Ecommerce & CRO
ROAS
Return on Ad Spend
ROAS, or return on ad spend, is the revenue produced per dollar of advertising — $4 back on $1 spent is a ROAS of 4. It measures campaign efficiency, not profitability: a 4× ROAS on a product with a 20% margin still loses money. Blended ROAS, measured across all spend and all revenue, is harder to game than platform-reported figures.
Why it matters
ROAS is the most quoted and least reliable number in ecommerce marketing, for two reasons that compound.
The first is that it measures revenue, not profit. A 4× ROAS on a product with 20% gross margin returns £0.80 of margin for every £1 spent — a campaign that is efficient by the reported metric and loses money in reality. The break-even ROAS depends entirely on margin, and every business should know its own figure before reading a dashboard.
The second is attribution. Every ad platform counts conversions it believes it influenced, using its own window and its own rules. Add the platform-reported revenue across channels and it routinely exceeds what the store actually sold. Each platform is not lying; they are all claiming the same customers.
How it works on Shopify
The defensible version is blended ROAS: total revenue over a period divided by total marketing spend in it. It cannot be gamed by attribution because it does not attribute — it simply asks what the business spent and what it earned.
Channel-level ROAS still has a job, as a directional signal for reallocating budget between campaigns. It should not be reported to finance as fact.
The reconciliation worth building is platform-reported revenue against Shopify's own sales for the same window. The gap is the attribution inflation, and knowing its size tells you how much to discount every platform figure afterwards.
Two adjustments make the number honest. Deduct returns, which arrive weeks later and hit some channels harder than others. And separate new-customer revenue from repeat, because a campaign retargeting existing customers reports a spectacular ROAS for orders that would largely have happened anyway.
Read alongside CAC and LTV: a low first-order ROAS can be correct if the customers acquired come back.
Common mistakes
- Not knowing break-even. Without the margin-derived threshold, "4× ROAS" is a number without a verdict.
- Summing platform figures. Double-counted conversions, presented as a total.
- Ignoring returns. Revenue that comes back reduces ROAS after the report was filed.
- Optimising retargeting. The highest-ROAS campaign is usually the one reaching people who already intended to buy.
- Judging brand and prospecting by the same bar. Upper-funnel work reports poor ROAS by design.
- No incrementality test. Pausing a campaign to see whether revenue actually falls is the only direct evidence available, and almost nobody runs it.
When you need help
The signal is the familiar meeting: the ad platforms report profitable growth and the bank balance disagrees. That gap is attribution inflation plus margin, and it is resolvable with data the business already has.
The work is building a blended, margin-aware view that everyone accepts, then testing incrementality on the channels claiming the most credit. It is not glamorous, and it usually changes where the budget goes.
Related terms
- CACCAC, or customer acquisition cost, is the total sales and marketing spend needed to win one new customer. It is read against LTV: a business whose CAC approaches its LTV is buying revenue at a loss. Rising ad costs are the main reason merchants invest in organic channels such as SEO, where the cost per customer falls over time instead of rising.
- GMVGMV, or gross merchandise value, is the total value of goods sold through a store over a period, before deducting discounts, returns, refunds, or fees. It measures the scale of a marketplace or store rather than its profitability, which is why it is quoted often and should be read carefully — net revenue can be far lower.