Attribution Drift
The conversions column and the model-attributed conversions column on this account differ by more than a tolerable margin. Large drift means the selected attribution model is redistributing credit across touchpoints enough to change optimisation and reporting conclusions. Drift can be valid, but it has to be intentional and documented so finance, media, and analytics are not arguing from different numbers.
Why It Matters
Large drift means the selected attribution model is redistributing credit enough to change optimization and reporting conclusions. That can be valid, but it needs to be intentional and understood by the team. Otherwise finance, media buyers, and analytics will argue from different numbers for the same conversion action.
How To Fix It
Compare the conversion action attribution model, reporting columns, and business sales cycle. If drift is expected, document which column is used for budget decisions and why. If it is not expected, standardize attribution models across similar actions and inspect whether view-through or cross-device conversions are driving the gap.
Example
Standard conversions: 120
Current model attributed conversions: 43
Drift: 64%This account shows a significant gap between standard conversion counts and model-attributed conversion counts on one or more conversion actions. Per Google Ads attribution documentation, model-attributed conversions redistribute credit across the user path based on the selected model (data-driven, position-based, time-decay, linear, last-click, or first-click), and large drift between this column and the standard conversions column indicates the model is materially reshaping reported performance. The risk is reporting inconsistency: finance, media, and analytics teams arguing from different numbers for the same action. Fix: review each flagged conversion action attribution model against the business sales cycle, document which column drives budget decisions and why, standardise attribution models across similar actions, and inspect whether view-through or cross-device conversions are the source of the drift. Source: support.google.com/google-ads/answer/6259715.
Drop this paragraph into your client deliverable. Sources back to the canonical platform documentation linked below.
References
Audit your own files for this check
AdLint runs this check (and 177 others) against your GTM, Google Ads, Meta, TikTok, LinkedIn, Pinterest, Twitter/X, and Snapchat exports. Everything stays in your browser. No uploads, no accounts.
Run a free audit