ROAS vs MER: Which Metric Should You Optimize?

You know the Monday call. Meta claims a 4x return, Google claims 6x, TikTok swears it drove another 3x, and the bank account barely moved. That gap is the whole roas vs mer problem in one meeting. Three platforms are all taking credit for the same dollars, and someone has to reconcile it. Usually that someone is you, at 11pm, with a spreadsheet.

This is one of the easiest ways for a good media buyer to get fooled by their own dashboards. So let's define both numbers honestly, show where platform ROAS lies to you, and sort out when each one is worth trusting.

What ROAS and MER actually measure

ROAS, return on ad spend, is the revenue a channel takes credit for divided by what you spent on that channel. Spend 10,000 on Meta, Meta's pixel reports 40,000 in attributed revenue, that's a 4x. Simple, and it's the number every platform puts front and center, because it makes the platform look good.

MER, marketing efficiency ratio, is total revenue divided by total marketing spend. Every dollar out, against every dollar the business actually collected. Do 200,000 in revenue, spend 50,000 across Meta, Google, TikTok, email tools, and agency fees, and your MER is 4x. Some people call it "blended ROAS." Same math.

The distinction sounds academic until you notice that channel ROAS is a claim and MER is a fact. One is what a platform's attribution model wants you to believe. The other is tied to your bank statement.

Why summing platform ROAS double-counts your revenue

Here's the mechanic that catches everyone. Each platform runs its own attribution window and its own view of the customer journey, and none of them talk to each other. A single purchase can get claimed by Meta as the last-click retargeting ad, by Google as the branded search the customer typed after seeing that Meta ad, and by TikTok as the top-of-funnel video that started the whole thing.

That isn't fraud. Each platform only sees the touchpoints it was part of, and assumes it earned the sale. Add up the attributed revenue from three dashboards and you're counting the same order two or three times.

We see this in almost every audit we run. Sum the platform-reported revenue and it lands at 130 percent of what the company actually made. It can't be right, because the platforms are fighting over the same conversions. That's why you need to understand how attribution models assign credit before you trust any single ROAS figure. And it's why server-side tracking helps at the channel level but doesn't fix the double-counting between channels. Cleaner signal per platform still leaves you with three platforms claiming overlapping wins.

MER is the honest number tied to the bank

MER can't double-count, because there's one numerator: total revenue. What the business made, divided by everything you spent to make it. No attribution model gets a vote. No pixel gets to inflate its own contribution.

That's the appeal. Scale spend, and if MER holds, you're genuinely growing efficiently. Pour money into a new channel and watch channel ROAS look great while MER drops, and that new channel is probably stealing credit for sales you'd have made anyway. MER catches the cannibalization that platform dashboards can't show you, because they're built not to.

This is also why MER pairs with a target for the whole account instead of per-campaign micromanagement. We wrote a fuller breakdown of ROAS versus MER as planning targets if you want the long version, but the short one is that MER is the number you steer the business by.

When each metric earns its place

MER is not always the right tool. It's blunt, and blunt is sometimes wrong.

• MER is for budget and overall health: setting total spend, judging whether the marketing engine is profitable, reporting to a CEO or a board. It maps to profit, so this is where it belongs.
• Channel ROAS is for optimizing inside a platform. Deciding which Meta ad sets to scale or cut, the platform's own ROAS is still your best signal for relative performance between campaigns in that account. You just can't trust the absolute value.

The trap is using the wrong one for the job. People make budget calls on channel ROAS and overspend on retargeting that was claiming easy conversions. Or they try to optimize ad sets on MER, which moves too slowly and too noisily to guide daily decisions. Match the metric to the decision.

Then there's timing. MER reacts slowly. Long consideration cycles, subscriptions, anything where revenue lands weeks after the click, all of it makes MER look ugly in the exact week you scaled. Read it over a sensible window, not day by day. If your cost per acquisition is creeping up while MER holds flat, that's usually your channel mix shifting, not something broken.

How we actually run it

We steer accounts on MER and diagnose with channel ROAS. Set a MER target that protects your margin, watch it weekly, and use each platform's ROAS only to rank winners against losers inside that platform. When MER and summed channel ROAS disagree, MER wins. Every time. It's the one number your customer's bank and yours both agree on.

If your dashboards are telling three different stories and you can't say which spend is driving profit, that's the knot we untangle in a free ad account audit. We'll pull your real MER, show you where the double-counting is hiding, and tell you which channels are pulling their weight.

FAQ

Is MER just blended ROAS? Yes, it's the same calculation: total revenue divided by total marketing spend. "Blended ROAS" is the common term in ecommerce and "MER" in subscription and DTC circles, but the number is identical.

What is a good MER? There's no universal benchmark, because it depends on your margins. A business at 80 percent gross margin can thrive at a 2x MER, while a low-margin retailer might need 5x or higher just to break even. Work backward from your contribution margin, not from someone else's number.

Can I stop looking at platform ROAS entirely? No. Channel ROAS is still your best signal for optimizing inside a single platform. Treat it as a relative ranking, not an absolute truth, and never sum it across channels to judge total performance.