A useful account audit starts with business outcomes and signal quality, not a preference for one campaign type. Historical performance can be interpreted only after duplicated conversions, mixed intent and attribution limits are understood.
What this guide covers
Review measurement, architecture, query quality, bidding inputs, message match and sales feedback in a sequence that protects useful account history.
Validate conversion definitions first
Bidding and reports are unreliable when primary outcomes include duplicate tags, page views or low-value actions.
Classify primary and secondary actions
Only outcomes the business wants bidding to pursue should be primary for the relevant goal.
- List every conversion source and owner.
- Separate sales outcomes from diagnostic actions.
- Check counting, value and attribution settings.
Test the implementation end to end
A green platform status does not prove the right event fires once in the right scenario.
- Complete real lead or purchase journeys.
- Check duplicate browser and imported events.
- Exclude personal data from URLs and event parameters.
Review architecture and budget allocation
Campaign structure should preserve meaningful control while concentrating enough data for learning.
Map campaigns to intent and economics
Brand, high-intent services, products and exploratory demand may need different controls.
- Document objective and audience for each campaign.
- Find overlapping targeting and internal competition.
- Separate materially different value or geography where justified.
Inspect budget constraints and waste
Daily budgets should reflect priority and opportunity rather than legacy structure.
- Review lost impression and budget diagnostics in context.
- Compare spend with qualified outcomes.
- Identify campaigns too fragmented to learn.
Audit queries, ads and landing pages
Keywords and targeting settings are inputs; actual search terms reveal what traffic was purchased.
Evaluate search-term quality
Classify terms by relevant intent, ambiguity, brand status and downstream value.
- Add evidence-based exclusions.
- Find missing high-value themes.
- Review close variants and broad-match behavior.
Check message and page alignment
The ad and page should answer the same need and make the next step credible on mobile.
- Match claims and offers across the journey.
- Review asset relevance and policy status.
- Test speed, clarity, forms and calls to action.
Connect platform data with business quality
Cost per lead can improve while sales quality declines if every submission is treated equally.
Add sales feedback where possible
Qualified lead, opportunity and sale stages give optimization a more useful target.
- Agree lead-quality definitions with sales.
- Use stable identifiers without exposing personal data.
- Import values only after reconciliation testing.
Prioritize changes and preserve evidence
Avoid changing tracking, bidding, targeting and pages simultaneously when the account lacks data.
- Rank findings by impact and confidence.
- Annotate major changes and learning periods.
- Define the review period before declaring a result.
Primary sources
Platform features and policies change. Review the current primary documentation before implementation.