Amazon ACOS: What It Is & How to Calculate It (2026 Guide)

What Is Amazon ACOS

Amazon ACOS stands for advertising cost of sales. It is an Amazon advertising metric that tells you how much you spent on ads for every dollar those ads earned. Amazon uses it to measure how cost-efficient your Sponsored Products, Sponsored Brands, and Sponsored Display campaigns are, and the formula is the same across all three.

Two things about that definition are worth getting right before you act on any ACOS number.

1. The sales in your ACOS come from shoppers who clicked your ad and bought inside Amazon’s attribution window. A shopper who clicks your ad on Monday and buys on Saturday counts. A shopper who scrolls past your ad and buys organically does not.
2. It is what shoppers paid Amazon at checkout, not what Amazon paid out to you. Coupons and deal prices push it down. Returns do not come back out of it. Keep this in mind when we get to the break-even math, because a lot of sellers calculate their margin from the wrong sales number.

How to Calculate Amazon ACOS

The formula:

For example you spent $150 on a Sponsored Products campaign last week, and that campaign generated $600 in sales.

Your ACOS is 25%. In plain terms, you spent 25 cents on ads for every dollar of sales those ads brought in.

When you check your ACOS, look at it campaign by campaign, not just for the whole account. Your account average can be misleading. You could have an account ACOS of 30% while one campaign runs at a healthy 12% and another quietly burns money at 90%. Checking per campaign is how you catch the second one.

Where to Find Your ACOS

You will find ACOS inside Campaign Manager under Advertising in Seller Central. It appears as a column next to spend, sales, and orders at every level: campaign, ad group, keyword, and product target. For ACOS by individual customer search term, download the Sponsored Products search term report under Measurement and Reporting.

The data may not be in real time. Amazon can take up to 48 hours to post attributed sales, and invalid click adjustments can change your spend figures for up to 3 days after the fact (as of July 2026). Sponsored Products also runs on a 7-day attribution window, so a click on Tuesday can still be credited with a sale the following Monday.

Always make bid decisions on 7 to 30 days of data, and never judge a keyword or campaign performance on daily or 2 to 3 days data.

What Is a Good Amazon ACOS

There is no single good ACOS, because it depends on your profit margins, your category, and what the campaign is built to do. As a general range, an ACOS under 30% is considered low, 25 to 40% is average, and anything above 40% is high. Most profitable, mature Sponsored Products campaigns settle between 15 and 30%.

A 28% ACOS is comfortable for a supplement brand at 65% gross margin and a slow loss for an electronics brand at 15%. Your break-even ACOS, calculated in the next section, is the only benchmark that applies to your account.

Accounts we manage, the mature campaigns that are left to hold a profit usually land between 18 and 28%, with launch campaigns running well above that on purpose.

Find Your Break-Even ACOS Before You Set Any Target

Your break-even ACOS is your profit margin before ads, expressed as a percentage of the selling price. At break-even, an order from an ad makes you $0. Above it, every ad order loses money. Below it, every ad order makes you profit. Every ACOS target you set should start with breakeven ACOS.

Here is how a $24.99 kitchen product looks like:

Round down to 36% to give yourself a buffer, then read your ACOS column against it. At 36.6% ACOS, you give the full $9.14 back to Amazon in ad costs. At 25%, the ads cost about $6.25 per order and you keep roughly $2.89. At 50%, you lose $3.36 on every single ad order. On this product, each one-point change in ACOS is about $0.25 per order, which makes bid change decisions much easier.

Break-even ACOS is not fixed. It moves every time your costs or your selling price change. Three situations invalidate your calculations:

Coupons and deals

Always calculate break-even at the price customers actually pay. Run that same unit at $22.49 with a 10% coupon and break-even drops from 36.6% to about 31%, partly because the referral fee recalculates on the lower price.

Returns

Ad sales count at the time of order, and a return 2 weeks later does not come back out of your ACOS. In categories like apparel, where return rates run 15% or higher, pad your target by three to five points to stay honest about your real margin.

Fee changes

Amazon re-measures products without warning, and a packaging tweak that moves you from small standard to large standard can add $1.50 or more per unit in fulfillment fees. Recheck your break-even after every fee changes.

ACOS Targets by Campaign Stage

Your break-even is the anchor. Your target moves around it depending on what you need the campaign to do.

During launch, a high ACOS is a cost you budget for, like inventory. Decide in advance how much you will spend to rank and for how long. On a typical launch we plan for 30 to 45 days above 60% ACOS before the review count can support conversion rate. After that, taper down as you start getting ranked and cut the terms that only spend.

One exception worth knowing about consumables. If your product reorders every 30 to 45 days through Subscribe & Save, you can afford to break even on the first order, because repeat orders carry no ad cost. Helium 10 has made this argument for years, and for replenishable products it holds up. For a one-time purchase like furniture, running at break-even is just losing money slowly.

Why Your ACOS Can Mislead You

ACOS numbers can sometime mislead you. Here are few situations to consider.

Branded campaigns inflate blended ACOS

Campaigns targeting your own brand name usually show 3 to 8% ACOS because those shoppers already typed your name into the search bar. Report branded and non-brand campaigns separately, always. Judge your advertising on non-brand ACOS. Brand campaigns are defense spend.

The attribution window distorts short date ranges

Sales credit can post up to seven days after the click, so ACOS almost always reads high early in the week. Compare full weeks against full weeks, and full months against full months.

ACOS never sees the organic sales your ads create

An exact match keyword running at 55% ACOS that lifts your product onto page one can pay for itself in organic orders that never appear in the ACOS column. Pause every keyword above break-even on a new product and you can stall the ranking those ads were building. This is the main reason experienced sellers set targets by stage instead of chasing one flat ACOS number.

How to Lower Your Amazon ACOS

Below are some of the strategies you can use to lower your ACOS.

1. Add Negative Keywords

Pull the Sponsored Products search term report for the last 30 days (60 for low-volume accounts) and sort by spend. Any search term that has spent more than 50% of your product cost, gets added as a negative exact. On the $24.99 product, any term past $12 is negative target. In our audits this single step recovers 8 to 15% of total account spend in the first week.

2. Harvest Converting Search Terms

Terms with 5 or more orders at or below your break-even ACOS get moved into an exact match manual campaign with a bid at which it was converted previously. Keep the auto campaign running at a lower budget so it keeps finding new terms for you.

3. Lower Bids Gradually

Cut 10 to 15% per adjustment, then wait 4 to 7 days or at least 10 more clicks before the next change. Sellers who slash bids 40 to 50% overnight watch impressions collapse, because Amazon’s auction effectively stops entering them. A 0% ACOS from zero impressions is not a win.

4. Adjust Placement Multipliers

Check the Placements tab in each campaign. If product page placements are taking 30% of your budget at a 70% ACOS while top of search converts at 22%, pull the product page multiplier down toward zero and let top of search do the work. Check your own placement report before cutting, because the split varies by product.

5. Fix Your Listing Before Blaming Ads

ACOS is spend divided by sales, so when clicks are not converting, the problem is the listing, not the ads. Compare your main image and price against the top of page one, and check your review count against the sellers next to you. A weak listing at any bid is expensive traffic.

Do not over-negate. An auto campaign with every weak term blocked stops finding new ones, so leave it some budget to keep exploring. And check what a keyword does for your organic rank before pausing it. Cut the highest-ACOS exact term on a launch product and you often give up the page-one position that spend was buying.

Frequently Asked Questions

What does ACOS stand for?

ACOS stands for advertising cost of sales. It is the percentage of ad revenue you spent to generate that revenue: ad spend ÷ attributed ad sales × 100. Amazon reports it for Sponsored Products, Sponsored Brands, and Sponsored Display campaigns.

What is a good ACOS on Amazon?

For most sellers, 15 to 30% is a healthy range on mature campaigns. But the real test is your break-even ACOS. Campaigns below break-even make money per order; campaigns above it lose money per order, unless the high ACOS is a deliberate launch spend.

Is a lower ACOS always better?

Not always. Below break-even, lower is better for profit per order. But an ACOS near zero usually comes from branded terms or almost no spend, and neither grows a product. Launch campaigns often run 60 to 100% ACOS on purpose to buy ranking and reviews.

What is the difference between break-even ACOS and target ACOS?

Break-even ACOS is where ad profit hits zero, and it equals your profit margin before ads. Target ACOS is the number you manage toward: below break-even for profit campaigns, at or above break-even during a launch.

How often should you check your ACOS?

Weekly for bid and budget changes, monthly for target reviews. Amazon takes up to 48 hours to post sales and uses a 7-day attribution window on Sponsored Products, so daily checks push you into decisions on unfinished data.