ROAS on its own tells you almost nothing
Return on ad spend is revenue divided by spend. A 4× ROAS means every rupee of advertising produced four rupees of revenue. It sounds like a scorecard, and platforms present it as one.
But revenue is not profit. If your gross margin is 20%, a 4× ROAS returns ₹4 of revenue containing ₹0.80 of gross profit against ₹1.00 of ad spend. That campaign is losing twenty paise on every rupee spent, while the dashboard shows a number most people would celebrate.
The formulas
Break-even ROAS = 100 ÷ gross margin %
Profit = ( Revenue × margin% / 100 ) − Ad spend
Break-even ROAS is the number worth memorising. It is set entirely by your margin, and it does not care which platform you advertise on.
Break-even ROAS by margin
| Gross margin | Break-even ROAS | Target for healthy profit |
|---|---|---|
| 10% | 10.0× | 15×+ |
| 20% | 5.0× | 7×+ |
| 30% | 3.33× | 5×+ |
| 40% | 2.5× | 4×+ |
| 50% | 2.0× | 3×+ |
| 70% | 1.43× | 2.5×+ |
| 90% | 1.11× | 2×+ |
This explains why a software company can run happily at 2× ROAS while a grocery delivery business at 4× is quietly burning cash. Never compare your ROAS to someone else's without comparing margins first.
Worked example
₹40,000 spend, ₹1,60,000 revenue, 35% margin
- Ad spend
- ₹40,000
- Revenue
- ₹1,60,000
- ROAS
- 4.00×
- Gross profit on revenue
- ₹56,000
- Profit after ad spend
- ₹16,000
- Break-even ROAS at 35%
- 2.86×
Here the 4× genuinely works: it clears the 2.86× break-even with room to spare. The campaign is profitable and there is a case for increasing spend as long as efficiency holds.
Where the number gets slippery
- Attribution windows. Meta's 7-day click window and Google's data-driven model will report different revenue for the same sales. Compare like with like.
- Double counting. Two platforms claiming the same conversion is normal. Blended ROAS — total revenue over total spend — is the honest cross-check.
- Returns and RTO. For Indian e-commerce with cash on delivery, return-to-origin rates can be substantial. Reported revenue is not collected revenue.
- Repeat purchase. A 1.5× first-order ROAS can be excellent if customers buy four more times. That requires lifetime value, not campaign ROAS.
Raising ROAS without raising spend
- Cut what is clearly not working, weekly. Most accounts have spend sitting in placements nobody has reviewed in months.
- Improve the landing page before the ad. Conversion rate multiplies through every campaign at once; creative only helps one.
- Raise average order value with bundles or free-shipping thresholds — it lifts ROAS without touching acquisition cost.
- Separate prospecting and retargeting budgets, and judge them against different targets.
Frequently asked questions
What is a good ROAS?
The one that beats your break-even, which is 100 divided by your gross margin percentage. At 30% margin, break-even is 3.33× — so 4× is good and 3× is losing money.
What is the difference between ROAS and ROI?
ROAS compares revenue to ad spend as a multiple. ROI compares profit to total investment as a percentage and accounts for cost of goods and other expenses.
Should ROAS include GST?
Use tax-exclusive revenue. GST collected is passed to the government and was never your income, so including it inflates ROAS artificially.
Why does Meta report a higher ROAS than my actual sales?
Attribution windows and modelled conversions. Platform-reported figures include view-through and modelled data. Blended ROAS — total revenue divided by total spend — is the reliable cross-check.
Is a 2× ROAS bad?
Not necessarily. At 60% gross margin, break-even is 1.67× and 2× is profitable. At 20% margin the same 2× loses money badly. Margin decides.