Short answer: Platform ROAS lies because every ad platform grades its own homework. Each one counts the sales it can claim, including orders another channel also claims and orders that would have happened anyway, and none of them subtract product costs. The number that actually matters is MER (total store revenue ÷ total ad spend), judged against a break-even line built from your contribution margin. Use ROAS to pick winners inside a platform; use MER to decide how much to spend.
Every figure in this section is illustrative, built to show the math. It is not a client result.
Picture a skincare brand spending $70,000 in a month: $50,000 on Meta and $20,000 on Google. Here is what the dashboards report:
| Source | Ad spend | Revenue claimed | Reported ROAS |
|---|---|---|---|
| Meta Ads Manager | $50,000 | $180,000 | 3.6x |
| Google Ads | $20,000 | $95,000 | 4.75x |
| Platforms added together | $70,000 | $275,000 | 3.9x |
| Shopify, actual | $70,000 | $168,000 | 2.4x MER |
The platforms claim $275,000 of revenue. The store only took in $168,000, and that includes email, organic and repeat customers who would have bought with no ads at all. Roughly $107,000 of "revenue" exists only in the dashboards because both platforms counted the same orders, and because view-through windows credited ads that people merely scrolled past.
Now run it to profit. Say the brand keeps 45% of each sale after product cost, shipping, payment fees and returns. That makes the break-even MER 1 ÷ 0.45 = 2.22x.
Same month, same ads, two opposite decisions. If this brand pushes spend 30% because "ROAS is 3.9," the extra dollars almost certainly land below break-even. That gap is what we mean when we say ROAS is lying.
None of this requires anyone to be dishonest. Each platform uses a reasonable-sounding rule that, added up across channels, overstates what ads did.
A shopper sees a Meta ad Monday, clicks a Google ad Wednesday and buys Thursday. Meta counts the sale through its view or click window. Google counts it through its click. Your store recorded one order. Add channel ROAS together and you get more revenue than exists.
Meta can credit a purchase to an ad the customer only saw in their feed. Some of those views genuinely nudged a sale. Many were people already heading to your site from an email, a podcast mention or a friend's recommendation.
Retargeting and branded search look excellent because they reach people who already intended to buy. They harvest demand more than they create it. Our guide to retargeting and catalog ads in the Advantage+ era shows how to cap and test that spend.
Since iOS privacy changes, a share of conversions are modeled rather than observed. Modeling is useful, but it is still an estimate produced by the party being graded. See why attribution is failing DTC brands for how this got worse in 2026.
If you cannot answer all three, treat the ROAS as a relative score, not a profit report.
ROAS measures revenue, not profit. If product, shipping, fees and fulfillment take 60% of each sale, you keep 40%, and you need a ROAS of 2.5 just to get your ad dollar back.
The figures below are illustrative, not client results.
| Reported ROAS | Contribution margin before ads | Profit per $1 of ads |
|---|---|---|
| 2.0 | 40% | −$0.20 |
| 3.0 | 40% | +$0.20 |
| 4.0 | 40% | +$0.60 |
Now stack the overstatement on top. If a platform's 3.0 ROAS is really a 2.2 once overlap is removed, that "+$0.20" row is actually losing money. Low-margin brands feel this first: a supplement brand at 70% gross margin can survive sloppy measurement far longer than an apparel brand at 45% with heavy returns.
MER = total revenue ÷ total ad spend, pulled from Shopify or your payment processor. Because it uses money that actually arrived, no platform can inflate it. Pair it with:
MER has a weakness: it does not tell you which ad worked. That is fine, because it is not supposed to. It tells you whether the whole system is profitable and whether more spend is buying more revenue. The full method, including how to set targets by growth stage, is in our pillar guide: MER vs ROAS: the marketing efficiency ratio explained.
ROAS still has a job. Inside one platform, using one attribution setting, it is a fine way to compare ads, ad sets and campaigns relative to each other. If concept A shows 3.1 and concept B shows 1.8 under the same rules, A is probably the better ad.
What ROAS should not do is set total budget or prove that ads are profitable. When ROAS looks solid but spend refuses to grow profitably, the cause is usually creative fatigue or audience saturation, which we unpack in why your Meta ads aren't scaling even with a solid ROAS. Advantage+ Shopping makes the gap wider, so read Advantage+ in 2026 for MER-focused brands before you hand it more budget.
This split, MER for budget and ROAS for tactics, is the core of our Engineered Growth System and how our performance marketing team reports to clients. It is how we have managed $50M+ in ad spend at a 4.9x average ROAS while keeping every budget decision tied to store revenue.
We'll compare your dashboards with your store data and show you the highest-leverage fixes. Book a free growth audit with Lion Media.
No. It is useful for comparing ads and campaigns inside one platform under the same attribution setting. It just should not decide your total budget.
The ROAS at which ad spend equals contribution margin. It is 1 divided by your contribution margin before ads, so a 40% margin means a break-even of 2.5.
Because each platform credits itself for shared orders and view-through conversions. That overlap makes platform totals add up to more than real store revenue.
Weekly for most brands. Daily numbers are noisy because of promotions, payday cycles and shipping timing, so trends matter more than single days.
Yes, the terms are used interchangeably. Both divide total store revenue by total ad spend across every channel, instead of trusting any single platform's report.
Book a free 30-minute growth audit. We'll review your ads, tracking and site, and show you where the profit is leaking.
No commitment. No pitch deck. A real conversation about your growth architecture.
.png)
