Kamal Noori
ROAS Is Lying to You. Watch MER Instead.
Platform ROAS measures the ad. MER measures the business.
Those are not the same thing. Founders confuse them and quietly go broke at a “profitable” 4x.
The short answer
Platform ROAS is the number Meta or Google reports for the conversions they claim. It double counts, it cherry picks, and it ignores everything outside the ad account. MER, or marketing efficiency ratio, is total revenue divided by total marketing spend across every channel. It can’t lie, because it pulls from your bank, not the platform’s. Scale on MER and new-customer economics. Treat platform ROAS as a steering input, not a scoreboard.
Why platform ROAS lies
Every ad platform is graded on its own homework. Meta counts a sale if someone saw an ad and bought within the attribution window, even if Google, your email list, and an influencer all touched that same order. Google does the same. So does Klaviyo.
Add up the reported revenue from all your channels and you’ll often get a number bigger than your actual sales. I’ve seen accounts where the platforms together claimed 130% of total revenue. The extra 30% was the same orders, counted twice.
Then there’s the part platform ROAS never sees. Agency fees. Creative costs. The influencer you paid in cash. Your own time. None of it shows up in the ad account, so none of it touches your reported ROAS. The number looks clean because half the costs are invisible to it.
What MER actually is
MER is brutal in its simplicity.
MER = total revenue / total marketing spend.
Total revenue is what hit your Shopify or bank account. Total marketing spend is everything: ad spend, agency retainers, software, creative, affiliate payouts, the lot. One number, top down, no attribution voodoo.
Say you did 200,000 in revenue and spent 50,000 to get it. Your MER is 4. Every marketing dollar returned four. The platforms can argue about who deserves credit. MER doesn’t care. It just tells you what the whole machine produced.
This is why I call it the business metric. It survives an iOS update, a tracking break, a cookie deprecation. When your pixel goes dark for a week, your blended numbers still tell the truth because they come from the till.
Blended ROAS, and where it sits between the two
Blended ROAS usually means total revenue divided by total ad spend. It’s MER’s close cousin, just narrower on the cost side. MER folds in all marketing costs. Blended ROAS often counts only media.
I use both. Blended ROAS catches a platform inflating its numbers. If Meta reports 5x but your blended ROAS across all ad spend is 2.1x, Meta is taking credit for sales it didn’t drive. The gap is the lie, measured. MER then tells you if the whole operation, fees included, is actually paying for itself.
The number that breaks founders: new-customer economics
Here’s where a great-looking ROAS hides a dying business.
A brand runs at 4x blended ROAS. Looks healthy. But split the revenue: returning customers are 65% of it, at a 9x ROAS, because retargeting and email are cheap. New customers come in at 1.4x.
You’re spending a dollar to win a new customer back at 1.40 in revenue. After cost of goods at 35% and shipping, that first order loses money. The 4x is a mirage propped up by people who already knew the brand.
So you scale, because 4x told you to. You pour budget into top of funnel. CAC climbs, new-customer ROAS slips to 1.1x, and the blended number sags as the existing-customer well runs dry. You were never profitable on growth. You were harvesting.
The fix is measuring new-customer acquisition cost against first-order margin and lifetime value, separately from the blended figure. Scale only when a new customer pays back inside a window you can fund. I’ve watched a brand cut spend 20% and grow profit, because the cut spend was buying losing first orders.
FAQ
Is MER better than ROAS?
For deciding how much to spend, yes. MER tells you if the business is winning. ROAS is still useful inside a platform to compare campaigns and creatives against each other. Use platform ROAS to steer, use MER to decide scale.
What is a good MER?
It depends entirely on your margins. A brand at 75% gross margin can thrive at a 3x MER. A brand at 30% margin can go bankrupt at 4x. Calculate your breakeven MER first: 1 divided by your gross margin. At 40% margin, breakeven MER is 2.5. Below it on new customers, you’re paying to lose money.
How do I track MER if attribution is broken?
That’s the point. MER doesn’t need attribution. Pull total revenue from your store and total marketing spend from your invoices and ad accounts. Divide. You can build it in a spreadsheet pulling two numbers a week. Cleaner than any attribution model.
Does blended ROAS replace platform ROAS?
No. They answer different questions. Platform ROAS tells you which ad worked relative to other ads. Blended ROAS tells you whether the platform is inflating its claims. You want both on the same dashboard so you can see the gap.
The payoff
Platform ROAS is a vote the platform casts for itself. MER is the receipt from your bank. One is marketing. The other is math you can’t argue with.
Build your dashboard around MER and new-customer payback, keep platform ROAS as a steering input, and you’ll stop scaling things that look profitable and aren’t.
If you want a second set of eyes on whether your real numbers say what your ad accounts claim, Book a free audit.