Google Ads audit: how to check your account yourself

A Google Ads audit checks whether your account is actually working toward the right business goals: is the conversion data correct, are you buying the right search intent, and can the budget be justified economically? The optimization score alone doesn’t answer any of that. This checklist takes you from measurement to account structure, and only then to ads and testing.

Want your account checked independently? My Google Ads audit (in German, from €1,900) connects an Ads account, a GA4/GTM setup and an agreed lead channel. You get written findings and a prioritized action plan your existing team can implement. Delivered within seven working days of full access and a briefing; scope and price are fixed upfront.

1. Is the account even measuring correctly?

Before anything else makes sense, this question needs an answer: does the account know what a success actually is? Everything after this, bidding strategy, structure, assets, optimizes against the wrong number otherwise.

• Conversion tracking. Does the account count real inquiries, purchases or calls, or also page views, form clicks without a submit, or confirmation pages people load twice by accident? I’ve seen accounts reporting “conversions” that were really bot traffic. A 15-minute check is enough to verify this yourself.
• GA4 or data manager connection. Is Google Ads cleanly linked to GA4, or are there two separate sets of numbers running side by side that never match up at month’s end?
• Google Business Profile linked. For local businesses this is the foundation for location assets and local campaign signals. Skip the link and you’re giving away visibility that costs nothing.

If the foundation is wrong, close the measurement gaps before drawing any reliable conclusion about performance. Cross-check reported inquiries against real lead records: a conversion inside the ad account is not yet a qualified prospect.

2. Can the account structure actually support good decisions?

Structure should enable decisions: which search intent buys budget, which campaign delivers the right inquiries, and which goals steer the bids? Extra campaigns only help when the split answers one of those questions.

• Keywords and the negative keyword count. How granular are the keywords, and how thorough is the negative list? Too few negative keywords mean wasted spend on searches that never convert. Careless negative keywords, on the other hand, can block your own best search terms. I’ve audited accounts that had accidentally negated their strongest ad group.
• Semantic grouping of keywords into ad groups. Are thematically related keywords sitting in the same ad group with a matching ad, or was everything thrown into one big group? Clean semantics are the precondition for a good Quality Score, more on that below.
• Bidding strategy. Do the conversion goal, data volume, time-to-close and budget actually fit the bidding strategy? Thin history isn’t an automatic disqualifier for Smart Bidding, but it raises the uncertainty in the assessment. Check first whether the signals being fed back reflect the lead quality you actually want.
• Campaign type mix. How much of the budget sits in Performance Max? Google provides a search terms report for PMax too. Low-volume search terms can be missing for privacy reasons, and search term insights partly bundle that demand together. The report also doesn’t cover the full delivery across every PMax channel. Check the data you do have and name the remaining gap. When a Performance Max campaign works, and when it just costs you visibility, is a separate question.
• Location setting: presence or interest. The default “presence or interest” setting includes people who are merely interested in your target region without actually being there. For a local service business, that’s the fastest way to spread budget across Europe.
• Search and Display Network partner boxes. Are these ticked even though the campaign is meant to be pure search? Partner networks often bring cheaper clicks, but also lower-intent ones.

Several problems at once? In the paid audit with a sample finding (in German) I separate confirmed defects from hypotheses and decide which check or change makes sense first. Implementation stays a separate engagement.

3. Is the potential that’s already there being used?

Even with clean tracking and a solid structure, potential often gets left on the table, in places nobody checks regularly because they don’t show up in the standard dashboard.

• Sitelinks and other assets. Fully filled in, current, and actually used? Call assets, price assets, snippets: assets can show extra relevant information, but their display is never guaranteed. The usual costs for ad interactions still apply.
• Quality Score. Use the three components, expected CTR, ad relevance and landing page experience, to diagnose individual keywords. The score you see is not a direct auction factor and not an economic target metric in its own right. A low score by itself proves neither a specific CPC markup nor wasted budget.
• A/B test opportunity cost. How many months has the same ad run unchanged, with no variant tested against it? Write a concrete hypothesis and a decision threshold. With few conversions a test can stay undecided for a long time; extra variants aren’t an end in themselves at that point.

The bottom line

A real audit isn’t a snapshot of the optimization score. It’s a sequence of three questions: are the numbers right? Can the structure do anything with them? Is the remaining potential being used? If even the first answer is uncertain, measurement belongs at the top of the action plan.

Want to see in two minutes where your own account stands? The Google Ads settings check (in German) covers part of this list in a short self-test. And if you want to know the five typical traps that these exact structural mistakes lead to: the five lies Google Ads tells you.

If you’d like a second pair of eyes on your account, write to me.