PPC audit checklist: nine checks before the next budget decision
A PPC audit should leave the team able to decide what to fix, what to test and whether to keep spending. The checklist below starts with the business outcome and follows the customer's path through measurement, delivery and the product. Its depth depends on the account and the available evidence, not a fixed timer.
Set the scope and preserve the starting point
Write down the question the audit must answer, the accounts and markets in scope, and the outcome each campaign is meant to produce. Include the business constraints: margin, stock, sales capacity or onboarding capacity. A high cost per lead means little until the team knows what kind of lead is useful.
Use read-only access for diagnosis and save the settings and reports relevant to the review. Select a period that allows for conversion delay and mark product releases, pricing changes, outages and seasonal demand. Record unavailable data as unknown. A rapid first pass can identify priorities, but it cannot certify parts of the account that were never inspected.
Trace one outcome before interpreting the totals
Follow a permitted test action through the page, product or CRM and the advertising system. Inspect the trigger, event identifier, value, currency and status. Then check which events the campaign actually uses for optimization. In Google Ads, primary and secondary labels matter, but actions in a selected custom goal can be used for bidding even when labelled secondary.
Do not expect different systems to produce identical totals without aligning their definitions. They may use different attribution windows, dates, models and inclusion rules. One business outcome can also be credited by several ad platforms, so adding their attributed revenue can overstate total revenue.
Where browser and server integrations report the same event, test the platform's deduplication setup rather than assuming two integrations are automatically better or automatically double-counted. A measurement defect needs repair before the affected metric can support a budget decision. The server-to-server tracking guide explains why delivery of an event is different from proving advertising caused it.
Use the nine checks as a working record
For each check, record the result, the inspected campaigns or paths, the evidence and the next action. Use "unknown" when access or event history is missing, and "not applicable" only when the check genuinely does not apply. A clean sample does not prove every untested location, page or campaign is correct.
Choose controls that exist in the current campaign type. Keyword exclusions belong to search intent, while placement and audience controls answer different questions. Automated audience expansion may make a seed audience a suggestion rather than a fixed boundary. For Google-specific reporting and PMax controls, use the Google Ads audit walkthrough.
| Check | Question to answer | Evidence to inspect |
|---|---|---|
| 1. Business outcome | Which customer action is worth acquiring? | Campaign goal and product or sales definition |
| 2. Conversion recording | Is that action recorded once with the right value and status? | Test event and matching business record |
| 3. Reporting basis | Are dates, windows and attribution rules comparable? | Report settings, delays and reconciliation |
| 4. Customer quality | Do acquired people qualify, activate, pay or stay? | Comparable CRM or product cohorts |
| 5. Targeting and eligibility | Can the intended audience receive and use the offer? | Queries, delivery, markets, policy and exclusions |
| 6. Ad promise | Are the delivered assets accurate and appropriate? | Actual ad combinations and destination |
| 7. Product journey | Can the visitor complete the promised next step? | Mobile and desktop paths, stock and form behavior |
| 8. Spend and bidding | Are the target, spend exposure and capacity acceptable? | Budget rules, goals, economics and change history |
| 9. Next decision | Is this a proven defect, a testable hypothesis or an unknown? | Ranked action, owner and review condition |
Prioritize changes by consequence, not by a generic score
A broken checkout, invalid destination or materially incorrect conversion signal can warrant immediate action by the authorized owner. A suggestion to replace an old creative is a hypothesis unless the evidence supports it. Age, frequency and a low click-through rate alone do not establish fatigue or poor customer value.
Read the actual spend rules before recommending a budget change. For most Google campaigns using an average daily budget, the daily billed limit is twice that budget and the monthly limit is 30.4 times it, subject to the relevant budget-change rules. Other budget types have different limits. The number labelled "daily budget" is not always a hard daily ceiling.
For each proposed change, name the person responsible and the observation that would justify the next decision. A tracking repair may change reported CPA without changing customer behavior, so mark the break in the measurement series. Use a suitable experiment for uncertain performance improvements and check that any gain survives into qualified customers or product use. Do not promise a fixed savings percentage merely because the checklist found several issues.
FAQ
Can this PPC audit checklist cover Google, Meta and TikTok?
The business, measurement and product-path questions transfer. The available targeting, attribution, exclusions and budget controls differ by platform and campaign type, so verify the relevant implementation rather than copying settings.
How long should a PPC audit take?
A first pass can be time-boxed, but completion depends on the scope, access, event history and tests required. State what was examined and what remains unknown instead of promising a complete audit in a fixed number of minutes.
What if conversion data is unavailable?
You can still inspect delivery, assets and visible customer paths. You cannot confidently judge customer acquisition economics from clicks alone. Record that limitation and agree how the missing outcomes will be measured.
Sources
- Google Ads: primary and secondary conversion actions
- Google Ads: spending limits
- Google Ads: campaign total budgets
- Google Ads: Performance Max audience signals
Find where your funnel loses users
We can review acquisition, onboarding and measurement together, then decide what to improve first.
