If your business is making paid media decisions on Marketing Efficiency Ratio without running incrementality tests, you’ve replaced one broken measurement system with a better-looking one. You haven’t fixed the problem.
Most DTC operators made the right move in the last two years. In-platform ROAS got harder to defend after iOS 14. Attribution windows started telling stories the P&L didn’t support. The platforms started taking credit for conversions they couldn’t have meaningfully caused, and the gap between what Meta and Google reported and what the bank account showed got too wide to explain away.
This is where teams are starting to (or have already), built MER into their dashboards, their reporting cadences, their board decks. They started anchoring budget conversations on a metric that ties cleanly to the income statement.
MER is a better number than platform-reported ROAS, and the teams that adopted it have an easier time talking to their founders and/or CFO about paid media than the ones still defending platform attribution every quarter and getting pushback every time.
Here’s where it gets uncomfortable. If MER is the only metric your team is reporting on, you’re about to find yourself in the same trap you just climbed out of. The math is cleaner. The decisions could still be wrong.
Why MER won the measurement argument
MER deserves the position it’s earned because it focuses on the bottom line. It ties directly to the P&L. Total ad spend in, total revenue out.
The math is simple enough that a founder, a CFO, and a performance marketer can all look at the same number and agree on what it says.
It also doesn’t move when the platforms change their reporting rules. When Meta updates its attribution model or Google rolls out a new bidding strategy that changes conversion logic, MER doesn’t change anything, it stays the same. It is shielded from the platform-level chaos that made in-platform ROAS unreliable in the first place.
That’s why MER is my biggest focus when we work with companies and it’s why people are looking at MER more and more. We just have to be careful not to make the same mistake again, of only looking at one Metric for business health.
The question MER can’t answer
MER measures efficiency. It does not measure causality. That is the whole argument and everyone needs to understand the difference before you treat MER as the answer.
A 4.5x MER tells you the business produced $4.50 in revenue for every $1 it spent on advertising. It does not tell you how much of that revenue would have happened without the ad spend. It cannot separate revenue your paid media caused from revenue that would have come in anyway through brand search, organic traffic, email, repeat customers, or word of mouth.
This isn’t a data problem. It’s the metric’s structural limit. MER is a ratio of two real numbers, cost of marketing and the revenue that came from it. What it can’t capture is if you had spent less or spent nothing, what would revenue have looked like?
The metric has no way to answer that, because there is not version of those metrics anywhere.
This question matters most at two moments, and both of them are moments where executive leadership is being asked to commit capital.
The first is when MER looks stable and the team wants to scale spend. The natural read is that the new spend is working, that it’s producing proportional revenue, that the system is healthy. Sometimes that’s true. Sometimes the new spend is buying conversions that would have happened anyway, and the MER stays steady because the denominator is moving for reasons that have nothing to do with the ads. You can’t tell the difference from the number alone. You’re being asked to commit more capital based on a metric that can’t tell you whether the last dollar produced a return.
The second is when MER drops and the team wants to cut spend. If the spend was driving genuinely incremental revenue, the cut will hurt the business more than the MER drop suggested. If the spend wasn’t incremental, the cut is fine and the MER recovers. Same MER drop, opposite responses warranted, no way to tell which scenario you’re in. You’re being asked to defund a channel based on a metric that can’t tell you whether that channel was actually working.
For mature brands, this gap gets bigger, not smaller.
The more brand demand, organic traffic, and repeat customer behavior a business has, the more revenue is being produced that gets credited to whatever ad spend is running at the same time. The MER stays healthy. The actual paid contribution shrinks. Eventually the brand is paying for revenue it would have earned anyway, and the metric can’t see it.
Where this breaks the business, not just the dashboard
The structural limitation isn’t theoretical. It shows up in three places that matter to anyone running a business, not just anyone running an ad account.
Budget allocation gets made on numbers that can’t predict returns. “MER is healthy, let’s increase the paid budget” is the most common decision. But if a meaningful portion of that healthy MER is non-incremental revenue, scaling spend won’t produce a proportional revenue lift. This is when the team commits to higher PPC spend, MER trails off, and revenue lift isn’t there. Then, leadership is surprised by a result the metric couldn’t have predicted, and trust erodes the trust in the next paid media forecast erodes. The same dynamic shows up when you look at it through the lens of unit economics — the LTV-to-CAC math gets worse before the headline metric reflects it.
Forecasting accuracy collapses without warning. Boards expect paid media spend to produce predictable revenue at predictable efficiency. Because MER can’t separate caused revenue from non-incremental revenue, the relationship between spend and revenue stops behaving the way the historical MER suggested it would. The forecast misses, often badly, and nobody can explain why because the metric they were forecasting on can’t see the problem.
The defense of the paid media budget gets shakier every quarter. “MER was 4.5x” is a clean answer until a founder or board member asks the right question: “If we hadn’t run paid media at all, what would have happened?” The team that’s only looking at MER can’t answer it. The team running incrementality tests can. The first conversation ends with budget cuts. The second one ends with budget commitment.
We use MER frequently and have run into some version of this where we thought advertising would drive more incremental revenue but the truth was as we spent more, we didn’t see the revenue increase proportionally. This is where you need to be cautious.
Efficiency and causality are not the same investment
The fix isn’t to replace MER, because MER is absolutely the right metric for what it measures.
The real fix is to commit to measuring what it can’t.
Incrementality testing answers the causality question MER can’t. Geo holdouts on Meta turn off ads in matched regions and measure what happens to revenue without them. Conversion lift studies on Google compare exposed audiences to control audiences. Structured incrementality testing across the funnel isolates portions of spend and tests them against a no-ad baseline. None of these techniques are new. What’s changed is the cost of running them has dropped while the cost of ignoring them has gone up.
For an executive team, this is a capital allocation question. Are you willing to give up some short-term efficiency to know whether your spend is actually causing your revenue? The teams that say yes get an honest answer to the question “what would happen if we cut paid media in half?” The teams that say no keep operating on a number that sounds confident and can’t survive a serious follow-up question.
A future piece will get into how to actually run an incrementality test without breaking the business. The point of this one is to land the principle before the tactics. MER is the foundation. Incrementality is the validation. A measurement system without both is a measurement system that doesn’t actually measure what leadership thinks it does.
The decision in front of you
If you’re a founder, a CFO, or a Head of Growth, the question to bring back to your team is direct. “Show me the math on how much of last quarter’s revenue was caused by our paid media, not just correlated with it.” If the answer is some version of “MER was 5x,” the team is reporting on a metric that can’t answer the question you just asked.
This isn’t the team’s fault and is a measurement gap most in DTC. But, you have to make a real decision to close the loop on it because if you don’t, the rest of the market will be the ones making better capital allocation decisions for the next few years.
If you’re running ecommerce PPC at any meaningful scale, this is the gap to close next. MER won the argument against in-platform ROAS. It deserved to. The next argument is whether MER alone is enough, and it isn’t.
About the Author: Tony Adam
