Use this before changing bids or creative. If the conversion signal is wrong, every optimisation made after it is built on bad evidence.
Use this resource when
Download the working file.
This is the complete editable template—not a PDF preview. Add your own account data, owners and decisions.
Start with commercial truth, not the ad platform
Export 90 days of spend, leads, qualified leads, opportunities and won revenue. Reconcile those totals against the CRM before opening campaign settings. The scorecard is deliberately weighted toward measurement and lead quality because a low CPL can hide a high cost per SQL.
- Use one date range and one reporting timezone
- Document the exact CRM definition of MQL, SQL and Opportunity
- Flag any source with more than a 10% lead-count mismatch
Audit conversion integrity
List every primary conversion action and the campaign goals using it. Page views, button clicks and duplicate thank-you events should not compete with qualified outcomes in the same bidding portfolio.
- Check tag duplication in GTM and platform diagnostics
- Confirm value, currency, attribution window and counting method
- Separate form submit, qualified lead and closed sale events
Separate captured demand from incremental demand
Brand Search, retargeting and Performance Max can look efficient because they harvest demand created elsewhere. Report them separately from prospecting and non-brand acquisition so the blended result cannot hide weak incrementality.
- Split brand and non-brand reporting
- Review PMax brand exclusions and search categories
- Compare new-customer share and total-account conversions
Close the sales-feedback loop
For every source, calculate lead-to-SQL, SQL-to-opportunity and win rate. Add the top three rejection reasons. This turns a media review into a commercial review and prevents platforms from learning from low-value form fills.
- Import qualified or converted lead outcomes
- Review rejection reasons weekly
- Assign an owner and due date to every failed check
Practical questions
Questions that change the decision
How often should I run the full audit?
Quarterly, after a major tracking change, or before a material budget increase. Run the shorter weekly review between full audits.
Does a failed check mean I should pause campaigns?
No. Measurement failures may justify holding scale, but pausing should depend on risk, spend and whether the issue corrupts bidding.
Can this be used for Meta and LinkedIn?
Yes. The commercial checks are channel-neutral; the workbook includes space for platform-specific evidence and actions.
Sources and platform documentation
Platform features change. These links point to the primary documentation used for the implementation details above.