There is a specific meeting that happens in most growing ecommerce businesses about twice a year. Marketing presents a strong ROAS. Finance points out that profit has not moved. Both sides have correct data and neither can explain the other's number.
That meeting is a measurement problem, and it usually resolves the moment someone puts MER on the table.
What each number actually measures
ROAS is platform-reported revenue divided by platform-reported spend, per channel. Every platform calculates it using its own attribution model, and every platform is incentivised to claim as much credit as it defensibly can.
MER, or marketing efficiency ratio, is total business revenue divided by total marketing spend. It does not care which channel claims the sale. There is no attribution model to argue with because there is no attribution happening.
The difference matters because the first number is a claim and the second is an outcome.
Why they disagree
Add up the revenue your platforms individually report and compare it to what your ecommerce backend recorded. In most accounts the platform total is somewhere between 120% and 200% of actual revenue.
Nothing is broken. Meta counts a conversion it influenced. Google counts the same conversion. Your email platform counts it too. Each is applying its own rules to the same order, and none of them is wrong from inside its own model. They just cannot all be right at once.
The practical consequence: a channel can show excellent ROAS while contributing almost nothing incremental, because it is claiming credit for demand that would have converted anyway.
The branded search example
The clearest case is branded search.
Someone searches your brand name. They already intended to buy. Your branded campaign serves an ad, they click, they convert. The campaign reports an outstanding ROAS, often 10× or higher.
But what did that spend create? In most cases, a click you would have received for free from the organic result immediately below. The reported return is real as arithmetic and largely fictional as a growth signal.
We worked with a fashion brand where branded search carried more than 80% of PPC revenue. On paper the account looked exceptional. In practice, new customer acquisition was barely functioning, because the number everyone was optimising toward was measuring loyalty rather than acquisition.
What MER does differently
MER refuses to engage with any of that. Total revenue, total spend, one ratio. It cannot be inflated by double-counting because it never counts at the channel level.
That makes it useful for one specific decision, which happens to be the most important one: should total marketing spend go up or down?
If you increase spend by 20% and MER holds, the additional spend produced proportional revenue. If MER falls, it did not. No attribution argument required.
Where MER falls short
MER is a blunt instrument, and it is worth being honest about that.
It cannot tell you which channel to cut. It moves for reasons unrelated to marketing, including pricing, seasonality and stock availability. And it is a revenue ratio, not a profit ratio, so a strong MER on low-margin products can still be unprofitable.
This is why we do not use MER alone. We pair it with contribution margin, so we can see not just how efficiently spend generated revenue, but how much of that revenue survived cost of goods, shipping and fees.
How to use both
A practical split:
MER for budget-level decisions. Total spend up or down, month to month. This is the number that goes in front of the board.
Contribution margin for what to sell. Which categories and products deserve media weight, based on what they leave behind rather than what they turn over.
Platform ROAS for within-channel decisions only. Which campaign, which audience, which creative. Directionally useful inside a channel, misleading the moment you compare across channels.
Non-brand ROAS as your acquisition read. Strip branded search out entirely and look at what remains. That number tells you whether your marketing is creating demand or harvesting it.
Where to start
Calculate your MER for the last twelve months, monthly. Total revenue from your ecommerce backend, total marketing spend including agency fees and tooling.
Then plot it against total spend. Most brands discover a range where MER is stable and a point where it starts to degrade. That inflection is your current efficient ceiling, and finding it is usually worth more than another quarter of campaign optimisation.
It also settles the meeting.
