You are hitting your ROAS target. Maybe exceeding it. The agency reports look healthy. But when you reconcile with your finance team, profit is flat or falling. Something does not add up.
This is one of the most common issues we diagnose in ecommerce account audits. And it is almost never a tracking issue. It is a structural issue with how ROAS is being used as a success metric.
ROAS Measures Revenue, Not Profit
This sounds obvious, but the implications are not. ROAS tells you how much revenue you generated per pound of ad spend. It does not tell you whether that revenue was profitable.
A 5.0 ROAS means you generated £5 in revenue for every £1 spent on ads. If your gross margin is 60%, you kept £3. If your gross margin is 30%, you kept £1.50. Same ROAS. Very different commercial outcomes.
This is why we focus on profit on ad spend (POAS) rather than revenue-based metrics. ROAS can stay flat while profit collapses. POAS cannot.
The Margin Mix Problem
Most ecommerce catalogues have products at different margin levels. Some items generate 70% gross margin. Others generate 25%. When you optimise for ROAS, you are optimising for revenue, which means the algorithm will push whichever products convert most efficiently.
Efficient converters are often low-margin products. They have lower price points, lower consideration time, and higher purchase intent. They look great in the dashboard. They look terrible on the P&L.
We see this constantly in fashion and home and living accounts. The algorithm learns to favour sale items, accessories, and entry-level SKUs because they convert. Meanwhile, your hero products with 65% margins barely get impressions.
The Attribution Mirage
Google Ads takes credit for conversions it may not have caused. Someone searches your brand name, clicks an ad, and buys. The ad gets the conversion. But would they have bought anyway?
Brand traffic inflates ROAS. It makes the account look more efficient than it is. And when you scale spend, you are often scaling into prospecting traffic that converts at a much lower rate than the blended number suggests.
This is especially problematic in Performance Max campaigns, where brand and non-brand traffic are blended by default. A 6.0 ROAS in PMax might be 12.0 on brand and 2.5 on prospecting. You cannot tell without segmentation.
CPC Inflation and Margin Compression
Cost per click rises over time. Competition increases. Auction dynamics shift. If your ROAS target stays fixed while CPCs rise, your margin gets squeezed.
A 4.0 ROAS at £0.80 CPC might be profitable. A 4.0 ROAS at £1.40 CPC might not be. The ROAS looks identical. The economics are completely different.
This is why spend governance matters. ROAS targets should not be static. They should flex with market conditions, margin shifts, and business objectives.
The Discount Trap
Promotions improve conversion rates. Conversion rates improve ROAS. But promotions also reduce margin. A 20% discount on a 50% margin product leaves you with 30% gross margin.
When you run promotions and your ROAS improves, you are often seeing an illusion. The revenue looks better. The profit is worse. And if your agency is optimising toward ROAS, they will push more spend during promotions because the efficiency looks better.
This compounds during peak trading periods. Black Friday accounts often show record ROAS alongside record margin erosion.
The Returns Reality
ROAS is measured at point of sale. It does not account for returns. A product that converts well but returns at 40% will show strong ROAS and deliver weak profit.
This is especially relevant for fashionand any category with high return rates. If your agency is not factoring return rates into their optimisation decisions, they are optimising for the wrong outcome.
What To Do About It
The fix is not complicated, but it requires a shift in how you measure success.
Calculate your breakeven ROAS by product category.
Not a blended target. A category-level or SKU-level target that accounts for actual margin. We explain how to do this in our Performance Max Commercial Framework.
Separate brand and non-brand performance.
If you are running Performance Max or Smart Shopping, insist on brand exclusions or use search term analysis to understand what proportion of conversions are brand-driven.
Factor in returns.
Adjust your ROAS targets for categories with high return rates. A 4.0 ROAS with 35% returns is effectively a 2.6 ROAS.
Review promotional periods separately.
Do not blend promotional performance with BAU performance. Evaluate whether increased efficiency during promotions was offset by margin reduction.
Move to contribution-based targets.
The ultimate goal is not ROAS. It is contribution margin. Revenue minus COGS minus ad spend minus returns minus fulfilment. If that number is positive and growing, your advertising is working. If it is not, your ROAS target is misleading you.
Who This Is For
If you are spending £10k or more per month on Google Ads and your finance team keeps asking why profit is not matching ROAS, this is the most likely explanation. It is not a bug. It is how the platform works when left unchecked.
Our ecommerce audit process starts here. We reconcile platform metrics with commercial reality. If there is a gap, we find it and explain exactly where the profit is leaking.
If you are not ready for a conversation, our ROAS vs Profit Reference Guidewalks through the calculations in more detail.
ROAS is a platform metric. Profit is a business metric. When they diverge, the platform is not lying. It is just measuring something different from what you need.