A good ROAS is any return above your break even ROAS, and your break even ROAS is one divided by your gross margin. At a sixty per cent margin it is 1.67. At thirty per cent it is 3.33. There is no industry figure underneath that arithmetic, which is why the benchmark somebody quoted you was useless before you heard it.
That is the entire answer. The rest of this is how to do the division properly, and the three places where the arithmetic quietly stops working.
How do you calculate break even ROAS?
Divide one by your gross margin, written as a decimal. Gross margin here means what is left of an order after cost of goods, inbound freight, pick and pack, the shipping you actually pay for, payment processing and your expected return rate. Not the margin on the product page. The margin on the order that ships.
Break even ROAS by gross margin
| Gross margin | Break even ROAS | Revenue needed |
|---|---|---|
| 20 per cent | 5.00x | 50 000 |
| 25 per cent | 4.00x | 40 000 |
| 30 per cent | 3.33x | 33 300 |
| 40 per cent | 2.50x | 25 000 |
| 50 per cent | 2.00x | 20 000 |
| 60 per cent | 1.67x | 16 700 |
| 70 per cent | 1.43x | 14 300 |
| 80 per cent | 1.25x | 12 500 |
Read it as a floor and not as a target. At break even the campaign has paid for itself and for nothing else: not the salary, not the rent, not the software, not the discount you ran to win the order in the first place. A business advertising at break even is running a customer acquisition programme and filing it under profit.
What is a good ROAS for ecommerce?
For a store it is your break even plus enough to carry the fixed costs those orders have to pay for, which for most of the brands we see lands somewhere between 2x and 4x. That range describes margins, not quality. A jewellery brand at seventy per cent can grow all year at 1.8x. A shop reselling other people's electronics at twelve per cent will not survive at 6x.
So work out your own floor before you compare yourself to anybody. After that the only interesting question is how far above it you are, and whether that distance is growing at the same spend.
Is a 3x ROAS good?
Only if your gross margin is above thirty three per cent. Below that, a 3x campaign loses money on every order it brings in, and it loses more of it the more you spend, which is the failure that nobody catches because the dashboard is green the whole way down.
What is a good cost per acquisition?
A good cost per acquisition is below the gross profit on the order it buys. It is the same test as break even ROAS, written per order instead of per euro. An 80 euro average order at fifty per cent margin carries 40 euro of gross profit, so 40 euro is the ceiling on what that first purchase can cost you.
Keep both numbers, because they fail in opposite directions. ROAS hides basket size, so a campaign chasing cheap small orders can post a handsome ratio while the acquisition cost eats every one of them. Cost per acquisition hides revenue, so a campaign buying expensive customers who spend heavily can look wasteful. Read one without the other and you will optimise it straight into the other one's ditch.
Where the arithmetic stops working
- The revenue in the ratio belongs to the platform. It is whatever the ad account decided to claim, on its own attribution window, and it is not the money in your bank. On a brand we owned outright, Meta reported 140 840 euro of revenue against 48 242 euro of spend, a 2.92x return, while claiming 26.6 per cent of what the store actually took. Both figures were true and neither one was the answer. The attribution note works through what to do with a gap that size.
- The first order is not the customer. Break even on a single purchase is the right floor for a business with no repeat. If a third of your buyers order again inside a year you can pay at or under break even for the first order on purpose and take the margin on the second. That only works if you can see the second one, which means measuring email against the store rather than against the email tool. There is a note on that.
- A ratio has two halves and one of them is easy to move. ROAS rises when revenue rises and it rises when spend falls. An account cut back to its best performing slice will post the best return of its life while the business gets smaller. Improving it without shrinking the business is a different job.
The number we put next to it
Blended return: everything the store took in the period, divided by every euro spent on advertising across every channel, for the same dates. It is blunt, it flatters nobody and it is very hard to fool, because there is one numerator and every channel has to share it. Beside it goes each platform's own claim and the gap between the two, named in a sentence rather than left for somebody to find. The format is shown in full on the Monday report, and the account these figures come from is written up in the Curated Chrome case study.