Data-Driven Attribution Volume Check

Data-Driven Attribution is Google's ML-based alternative to fixed rules like last-click or linear. It learns from observed conversion paths in the account and assigns fractional credit to each touchpoint. The model is powerful when given enough data. Google has historically required several hundred conversions and several thousand ad interactions within 30 days for DDA to fully activate, though the exact thresholds have evolved. Below threshold, DDA either falls back to a simpler model behind the scenes or produces unstable credit assignments that swing between reporting periods. The risk is interpretation: teams discuss DDA-attributed credit as if it is precise ("Campaign X gets 35% credit") when the underlying model may be running on too little signal to support that precision. This check is info-level because eligibility status is only visible in the Google Ads UI itself; AdLint can flag the configuration but not verify the model state.

1. In Google Ads, open Tools & Settings → Measurement → Attribution.
2. For each flagged conversion action, check the model status and recent volume. DDA-eligible actions show a green status; ineligible or low-volume actions show a warning.
3. If a flagged action is showing insufficient volume: (a) consolidate duplicate conversion actions (one canonical Primary per business event), or (b) temporarily switch to a simpler attribution model (Position-based, Linear, or Last-click) while volume builds.
4. Document the attribution model chosen for each action. Reporting, bidding, and client-facing decks should all use the same attribution assumption to avoid confusion.