30–50% Daily ROAS Swings? Use MER to Set Your Ecommerce Budget

August 29, 2026

MER tells you whether your marketing engine is healthy overall; ROAS tells you where inside that engine to push or pull. Use MER (total revenue divided by total marketing spend) to size your budget and report to leadership. Use ROAS (attributed revenue divided by platform ad spend) to decide which campaigns, ad sets, or creatives deserve the next dollar. Neither replaces the other. You need both, checked on different clocks.

TL;DR:

Blended MER, including all revenue and marketing costs, guides overall budget size and indicates if the business is profitable, regardless of channel performance.

Platform-specific ROAS measures attribution within a single campaign or platform but can be distorted by modeling assumptions and attribution overlaps.

Use MER to assess overall health and set spending ceilings, then rely on ROAS to optimize allocation among campaigns, especially when adjusting creative or audiences.

Quarterly incrementality tests, such as geo holdouts, verify what actual lift platforms contribute beyond claimed attribution, preventing misdiagnosis.

Combining fast-reacting ROAS signals with slower, comprehensive MER analysis ensures brands avoid reactive overreaction and maintain long-term profitability.

ROAS vs MER: What MER Measures and Why It Matters
ROAS vs MER: How Platform Attribution Shapes Return
MER vs ROAS Comparison: Strengths, Weaknesses, and What Each Answers
When to Use MER vs When to Use ROAS: A Decision Guide
How to Calculate MER and ROAS Variants: Worked Examples
Common Traps: Diagnosing ROAS vs MER Mismatches
Operating Cadence: A Weekly, Monthly, Quarterly Playbook
Practitioner Checklist: Rules I Use on Meta Programs
Balancing Fast Signals Against Long-Term Health
How My Team at Cosma Builds Measurement Systems That Hold Up
Sources
FAQ

ROAS vs MER: What MER Measures and Why It Matters

MER strips away platform noise and answers one question: is the business making money relative to what it spends on marketing? Period. The formula is simple: total revenue ÷ total marketing spend. If your store did significant revenue last month and you spent a notable amount across Meta, Google, affiliates, and email tools, your MER reflects how many dollars you earned per dollar spent. No attribution model, no click windows, no platform dashboard involved. Just the money in and the money out, which is exactly why MER is the number that should run your budget rather than any single channel’s reported return.

That simplicity is also MER’s limitation on its own. A blended MER of 5.0 doesn’t tell you whether Meta prospecting is dragging performance while branded search carries it. That’s a job for other metrics, which is why practitioners lean on a few variants depending on the question they’re asking.

Blended MER covers all revenue and all marketing costs, including tools, agency fees, and influencer payouts. This is the number your CFO wants.
a-MER (acquisition-only MER) isolates new-customer revenue against acquisition spend, stripping out retention and email revenue that would otherwise flatter the ratio. Use it when you want to know if your prospecting engine, not your existing customer base, is actually profitable.
Contribution-margin MER replaces revenue with contribution margin (revenue minus cost of goods, shipping, and payment fees) in the numerator. For businesses with wide margin variance across SKUs, revenue-based MER can look great while the business bleeds cash. Contribution-margin MER is the version that ties directly to profit, and I push most of my ecommerce clients toward it once they have clean COGS data.

Blended revenue MER would tell that brand it’s healthy while it quietly funds unprofitable growth.

ROAS vs MER: How Platform Attribution Shapes Return

ROAS answers a narrower, faster-moving question: for the dollars I put into this specific platform or campaign, how much attributed revenue came back? The formula is attributed revenue ÷ ad spend, calculated inside a single account. Spend $10,000 on a Meta campaign and Meta’s ads manager reports $35,000 in attributed purchases, and you have a 3.5x ROAS on that campaign, by that platform’s own accounting.

The catch is the phrase “by that platform’s own accounting.” ROAS is platform-attributed revenue divided by platform ad spend, and every platform decides differently what counts as attributed. Meta’s reporting still leans on modeled conversions since Apple’s iOS 14 tracking changes limited direct pixel visibility, meaning a chunk of the revenue in your ROAS number is Meta’s statistical best guess, not a confirmed click-to-purchase path. Meta’s reported ROAS is partly modeled rather than directly observed, which is exactly why two platforms can each claim credit for the same sale.

That doesn’t make ROAS useless. It makes it a relative signal, not an absolute one.

Use ROAS to compare one campaign against another inside the same platform, where the attribution rules stay consistent.
Use ROAS to judge whether a creative refresh, a new audience, or a bid strategy change moved performance week over week.
Don’t use raw ROAS to compare Meta against Google against TikTok, since each platform’s attribution window and modeling assumptions differ enough to make the comparison meaningless.

Treat platform ROAS the way Shopify’s own guidance frames it: watch the trend, not the exact number, and dig into the underlying levers (CTR, conversion rate, CPC) when it moves.

MER vs ROAS Comparison: Strengths, Weaknesses, and What Each Answers

Put side by side, the two metrics aren’t competing for the same job. MER answers “should we spend more or less overall?” ROAS answers “which campaign or ad set earns the next dollar?” Confusing the two is how brands end up cutting a profitable channel because its platform ROAS dipped, or overspending because blended MER looked fine while one channel quietly burned cash.

Scope: MER covers the whole business; ROAS covers a single platform or campaign.
Sensitivity to tracking changes: MER is largely immune to attribution shifts, since it’s calculated from actual revenue and actual spend. ROAS moves whenever a platform updates its attribution model or measurement window.
Speed: ROAS updates daily and reacts fast to creative fatigue or audience saturation. MER lags, since it depends on full revenue reconciliation across every channel.
Reporting audience: MER is the number for founders, CFOs, and board decks. ROAS is the number for media buyers making Tuesday-morning budget shifts.
Risk of misuse: MER can mask a struggling channel inside a healthy blend. ROAS can mask overlap, where two platforms both take credit for the same conversion.

Neither metric is “better.” They’re built for different decisions, and treating one as a universal scoreboard is the most common measurement mistake I see in ecommerce accounts.

When to Use MER vs When to Use ROAS: A Decision Guide

Here’s the order I walk through with clients, and it holds regardless of account size.

Start with MER against your margin target. If contribution-margin MER is above your breakeven threshold, you have room to invest. If it’s below, you have a spending problem before you have a channel problem.
Check platform ROAS by campaign, not by account. Account-level ROAS averages hide winners and losers. Break it out by campaign objective (prospecting vs retargeting) at minimum.
If MER and ROAS disagree, run diagnostics before touching budget. A campaign showing 4x ROAS while blended MER slides usually points to overlap between platforms, not a genuinely underperforming account.
Use MER to authorize the total number. Boards and finance teams should see MER trends monthly, not a spreadsheet of 12 campaign ROAS figures they can’t contextualize.
Use ROAS to decide which ad sets get that authorized budget. Once the total is set, ROAS (and its underlying levers) governs allocation inside the platforms.

Pro Tip: Never cut a campaign on a single day’s ROAS dip. Pull a 7-day trailing average first. Daily ROAS swings 30% to 50% on normal accounts just from purchase timing and platform reporting lag.

The rule of thumb I give founders: MER sets the ceiling, ROAS decides who gets to spend inside it.

How to Calculate MER and ROAS Variants: Worked Examples

Numbers make this concrete. Say your Shopify store recorded $420,000 in revenue last month, and your marketing costs (all paid platforms, agency fees, and software subscriptions) totaled $84,000.

Blended MER = $420,000 ÷ $84,000 = 5.0
If $60,000 of that revenue came from returning customers and email flows, new-customer revenue is $360,000. If acquisition spend (excluding retention tools) was $78,000, a-MER = $360,000 ÷ $78,000 = 4.6
If your average contribution margin is 55%, contribution dollars from that $420,000 total roughly $231,000. Contribution-margin MER = $231,000 ÷ $84,000 = 2.75, a very different picture from the 5.0 headline number.

For the overlap check: sum every platform’s self-reported ROAS times its spend to get “claimed revenue,” then compare that total against actual Shopify revenue for the same period. Platform-reported ROAS commonly sums to 1.4x to 1.8x true blended MER on mature accounts because Meta, Google, and TikTok each claim credit for shared conversion paths. If your claimed revenue is running 1.6x actual revenue, that’s normal overlap, not a fraud alarm, but it’s exactly why platform ROAS alone should never set your total budget.

Common Traps: Diagnosing ROAS vs MER Mismatches

When MER climbs but a specific platform ROAS drops, check attribution and overlap first, not creative fatigue. Something is redistributing credit, not necessarily hurting performance.

Hands adjusting budget tokens on desk

When platform ROAS looks strong but blended MER stays flat or falls, that’s the classic overlap tax showing up. You’re likely paying multiple platforms to fight over the same customer, and the “growth” one dashboard shows isn’t incremental at all.

Work through these checks in order:

Attribution window: confirm you’re comparing the same window (1-day click, 7-day click, etc.) across time periods before concluding a platform’s performance actually changed.
Cross-channel overlap: pull retargeting spend separately from prospecting spend. Retargeting ROAS is almost always inflated because it’s converting demand you already created elsewhere.
Conversion event integrity: verify your pixel or conversion API is firing on the right event, at the right value, especially after any site or checkout change.
Lever check before budget check: most ROAS drops trace back to a broken CTR, conversion rate, or CPC, not a fundamentally broken channel. Fix the lever before you cut the spend.

Pro Tip: When numbers won’t reconcile, split acquisition and retention into separate reporting views. Blending them into one ROAS or MER figure is the single most common reason teams misdiagnose a healthy account as failing.

The fastest way to settle a genuine dispute between what the platform claims and what your bank account shows is a short geo holdout or conversion lift test, which isolates real incremental revenue from claimed credit.

Operating Cadence: A Weekly, Monthly, Quarterly Playbook

Measurement only works if it runs on a schedule, not just when something looks off. Here’s the cadence I run on client accounts, with an owner attached to each check.

Weekly (media buyer or founder): Check blended MER against your margin target, then triage platform ROAS by campaign. Any campaign showing a ROAS drop of more than 20% week over week gets a lever check (CTR, CVR, CPC) before any budget change. Green MER plus red campaign ROAS means investigate the campaign, not the total budget.
Monthly (growth lead or CMO): Recalculate a-MER to separate new-customer acquisition efficiency from retention-driven revenue. Run a cohort check on customers acquired 60 to 90 days ago to see if their repeat purchase rate matches historical patterns, since a-MER can look fine while acquiring lower-quality customers who never come back.
Quarterly (leadership, with agency or analytics support): Run an incrementality test, ideally a geo holdout, where you pause spend in a matched set of regions and compare resulting revenue against markets running as normal. This is the only reliable way to confirm what platform ROAS is actually adding versus what it’s claiming. Use the results to rebase your contribution-margin MER target for the next quarter.

Cadence	Metric to check	Owner	Trigger for action
Weekly	Blended MER + campaign-level ROAS	Media buyer / founder	ROAS drop over 20% week over week
Monthly	a-MER + new customer cohort behavior	Growth lead / CMO	a-MER declining while blended MER stays flat
Quarterly	Incrementality (geo/holdout) test	Leadership + agency	Rebase contribution-margin MER target

This layered approach, MER weekly, ROAS for allocation, incrementality testing periodically, is the structure that keeps teams from overreacting to a single platform’s Tuesday numbers while still giving them a fast enough signal to catch real problems. Geo holdout and lift tests remain the closest thing this industry has to a control group, and quarterly is often enough unless you’re scaling spend aggressively.

Practitioner Checklist: Rules I Use on Meta Programs

I set MER targets by working backward from contribution margin, not revenue. If a brand’s margin supports a 3.0 contribution-margin MER at breakeven, I build in a buffer for retention assumptions before authorizing new spend. Inside that ceiling, ROAS drives every creative and audience decision, since it moves faster and shows which specific asset is earning its keep. When platform ROAS and real revenue disagree, I run three checks: a short geo holdout to isolate true lift, a side-by-side of Meta’s reported ROAS against Shopify’s actual revenue for the same window, and a look at whether retargeting is inflating the account average. Most disagreements resolve in one of those three checks.

Practitioner Checklist: Rules I Use on Meta Programs — overview diagram

Balancing Fast Signals Against Long-Term Health

The mistake I see most often isn’t picking the wrong metric. It’s reacting to whichever one moved today. Platform ROAS updates fast and demands attention, but treating a single dashboard as gospel without running a diagnostic first leads to cutting campaigns that were actually working. MER is slower and less exciting, but it’s the number that keeps the business solvent while ROAS tells you where to steer.

— Stefano Mazzei

How My Team at Cosma Builds Measurement Systems That Hold Up

Most brands don’t have a MER problem or a ROAS problem. They have a measurement system that was never built to separate the two, which is exactly what my team at Cosma fixes for DTC and ecommerce brands scaling paid acquisition in the U.S. and Canada.

We build the reporting layer first: blended MER, contribution-margin MER, and clean platform ROAS by campaign, so budget decisions and creative decisions stop competing for the same dashboard. Then we run the incrementality tests, geo holdouts and lift studies, that tell you what your Meta spend is actually adding versus what it’s claiming credit for. You can see the approach in our case studies or browse creative examples in our portfolio. If your MER and ROAS numbers keep disagreeing and you want a second set of eyes on the diagnosis, book a call with my team and we’ll walk through your account together.

FAQ

How is MER different from ROAS?

MER divides total revenue by total marketing spend across the whole business, while ROAS divides attributed revenue by ad spend inside a single platform. MER measures overall health; ROAS measures in-platform performance.

What ROAS equals a 25% ACoS?

A 25% ACoS (advertising cost of sale) converts directly to a 4.0 ROAS, since ACoS is the inverse of ROAS expressed as a percentage (ad spend ÷ revenue). Divide 1 by the ACoS as a decimal (1 ÷ 0.25) to get the ROAS figure.

Are MER and ROAS the same thing?

No. They’re related but distinct: MER is platform-agnostic and covers total spend and total revenue, while ROAS is calculated inside a single platform and depends on that platform’s own attribution model. Using them interchangeably is a common reason teams misread account health.

Is a 20% ROAS good?

If you mean a low ROAS, that’s often break-even to thin margin territory depending on your contribution margin, which is why pairing ROAS with contribution-margin MER matters more than judging ROAS in isolation.

When should I run an incrementality test?

Run a geo holdout or lift test quarterly, or any time platform ROAS and blended MER disagree enough that you’re considering a major budget change. It’s the most reliable way to separate real revenue lift from claimed platform credit.