Reference Guide
ROAS vs Profit
A Reference Guide for Ecommerce
ROAS (Return on Ad Spend) is perhaps the most cited metric in ecommerce advertising. It's also one of the most misunderstood. This guide provides a neutral, clear-eyed explanation of what ROAS actually measures, when it serves you well, when it misleads, and what alternatives exist.
No evangelism. No spreadsheet models. Just clarity.
Why ROAS Exists
ROAS emerged as a simple way to express advertising efficiency: for every £1 spent, how many pounds of revenue came back? It became popular because:
- Universal comparability - Every channel, campaign, and platform can report ROAS
- Immediate calculation - Revenue ÷ Spend = a number you can act on today
- Client-friendly - "Your ROAS is 4x" sounds reassuring in a report
- Platform alignment - Google, Meta, and others optimise toward it directly
ROAS answered a real need: a quick, standardised way to compare advertising efficiency across different contexts. For this purpose, it works.
When ROAS Works
ROAS is a reliable indicator when certain conditions hold true:
Consistent margins across products
If your margin is roughly the same on everything you sell, a higher ROAS genuinely means more profit.
Low return rates
If what's purchased stays purchased, revenue in equals revenue retained.
Simple attribution
If most customers convert in one session, ROAS captures most of the picture.
Single channel dominance
If paid search is your primary acquisition channel, cross-channel bleed is minimal.
Relative comparison only
When you're comparing campaign A vs campaign B within the same account, ROAS can highlight which is more efficient.
In these scenarios, ROAS functions as intended: a simple efficiency ratio that helps you allocate budget toward better-performing segments.
When ROAS Lies
ROAS becomes misleading-sometimes dangerously so-when the underlying assumptions break down:
Variable margins
A 5x ROAS on a 20% margin product is worse than a 3x ROAS on a 60% margin product. ROAS ignores this entirely.
Returns and refunds
ROAS counts revenue at checkout. It doesn't account for the 40% of fashion orders that come back. That "6x ROAS" might be 2x after returns.
BNPL distortion
Buy Now Pay Later inflates AOV but the cash arrives in instalments-and defaults aren't captured in ad platform revenue.
Cross-channel cannibalisation
That "new customer" from Google might have converted through email anyway. ROAS claims the full credit.
Platform self-reporting
Google reports Google conversions. It's measuring itself. Server-side tracking reveals gaps the platform doesn't show.
Blended reporting hides variance
Account-level ROAS can mask that 80% of your profit comes from 20% of your campaigns-and that several campaigns are destroying value.
The core problem: ROAS measures revenue, not profit. It counts what arrives, not what stays. It ignores costs of goods, fulfilment, returns, and fraud. It's a top-line metric in a bottom-line world.
What Replaces It
No single metric replaces ROAS. The goal is to build a clearer picture using multiple lenses:
Contribution Margin per Order
Revenue minus COGS, fulfilment, and ad cost. This tells you what each order actually contributes to the business.
Contribution = Revenue − COGS − Fulfilment − Ad CostBreakeven ROAS
The ROAS at which you make zero contribution. Below this, you're losing money on each sale. This varies by product, category, and channel.
Breakeven ROAS = 1 ÷ Gross Margin %Net Revenue (Post-Returns)
Track revenue after returns are processed, not at checkout. In high-return categories, this can differ by 30-50% from reported revenue.
Marketing Efficiency Ratio (MER)
Total revenue ÷ total marketing spend across all channels. Less granular than ROAS, but harder to game and more reflective of true business efficiency.
Customer Acquisition Cost (CAC)
Total marketing spend ÷ new customers acquired. Paired with LTV, this shows whether your acquisition engine is sustainable.
The shift isn't from one metric to another-it's from relying on a single number to understanding the commercial reality of your ad spend.
Summary
ROAS is a tool, not a truth. It was designed for quick comparisons, not strategic decisions.
It works when margins are consistent and attribution is clean. It fails when products, returns, and customer journeys are complex.
The alternative isn't to abandon ROAS-it's to pair it with profit-based metrics that reflect what actually matters to the business.