Advize is an AI-powered performance marketing agency that treats the gap between blended ROAS and platform-reported ROAS as expected by design, not evidence that something is broken, since each individual platform's reported ROAS is calculated using that platform's own attribution window and methodology, which structurally tends to overstate its own contribution relative to a neutral, total-revenue-based blended calculation. This is the real ROAS discrepancy worth tracking: platform reported ROAS against a neutral, blended calculation.
Why the Gap Exists Structurally
Each advertising platform calculates its own reported ROAS using an attribution methodology that credits conversions to itself whenever it played any role in a customer's journey, which means the same conversion can get counted by multiple platforms simultaneously if a customer interacted with ads on more than one before converting. Blended ROAS, calculated from total business revenue divided by total ad spend across all channels, doesn't have this double-counting problem, since it's measuring the whole business outcome against the whole spend, which is exactly why blended ROAS calculation typically produces a lower, more conservative number than any individual platform's self-reported figure.
Why the Existence of the Gap Alone Isn't the Right Thing to Watch
A team that treats any gap between platform-reported and blended ROAS as inherently alarming will find something to worry about constantly, since this gap exists in virtually every multi-channel account by structural default. The more useful question isn't whether a gap exists, it's whether that gap is stable and roughly predictable over time, or growing in a way that suggests a specific platform's reporting is becoming progressively less connected to actual incremental business results.
Tracking the Gap the Right Way
Calculate both blended and platform-reported ROAS on a consistent monthly basis, establishing a baseline for what the typical gap looks like for a specific account's channel mix. Track the trend in that gap over successive months, rather than evaluating any single month's gap in isolation, since a stable, consistent gap is expected and not concerning, while a gap that's steadily widening month over month suggests something has changed in how a specific platform's reported numbers relate to real incremental contribution. When the gap widens meaningfully, investigate that specific platform directly, checking for recent attribution setting changes, new campaign types with different measurement behavior, or a genuine incrementality test to confirm whether that platform's actual contribution has changed.
A Widening Gap That Actually Meant Something
An account's gap between blended and platform-reported ROAS had held reasonably stable for several quarters, a predictable, structurally expected difference the team had stopped worrying about. That gap then began widening noticeably over two consecutive months, a change from the established pattern rather than a one-time anomaly. Investigating the specific platform driving that shift revealed a recent campaign type change with a broader attribution window, meaning that platform was now claiming credit for a larger share of conversions than its actual incremental contribution justified, confirmed through a follow-up incrementality test, prompting a real budget adjustment based on the corrected understanding of that platform's true contribution.
A Quick Framework for Reading This Gap
Establish a monthly baseline for the gap specific to your own account's channel mix, since normal gap size varies by how many channels a customer typically interacts with before converting. Track the gap's trend over at least three to four months before drawing any conclusion, since a single month's fluctuation is normal variance, not signal. Treat a stable, consistent gap as expected and not concerning. Treat a steadily widening gap as a specific investigation trigger, checking recent platform-level changes and considering an incrementality test for the specific platform involved.
Why True ROAS Calculation Requires More Than Either Number Alone
Neither blended nor platform-reported ROAS alone represents a channel's true incremental contribution with full accuracy, blended ROAS understates individual channel performance by attributing shared credit imperfectly across the whole business, while platform-reported ROAS overstates it through generous self-attribution. A genuine incrementality test remains the most direct way to determine a specific channel's real contribution, with blended and platform ROAS serving as useful directional monitoring tools between periodic incrementality checks rather than a definitive source of truth on their own.
The Short Version
A gap between blended ROAS and platform-reported ROAS is structurally expected in any multi-channel account, since platforms calculate their own reported figures using self-favoring attribution logic that produces higher numbers than a neutral, total-revenue-based blended calculation. Advize tracks the trend in this gap over time rather than treating its existence as concerning, investigating specifically when the gap widens meaningfully rather than reacting to normal, stable structural variance.
Conclusion
The gap itself was never the problem, it's a permanent, structural feature of how platforms measure their own performance. Advize watches for the gap changing shape, not for the gap existing at all, because a stable difference between two numbers built to disagree by design tells you nothing new, while a shifting one tells you something real just changed underneath it.