACoS. Comparing December to September and concluding the account broke is a common and expensive mistake.
The rule is not that high ACoS is bad. It is that a high ACoS you did not choose is bad.
What to do with this
Work out your real contribution margin per ASIN this week. Set break-even ACoS as a kill line rather than a target, so it tells you what to stop rather than what to aim at. Then track TACoS monthly and use it to judge whether the whole programme earns its place.
The mechanical half of this is search term work: refreshing the report, finding terms that spend without converting, adding negatives, moving bids. Muze exposes 48 Amazon Ads tools covering search term reports, negative keywords, bid changes and campaign creation, with every write previewing before it runs. The topic pages go deeper on ACoS optimization and negative keywords.
Related questions
- Is 25% a good ACoS?
- Only if your contribution margin is above 25%. At exactly 25% the campaign covers its own cost and contributes nothing. At a 20% margin it loses money on every order it generates.
- What is the difference between ACoS and TACoS?
- ACoS divides ad spend by ad-attributed sales. TACoS divides ad spend by total sales, organic included. ACoS tells you whether a campaign is efficient. TACoS tells you whether advertising is growing the business.
- How do I convert ACoS to ROAS?
- Divide 1 by the ACoS expressed as a decimal. A 25% ACoS is a 4.0x ROAS. A 50% ACoS is 2.0x. A 100% ACoS is 1.0x.
- Should my branded campaigns have a low ACoS?
- They almost always will, because you are bidding on people who already searched for you. That low number is not evidence the campaign is working. The honest test is what happens to total branded revenue when you turn it off, which is worth running once a year.
- Does ACoS include organic sales?
- No. That is its main limitation and the reason TACoS exists.
Keep reading
Other answers
Ask your own account instead
Muze connects Meta Ads, Google Ads, Amazon Ads and Shopify to ChatGPT or Claude, so the answer comes from your numbers rather than a general one. Free to start, and every write previews before it runs.
See how Muze works