Enter ad spend and the sales those ads produced. You get ACoS, the equivalent ROAS, the break-even ACoS your gross margin allows, and how much room is left.
The term comes from Amazon Advertising, where it is the headline efficiency figure in the reporting, but the arithmetic applies to any channel that attributes revenue back to spend.
Note which way it points. Lower ACoS means each sale cost less to buy, so the number improving means it going down. Every other efficiency metric on this site improves by going up, which is exactly why ACoS and ROAS get mixed up in the same sentence.
They are not related metrics. They are the same fraction inverted, so one can always be converted into the other exactly.
| ACoS | Equals a ROAS of |
|---|---|
| 10% | 10.00x |
| 15% | 6.67x |
| 20% | 5.00x |
| 25% | 4.00x |
| 30% | 3.33x |
| 40% | 2.50x |
| 50% | 2.00x |
| 75% | 1.33x |
| 100% | 1.00x |
Which one you use is a matter of what the room is used to. Amazon sellers think in ACoS because the console reports it. Everyone else thinks in ROAS. Reporting both to the same audience adds no information and doubles the arguments.
This is the most useful thing on the page and the least widely known. Break-even ROAS is one divided by gross margin. Since ACoS is one divided by ROAS, the two inversions cancel and break-even ACoS is the gross margin itself.
A product at 42 percent gross margin breaks even at 42 percent ACoS. At 30 percent margin it breaks even at 30 percent ACoS. There is no conversion to do, which makes ACoS genuinely easier to reason about than ROAS for anyone who already knows their margin.
One correction is worth making before trusting it. On marketplaces the referral fee, fulfilment fee and storage all come out before you see a cent, so the margin that matters is the one after those, not the one in your supplier spreadsheet. Working from the pre-fee margin will set a break-even target ten to fifteen points too generous.
TACoS divides the same ad spend by total sales from every source, organic included.
In the worked example, $10,000 against $37,410 of ad-attributed sales is a 26.73 percent ACoS. The same $10,000 against $52,000 of total sales is a 19.23 percent TACoS. Both are correct, and they answer different questions.
The pattern worth watching is the two moving apart. ACoS holding steady while TACoS falls means organic sales are growing underneath the advertising, which is usually what a healthy listing looks like as rank improves. Both climbing together means spend is rising and nothing underneath it is compounding.
TACoS is a trend metric. A single month of it tells you almost nothing; six months of it tells you whether the advertising is building anything.
A low ACoS target is not an achievement. It is usually a decision to leave volume on the table.
Someone reads that a good ACoS is fifteen percent, sets that as the target, and pauses everything above it. On a 42 percent margin that instruction throws away every sale between 15 and 42 percent, all of which were making money. Efficiency improves on the dashboard while gross profit falls, and the report shows only the first half of that.
The number to optimise is profit, not the ratio. Spending $10,000 at 27 percent ACoS earns more than spending $3,000 at 15 percent, unless the margin says otherwise. The right question at every level of spend is whether the next dollar still comes back with more than a dollar of gross profit attached.
The second version of the trap is judging campaigns against a single account-wide target. Brand defence and broad discovery do different jobs and cannot share a threshold. The chart above shows five campaigns against one break-even line: the two above it are losing money, and the one at 8 percent is not the winner, it is a campaign that is almost certainly underspending.