Why platform ROAS and causal ROAS disagree
Platform ROAS and causal ROAS are not two attempts at the same number. They answer different questions, and the gap between them is structural rather than accidental.
Three platforms, three answers, one budget to defend. Your GA4 export already holds the answer, and the first finding is free and never leaves your browser. Get your first finding free
By Joris van Huët, Founder & CEOUpdated 3 min read
Run the numbers for your store: the free ad platform over-reporting checker.
The gap between platform-reported and causal ROAS is not a bug in either number. They are answers to different questions, and the difference is predictable. Once you can predict it, the gap becomes diagnostic rather than annoying.
The two questions
| Platform-reported ROAS | Causal ROAS | |
|---|---|---|
| Question | Of the conversions I can see, how many touched my ads? | Would these conversions have happened without the channel? |
| Total | Fixed by conversions observed | Depends on whether the channel ran |
| Who defines the rule | The platform selling the media | You, or an independent method |
| Typical direction | Higher | Lower, for proximity-heavy channels |
The third row is the one worth sitting with. Every ad platform sets its own attribution window and its own view-through rules, and every platform benefits when the number is high. That is not corruption, it is an incentive, and it is the reason a second read exists at all.
Where the gap concentrates
Predictably, in three places. Branded search, because it captures demand created elsewhere. Retargeting, because it reaches people already intending to buy. And anything with generous view-through crediting, because a view is a weak signal that someone was influenced.
Conversely the gap tends to be small on channels that reach genuinely new audiences at scale, because for those the proximity story and the causal story are closer together. The general mechanism is set out in platform attribution overcounting.
The simplest possible check
Sum what every platform claims for one window and compare it against what your store actually took. If the platforms collectively claim more revenue than exists, that difference is the overlap, and it sizes the problem in euros without any modelling at all.
The method is in the claim ratio and the store-side comparison in platform-reported ROAS against orders.
What the gap is not
It is not evidence that your ads do not work. A channel can have inflated reported ROAS and still be strongly profitable; the two facts are independent. The mistake in both directions is treating the reported number as either accurate or worthless, when it is a specific, knowable overstatement.
It is also not something a better attribution model fixes, because the platform number and the model number are answering different questions rather than measuring the same thing at different qualities. That distinction is in causal inference versus rule-based attribution.
Using the gap deliberately
Track the gap per channel over time rather than trying to eliminate it. A stable gap is a calibration you can apply. A gap that suddenly widens usually means something changed in the platform's rules or in your own coverage, and that is worth investigating.
Causality Engine produces the causal side of the comparison from a Google Analytics export: what each channel caused, a data-health score rather than a confidence interval, and a next step, at €99 for a first read, refundable within 30 days, no questions asked.
The framing that helps
Reported ROAS is a claim made by a party with an interest. Causal ROAS is an estimate made with a stated method and a stated uncertainty. Neither is the truth, and only one of them tells you how sure it is.
Related answers
Frequently asked questions
Why is my platform ROAS higher than my causal ROAS?
Because platform ROAS counts conversions the channel touched under rules the platform sets, while a causal estimate asks whether those conversions would have happened anyway. Channels close to the conversion, like branded search and retargeting, show the largest gaps.Does a large gap mean my ads are not working?
No. Inflated reported ROAS and genuine profitability are independent facts. A channel can be over-credited in platform reporting and still be worth funding. The gap tells you how much of the reported figure is overlap, not whether the channel earns its budget.Can a better attribution model close the gap?
Not really, because the two numbers answer different questions rather than measuring the same quantity at different qualities. A better model gives you a better causal estimate; the platform number keeps answering its own question.
Go deeper: Causal attribution, explained.
Sixty-second versions of these ideas: Causality Engine on YouTube Shorts.
Have an audience of ecommerce operators? Earn 20% recurring commission by recommending the €99 read.
Keep reading
Terms in this article
- AttributionAttribution identifies user actions that contribute to a desired outcome and assigns value to each. It reveals which marketing touchpoints drive conversions.
- Attribution ModelAn Attribution Model defines how credit for conversions is assigned to marketing touchpoints. It dictates how marketing channels receive credit for sales.
- Attribution WindowAttribution Window is the defined period after a user interacts with a marketing touchpoint, during which a conversion can be credited to that ad. It sets the timeframe for assigning conversion credit.
- Causal InferenceCausal inference is the set of methods for estimating how much an outcome changes because of a cause, rather than alongside it. It works by comparing what happened with an estimate of what would have happened without the cause.
- Confidence IntervalConfidence Interval is a statistical range of values that likely contains the true value of a metric. In marketing analytics, it quantifies uncertainty around estimates, indicating the precision of an outcome or causal effect.
- ConversionConversion is a specific, desired action a user takes in response to a marketing message, such as a purchase or a sign-up.
- Google AnalyticsGoogle Analytics is a web analytics service that tracks and reports website traffic.
- RetargetingRetargeting is online advertising that targets users who have previously interacted with your website or content. Attribution analysis shows the causal role of retargeting in driving conversions and improving ad spend.