Platform attribution is not measurement. It is marketing. Every ad platform reports the number that makes it look essential, and when you add those numbers together, they routinely exceed your actual revenue. That gap is a rounding error only if you ignore the structural incentive of closed ecosystems grading their own homework.
A different anchor solves this: one that lives in your own books and cannot be touched by platform measurement choices.
In brief: Marketing Efficiency Ratio (MER) is total revenue divided by total marketing spend, pulled from your own books, with no platform pixels involved. Because it uses a single numerator that cannot be double-counted, it gives growth-stage brands a stable read on whether the whole system is working. Used alongside incrementality tests, MER tells you both the score and which levers actually moved it. Brands that build around this anchor make budget decisions from evidence rather than from whichever platform sent the most convincing PDF.
Marketing Efficiency Ratio is total revenue divided by total marketing spend, calculated from first-party financial records rather than platform-reported conversions.
1. Platform ROAS is structurally incapable of telling you the truth
Every major ad platform attributes conversions using its own measurement window, its own identity graph, and its own definition of what counts as an assist. When a customer sees a Meta ad on Tuesday, a Google search ad on Thursday, and converts on Friday, both platforms claim the sale. Your revenue line counts it once.
According to LatentView Analytics, the core problem in retail and consumer marketing measurement is proving incremental impact rather than simply reporting attributed conversions. The distinction matters because attributed ROAS answers "did someone who saw our ad buy something?" MER answers "did spending this money grow the business?" Those are different questions, and only one of them compounds.
If your blended ROAS across channels is 4.2 but your MER is 2.1, the math is telling you something. The platforms are each telling a partial truth that adds up to a fiction.
2. MER gives you a number that cannot be gamed
The formula is straightforward: take total revenue from your accounting system, divide by total marketing spend from your budget, and you have MER. No pixels. No attribution windows. No last-click versus data-driven arguments.
This simplicity is the point. Because the numerator comes from your books, no platform can influence it by widening a conversion window or changing a default attribution model. When Chobani leaned on media mix modeling to optimize its commerce media allocation, the underlying discipline was the same: measure from the outcome backward, not from the platform outward. MER is the fastest version of that discipline available to a brand that does not yet have a full MMM infrastructure.
For B2B brands measuring in pipeline rather than carts, the same logic applies. Replace revenue with sourced pipeline or closed-won ARR, keep the denominator as total marketing spend, and the ratio still gives you a platform-agnostic read on system efficiency.
3. Incrementality tests are what MER cannot do alone
MER tells you the score but leaves the question of which player scored unanswered. If you cut paid social spend by 30 percent and MER holds steady, you have learned something. If it drops proportionally, you have learned something different. But running the experiment blind is slow and expensive.
Incrementality testing, whether geo-holdout, matched market, or intent-to-treat design, isolates the causal contribution of a specific channel or campaign. According to Osmos AI's 2026 retail media measurement guide, closed-loop attribution in retail media depends on matching impressions to purchases through first-party transaction records, which is structurally similar to how a proper incrementality test is constructed: you need a control group that did not see the ad, not just a population that did.
The practical workflow: use MER as your weekly operating metric to catch system-level problems early, and run incrementality tests quarterly to calibrate which channels are actually moving the numerator.
4. The privacy shift makes platform attribution less reliable every year
Third-party cookies are largely gone. Signal loss from iOS privacy changes has been compounding since 2021. The tracking architecture that made last-click attribution plausible is being dismantled by browser policy, consent law, and platform competition simultaneously.
Recent analysis of attribution technology trends notes that the direction of travel is away from single-platform pixels and toward blended, probabilistic systems. Those systems are better than what came before, but they are still estimating. MER reads from the ledger rather than estimating.
As signal degrades further, the brands anchored to MER will have a stable operating metric while brands anchored to platform ROAS chase a number that is increasingly modeled by the platform with the most to gain from a favorable result.
5. Commerce media scale makes the double-counting problem worse, not better
According to EMARKETER's June 2026 forecast, commerce media ad spending will reach $83.71 billion in the US this year, accounting for 20.5 percent of all digital ad spending. That is a lot of closed-loop attribution claims from a lot of retailers, each with their own measurement methodology and their own incentive to show strong ROAS to their advertiser partners.
Retail media networks are under pressure to prove that their data improves advertising outcomes beyond their own owned channels. That pressure produces optimistic measurement. When you are buying across three retail media networks and two social platforms, the sum of their reported ROAS will not reconcile to your revenue. MER will.
6. MER compounds because it changes how you make decisions
The measurement anchor you choose shapes the decisions you make, and those decisions compound over time. A team optimizing to platform ROAS will shift budget toward channels that report well. A team optimizing to MER will shift budget toward channels that actually grow revenue, confirmed by incrementality tests.
The unified measurement frameworks gaining traction in 2026 treat MER, incrementality, and media mix modeling as a layered stack, not competing alternatives. MER is the top-line efficiency signal. Incrementality testing is the causal layer beneath it. MMM is the long-run planning layer beneath that. You do not need all three on day one, but you do need to start with an anchor that cannot be manipulated by the platforms you are measuring.
Begin with MER, add incrementality tests once you have enough spend to run clean experiments, and build toward MMM when the business warrants it. Each layer makes the one above it more useful for decisions.
The brands that will look back at 2026 as the year they got measurement right are the ones who stopped letting platforms set the terms of the conversation and built their operating cadence around a number that lives in their own books. MER is a straightforward ratio, but choosing it consistently means every budget decision starts from your revenue ledger rather than from whichever platform dashboard tells the most flattering story. Over time, that starting point is the difference.
For more on how growth-stage brands build durable marketing systems, the Method blog covers the decisions that separate compounding programs from expensive experiments.
Frequently asked questions
What is Marketing Efficiency Ratio and how do I calculate it?
Marketing Efficiency Ratio (MER) is total revenue divided by total marketing spend, where both numbers come from your own financial records rather than platform dashboards. To calculate it, take your revenue for a given period from your accounting system and divide by every dollar of marketing spend in that same period. The result is a platform-agnostic ratio that shows whether your overall marketing system is generating returns, with no attribution modeling involved.
Why is MER better than ROAS for measuring marketing performance?
ROAS is reported by individual platforms using their own attribution windows and identity graphs, which means multiple platforms routinely claim credit for the same sale. MER uses a single revenue number from your books that cannot be double-counted. It does not replace channel-level data, but it gives you a stable top-line signal that no platform can inflate by changing its default measurement settings.
How do I use MER alongside incrementality testing?
Use MER as your weekly or monthly operating metric to track whether the whole system is working. Run incrementality tests, such as geo-holdout or matched-market experiments, to identify which specific channels are causally driving revenue. MER tells you if the score changed; incrementality testing tells you which lever moved it. Together they give you both the outcome and the explanation.
Can MER work for B2B brands that measure pipeline instead of revenue?
Yes. Replace revenue with sourced pipeline or closed-won ARR in the numerator, keep total marketing spend in the denominator, and the ratio functions the same way. The logic holds for any business where a single financial outcome can be read from internal records and compared to total spend, regardless of whether the buying unit is a consumer or a committee.
What is the right MER benchmark for a growth-stage brand?
There is no universal benchmark because MER varies by category, margin structure, and growth stage. A brand with high gross margins can sustain a lower MER than one with thin margins. The more useful practice is to track your own MER over time, run incrementality tests to understand what moves it, and use the trend rather than an external number as your decision signal. Your MER from six months ago is a more honest benchmark than an industry average.