Profit over ROAS

Open Google Ads and it will tell you that you made eight pounds for every one you spent. Open Meta and it claims five. Add those up across a busy month and, on paper, you have made more money than ever landed in the bank. Something is clearly off.

The number everyone reports, and nobody quite trusts

The problem is simple once you see it. When a customer sees a Meta ad on Monday, clicks a Google ad on Thursday and buys on Friday, both platforms record that sale as theirs. Neither is lying exactly; they just both want the credit. So the returns they report overlap, and the overlap grows the more channels you run.

You can make ROAS go up by shrinking

Here is the trap that catches a lot of advertisers. Pull spend back to only the cheapest, highest-intent searches, mostly your own brand name, and your reported ROAS will rocket. It looks like a triumph. Meanwhile total revenue is falling, because you have stopped reaching anyone new. A rising return on a shrinking business is not success. It is a slow decline with a flattering chart on top.

MER is much harder to argue with

MER, the marketing efficiency ratio, is your total revenue divided by your total advertising spend across every channel. It does not care which platform claims what. If your MER improves while revenue is also growing, you are genuinely getting more efficient. If MER only improves because revenue dropped and spend dropped faster, the single number exposes it straight away.

It is not a perfect measure of incremental impact on its own, but as a north star it is honest, and honesty is the whole point.

Then bring margin into the room

Even MER has a blind spot: it treats all revenue as equal. A five-times return on a product that keeps seventy pence of every pound is a completely different business outcome from a five-times return on one that keeps fifteen. The first prints money. The second can lose it once you count packaging, shipping and the cost of the goods.

That is why we optimise to contribution, the profit left after the cost of each sale, rather than to a headline return. It changes which products get the budget, which campaigns get scaled, and which “winners” turn out to be quietly unprofitable.

How we run it in practice

We feed the platforms real value data, so they bid harder for the orders that actually make money. We judge performance on blended efficiency and contribution, not on whichever dashboard looks best that week. And we report on revenue and profit first, because those are the numbers that pay wages and fund the next stage of growth.

The proof