Use marketing efficiency ratio to run the business and campaign return to run the campaigns. They are not rivals and they are not two versions of the same measure. They answer different questions, at different altitudes, for different decisions.
MER is total store revenue divided by total marketing cost. Everything on the top, everything on the bottom, no attribution involved. It tells you whether the whole marketing operation is paying for itself at the size you currently are.
Campaign return is one platform's reported revenue divided by that platform's spend. It tells you how one campaign performed against another campaign, measured by the same slightly biased ruler, which is fine as long as you only compare like with like.
Stores get into trouble when they use the second to answer the first question. The result is an ad account full of efficient campaigns and a business that is not growing, or a healthy blended picture nobody can act on because it hides which part is dragging.
What each ratio can and cannot see
MER sees everything, including the things you did not pay a platform for. Organic search, direct traffic, your email list, word of mouth and the sale someone made because a friend recommended you all land in the top line. That is its strength and its weakness.
Its strength is that nothing can hide from it. If a platform claims a large improvement and MER does not move, the improvement did not reach your bank account. If you double the budget and MER falls, the extra spend is buying less than the old spend did, whatever the platform reports.
Its weakness is that it cannot tell you what to change. A falling MER says something is wrong, not which campaign is wrong, and it moves for reasons outside marketing entirely, like a stock outage or a price change or a seasonal slump.
Campaign return sees a narrow slice in high resolution. It knows which of two ad groups produced more revenue per dollar, and that is precisely the decision it should be used for. It cannot see cost of goods, other channels, or whether the customer would have bought anyway.
A rule built on MER cannot be satisfied by shrinking, which is exactly why we set the budget with it.
Reading the two together tells you what is happening
The interesting information is in the relationship between them, not in either number alone. Four patterns cover most of what we see.
Campaign return up, MER flat. The classic sign that spend has moved toward audiences that were already going to buy, usually brand searches or retargeting. Efficiency improved on paper and the business did not grow.
Campaign return down, MER up. This looks like failure and is often success. You are buying harder, more expensive customers, and the total is growing. This is what scaling looks like from inside the ad account, and it is the moment stores most often panic and cut.
Both falling. A real problem, and it is usually not in the ads. Check conversion rate, check for stock issues, check whether a price change or a shipping change landed recently.
Both rising. Either you found demand you were not serving, or the season is doing it for you. Compare with the same period last year before you take credit.
A worked example of the first pattern. Say marketing cost is $20,000 and store revenue is $100,000, so MER is 5. Next month the reported campaign return improves nicely, marketing cost is still $20,000, and revenue is $101,000. The campaigns got better at claiming credit. The business stayed still.
Which one sets the budget
MER sets the budget, because MER is the only one of the two that knows what the business can afford.
Work backwards from your own economics. Take your gross margin, decide what share of revenue you can spend on marketing while still covering overheads and leaving profit, and that gives you a MER floor. A store with a gross margin of five in ten that can afford to spend a fifth of revenue on marketing needs a MER of 5 to hold. That number is your line.
Then the rule is simple. While MER sits comfortably above the floor and orders are growing, spend more. When MER approaches the floor, stop adding and start fixing the other three levers, because they are what create room to spend.
This is a better growth rule than any campaign target, because it cannot be satisfied by shrinking. It also puts the conversation where it belongs, on whether the store can convert and retain the traffic it is buying, rather than on whether a platform is reporting favourably this week.
Which one decides where money goes
Campaign return decides the allocation inside the budget MER set. Once you know you can spend $20,000 this month, something has to say how much goes to Shopping, how much to social, and how much to the campaign selling your thin-margin range.
For that job, platform reporting is adequate and convenient, with two conditions. Compare campaigns within a platform rather than across platforms, because the measurement rules differ. And set the target for each group of products from its own break-even, not from a single store-wide figure, so that a low-margin product cannot quietly consume the budget a high-margin one should have had.
We also keep one guard rail on this. Any campaign whose reported result depends heavily on people who already know the brand gets judged on new customers acquired rather than on revenue claimed. Otherwise the allocation drifts, month after month, toward the campaigns that are best at taking credit rather than the ones that are best at finding buyers.
The reporting we would build
One page, monthly, with both altitudes visible.
At the top, the business view. Store revenue, total marketing cost across every platform and tool, MER, and new customer count. This is the section a founder reads to know whether the machine is working.
Underneath, the channel view. Spend and reported return by platform, with the same period last year alongside it, so seasonality is visible rather than argued about.
Underneath that, the four levers. Visitors, conversion rate, average order and repeat rate, each with the direction it moved. This is what explains the top section, because a MER that fell while spend held steady is almost always one of these four sliding.
What we leave out is as important. No blended figure that mixes attributed and unattributed revenue without saying so. No metric that only exists inside one platform. If a number cannot be reconciled with the store's own order data, it does not go on the page a decision gets made from.