Before putting more money into a campaign, I would want to know what the current spend is actually doing.

Are the conversions useful? Are the searches relevant? Can someone follow the promise in the ad through to the landing page and the next step? A lower cost per conversion does not answer all of those questions.

A PPC audit is a structured review of measurement, traffic, ads, landing pages, and budget decisions. The useful output is not a long list of settings to change. It is a short list of problems to fix, questions to test, and things that need more evidence.

This checklist uses Google Ads examples, particularly for lead generation. The same business questions can guide a wider paid-media review, but each platform’s measurement and reporting need their own checks.

1. Write down what a conversion represents

Start with the action behind the number. A contact-page visit, a submitted form, a qualified enquiry, and a signed customer are different events. Do not treat them as interchangeable just because they appear in the same report.

In Google Ads, check both the conversion action and the goals each campaign actually uses. Primary actions can be used for bidding when the campaign uses their goal. Secondary actions normally appear in All conversions for observation. There is an important exception: actions included in a custom goal can be used for bidding even when marked secondary. Google’s primary and secondary conversion guidance

  • What triggers each action, and could the same enquiry trigger more than one?
  • Which actions are included in the campaign’s bidding goals?
  • Are values real transaction values, estimates, or simply assigned numbers?
  • Can the business identify what happens after the tracked action?

Then test the journey. Submit a clearly labeled test enquiry with the team’s permission, check the tracking, and confirm it reaches the right place. Google’s conversion summary includes tracking statuses and troubleshooting options, including Tag Assistant where available. A configured action is not the same as a verified end-to-end journey. Google’s conversion-summary guidance

2. Make the comparison fair

I would compare similar periods and note anything that changed: budget, bidding, offers, landing pages, tracking, or the sales process. A result that changed after an edit is a reason to investigate, not proof that the edit caused it.

Do not compare a partial week with a full one. Also allow for conversions that happen after the initial click. Google documents conversion delay as a reason recent results can be incomplete. Check the account’s actual timing rather than assuming every business converts on the same day. Google’s data-discrepancy guidance

Record the date range, attribution settings, and what remains unreported. If you cannot yet evaluate qualified leads or revenue, say that plainly. Clicks and submitted forms may still help diagnose the journey, but they do not settle the business case.

3. Review the searches, not just the keywords

A keyword is your targeting input. The search term is what someone actually searched. Review the terms alongside the matched keyword, spend, conversions, and any available lead-quality information.

Separate clearly irrelevant intent from relevant but unproven intent. For a service business, a job seeker and a potential buyer may use similar words. That does not mean every broad query should be excluded, or that a query with no conversion after a few clicks has failed.

Google notes that some low-activity search terms are omitted for privacy. Treat the report as useful evidence, not a complete list of every query. Google’s search terms report documentation

For each finding, write down whether you would exclude it, improve the ad or landing-page match, or keep watching. Review negative-keyword scope before applying it so a cleanup does not block useful demand.

4. Follow the promise through the landing page

Open the ad’s destination on a phone. Can you quickly find the thing the ad promised? Is the next action obvious? Does the form work? What happens after submission?

I would check the specific path that is spending money, not just the homepage. An ad offering a consultation should not leave visitors searching a general services page for the booking link.

Keep the diagnosis specific. If people click but do not submit, review the page, offer, and form. If they submit but are not a fit, review the promise, qualification, and targeting. Neither pattern proves a single cause, but each gives you a different next test.

5. Separate performance from budget mix

An account average can hide very different jobs. Existing-brand demand, new-customer searches, remarketing, and broader discovery should not automatically get the same efficiency expectation.

Break the result into meaningful groups before recommending more spend. Ask where the additional budget would go and what evidence supports that choice. A strong average does not establish that the next dollar will perform the same way.

For Performance Max, the channel reporting guide can help you inspect the delivery mix. Use that reporting to ask better questions, not to assume every channel can be assessed in isolation.

6. Turn the audit into a decision list

I would ask an audit provider for evidence, an owner, and a next action for each material finding. Screenshots without a recommendation leave the implementation work unresolved.

  • Fix: a verified problem, such as a broken form or an incorrectly triggered event.
  • Test: a plausible improvement that needs a comparison, such as a clearer offer or different landing-page message.
  • Watch: an unresolved question where delivery, conversion lag, or business feedback is still insufficient.

A hypothetical audit finding

Suppose a campaign records both a contact-page visit and a submitted enquiry as primary actions in the goal it uses. The headline conversion total looks healthy, but the business cannot tell how many useful enquiries arrived.

The next step is to verify the event setup and agree which action should guide bidding, not immediately raise the budget. Keep the page visit available as a diagnostic signal if useful, then review the actual enquiry journey and quality. This is an illustrative scenario, not a client result or a guarantee that changing goals will improve performance.

Copy this PPC audit worksheet

Campaign and business objective:
Review period and comparison period:
Recent changes and known reporting delays:

Finding:
Evidence and exact report or test:
What the tracked conversion means:
What we know about the business outcome:
What remains unknown:

Decision: Fix / Test / Watch
Recommended action and reason:
Owner and implementation approval:
What stays unchanged:
Success measure and review date:

Start with the finding most likely to change your next decision. Resolve measurement problems before using those numbers to justify scaling. Then test a focused improvement and check what happens at the next funnel step.

That is what I would want from a PPC audit: a clearer answer about where the current spend is going, and a practical reason for what to do next.

Sources

Platform documentation reviewed November 2, 2026. Recommendations should be checked against your account setup and business objectives.