POAS vs ROAS: What's the Real Difference?
The real difference comes down to one thing: ROAS measures revenue generated by ads, while POAS measures profit. ROAS can look spectacular while you lose money on every order; POAS tells you what is actually left. At low margins or in campaigns with discounts, the two can point in completely opposite directions.
ROAS (Return on Ad Spend) = revenue ÷ ad spend. POAS (Profit on Ad Spend) = gross profit ÷ ad spend. The only difference, but an essential one: POAS subtracts the cost of goods, ROAS does not.
For years, ROAS has been the king metric in performance marketing. It is simple to calculate and easy to report. The problem is that it measures the sale, not what remains after it: the profit. And for an online store with products at different margins, that can be the difference between a sound decision and one that costs you money.
Side-by-side comparison
The two metrics answer different questions. The table below shows exactly where they part ways:
| Aspect | ROAS | POAS |
|---|---|---|
| What it measures | Revenue generated | Profit generated |
| Formula | Revenue ÷ Ad spend | Gross profit ÷ Ad spend |
| What it ignores | Cost of goods (COGS) | Nothing - it includes it |
| When it lies | At low margins and with discounts | Rarely, if the cost data is correct |
| The question | How much am I selling? | How much am I actually earning? |
The formula that links them
POAS and ROAS are not rivals - they are relatives. A single variable links them: gross margin.
That is why the same ROAS value can mean profit or loss. It all depends on how high your margin is.
A ROAS of 5× at a 30% margin means a POAS of 1.5× - profit. The same ROAS of 5×, but at a 15% margin, means a POAS of 0.75× - a loss. The same number, two opposite realities.
An example that shows the difference
Two products, exactly the same ROAS of 5×. The only variable that changes is the margin:
| Product A | Product B | |
|---|---|---|
| Gross margin | 40% | 15% |
| ROAS | 5× | 5× |
| POAS (= ROAS × margin) | 2.0× | 0.75× |
| Result | Profit | Loss |
On ROAS, both products look identical and seem like a success. On POAS, one brings in money while the other loses it. If you only looked at ROAS, you would increase the budget on both - including the one that is eroding your profit.
And that is exactly what happens in real accounts: low-margin products often get the most budget, because their attractive price generates the most conversions. The more aggressively you scale on ROAS, the deeper you push a loss that the report does not show you.
Why ROAS lies
ROAS isn't “wrong”, it's just incomplete. It becomes misleading in exactly the situations where it matters most:
- Low margins. The lower the margin, the higher your ROAS has to be just to break even.
- Discount campaigns. The algorithm delivers on the discounted products - which are exactly the ones with an already thin margin.
- Catalogs with mixed margins. A good account-level ROAS can hide products that consistently sell at a loss.
- Returns. ROAS counts the sale, not the return that wipes out the profit.
The classic scenario is the discount campaign. You put a 20% discount on your bestsellers, the algorithm sees that they sell and pushes the budget right there. ROAS stays nice, because revenue goes up - but the discount ate your margin, and real profit drops even while the report looks better than ever. On ROAS everything looks perfect; on POAS, that is exactly when you lose the most.
What ROAS you need to be profitable
Because POAS = ROAS × margin, the profitability threshold (break-even) is simple to calculate: break-even ROAS = 1 ÷ gross margin. Below it, you lose money no matter how “good” the ratio looks.
| Gross margin | Break-even ROAS (POAS = 1) |
|---|---|
| 10% | 10× |
| 20% | 5× |
| 30% | 3.3× |
| 40% | 2.5× |
| 50% | 2× |
You can have a ROAS of 8 and lose money. You can have a ROAS of 2 and be profitable. POAS makes the difference.
Does ROAS have any use?
Yes. ROAS remains useful as a quick indicator, especially when your margins are uniform and high - then a good ROAS almost certainly translates into a good POAS. It is also easier to track, because it does not require cost data. The problem appears when you use it alone, as an optimization target, on a catalog with varied margins. Then you need POAS to see the truth behind the number.
In short, ROAS and POAS do not exclude each other. The best marketers look at both - ROAS for speed and volume, POAS for profit. The difference is that when the two contradict each other, the final decision always goes to POAS, because it measures the thing your business depends on: profit.
How to move from ROAS to POAS
Moving over does not mean giving up ROAS, but adding a profit layer on top of it:
- Calculate your real margin per product or category (COGS, shipping, fees, taxes, returns - case study).
- Send profit, not revenue, as the conversion value to Google and Meta.
- Optimize for POAS, keeping ROAS only as a secondary reference.
The result: the same campaigns, but optimized for the money that stays, not the money that just passes through your account. ROAS does not disappear - it just moves to second place.
Key takeaways
- ROAS measures revenue, POAS measures profit - that is the whole difference.
- POAS = ROAS × gross margin; the same ROAS figure can be profit or loss.
- Break-even ROAS = 1 ÷ margin; below it, you lose money however good the ratio looks.
- Use ROAS as a quick reference, but make budget decisions on POAS.
Conclusion
POAS vs ROAS is not a competition where one eliminates the other - it is a matter of depth. ROAS shows you the surface, POAS shows you the depth. For an eCommerce business with varied margins or discount campaigns, the difference between them is exactly the difference between thinking you are profitable and being profitable. And the only metric you can truly base your budget decisions on is the one that measures the money you keep, not the money you collect.