What is ACoS?

ACoS (advertising cost of sale) is your ad spend expressed as a percentage of the sales Amazon attributes to your ads. Amazon Ads spells it ACOS. A 25% ACoS means that for every dollar of sales credited to your ads, you paid 25 cents for the clicks that produced them.

ACoS measures efficiency, not profit. It tells you what the ad-driven sales cost, but whether that cost was worth paying depends on your margin. That is why the most useful thing to do with an ACoS is to put it next to your break-even ACoS.

The formula

ACoS = Ad Spend ÷ Attributed Sales × 100

Amazon Ads gives the same calculation, "ACOS = (ad spend ÷ ad revenue) x 100", in its guide to ACOS (checked 2026-10-03).

Spend is clicks multiplied by cost per click, and attributed sales are clicks multiplied by conversion rate and order value. The clicks cancel, which splits ACoS into three levers:

ACoS = CPC ÷ (CVR × AOV)

Keep the second form in mind whenever ACoS moves. A higher ACoS comes from paying more per click, converting fewer clicks, or selling less per order, and each one has a different fix.

Worked example

The figures below are example inputs chosen for round arithmetic. They are not taken from any account.

Suppose a Sponsored Products campaign receives 200 clicks at an average CPC of $0.75, and 20 of those clicks become orders worth $30 each.

• Ad spend = 200 × $0.75 = $150
• Attributed sales = 20 × $30 = $600
• ACoS = $150 ÷ $600 × 100 = 25%

The lever form agrees. Conversion rate is 20 ÷ 200 = 10%, so ACoS = $0.75 ÷ (0.10 × $30) = $0.75 ÷ $3.00 = 25%.

Now add the product's economics, again as example inputs: a $30 price, $8 of Amazon fees and a $7 landed cost leave $15 of contribution margin per order before advertising.

• Break-even ACoS = $15 ÷ $30 = 50%
• At 25% ACoS, each $30 order carries $7.50 of ad cost and keeps $7.50 of its margin
• At 50% ACoS, advertising takes the whole $15
• At 60% ACoS, each advertised order costs $18 in ads and loses $3

How to use ACoS to make decisions

Derive your ceiling, don't borrow one. Amazon's guide is direct about it: "There isn't a definitive number for a good Amazon ACOS" (same source, checked 2026-10-03). Your ceiling is your break-even ACoS, which is contribution margin before advertising divided by price. A product with a 50% break-even can carry an ACoS that would sink a product with a 20% one. Work yours out per product with the break-even ACoS calculator.

Choose a target on purpose. The target is a business decision about how much of the margin to keep and how much to reinvest in volume. A common way to ladder it is by the job a keyword does: brand defence well below break-even, core non-brand terms at or near it, and a time-limited ranking push above it, with an end date set before the push starts.

Turn the target into a bid. Rearranging the lever form gives the most a click is worth at your target:

Max CPC = Target ACoS × CVR × AOV

With the example's 10% conversion rate and $30 order value, a 40% target allows up to 0.40 × 0.10 × $30 = $1.20 per click. The max CPC calculator runs this from your price, conversion rate and target. When a keyword sits above target with enough clicks to trust its conversion rate, move its bid toward that number. When it sits well under target and is short of impressions, there may be room to bid higher. The bid optimization guide covers that loop in full.

Decide at the grain where bids live. Account-level ACoS blends brand defence, launches and core terms into one number. Read it to spot a trend, then act on the campaign, keyword or target.

Common mistakes

• Copying a benchmark. A target lifted from a forum or another seller ignores your price, your fees and your cost of goods. Derive it from your own margin.
• Judging the last few days. Amazon reports ad sales on the date of the ad click, not the date of the purchase, and its help centre says attribution metrics for a date "remain incomplete until the lookback window has ended" (Amazon Ads attribution help, checked 2026-10-03). The most recent days therefore set spend against sales that are still arriving. See attribution window.
• Averaging percentages. A campaign at 10% and another at 50% do not blend to 30% unless they spent the same amount. Add up the spend, add up the sales, then divide.
• Treating a low ACoS as automatically good. A very low ACoS can mean bids are too timid to win volume you could profitably buy. Under target and short of impressions is a reason to test a higher bid.
• Comparing ad types as if they counted the same sales. Amazon describes ACOS as the ratio of ad spend to "promoted product sales (Sponsored Products) or overall brand sales (Sponsored Brands)" in its ad products guide (checked 2026-10-03).
• Calculating break-even with no cost entered. Without a cost of goods the margin is overstated, and so is the ceiling. Treat the break-even as unknown until the cost is on file.
• Writing off a new campaign too soon. Amazon's ACOS guide notes that "new campaigns tend to have a high ACOS, merely because of their newness."

How ACoS relates to other metrics

ACoS only sees sales that Amazon attributes to an ad. To see what advertising costs the whole business, organic sales included, read it next to TACoS.

How Shurq helps

• ACoS and TACoS side by side. Shurq's dashboard tables report both for each marketplace, parent ASIN and child ASIN, so a high ACoS can be read against what it did to total sales.
• A break-even from your own numbers. Ad Manager measures break-even ACoS from your own profit and loss, only for products with a cost of goods on file, shows the working, and lets you replace it with your own figure.
• Targets from break-even. Campaign targets can be set from each campaign's measured break-even with a Profit-First, Balanced or Growth strategy, or typed in directly. Campaigns with no measured break-even are skipped rather than given a guessed one.
• Bids priced toward the target. The Bid Optimizer prices every bid toward your target ACoS, and when a target sits above a measured break-even it warns you and shows what each sale would lose.