Fix: Primary conversion actions use mixed attribution models (DDA + Last Click)

Your audit surfaced primary conversion actions that aren't all using the same attribution model. Some are on Data-Driven Attribution (DDA), others are still on Last Click. Smart Bidding now receives contradictory credit signals from the same account — and the contradiction is most painful exactly where it matters most: the conversion actions you've marked Primary, which feed bid optimization.

Why this matters

Attribution model selection determines how Google Ads distributes credit for a conversion across the touchpoints that preceded it. Last Click hands 100% of credit to the final paid interaction. DDA uses your account's own conversion-path data to assign fractional credit across each ad interaction based on its measured contribution [1]. These two models tell Smart Bidding fundamentally different stories about which keyword, ad, or campaign deserves the next dollar.

When primary conversion actions disagree on which story to tell, Smart Bidding can't reconcile them. A Lead action on Last Click tells the algorithm "the final-click keyword is the hero." A Purchase action on DDA tells the same algorithm "credit is spread across the funnel." The bid models for Target CPA and Target ROAS that consume those primary signals end up optimizing toward conflicting objectives within a single account. Brad Geddes frames it as the root cause of "bad bids being made for your account" — inconsistency in attribution, counts, and conversion windows cascades into faulty optimization [2].

The fragmentation also breaks cross-channel learning. Google's automated bidding is most effective when it can reason about the full conversion path. Last-Click conversion actions provide a single-touchpoint slice of that path; DDA actions provide the full distribution. Mixing them within the primary set means the model is averaging across two incompatible representations of the same customer journey. ALM Corp's 2026 attribution guide notes that switching attribution models without recalibrating bidding targets creates "a misalignment between the model's conversion values and the bidding algorithm's optimization targets" [4] — the same misalignment exists within an account when primary actions don't agree.

How to verify the issue

1. Open Google Ads. Navigate to Tools → Conversions → Summary.
2. Add the Attribution model column if it isn't already visible (column picker → Attributes → Attribution model).
3. Filter or sort by Include in "Conversions" = Yes. These are your primary actions — the ones Smart Bidding optimizes toward.
4. Compare the Attribution model column across all primary rows. If you see a mix of Data-driven and Last click, the rule has fired correctly.
5. Cross-check with the Attribution report at Tools → Attribution. If Google has flagged any primary action with a "Switch to DDA" recommendation, that action is eligible (meets the 3,000 ad interactions / 300 conversions threshold for the initial switch) [5].

Time required: 5 minutes.

How to fix it

1. List every primary conversion action that's currently on Last Click. From Tools → Conversions → Summary, click each row to confirm it's marked primary (Include in "Conversions" = Yes).
2. For each Last-Click primary action, click the action name → Edit settings → Attribution model → select Data-driven → Save. Google's Help center confirms this as the supported migration path [1].
3. If Google reports a primary action as ineligible for DDA (insufficient volume — fewer than 3,000 ad interactions or 300 conversions over 30 days), do not force it. Either:
   • Consolidate it with a higher-volume action (e.g., promote a single Lead action and demote duplicates to secondary), or
   • Leave it on Last Click as a secondary action (Include in "Conversions" = No) so it still records but doesn't feed Smart Bidding.
4. If your account uses conversion value tracking, audit your Target ROAS and Target CPA settings immediately after the switch. DDA distributes credit more broadly than Last Click, so reported conversion counts often rise modestly (Google has historically cited a 6% lift on average [3]). Tighten your targets only after the new baseline stabilizes — typically 2-4 weeks.
5. Document the change. Note the switch date in your account changelog so downstream metric anomalies during the recalibration window aren't misattributed to creative, budget, or audience changes.

Time required: 15-30 minutes for the switches plus a 2-4 week recalibration window before further structural changes.

How to confirm the fix worked

• All primary conversion actions show Data-driven in the Attribution model column.
• No primary action displays a "Switch to DDA" recommendation banner in the Attribution report.
• Conversion volume reported in the account has stabilized (no continued week-over-week drift attributable to the model change) — wait 14-28 days.
• Smart Bidding strategies tied to those primary actions remain in Active status (not Learning for more than 7-14 days after the switch).
• If any primary action was demoted to secondary, confirm Smart Bidding campaigns reference the remaining primary set correctly via the bid strategy's conversion goal mapping.

Verification window: 14-28 days. Don't change bid targets, structures, or other attribution settings during this window — you'll lose the ability to attribute downstream metric movement to the model change.

Related rules + concepts

• Data-Driven Attribution — concept article: what DDA is, eligibility, and how credit is distributed
• Last-Click to DDA migration — full migration playbook including bid-target recalibration
• Smart Bidding — how automated bid strategies consume primary conversion signals
• Conversion tracking — foundation: how conversion actions are configured and counted

Sources

1. Google Ads Help — About data-driven attribution. https://support.google.com/google-ads/answer/6394265 (accessed 2026-05-25)
2. Adalysis — The Top 10 Google Ads Conversion Tracking Mistakes. https://adalysis.com/blog/the-top-10-google-ads-conversion-tracking-mistakes/ (accessed 2026-05-25)
3. Search Engine Land — Google confirms sunset details for 4 attribution models in Ads and Analytics. https://searchengineland.com/google-confirms-sunset-details-for-4-attribution-models-in-ads-and-analytics-433352 (2023, accessed 2026-05-25)
4. ALM Corp — Attribution Modeling in Google Ads: The Complete 2026 Guide. https://almcorp.com/blog/attribution-modeling-google-ads/ (2026, accessed 2026-05-25)
5. Google Ads Help — About "Switch to DDA". https://support.google.com/google-ads/answer/10762625 (accessed 2026-05-25)
6. Google Ads Help — About attribution models. https://support.google.com/google-ads/answer/6259715 (accessed 2026-05-25)