Est. reading time: 5 minutes
Attribution has become a knife fight nobody wins. Meta claims a purchase, GA4 credits it elsewhere, the platforms’ numbers sum to more revenue than the store actually made, and the weekly meeting spends its energy adjudicating instead of deciding. MER exists to end that meeting. Total revenue divided by total marketing spend, one blunt ratio that ignores platform credit entirely and answers the question underneath all the squabbling, which is whether the business’s whole marketing motion is turning money into revenue at a rate worth continuing. Blunt is the feature. The catch is that a blunt instrument used sloppily is just a wrong number with confidence, so the definitions have to be exact.
What MER is, and the standards that keep it meaningful
MER is a top-down portfolio metric. It isn’t channel ROAS, which is attributed and partial. It isn’t CAC. It captures the all-in effect of the ecosystem, paid, organic ripple, email harvesting demand that ads created, precisely because it refuses to assign credit. Its job is one question: at this level of total spend, is total revenue sufficient?
That only works if the two inputs are standardized, because “revenue” and “spend” are both negotiable terms until someone writes them down. Our recommended definitions: revenue means net revenue, gross sales minus discounts, refunds and returns, sales tax, and gift card redemptions counted at redemption, with shipping collected excluded. Spend means paid media plus the variable marketing costs that scale with it, platform fees, affiliate commissions, creator payments including product at cost, and agency fees, with fixed headcount and brand overhead excluded unless you deliberately report a second, broader version. Name the standard you’re using, MER-Media versus MER-Total, document it, and never mix the two in one report, because a MER that quietly changed definitions between quarters isn’t a trend, it’s an accident wearing one.
The math, including the number most teams skip
The equation is net revenue divided by marketing spend, and its inverse, spend divided by revenue, reads as the marketing cost per revenue dollar. A MER of 4 means 25 cents of marketing bought each dollar of sales. Simple enough that the real question isn’t the formula, it’s the threshold, and the threshold comes from your margin structure, not from a benchmark article.
Break-even MER equals one divided by contribution margin, where contribution margin is what’s left after variable costs, COGS, payment fees, pick-pack-ship, but before marketing. At 40 percent contribution margin, break-even MER is 2.5, meaning a 3.0 that would thrill a software company is a thin quarter for a brand at 30 percent margin, whose break-even sits at 3.33. Covering fixed overhead has no universal formula, so model it directly, profit equals contribution margin times revenue, minus marketing, minus fixed costs, and solve scenarios. And retire the myths while you’re at it: MER is not platform ROAS, it is not “sales over all costs,” and it should never be computed on gross sales with tax and unnetted returns inside, because each of those errors produces a number that looks like MER and steers like a rumor.
Audit the inputs before trusting the gauge
MER dodges attribution bias and remains fully exposed to data hygiene. Align time zones, booking logic (order date versus shipped date), and currency conversion identically across revenue and spend, and learn your calendar’s rhythm, since real-time spend against delayed revenue makes MER sag midweek and recover later, a pattern to recognize rather than react to.
Sanitize the revenue side. Net revenue as defined above, with wholesale, B2B, and one-off revenue that marketing didn’t influence excluded, and a return-adjusted factor by cohort when returns land weeks after the orders they undo. Complete the spend side with everything variable, and then flag the shocks the ratio can’t see, press hits, product drops, stockouts, viral moments, because MER will happily credit marketing for a TV mention or blame it for a warehouse outage, and an annotated timeline is the difference between reading the gauge and being fooled by it.
Run it weekly, and watch the margin, not just the average
Weekly is the operating cadence, since daily MER is noise and monthly is autopsy. Pair the weekly number with a trailing four-to-eight-week average and the same week last year, so trend, seasonality, and momentum are all visible in one view, and set an explicit guardrail, something like never below break-even on the four-week average, so the ratio has a tripwire instead of just a history.
The sharper tool is marginal MER, incremental revenue divided by incremental spend, because averages hide the decision that matters. An account can hold a healthy average MER while the last dollars spent return below break-even, which means the average is being subsidized by the early, efficient spend and the marginal budget is quietly destroying profit. A simple simulator, budget in, contribution margin and an expected revenue curve, projected MER and profit out, turns budget planning from vibes into scenarios, and the marginal read is what tells you whether the next dollar is accretive or decorative. The channel-level version of this profitability lens is one we’ve built before in measuring campaign profitability in Google Analytics.
Segment where the answer changes the action. New-customer MER, new-customer revenue over spend, is the growth-quality gauge, since a blended MER propped up by returning customers can mask an acquisition engine that stopped working. Returning MER reads retention efficiency. Compare against prior-year weeks and around promotions, use alert bands to catch drift early, and when MER moves in a way that matters, confirm the cause with incrementality tests before reallocating, because the ratio tells you that something changed, never by itself why.
Defined once, standardized in writing, and enforced weekly with a margin-derived threshold, MER becomes the number that ends arguments instead of starting them, the gauge that says push, hold, or trim while the attributed metrics fight about credit underneath it. The platforms will keep disagreeing about who caused what. The bank account and the ad invoices, which is all MER reads, never do.
