ROAS (return on ad spend) answers one question: for every euro you give the ad platforms, how many euros of revenue come back? Spend €100 on ads, get €400 in sales — your ROAS is 4 (sometimes written 400%). That's the whole formula. The complicated part is everything around it, and that's where this page earns its keep. (ROAS is a metric — a number — not a dimension, which is a category you slice numbers by.)
The formula (with actual numbers)
ROAS = Revenue from Ads ÷ Cost of Ads
€2,500 in attributed revenue from €500 of spend → 2,500 ÷ 500 = ROAS of 5.
Simple. Deceptively simple, in fact, because of one word hiding in there: revenue.
The trap everyone falls into once
ROAS counts revenue, not profit. And the difference is your margin.
Two shops, both with a ROAS of 4. The first sells products with a 60% gross margin: €100 of ads brings €400 of revenue containing €240 of margin — comfortably profitable. The second sells at a 20% margin: the same €400 of revenue contains only €80 of margin, against €100 of ad cost. Same ROAS. One shop is making money; the other is paying for the privilege of being busy.
This is why every account needs to know its break-even ROAS:
Break-even ROAS = 1 ÷ Gross Margin
At a 40% margin, break-even is 2.5. At 25%, it's 4. Below that line, a "great" ROAS is just a well-decorated loss. In our experience, surprisingly few shops have this number written down — and it changes everything about which campaigns deserve budget.
Why should I care about ROAS?
It's the standard profitability proxy for e-commerce advertising, and once you've anchored it to your margin, it's a genuinely good one. It lets you compare campaigns spending different amounts, set targets for automated bidding, and answer the eternal question "is the marketing working?" with a number instead of a feeling. Feelings are wonderful for choosing restaurants; for allocating ad budget, numbers are better.
What's a good ROAS?
E-commerce accounts commonly target a blended ROAS somewhere in the 3–6 range — but the spread around that is enormous, and three honest complications matter more than the range:
Brand campaigns flatter themselves. Ads on your own brand name often show ROAS above 10, partly because they harvest demand that already existed — many of those people were coming anyway. If your blended ROAS rose because brand traffic grew, that's a mix shift wearing a growth costume.
Prospecting can run "low" and still be right. Cold-audience campaigns below a ROAS of 2 can be perfectly sound if they bring in new customers who reorder. First-order ROAS doesn't see the second order.
ROAS and scale fight each other. Pushing spend up almost always pulls marginal ROAS down — you exhaust the easy buyers first. The right target depends on whether you're optimizing for efficiency or growth; no universal number settles that for you.
One more, on measurement: the ad platform's ROAS and GA4's ROAS will disagree, because each claims credit for sales differently. Neither is lying; they're using different rules. Label which source each report uses and the arguments stop.
How do I improve my ROAS?
Feed accurate purchase values into your conversion tracking — value-based bidding is only as clever as the values it sees. Push high-margin, high-value products in your feeds and quietly suppress the low-margin ones. Raise average order value (bundles, free-shipping thresholds, cross-sells) — same ad cost, bigger basket, better ROAS by arithmetic alone. All the usual CPA levers apply too: cut wasted spend, improve conversion rate. And set different targets per campaign type instead of one account-wide number that's simultaneously too strict for prospecting and too lazy for brand.
Related metrics worth knowing: CPA (the cost view — ROAS = AOV ÷ CPA), AOV (basket size, a direct ROAS lever), break-even ROAS (the line that makes your ROAS mean something), and POAS (profit on ad spend, for when you're ready to count margin directly).
Key Idea: ROAS counts revenue, not profit. A ROAS of 4 on a 20%-margin product loses money; a ROAS of 2.5 on a 60%-margin product is doing fine. Know your break-even.
This week's homework: calculate your break-even ROAS (1 ÷ gross margin — thirty seconds, one division) and compare every active campaign against it. Anything below the line either gets fixed, gets a new-customer justification, or gets paused.
If you'd like ROAS shown next to margin, per category, with the gaps flagged for you — that's broadly what our dashboards at airdan.ai do all day. Stop by when you're curious.
FAQ
What is a good ROAS for e-commerce? Blended targets of 3–6 are common, but the only number that's truly yours is break-even ROAS = 1 ÷ gross margin. Above it you profit; below it you don't, however impressive the figure looks.
What's the difference between ROAS and ROI? ROAS compares ad revenue to ad cost only. ROI compares profit to total costs, including products, shipping, and overhead. ROAS is the campaign view; ROI is the business view.
Is a ROAS of 3 good? It depends entirely on margin: at 50% margin (break-even 2.0), a ROAS of 3 is healthy; at 25% margin (break-even 4.0), it's losing money.
Why does my ROAS differ between Meta and GA4? Each system attributes sales using different rules and windows, so they claim credit differently. The mismatch is normal — pick one source per report and label it.