What Is ACOS on Amazon? The Formula, a Worked Example, and Your Real Target

What ACOS Means and How to Calculate It

ACOS stands for Advertising Cost of Sales. It's the percentage of ad-attributed revenue that got eaten by the ad spend that generated it. The formula is: ACOS = (Ad Spend ÷ Ad Sales) × 100. Amazon calculates it automatically at four levels inside Campaign Manager — account, campaign, ad group, and keyword — so you can see where the spend is concentrated, not just the overall number.

Worked example: you spend $500 on Sponsored Products in a week and those campaigns generate $2,000 in attributed sales. ACOS = (500 ÷ 2,000) × 100 = 25%. That means 25 cents of every ad-generated sales dollar went straight back into Amazon's ad auction. Flip the same numbers and you get ROAS (return on ad spend): 2,000 ÷ 500 = 4. Same data, read in the opposite direction — ACOS tells you the cost, ROAS tells you the multiple.

Neither number tells you whether that 25% was good. That depends on one thing ACOS doesn't know about on its own: your margin.

The Only 'Good ACOS' That Matters: Your Break-Even Number

Break-even ACOS is the ACOS at which the campaign's cost exactly equals your profit margin before advertising. Below it, ads are adding profit on top of everything else. Above it, ads are being funded by the margin, sale by sale — you're buying visibility, not banking money.

Worked example: a product retails for $25. Manufacturing, shipping, and Amazon fees add up to $17, leaving $8 of margin before you spend a cent on ads. $8 ÷ $25 = 32%. That's your break-even ACOS. Run campaigns at 25% and you're profitable on the ad spend itself. Run them at 40% and you're spending margin to buy rank, which can be the right call during a launch — but only if you know that's what you're doing.

Every seller's break-even number is different because every product's cost structure is different. A benchmark someone quotes from another category tells you almost nothing about yours.

Why the Same ACOS Can Be a Win or a Problem

A new campaign almost always opens with a high ACOS — Amazon hasn't gathered enough click and conversion data to place bids efficiently, and you have no organic rank yet to lean on. That's expected. It's not a verdict on the product.

• Launch: ACOS deliberately runs above break-even to buy visibility and reviews while there's no organic sales yet to carry the listing.
• Growth: ACOS should be trending down toward break-even as organic rank starts doing some of the work ads were doing alone.
• Mature: ACOS should sit at or below break-even — ads defending rank that's already earned, not building it from scratch.
• Clearance: ACOS can spike again on purpose, if the goal is moving inventory rather than growing the listing.

There's also a limit to what ACOS itself can tell you. It only measures ad-attributed sales, not total sales. TACOS (total ACOS — ad spend divided by total sales, organic included) catches a case ACOS misses entirely: a campaign hitting its target ACOS while actually cannibalizing organic sales that would have happened anyway. If TACOS is climbing while ACOS looks fine, the ads aren't adding as much as the ACOS number suggests.

The Mistakes That Actually Hurt (Including Ones We've Made)

The most common mistake is importing someone else's benchmark instead of calculating your own break-even. A category average of 20% is meaningless if your margin only supports 12%, and just as meaningless if your margin supports 35%.

The second is reacting too fast to a new campaign's ACOS. We've turned off keywords in week one that were still gathering their first real data — cutting a term before it had enough clicks to mean anything, based on a number that was always going to look bad early. A week-old campaign's ACOS is closer to noise than signal.

The third is treating ACOS as the only KPI in the account. Click-through rate, conversion rate, and TACOS all explain why ACOS moved. Optimizing ACOS in isolation can mean chasing a good ratio on a shrinking pool of sales.

When Your ACOS Is Bad News, What to Actually Do

Work through it in order, not by instinct. First, check whether the number is even real: has the attribution window closed, and is it one keyword dragging the average or the whole campaign? Second, check relevance — is the traffic converting elsewhere on the listing but not through this term? Third, check the listing itself; a rising ACOS with a falling conversion rate is often a listing problem wearing an advertising costume. Only after that should a bid change.

Every proposed change carries three things before it runs: the evidence behind it, a measurement plan, and a rollback trigger. That's not a slogan — it's what stops a reasonable-looking fix from making next week's ACOS worse than this week's, because you find out in three days instead of three weeks.

That's the standard we build into Dr. PPC — autonomous Amazon ad management operated by Full Circle, an agency that has managed more than $500M in revenue across 100+ brands. Dr. PPC is $300 a month plus 3% of ad spend, capped, month-to-month, with the first 30 days free, and Orbit — the analytics suite with the trackers this kind of troubleshooting needs — is included at no extra cost. Whether or not that's the right fit for your account, the discipline above holds regardless of who's running it.