skill-077-data-driven-attribution

Skill 077: Data-Driven Attribution

Purpose

Verify that Data-Driven Attribution (DDA) is enabled for all eligible conversion actions. DDA uses machine learning to distribute credit across touchpoints based on actual conversion patterns, providing more accurate optimization signals for Smart Bidding.

Data Requirements

Data Source: Standard

Standard Data:

• data/account/conversion_actions.md - Attribution model settings
• data/performance/campaigns/*/campaign_metrics_30_days.md - Conversion volume for eligibility

Reference GAQL:

SELECT
  conversion_action.id,
  conversion_action.name,
  conversion_action.attribution_model_settings.attribution_model,
  conversion_action.attribution_model_settings.data_driven_model_status,
  conversion_action.status,
  conversion_action.include_in_conversions_metric
FROM conversion_action
WHERE conversion_action.status = 'ENABLED'

Use /google-ads:get-custom for DDA eligibility status.

Analysis Steps

1. Check DDA eligibility: Minimum 3,000 clicks and 300 conversions in 30 days per action.
2. Review current attribution models: Identify actions not using DDA.
3. Identify migration opportunities: Eligible actions using Last Click or other models.
4. Prioritize by impact: High-volume primary conversions first.

Thresholds

Condition	Severity
Primary conversion using Last Click with sufficient volume for DDA	Critical
DDA eligible but using First Click	Critical
DDA status STALE (volume dropped)	Warning
Near DDA eligibility (>200 conversions)	Info
Low volume, cannot use DDA	Info

Output

Short (default):

## Data-Driven Attribution Audit

**Account:** [Name] | **Conversion Actions:** [X] | **Using DDA:** [Y]

### Attribution Summary
| Action | Current Model | DDA Eligible | Status |
|--------|---------------|--------------|--------|
| [Name] | [Model] | [Yes/No] | [Migrate/OK] |

### Critical ([Count])
- **[Action]**: Using [Model] with [X] conversions -> Switch to DDA

### Recommendations
1. [Priority action]
2. [Secondary action]