Forget ROAS: Why Blended MER Is the Only Number That Matters for DTC Brands

Why platform-reported ROAS is structurally dishonest

Let's be blunt: every ad platform has a financial incentive to show you a high ROAS. Meta, Google, TikTok — they are all in the business of selling you more ad inventory. A flattering attribution model keeps budgets flowing. This isn't a conspiracy; it's just how incentive structures work. But understanding it should change how you read the numbers they hand you.

Platform-reported ROAS is calculated by dividing the revenue the platform claims credit for by the spend you gave it. The key word is "claims." Platforms use different attribution windows, different conversion events, and — critically — different rules about what counts as an assisted conversion versus a direct one. Meta's default 7-day click / 1-day view window will pick up customers who saw your ad, did nothing, searched your brand name on Google three days later, and bought. Meta counts that. So does Google. You have now paid for the same customer twice in your attribution reports, but only once in your bank account.

This double-counting is endemic. Studies across DTC brands consistently find that the sum of all platform-reported attributed revenue exceeds total actual Shopify revenue by 40–200%. A brand doing £100,000 in monthly revenue might see £170,000 in claimed attributed revenue across their channels. The platforms aren't lying exactly — they're each applying their own counting rules. But the aggregate picture they produce is fiction.

Beyond double-counting, platform ROAS ignores everything it didn't pay for. Organic search, email, word of mouth, PR, influencer posts you didn't pay for — none of it shows up in platform ROAS denominators or numerators in a way that helps you understand your real marketing leverage. When your email list drives a surge of repeat purchases during a sale, your Meta ROAS might spike because Meta's view-through attribution catches some of those buyers. You might increase your Meta budget as a result. The actual driver was your email list.

What the platform reports	What it actually means	The risk
Meta ROAS 4.2x (7-day click, 1-day view)	Revenue from users who clicked or viewed an ad in the last 7/1 days — including customers who would have bought anyway	Inflated by view-through; over-credits Meta for organic intent
Google Ads ROAS 5.8x (last click)	Revenue from users whose last click before purchase was a Google ad	Over-credits branded search; under-credits upper-funnel channels
TikTok ROAS 3.1x (7-day click)	Revenue from users who clicked a TikTok ad in the last 7 days	Often double-counts with Meta view-through on the same customers
Sum of all channels: 13.1x	A fiction — no brand generates 13x ROAS across all spend	Leads to systematic over-spending and false confidence
Blended MER (Shopify revenue / total spend)	Exactly what it says: real revenue divided by real spend	None — this is the ground truth

The attribution model disagreement problem

Even setting aside platform self-interest, attribution is a genuinely hard problem. A customer might discover you via a TikTok ad on Tuesday, click a Google Shopping ad on Thursday, open a promotional email on Saturday, and finally buy after a retargeting ad on Sunday. Who gets credit? The answer depends entirely on which model you use:

• Last-click: Sunday's retargeting ad gets 100% of the credit. TikTok, Google, and email get nothing. You cut TikTok spend. Top-of-funnel collapses three months later.
• First-click: Tuesday's TikTok ad gets 100% of the credit. You double TikTok spend. Retargeting ROI appears to collapse. You cut retargeting. Conversion rate tanks.
• Linear: All four touchpoints share equally. Tidier, but still arbitrary. Does a view really deserve equal weight to a click?
• Data-driven (Google's version): Google's algorithm decides, based on patterns in its own data. The algorithm is a black box. You cannot audit it. Coincidentally, it tends to credit Google touchpoints generously.

None of these models is right. They're all approximations. And crucially, each platform uses its own model on its own data, then hands you a number and calls it ROAS. You're making six- and seven-figure budget decisions based on four different models applied to four overlapping, uncoordinated datasets. This is not analytics. It's a guessing game with expensive consequences.

What blended MER is and how to calculate it

Blended MER — Marketing Efficiency Ratio — cuts through all of this noise with a formula so simple it almost feels naive:

Blended MER = Total Net Revenue ÷ Total Paid Ad Spend

That's it. No attribution windows. No model choices. No platform claims. You take your real Shopify revenue — gross revenue minus refunds — for a given period, and divide it by every pound you spent on paid advertising in that same period. Meta spend plus Google spend plus TikTok spend plus any other paid channel. Everything in. Total revenue out.

A blended MER of 3.5 means: for every £1 you spent on advertising, your business generated £3.50 in revenue. Not £3.50 attributed to ads by some algorithm. £3.50 actually deposited into your Shopify account.

This number cannot be gamed. No platform can inflate it. No attribution model can distort it. It is an accounting identity between two numbers you already have: your Shopify revenue report and your total ad spend invoice. The relationship between them is your blended MER.

What to include in "total ad spend"

• All Meta (Facebook + Instagram) ad spend
• All Google Ads spend (Search, Shopping, Display, YouTube)
• TikTok, Pinterest, Snapchat, and any other paid social
• Influencer fees where you paid for guaranteed posts
• Affiliate commissions (if performance-based and paid)
• Any paid partnership or sponsorship with a direct response goal

What you should not include: organic social time costs, SEO agency retainers (these are not direct response), or email platform SaaS fees. MER is specifically about paid acquisition and direct-response media.

MER target floors by gross margin tier

Knowing your MER is only useful if you know what it should be. The target depends on your gross margin, because MER is really a measure of whether your marketing is covering its cost within your unit economics. The breakeven MER formula is straightforward:

Breakeven MER = 1 ÷ Gross Margin

If your gross margin is 50%, your breakeven MER is 2.0. If your MER is below 2.0, every pound you spend on ads costs you more in variable margin than you recover. If your gross margin is 60%, your breakeven MER is 1.67. Higher margins give you more headroom; lower margins mean you need MER to do more work.

But breakeven is not the goal. You have overheads: payroll, software, rent, fulfilment ops. A healthy MER needs to cover not just the gross margin gap but also contribute to those fixed costs and ultimately to profit. Below are practical MER target floors by gross margin tier, based on typical DTC overhead structures:

Gross Margin	Breakeven MER	Healthy MER floor	Warning: review spend if below
45%	2.22	3.5 – 4.5	2.8
55%	1.82	2.8 – 3.8	2.2
65%	1.54	2.2 – 3.2	1.8

These ranges assume a DTC brand at moderate scale (£500k–£5m annual revenue) with typical overheads. If you are in a high-growth phase deliberately investing ahead of profitability, you might run a lower MER intentionally — but you should know you're doing it, and you should have a plan for when MER needs to recover.

Note also that MER will naturally fluctuate seasonally. Q4 tends to produce higher MER due to demand spikes, while January and July often see suppressed MER as CPMs remain elevated post-peak but organic demand cools. Your floor should be calibrated against your seasonal baseline, not a flat annual average.

How MER and MMM work together

Blended MER tells you the aggregate health of your marketing machine. It's the single best top-line signal for whether your spend is working. But it doesn't tell you which channel is driving it, or what happens if you shift £10,000 from Meta to Google.

That's where Marketing Mix Modelling (MMM) comes in. MMM uses statistical regression on your historical data — spend by channel, revenue, external signals like seasonality and promotions — to estimate the incremental contribution of each channel. Where MER is an accounting identity (unfakeable), MMM is a model (imperfect but directional).

Used together, they give you something neither can provide alone:

• MER as the guardrail: If your blended MER is healthy, you know the aggregate is working. Don't panic about any single channel's platform ROAS.
• MMM as the diagnostic: When MER deteriorates, use MMM to identify which channel's incremental contribution is declining. This separates signal from noise far better than platform dashboards can.
• MMM for budget allocation: Use MMM's channel response curves to find where the next marginal pound of spend generates the highest incremental return — then validate the change by watching MER in the following weeks.

This combination — MER as health signal, MMM as directional diagnostic — is how sophisticated DTC brands operate. You stop arguing about whether Meta's ROAS is real and start asking: "Did MER improve when we shifted budget? What does the model say should happen if we scale TikTok by 20%?"

The mindset shift is from attribution (who deserves credit for past conversions?) to forecasting (what will the next pound of spend do?). Attribution is backward-looking and contested. Forecasting is forward-looking and actionable.

Tracking blended MER in Nuso

Nuso's MER dashboard pulls your Shopify revenue and your connected ad channel spend into a single view, calculating blended MER automatically for any date range you choose. You can see daily, weekly, and monthly MER trends alongside your configured MER floor — so deterioration is visible the moment it starts, not weeks later when you're reviewing a spreadsheet.

The dashboard also breaks down spend by channel so you can see the composition of your denominator at a glance: if Meta's share of spend has crept up while MER has declined, that's a data point worth investigating. Pair this with Nuso's MMM tab, which runs incremental contribution modelling on your own historical data, and you have the full picture: ground truth on top, directional model underneath.

ROAS had a good run. It was useful when tracking was reliable and platforms weren't competing for attribution credit. That world is gone. Blended MER is what replaces it — and the brands already running on MER are making better decisions than the ones still arguing about attribution windows.