How ACoS works
ACoS turns spend efficiency into a percentage that sits naturally next to margin. If a product carries a 35 percent gross margin, then a 35 percent ACoS is break-even on the incremental ad-driven sale, and anything below that is profit. Because Amazon reports ACoS at the campaign, ad group, and keyword level, it becomes the everyday dial sellers turn: bids come down where ACoS runs above the margin line and go up where there is headroom. The number is only as honest as the attribution window behind it, which is why comparing like-for-like windows matters.
ACoS vs ROAS
ACoS and ROAS describe the same relationship from opposite ends. A 25 percent ACoS is a 4.0 ROAS, because 1 divided by 0.25 is 4. Amazon sellers tend to speak in ACoS while Google advertisers speak in ROAS, but the math is identical. The practical difference is framing: ACoS makes the break-even comparison against margin immediate, since you can read a target ACoS straight off a margin table, while ROAS makes scaling headroom feel more intuitive. Pick one and stay consistent so the whole account speaks the same language.
Setting a target ACoS from margin
A target ACoS is not a universal number, it is derived from the specific product's economics and the job the campaign is doing. For a defensive campaign protecting an already-ranking hero product, a target well under margin protects profit. For a launch that is buying rank and reviews, a target at or above break-even margin can be the correct short-term investment, because the return shows up later in organic sales rather than in that campaign's ACoS. The mistake is applying one blanket ACoS target across products with very different margins and lifecycle stages.
How iClick uses ACoS
As an amazon ads agency, iClick derives each campaign's target ACoS from that product's real margin and its role, then read the account through TACoS to make sure campaign efficiency is not quietly hollowing out organic demand. ACoS answers whether a campaign is efficient, and that is a question worth asking per keyword, but it never gets optimised in isolation. The approach is to set the target from margin, hold campaigns to it, and treat a launch's temporarily high ACoS as an investment to be judged by later organic lift rather than a leak to be plugged.
