MER vs Platform ROAS: Why a 4x in Ads Manager Loses Money

Ads Manager says 4x. The dashboard is green, the campaign is “scaling,” and the account looks like a win. Then the monthly P&L comes back flat — or worse. The gap between those two stories is the single most expensive blind spot in performance marketing: platform ROAS is a marketing-tool number, not a business number. It counts revenue it didn’t fully create, and it never once looks at what a sale actually costs you to deliver.

This is the heart of the mer vs roas debate. One measures how good a channel looks. The other measures whether the business makes money. They are not the same metric with different labels — they answer different questions, and confusing them is how a “profitable” account runs at a loss for months.

For the neighboring economics, compare The Honest Profit Dashboard: 5 Metrics Beyond Platform ROAS and use Owned Channels Are an MER Lever: Shift Spend When CAC Rises to validate the measurement decision.

What each number actually measures

Platform ROAS = attributed revenue ÷ ad spend, as reported inside the ad platform. The word doing all the damage there is attributed. The platform decides which sales to claim, using its own attribution window (clicks plus, frequently, view-throughs) and its own self-interested model. It has every incentive to take credit generously.

MER (Marketing Efficiency Ratio, sometimes “blended ROAS”) = total revenue ÷ total marketing spend, taken straight from your store and your bank, across every channel. Nobody is grading their own homework. If you spend across several channels and pull total orders from the backend, MER is what actually happened.

The first number is a claim. The second is a fact. When they disagree, the platform is almost always the optimist.

The two lies hiding inside a 4x

A reported 4x carries two distortions, and they stack.

Lie one: double-counted revenue. Platforms claim conversions that would have happened anyway and conversions other channels also claim. A shopper sees a prospecting ad, gets a retargeting impression, clicks a branded search, opens an email, then buys. In siloed reporting, several channels each book that one order at full value. View-through attribution makes it worse — an unclicked impression in the lookback window still books the sale. Add returning customers who were always going to repurchase, and a meaningful slice of “attributed” revenue is not incremental. You paid for all of it; you caused only some of it. A useful planning assumption is that platform-attributed revenue overstates true incremental revenue by something on the order of 15–30% for a scaling account — illustrative, not a law, and worth measuring for yourself with holdouts or geo tests.

Lie two: margin blindness. ROAS treats revenue as if it were profit. It isn’t. Between the top line and your bank balance sit product cost, fulfilment, transaction fees, and returns. A 4x on a low-margin product and a 4x on a high-margin product are completely different outcomes, and the platform cannot tell them apart because it has never seen your COGS.

A worked example: the 4x that quietly loses money

Take a brand with realistic-but-tight DTC economics. Numbers below are an illustrative planning model, not a benchmark to copy.

Per $100 of platform-reported revenue at a 4x ROAS:

Line	% of revenue	Amount
Reported revenue	100%	$100
Ad spend (at 4x)	25%	$25
Product cost (COGS)	50%	$50
Shipping & fulfilment	7%	$7
Payment & transaction fees	3%	$3
Returns & refunds allowance	10%	$10

Stack the costs and the margin picture is clear. Strip ad spend out first to find your contribution margin — what each sale contributes before you pay to acquire it:

100% − 50% − 7% − 3% − 10% = 30% contribution margin.

That 30% sets your true breakeven ROAS: 1 ÷ 0.30 ≈ 3.33x. Below 3.33x, every incremental order loses contribution. The reported 4x sits comfortably above it — so far, so green.

Now apply lie one. If roughly a quarter of that attributed revenue isn’t incremental — double-claimed, view-through, or always-going-to-happen — your true return on what the ads actually drove is closer to 4x × 0.75 = 3.0x.

3.0x against a 3.33x breakeven. The campaign that reads as a 4x is running below the line. It looks like a 33% cushion; it’s actually a quiet bleed. Scale it, and you scale the loss — faster spend, redder P&L, greener dashboard.

That’s the trap in one number. The 4x was never wrong arithmetically. It was answering the wrong question.

Your real number: breakeven ROAS on contribution margin

Every brand has a breakeven ROAS, and almost nobody has it written on the wall. Calculate it once:

1. Start at 100% revenue.
2. Subtract product cost, fulfilment, payment fees, and a realistic returns allowance.
3. What’s left is contribution margin %.
4. Breakeven ROAS = 1 ÷ contribution margin %.

A 50% contribution margin needs 2.0x to break even. A 30% margin needs 3.33x. A 25% margin needs 4.0x — meaning that exact 4x dashboard is literally breakeven before you account for any attribution inflation at all. The thinner your margin, the more your “good” ROAS is an illusion, and the more attribution overlap pushes you underwater without a single red cell appearing anywhere.

Two practical refinements:

• Use contribution margin, not gross margin. Gross margin flatters you by ignoring shipping, fees, and returns — the costs that actually move with volume.
• Weight by new vs returning. If acquisition and retention are blended in one ROAS, a wall of cheap repeat orders can mask brutal new-customer economics. Look at first-order contribution separately; that’s where scaling decisions are really made or lost.

How to operate on MER without flying blind

You don’t have to abandon platform ROAS — it’s still useful for relative in-platform decisions (which ad set, which creative, which audience). The fix is to stop treating it as a P&L and put a business-level number above it:

• Set MER as the headline KPI. Total revenue ÷ total spend, pulled from the store, reviewed at the cadence you actually make budget calls — weekly, not by the hour, so daily noise doesn’t whipsaw you.
• Define a target MER from margin, not vibes. Anchor it to your contribution-margin breakeven plus the profit you require, then hold spend decisions to that line.
• Watch the wedge. Track summed platform-attributed revenue against true store revenue. When the gap widens, double-counting is creeping in — in many cases right when you scale retargeting or stretch attribution windows.
• Sanity-check incrementality periodically. A geo holdout or a controlled spend-down tells you what the platform never will: how much revenue actually disappears when the ads stop.

This is exactly the layer worth automating. An operator like Bach AI can hold blended MER against your real contribution margin continuously, flag the campaigns that look green but sit below breakeven, and surface the recommended cut — read-only until you approve the change, never acting on your account on its own.

The takeaway

Platform ROAS measures how good your ads look. MER measures whether your business makes money. A 4x in Ads Manager can — and routinely does — lose cash once you subtract the revenue the platform didn’t truly create and the costs it never saw.

Do the one calculation that ends the argument: build your breakeven ROAS from contribution margin, set MER as the number you actually steer by, and treat every platform ROAS as a claim to be checked against the bank — not a verdict. Green dashboards don’t pay the bills. Contribution does.