MER Instead of ROAS: How to Measure Paid Social When Attribution Crumbles

For years the case was clear: you look in the Ads Manager, read off the ROAS and know whether a campaign works. This reflex now leads you astray. The more strongly campaigns are steered by AI, the more unreliable platform ROAS becomes as the sole yardstick. Whoever keeps looking only at this one figure makes budget decisions on a shaky foundation.

The good news: there is a more robust way to measure. It combines MER as a north star, a profit view and simple incrementality tests. This guide shows why your ROAS lies, how MER works, when which metric counts and how you can test incrementality even as a mid-sized business.

Why your ROAS lies

Platform ROAS measures how much revenue a platform attributes to itself – divided by the spend on that platform. The problem lies in the first part: the attribution. Under automated campaigns the attribution boundaries blur. The systems deliver more broadly, capture users across various touchpoints and claim conversions for themselves that would have happened even without the ad.

This leads to a paradoxical effect: the reported ROAS can rise while your actual business success stagnates. The platform gets better at attributing revenue to itself – not necessarily at generating additional revenue. Whoever confuses these two things may scale a budget that isn’t working nearly as efficiently as the figure suggests.

An everyday example illustrates this: a loyal customer who wanted to reorder anyway sees a retargeting ad shortly before purchase and clicks on it. The platform books the entire revenue as its success – even though the purchase would very probably have happened without the ad. Multiplied by thousands of such cases, this creates a ROAS that impresses but says little about the actual contribution of the advertising.

A second blind spot is added: AI referral traffic is often misattributed in standard analytics. Visitors who come to your shop via AI systems wrongly end up under direct or organic traffic – so their contribution is systematically booked incorrectly. Why this source is growing and how to make it visible, we go into in the article ChatGPT instead of Google?.

MER explained

The answer to the attribution problem is to measure one level higher. Instead of asking what an individual channel claims for itself, you ask: how efficiently does my marketing work as a whole? That is exactly what the Marketing Efficiency Ratio (MER) measures:

MER = total revenue / total ad spend

MER is deliberately coarse. It ignores the question of which channel earned which conversion and looks only at the overall picture. This makes it immune to the attribution games of the individual platforms.

An illustrative worked example with fictitious figures: suppose in one month you make €100,000 in total revenue and spend €25,000 in ad spend across all channels together. Then your MER is 4.0 – for every euro spent, four euros of revenue come back. If in the following month you increase spend to €40,000 and revenue rises to €140,000, the MER falls to 3.5. The platform ROAS values may shine unchanged in the process – the MER honestly shows you that the additional spend worked less efficiently. (All figures are invented example values for illustration.)

This exact effect – falling marginal efficiency as budget rises – often stays invisible in platform ROAS and becomes immediately tangible in the MER.

When which metric counts

MER is the north star, but not the only figure you need. Three metrics complement each other:

• MER as an overarching efficiency compass for all marketing. Ideal for making budget decisions at the business level.
• Profit ROAS instead of pure revenue ROAS. Revenue is not profit. Whoever factors in margin, returns and variable costs sees whether a campaign is actually profitable – not just revenue-strong. Especially with thin margins, this view decides on survival.
• Incrementality tests, to answer the actual core question: how much of this revenue would have happened even without the advertising? Only the additional, incremental revenue justifies the spend.

The art lies in not drowning in one figure. MER sets the direction, profit ROAS grounds it in the reality of your margin, and incrementality checks whether the effect is real.

A practical side effect of the MER view: it uncovers double-counting. When you add up the self-reported revenues of all platforms, you often arrive at a value that exceeds your actual total revenue – because several channels attribute the same conversion to themselves. The MER can’t do that, because it starts from the real total revenue. Even this reconciliation – the sum of platform revenues against the real revenue in the shop backend – is a sobering but salutary first step. It shows in black and white how much “revenue” in the channel reports is double-booked.

Simple incrementality tests for mid-sized businesses

Incrementality sounds like a big data-science apparatus – but it doesn’t have to be. Even with a limited budget you can gain reliable directional indications:

Geo comparisons. Advertise more or less strongly in comparable regions and observe whether total revenue develops measurably differently in the more heavily served regions. This approaches a clean experiment without complex technology.

Controlled pauses. Deliberately switch off a campaign or a channel for a defined period and observe what happens to total revenue. If it stays stable, the attributed revenue was perhaps less incremental than the ROAS claimed. This exact approach also helps to honestly assess the contribution of a complementary channel like Snapchat, instead of relying on its in-platform ROAS.

Stepwise budget changes. Raise or lower the spend in clear steps and read the MER development alongside. That way you recognise at which point additional spend loses efficiency.

None of these tests is perfect. But all are better than blindly trusting a platform figure. It’s about a reliable direction, not scientific precision.

Reporting setup in practice

For this view to become routine, you need lean reporting that places the business figures next to the platform figures. In practice this means: keep MER and profit ROAS as fixed metrics alongside platform ROAS, pull together the total spend across all channels and take the revenue from the most reliable source – usually the shop backend – as the reference figure.

A detail that is often overlooked: the quality of your landing page feeds into the delivery. The post-click behaviour of users is a signal the systems read along with. A weak landing page therefore worsens not only the conversion but also the delivery – another reason why pure channel thinking falls short.

Just as important as the metrics is the rhythm in which you read them. Daily glances at the MER lead you astray, because individual days fluctuate strongly. A weekly and monthly view makes more sense, in which trends show rather than daily noise. And with every larger budget decision you ask the same question: how has the MER developed since I last changed something – and did the additional revenue come with stable or with falling efficiency? This one question disciplines scaling more than any dashboard.

Clean reporting also depends on clean signals. How closely measurement, tracking and AI-driven delivery are connected, we show in the article Understanding Meta Andromeda. And because your product data also pays into efficiency, a look at our Feed Hacking is worthwhile.

Conclusion

Platform ROAS is no longer a reliable north star under automated campaigns. Whoever wants to steer modern paid social budgets measures one level higher: MER as an efficiency compass, profit ROAS for the margin truth and simple incrementality tests for the question of what is really additional. This combination is more uncomfortable than a glance in the Ads Manager – but it protects you from putting budget into a nice-looking figure that has little to do with your real growth. Whoever internalises this way of thinking makes better decisions and is also better positioned when it comes to the question of whether an agency or an in-house team should be responsible for the budget. This is exactly how we steer accounts in our Meta Ads agency.