Platform ROAS measures how much revenue happened near your ads. Incremental ROAS measures how much revenue your ads actually caused. The gap between those two numbers is usually significant.
Almost every marketing team we have spoken to has the same experience: your Meta dashboard shows a strong ROAS, your Google dashboard shows a strong ROAS, and yet when you pause a campaign to test it, sales barely move. What is going on?
The credit-claiming problem
Imagine you own a shop on a busy high street. Every day, hundreds of people walk past a sign outside your door. When a customer comes in and buys something, your sign immediately claims credit for the sale.
Was the sign responsible? Maybe. But many of those customers were already coming to you. They searched for your brand name, found your address, and walked in. The sign was not the reason. It was just present.
Digital platforms work exactly the same way. Every ad platform tracks conversions that happened after someone saw or clicked an ad within a certain time window. If someone searched for your brand on Google, clicked a branded search ad, and bought, Google counts that as an ad-driven conversion. But that customer was already looking for you.
Why every platform overclaims
Each platform only sees its own slice of the customer journey. Meta sees Meta. Google sees Google. Neither can see what the other is showing the same customer. When both platforms are running simultaneously, both claim credit for the same conversion.
Add up the attributed revenue from all your platforms and it will almost certainly exceed your actual total revenue. This is called double-counting, and it is a structural feature of how attribution works, not a bug any platform is rushing to fix.
In our experience working with UK brands, the sum of platform-attributed revenue is typically 1.5 to 3 times the actual revenue recorded. That gap represents the overclaim.
What incrementality actually measures
Incremental ROAS answers a different question: if this campaign had not existed, how much less revenue would we have made?
That is a harder question to answer, but it is the right one. The revenue that would have happened anyway is not a marketing win. It is just your baseline. The only revenue that counts as a return on your media investment is the revenue that would not have existed without it.
A real-world example
Consider a brand running a retargeting campaign. The platform reports a ROAS of 8x. The targeting is set to people who visited the website in the last 30 days. These are people who already showed intent to buy. Many of them would have returned and purchased without being shown the retargeting ad.
When the brand runs a holdout test, switching off retargeting for a random group of qualified visitors, they find that conversion rates in the holdout group are only slightly lower than in the group being retargeted. The incremental ROAS drops to 1.4x. The campaign is barely breaking even on a true incremental basis.
How to get to the real number
- Run a holdout experiment: turn off a campaign for a randomly selected geographic area or audience segment and compare the results to a control group
- Use Marketing Mix Modelling to estimate the contribution of each channel across your full media mix, controlling for baseline demand
- Look for signals that do not come from the platforms themselves: your own revenue data, your CRM, your baseline sales trends
What this means for your budget
If you are allocating budget based on platform-reported ROAS, you are almost certainly overspending on channels that are claiming credit for organic demand, and underspending on channels that are actually driving new customers but getting less of the attribution credit.
Fixing this does not always mean spending less. It means spending differently. And for most brands, getting to the real incremental numbers reveals at least one channel that is dramatically undervalued and one that is significantly overvalued.
Can I trust my platform ROAS at all?
Platform ROAS is useful as a relative indicator for comparing performance within a single platform over time. What it cannot tell you is the absolute return your marketing is generating, or how it compares to doing nothing at all.
Which platforms overclaim the most?
All platforms overclaim to some degree, but retargeting campaigns, branded search campaigns, and any campaign targeting people who already showed purchase intent tend to show the largest gap between attributed and incremental ROAS.