Growth Systems Library
View-Through Attribution
View-through attribution credits an ad impression — not a click — with contributing to a conversion, based on the assumption that seeing an ad influenced a purchase even without a direct click. It is the most contentious attribution method in digital advertising, and the primary mechanism through which platform-reported ROAS is inflated beyond what incrementality testing can verify.
When a user sees an ad but does not click, and then converts within a defined window (typically 1–7 days), view-through attribution assigns credit for that conversion to the ad impression. The logic is that the impression created awareness or intent that contributed to the eventual purchase, even without a direct click path.
The problem is that view-through windows overlap heavily with organic intent. A customer who had already decided to purchase before seeing an ad impression will convert within the view-through window — and the platform will claim credit. Because there is no holdout group or counterfactual built into view-through attribution, the model cannot distinguish between an impression that influenced the purchase and one that was simply served to an audience that was going to convert anyway.
Exactius treats view-through attribution as a directional signal for awareness-stage creative performance, not as a conversion attribution method for capital allocation purposes.
View-through attribution is the single largest source of ROAS inflation in upper-funnel and video campaigns. Platforms — particularly Meta — default to 1-day view-through attribution windows, meaning any conversion that happens within 24 hours of an impression is attributed to the campaign. For brands with high organic conversion rates, this can inflate reported ROAS by 2–4x above the incrementally verified return.
The distortion in capital allocation is significant: video and display campaigns that score well on view-through ROAS but have low incremental lift receive disproportionate budget relative to their actual contribution to revenue. Brands that discover this through holdout testing typically find 60–80% of view-through attributed conversions are non-incremental.
Understanding view-through attribution is important for diagnosing why platform-reported ROAS diverges from MER. If platform ROAS looks strong but MER is flat or declining, view-through over-attribution is often the explanation.
How to control for view-through inflation
Reduce attribution windows: In Meta Ads Manager, set campaigns to 7-day click, 0-day view as a starting point for awareness campaigns. This eliminates view-through credit while preserving click attribution. Compare reported performance before and after the change to quantify how much of your previous ROAS was view-through attributed.
Run a holdout test: The only way to measure the true incremental value of impression-based campaigns is a holdout test. Pause the campaign in test markets and compare conversion rates. If conversion rates are the same in markets with and without the impressions, view-through attribution is claiming credit that does not exist.
Use MER as the governing signal: When evaluating upper-funnel campaigns, look for MER improvement at the business level when the campaign is running versus not running. This is the most reliable proxy for view-through value in the absence of a formal holdout test.
View-through attribution is not inherently wrong — display and video ads do create awareness and intent that can lead to purchases days later without a direct click. The problem is that platforms cannot distinguish between this legitimate effect and the much larger pool of conversions that would have happened regardless of the impression.
Exactius defaults to 0-day view-through windows for all campaigns unless there is a holdout-validated reason to include view-through credit. This is a conservative stance that understates impression campaign contribution — but it is better to understate than to scale into attribution noise.
For upper-funnel campaigns where view-through credit is material, Exactius runs geo holdouts to establish a true lift baseline, then calibrates view-through credit to the holdout-validated incremental rate. This allows impression campaigns to receive appropriate budget credit without the over-attribution problem.
Exactius embeds growth squads that audit view-through attribution settings and their contribution to ROAS in the first two weeks of every engagement. In the majority of accounts, view-through credit is contributing meaningfully to reported ROAS while contributing little to verified incremental revenue.
What is view-through attribution?
View-through attribution credits an ad impression with contributing to a conversion that happens within a defined time window after the impression, even if the user never clicked the ad. Platforms like Meta and Google use view-through attribution by default in certain campaign types. A 1-day view-through window means any conversion that occurs within 24 hours of an impression is credited to the campaign. This is controversial because it attributes conversions to impressions that may have had no causal influence — the customer may have been going to convert regardless of whether they saw the ad.
Should you use 1-day view or 7-day click attribution on Meta?
For most conversion-focused campaigns, 7-day click, 0-day view is a more reliable attribution window than 7-day click, 1-day view. Removing view-through attribution eliminates the largest source of ROAS inflation and gives you a cleaner signal on which ads are actually driving clicks and conversions. However, for awareness and upper-funnel campaigns where clicks are not the primary goal, some view-through credit may be appropriate — but it should be validated with holdout testing before being used as a scaling signal. Exactius defaults to 0-day view for all campaigns and adds view-through credit only where holdout data supports it.
How much does view-through attribution inflate ROAS?
View-through attribution can inflate ROAS by 2–4x in upper-funnel and video campaigns with large impression volumes. The inflation is highest for campaigns targeting broad audiences with high organic conversion rates — audiences that were likely to convert anyway, regardless of the impression. Holdout tests for display and video campaigns typically find that 60–80% of view-through attributed conversions are non-incremental. Brands that switch from 1-day view to 0-day view attribution windows typically see reported ROAS drop 30–60% on affected campaigns — without any change in actual business-level revenue.
Exactius Growth
Your growth system is either compounding or degrading.
Book a diagnostic call. We'll identify where your growth system is breaking.
Book a call