Performance Marketing KPIs in 2026: MER, POAS and Why ROAS Alone Lies to You

A client once showed us a dashboard with 6.2x ROAS on their top campaign and asked why the business still was not profitable. This happens more often than you would think, and it is rarely a reporting error - it is ROAS measuring the wrong thing. I am Manav Gupta, and here is the metric hierarchy we actually use to judge whether performance marketing is working, plus the specific ways ROAS misleads even sophisticated teams.

The citable answer: ROAS measures platform-reported return on a single channel's spend and ignores blended reality, discounts, returns, and overlapping attribution, so profitable brands in 2026 lead with MER (marketing efficiency ratio: total revenue ÷ total marketing spend) and POAS (profit on ad spend: contribution margin ÷ ad spend), reserving ROAS for in-platform optimisation only. Here is how the three metrics fit together, and how to actually build the reporting layer that connects them.

The Problem With ROAS Alone

Platform-reported ROAS has three structural blind spots. First, attribution overlap: if a customer sees your Meta ad, then converts from a Google search ad, both platforms can claim credit, inflating combined ROAS beyond what actually happened. Second, it ignores discounts and returns - a sale reported at full price that gets returned or was made at 30% off still counts as a "win" in ROAS math. Third, and most importantly, ROAS uses revenue, not profit - a 5x ROAS on a product with 20% margin can be less profitable than a 3x ROAS on a product with 60% margin.

None of this means ROAS is useless. It remains the fastest signal for in-platform optimisation decisions, like which ad set or creative to scale within a single channel. The mistake is using it as the business-level truth metric it was never built to be - it was designed as a media-buying tool, not a profitability tool, and most of the confusion we see comes from asking it to answer a question it was never built to answer.

The Three-Metric Hierarchy

Metric	Formula	Best used for
ROAS	Platform-attributed revenue ÷ platform ad spend	In-platform decisions: which creative/ad set to scale
MER	Total business revenue ÷ total marketing spend (all channels)	Business-level efficiency, immune to attribution overlap
POAS	Contribution margin ÷ ad spend	The truest read on whether marketing is actually profitable

We report all three to clients every week, specifically because they answer different questions. A campaign can show great ROAS, mediocre MER (because other channels are underperforming or overlapping attribution is inflating it), and weak POAS (because the product category has thin margins) - all three things can be true at once, and each tells you something the others do not. Reading only one of the three is like judging a business on a single line of its P&L.

Why POAS Is the One That Should Set Budget

POAS is the metric closest to the number a founder actually cares about: did this spend make money after real costs. Calculating it requires knowing your true contribution margin - revenue minus cost of goods, payment processing, shipping, and returns - not just gross margin off a spreadsheet. Once you have that number, POAS becomes a direct decision tool: a POAS above 1.0 means the marketing spend generated more profit than it cost; below 1.0, you are buying revenue at a loss even if ROAS looks strong.

Building this properly usually requires connecting ad platform data to real order-level profit data, which is exactly the kind of pipeline we set up in every data automation engagement - most brands are flying on ROAS alone simply because nobody has wired the profit data through yet, not because they do not care about it. The technical lift is smaller than most founders assume: it typically means pulling order-level cost data from an ecommerce platform or ERP and joining it against ad spend by date and campaign, which most modern stacks can support without custom engineering.

MER as the Sanity Check

MER is useful specifically because it cannot be gamed by attribution overlap between channels - it only cares about total revenue against total spend, full stop. If your combined platform-reported ROAS suggests the business should be growing fast but MER stays flat quarter over quarter, that gap is telling you attribution is overstating channel performance somewhere. We use MER as the check that keeps individual-channel ROAS honest, and it's often the first metric that reveals a problem before anyone notices it in the channel-level dashboards.

How the Three Metrics Interact in Practice

Scenario	What it means	Right response
High ROAS, flat MER	Attribution overlap is likely inflating channel-level numbers	Investigate cross-channel attribution before scaling that channel further
Good MER, weak POAS	Revenue is efficient but margin is thin on what's selling	Shift budget toward higher-margin SKUs, not just more volume
Strong POAS, declining MER	Profitable per-order but overall efficiency slipping (rising CAC elsewhere)	Check for a specific underperforming channel dragging the blended number down
All three trending together	A genuinely healthy, consistent read across the business	Safe to scale with confidence

A Worked Example: Building the Reporting Layer

For a client with only ROAS reporting in place, the build sequence usually looks like this: first, pull total business revenue (from the ecommerce platform, not ad platforms) and total marketing spend across every channel into one weekly view - that's MER, and it can usually be stood up within days since it needs no new cost data, just consolidation of numbers that already exist in different places. Second, work with finance or the founder to establish true contribution margin per SKU or product line, including payment processing fees and average return rate - this takes longer, often two to three weeks, because it requires pulling together cost data that's frequently scattered across supplier invoices, shipping accounts, and payment processor statements.

Once both are in place, POAS becomes a straightforward calculation layered on top. The reason we sequence it this way - MER first, POAS second - is that MER alone already catches a meaningful share of reporting blind spots, so a client gets real value within the first week while the more involved POAS groundwork is still being assembled.

Common Mistakes When Introducing MER and POAS

• Introducing all three metrics to a team at once without explaining the hierarchy. Teams used to ROAS alone can get confused seeing three numbers move differently; explaining upfront that they answer different questions prevents this.
• Using an estimated or rough contribution margin instead of the real number. A POAS calculation built on a guessed margin gives false confidence - it's worth the few weeks of work to get the real figure before trusting the metric for budget decisions.
• Abandoning ROAS reporting entirely once MER and POAS exist. ROAS still answers the specific, granular question of which ad set or creative to scale within a channel - it just shouldn't be the metric that sets the overall budget.
• Recalculating contribution margin infrequently. Cost of goods, shipping rates, and return rates all drift over time; a POAS model built on a year-old margin figure will quietly mislead budget decisions.

A Real Example

A client's Meta dashboard showed a healthy 4.8x ROAS, and they were preparing to scale spend another 50%. Our POAS analysis showed the actual contribution margin on their bestselling SKU was 22%, after true landed cost and return rate, meaning the real POAS sat at 1.05 - barely profitable, not the clear win the ROAS number implied. We redirected the planned scale-up toward a higher-margin SKU instead, and profit grew 31% over the next quarter on roughly the same total ad spend.

What made this catch possible wasn't a sophisticated model - it was simply having someone ask what the actual landed cost and return rate were for that specific SKU, a question the existing ROAS-only dashboard had no way to surface. The team had been scaling based on the metric they had, not the metric they needed, and neither was aware of the gap until the contribution margin data was pulled together.

FAQ

What is the difference between ROAS and MER?

ROAS measures return on a single platform's ad spend using that platform's own attribution, which can overlap with other channels and inflate the real picture. MER measures total business revenue against total marketing spend across every channel, giving an attribution-proof view of overall efficiency.

What is POAS and why does it matter more than ROAS?

POAS (profit on ad spend) divides contribution margin by ad spend, rather than revenue by ad spend. It answers the question ROAS cannot: after real costs, did this ad spend actually generate profit. A campaign can have strong ROAS and weak POAS if margins are thin.

Should I stop looking at ROAS entirely?

No - ROAS is still the fastest, most granular signal for in-platform decisions like which creative or ad set to scale. The fix is not abandoning ROAS, it is adding MER and POAS above it so budget-level decisions are made on profit, not platform-reported revenue.

How long does it take to build a MER and POAS reporting layer?

MER can typically be stood up within days, since it only requires consolidating revenue and spend data that already exists. POAS takes longer, often two to three weeks, since it requires gathering true contribution margin data (cost of goods, processing fees, returns) that's often scattered across different systems.

What's the biggest reason brands only track ROAS?

Usually it's not a lack of interest in profitability - it's that nobody has connected ad platform data to real order-level cost and margin data yet. Once that pipeline exists, POAS becomes a straightforward calculation rather than a major undertaking.

Get a KPI Stack That Reflects Real Profit

If your reporting stops at ROAS, you are likely making budget decisions on an incomplete picture. Book a call with Balistro and we will build a MER and POAS reporting layer that shows you what your marketing is actually doing to profit.