Because return on ad spend means nothing without a margin next to it.
ROAS is the most quoted and most misleading number in paid media, because it ignores what the product costs to make.
This is the line you have to clear. At a 40 percent margin, break-even is 2.5×, every dollar below that is buying revenue you cannot afford.
Feeding profit rather than revenue into your bidding changes which products the algorithm pushes. High-revenue, low-margin lines stop dominating the budget and the account starts optimising toward what the business keeps.
If your platform accepts a custom conversion value, sending profit instead of revenue is usually the single highest-leverage change available to an ecommerce account. It requires product-level margin data in the feed, which is the reason most accounts never do it, and the reason the ones that do pull ahead.
Channel ROAS is what a platform reports for itself, and every platform claims credit for the same conversions.
Channel ROAS is what a platform reports for itself, and every platform claims credit for the same conversions. Adding up reported revenue across Google, Meta and email routinely produces more revenue than the business made.
The ROAS a platform reports is an average across every dollar spent. The decision you actually face is about the next dollar, and the next dollar almost always returns less than the average, because the cheapest and most eager buyers are reached first.
| Monthly spend | Revenue | Average ROAS | ROAS on the extra spend | Profit at 40% margin |
|---|---|---|---|---|
| $2,000 | $10,000 | 5.00x | 5.00x | $2,000 |
| $4,000 | $16,000 | 4.00x | 3.00x | $2,400 |
| $6,000 | $20,000 | 3.33x | 2.00x | $2,000 |
| $8,000 | $22,800 | 2.85x | 1.40x | $1,120 |
At a 40 percent margin, break-even ROAS is 2.5x. Every row in the table shows an average above it, so every level of spend looks profitable in the platform report. It is not. Profit peaks at $4,000 and falls after that, because the step from $4,000 to $6,000 returned only 2.0x, below break-even. Those extra $2,000 bought $4,000 of revenue and lost $400.
The practical rule: raise budget only while the ROAS on the increment clears break-even. Measure it by comparing two periods at different spend levels, not by reading the blended figure after the change.
The most expensive habit in paid media is setting a single ROAS target and scaling until the account stops meeting it. The average falls slowly, so it keeps clearing the target long after the marginal return has dropped below break-even. By the time the blended figure finally breaks, the account has been losing money on its last several thousand dollars for weeks.
The second trap runs the other way. An account held at a very high ROAS is often leaving profit on the table. A 5x return on $2,000 is less money than a 4x return on $4,000 at a 40 percent margin, as the table shows. Efficiency and profit are different goals, and a target that only measures efficiency will quietly choose the smaller business.
The third is trusting the revenue figure at all. Platforms credit themselves for sales that would have happened anyway, particularly from branded search and retargeting. The ROAS in the interface is usually higher than the ROAS the business actually earned, and the gap is widest in exactly the campaigns that look best. The MER calculator is the check against that.
Because platforms reach the most likely buyers first. Each additional dollar goes to people slightly less likely to convert, so the return on the extra spend is lower than the average. The average falls slowly and hides how quickly the marginal return is dropping.
We use Google Analytics to count visits, only if you allow it. Nothing you type into a calculator is ever sent. Privacy policy