Amazon Ads reports ACoS instead of ROAS: how do I convert between the two?

Where ACoS appears in the calculator

The ROAS calculator has no field called ACoS, but the figure is there under another name: 'Ad spend as a share of revenue'. That line works out spend ÷ revenue × 100, which is exactly how Amazon's sponsored ads define ACoS. Enter the ad-attributed sales and the spend from your Amazon report and you read ROAS on the first line and ACoS on the third, with no conversion at all.

A quick conversion table, calculated with the tool

• Sales 4,000, spend 1,000: ROAS 4x, ACoS 25%.
• Sales 2,500, spend 1,000: ROAS 2.5x, ACoS 40%.
• Sales 1,000, spend 1,000: ROAS 1x, ACoS 100%, meaning the ads swallowed every sale.

Notice they move in opposite directions: as ROAS rises, ACoS falls. So in Amazon reports lower is better, while in ROAS reports higher is better. Mixing up the direction is an easy slip when you move between Amazon and Meta or Google.

When is an ACoS profitable?

The rule is simple: the ads do not lose money as long as ACoS is below your margin before ads. It is the same test the calculator runs when it compares ROAS with break-even ROAS. Try the same campaign, 4,000 in sales from 1,000 in spend, at two margins:

1. At a 30% margin: break-even ROAS is 3.33x and yours is 4x, so the verdict is above break-even with an estimated profit after ads of 200. In Amazon terms, a 25% ACoS sits below your 30% margin.
2. At a 20% margin: break-even ROAS is 5x and yours is 4x, so the verdict is below break-even with an estimated loss of 200. A 25% ACoS has overshot your 20% margin.

The same campaign makes money for one product and loses it on another. On Amazon, referral fees and fulfilment fees (if you use FBA) also eat into the margin, so include them in your costs before ads; the Profit margin calculator helps with that.

Before you copy numbers from an Amazon report

• The sales in Amazon's advertising reports are sales Amazon attributes to the ads under its own rules, not all your sales. The calculator uses what you enter and does not check it.
• Make sure sales and spend cover the same period and the same currency.
• Another metric divides spend by all store sales, not just attributed ones. If you enter total sales you are calculating something different from ACoS, so be clear which one you are measuring.