The marketing efficiency ratio (MER) is your total revenue divided by your total marketing spend over the same period, so $1 million of revenue on $200,000 of spend is an MER of 5. Because it divides all sales by all spend, it’s also called blended ROAS, and its appeal is that it needs no attribution model at all: nobody can argue about which platform deserves credit when the numerator is simply what landed in the bank. Its weakness is the same thing seen from the other side, which is that MER counts every sale as if marketing caused it.
I’ve worked with a lot of direct-to-consumer brands, both at my agency and later building marketing mix models for brands like Breeo and Groove Life, and in my experience MER is the number many DTC founders actually run the business on. It tends to sit on the daily report, it’s what boards ask about, and it’s usually the first number cited when someone wants to cut or raise the budget. I think it’s a sensible guardrail. However, I’ve also watched it reward the wrong decisions, and it’s worth understanding exactly why before you put a target on it.
How to calculate the marketing efficiency ratio
The marketing efficiency ratio formula is:
over a matching period, usually a day, week or month. If a brand books $1.2 million of revenue in a month and spends $200,000 across Meta, Google, TikTok, affiliates and influencers, its MER is 6.0. Some teams express it the other way round, as marketing spend as a percentage of revenue, in which case the same brand is spending about 16.7 percent of revenue on marketing.
The arithmetic is trivial, but the definitions aren’t, and two brands quoting “MER” are often measuring different things. On the revenue side you need to decide whether it’s gross sales or net of discounts and returns, whether it includes marketplaces such as Amazon and wholesale, and whether it’s total store revenue or just the channels marketing can influence. On the spend side, the question is whether it’s ad spend only or all marketing costs, including agency fees, creator payments, content production and the marketing team’s salaries. None of these is wrong, but the definition should be written down and kept fixed, because changing it quietly can move MER by more than any campaign ever will.
Why DTC operators use MER instead of platform ROAS
The rise of MER among ecommerce operators was largely a reaction to attribution getting worse. Platform-reported ROAS has always over-claimed, since every ad platform credits itself with any conversion it touched and the totals across platforms routinely add up to more than the revenue the business actually made. Once Apple’s App Tracking Transparency rules and other privacy changes reduced the data platforms could see, those numbers also became noisier and more model-driven. A metric calculated from the store’s own revenue and the finance team’s own spend figure sidestepped the whole argument.
It also has a practical virtue: MER reconciles with the P&L. Finance can check it, it can’t be inflated by changing an attribution window, and it captures effects attribution misses, such as a TikTok campaign that drives people to search for the brand on Google a week later, or an influencer post that shows up as “direct” traffic. If your marketing is working in ways that are hard to track, MER will at least pick up the total effect somewhere.
| Metric | Numerator | Denominator | Needs attribution? | Counts sales ads didn’t cause? |
|---|---|---|---|---|
| Platform ROAS | Revenue the platform attributes to its ads | That platform’s spend | Yes | Yes, often heavily |
| MER (blended ROAS) | All revenue | All marketing spend | No | Yes, all of them |
| aMER | New-customer revenue | All marketing spend | No | Yes, organic new customers |
| Incremental ROAS | Revenue the ads caused | Spend | No, needs a model or test | No |
| Marginal ROAS | Change in incremental revenue | Change in spend | No, needs a model or test | No |
What MER hides
The biggest thing MER hides is baseline revenue, meaning the sales you’d make if you switched off all paid marketing tomorrow: repeat customers, word of mouth, organic search, people who already know the brand. For a mature brand with a loyal customer base, that baseline can be most of the revenue. MER divides all of it by the marketing budget, which makes marketing look far more productive than it is. In the course’s Northstar exercise, a channel with incremental ROAS of £1.25 shows a total-sales ROAS of £3.75 when baseline sales are £100k, and £5.75 when they’re £180k, even though the ads are doing exactly the same work in both cases. A rising MER can simply mean your baseline sales went up.
It also hides the split between new and returning customers. A business can hold a steady MER for a year while new customer acquisition quietly collapses, as long as repeat purchases from earlier cohorts keep the top line up, and by the time the problem shows up in total revenue it’s months old. And it hides channel mix: a blended number can’t tell you that one channel is doing all the work and another is wasting a third of the budget, which is the decision you actually need to make most weeks.
The most dangerous property is that MER rises when you cut spend. Revenue doesn’t fall one-for-one when the budget falls, partly because the baseline doesn’t depend on spend and partly because the dollars you remove first are the least efficient ones on a saturating response curve. So MER goes up after almost any cut, including cuts that destroy profit. If MER is the target, the easiest way to hit it is to spend less, and a team managed on MER will drift towards under-investment without anyone deciding that’s the strategy.
With the defaults, a 20 percent cut lifts MER from 5.86 to 6.98 and costs about $55k of incremental revenue. In this example the cut actually improves contribution after advertising, because the marginal ROAS at current spend (1.22) is well below the breakeven of 2.22 for a 45 percent margin, so the money being cut was losing money. Now set current spend to $100k and try the same cut: MER still rises, from 10.0 to about 11.85, but contribution falls, because those dollars were returning more than breakeven. MER moves the same way in both cases, which is exactly the problem.
aMER, nMER and other MER variants
The most common variant is aMER (acquisition MER), which divides revenue from new customers by total marketing spend. It’s a big improvement for growth decisions, because it removes repeat purchases from existing customers, which is often the largest part of the baseline, and it falls when acquisition gets less efficient even if total revenue holds up. Many DTC brands now report MER and aMER side by side. It’s still a blended number, though: it counts every first-time buyer, including people who found you through a friend, a press mention or organic search and would have bought without seeing an ad.
You’ll also see “nMER” and “new-customer MER” used, but I couldn’t find a consistent definition. Most people seem to use them as another name for aMER, while the agency Darkroom’s formula table defines new-customer MER as total revenue divided by new-customer acquisition spend, which is a different number altogether. If someone quotes nMER to you, ask what’s in the numerator and the denominator before comparing it with anything. The same goes for “contribution MER” or “MER after returns”: they’re all reasonable, as long as the definition travels with the number.
How MER relates to marketing mix modeling
A marketing mix model splits the same revenue that goes into MER into a baseline and an incremental contribution from each channel. That means you can write MER as two terms: . The first term has nothing to do with how well the marketing works. It rises when the baseline grows or when spend falls, and it’s why MER can be high while incremental ROAS is below breakeven. The second term is the average return the ads actually generate, and the marginal ROAS is the return on the next dollar, which is the one that should drive budget changes.
I think of MER and MMM as answering different questions. MER is a quick, reliable health check that the business can calculate every morning; MMM is a slower causal estimate that tells you what the marketing is doing and where the next dollar should go. They’re also useful checks on each other. If the model says a budget cut will reduce revenue by $50k and MER barely moves, that’s interesting; if MER drops sharply after a change the model said was harmless, the model may be missing something, such as a channel’s long adstock tail or a shift in the baseline. Where the two disagree, an incrementality test is usually the cheapest way to settle the argument.
Setting an MER target that means something
If you’re going to manage to MER, at least derive the target from your economics rather than a benchmark from someone else’s business. Total contribution after marketing is contribution margin times revenue minus marketing spend, so with a contribution margin of you need an MER of at least for marketing to be covered by the contribution of all sales, and a higher number to cover fixed costs and leave a profit. With a 45 percent margin, that floor is 2.22, but because MER includes baseline sales, clearing it says very little about whether the advertising itself pays back. A brand could clear it comfortably while spending half its budget on campaigns with an incremental return of 1.
So I’d use MER as a guardrail, not a steering wheel. Set a floor that protects the P&L, track aMER alongside it so you can see acquisition separately, and make the actual budget decisions on incremental and marginal returns from a model or an experiment, compared with your breakeven ROAS. If the budget conversation is about how to split money between channels, a budget allocator built on those curves will tell you far more than any blended ratio. And when MER improves after a cut, the useful question is what the revenue you gave up cost you in contribution, which is the question the Vexpower MMM course keeps coming back to.
