Most Google Ads audits aren't audits. They're a 60-page PDF that screenshots your settings, lists every keyword without a conversion, and ends with a generic "switch to Smart Bidding" recommendation. You pay $500 or $2,000 for a checklist that an intern with a template could run.

A real Google Ads audit is the opposite. It's short, it's specific, and it gives you a punch list you could act on Monday morning. The point isn't to catalogue everything that's wrong. It's to find the three or four issues that are actually moving the dial, in the order you should fix them.

If you're commissioning an audit, or auditing your own account, here's what to look for. Each one of these has cost an account I've audited this year either real money or real performance, and most accounts I see have at least three of them running at the same time.

1. Conversion Tracking Integrity (Look Here First)

This is the one most audits skim and most accounts get wrong. Every other finding in the audit depends on the conversion data being correct, because that's the signal Smart Bidding optimises on. Garbage in, garbage at scale.

What a real audit checks:

Is the Google Tag actually firing sitewide, or only on the thank-you page? You'd be amazed how often it's the latter.
What's set as the primary conversion? If it's a button click or a pageview, the account is being told "buy us more of these" for an action that doesn't correlate with revenue.
Is Enhanced Conversions enabled and passing hashed lead data?
For lead-gen accounts: is Offline Conversion Import flowing in the last 7 days? If the CRM knows which leads turned into customers and Google doesn't, every bid decision is being made on a half-broken signal.
For e-commerce: are conversion values dynamic (order total) or static (a placeholder like $50)? Static values are silently common and they wreck tROAS.

A UK home improvement account I audited in May had Enhanced Conversions on, looked clean on the surface. Their primary conversion was a "Get a Quote" button click, not a form submission. Click rate was 12 times the form submission rate. Smart Bidding had been optimising for clicks that didn't lead to leads for six months. Their CPL was 40% higher than it should have been and nobody had spotted it because the conversion count looked healthy.

If your audit doesn't dig into conversion tracking at this level, the rest of the audit is built on sand.

2. Brand vs Non-Brand Bleed in Performance Max

This is the finding that costs the most money and the one templated audits miss most often. Performance Max happily eats your brand traffic, claims credit for it, and inflates the campaign's ROAS while your non-brand acquisition quietly starves.

What to check:

Is brand traffic excluded from Performance Max via brand exclusions or a brand negative keyword list at the account level?
What's the share of PMAX conversions coming from branded queries? If it's over 30%, you're paying for traffic you'd have got for free.
Is there a separate brand search campaign protecting the term, or is PMAX the only thing bidding on it?

A German client I worked with had PMAX showing a 4.2 ROAS and Search showing 1.8. Looked like PMAX was the hero. Pulled the brand search term breakdown out of PMAX and roughly 60% of the conversions were people searching the company name. Blended ROAS was actually closer to 2.4 once you removed the brand cannibalisation, and the non-brand acquisition was loss-making. The fix wasn't "spend more on PMAX". It was the opposite.

3. Search Term Waste at the Keyword Level

Every audit looks at search terms. Most stop at "here are the top zero-conversion queries". A real audit goes further:

What's the n-gram pattern? If "jobs", "free", or "DIY" shows up across multiple search terms, you need a root-level negative, not 30 individual ones.
What share of total spend went to search terms that aren't a match for your service? In healthy accounts it's under 15%. In broken ones, it's 35-50%.
Are existing negatives blocking good traffic? This one's almost never checked. Aggressive historical negatives sometimes block queries that have since become commercial.

I audited an Australian services account in April. 41% of 30-day spend had gone to search terms that didn't match the client's service category. Most of it was traceable to a single broad-match keyword that nobody had pruned since the account was migrated to broad in late 2025. One negative fixed about $3,200 a month in waste.

If you're not sure how to read this part of the report, the search term report guide walks through what to actually look for.

4. Bidding Strategy Fit

Smart Bidding needs data. tCPA needs at least 30 conversions in the last 30 days, tROAS wants 50. Below that, the algorithm doesn't have enough to train on and you'll see CPA bounce around for no obvious reason.

What a real audit checks:

Are the bidding strategies appropriate for the conversion volume per campaign?
Are tCPA or tROAS targets realistic vs the campaign's historical performance, or were they set 18 months ago and never revisited?
Are any campaigns on Maximise Clicks where conversions are tracked? That's usually a leftover from a launch phase nobody changed back.
Is the seasonality adjustment being abused? Some accounts have permanent "seasonality" set, which isn't what the lever's for.

The bidding strategy doesn't need to be exotic. It needs to fit the data. Most accounts I audit have at least one campaign on a strategy the data can't support.

5. Schedule and Geo Signals That Don't Match the Business

Ad schedule and location targeting are usually set once at launch and never looked at again. Audits skip them because they look fine on the surface. They're not always fine.

What to look for:

Does the schedule match when the business can actually service the lead? A debt advice account I audited was getting 60% of its lead volume between 8pm and 11pm, with the call team off until 9am. Half those leads went cold before anyone called them.
Is the geo targeting set to "Presence in or interest in" or just "Presence in"? The default targets people searching about your locations from anywhere, not people in your locations. For local services, that's a problem.
Are there campaigns targeting countries you don't service? You'd be surprised - account migrations often duplicate location settings without thinking.

For the debt advice client, narrowing the schedule and shifting the budget to in-hours dropped CPL by 27% in three weeks. The audit finding wasn't "your ads are bad", it was "your ads are running when nobody can answer the phone".

6. Creative Depth, Not Just Ad Strength

Old audits cared about per-RSA "Ad Strength". New ones should care more about library depth. Google rotates and remixes assets across formats now, and AI Max will lean on that even harder.

What to check:

How many active RSAs per ad group? Two minimum, ideally three.
How many sitelinks, callouts, and structured snippets per campaign? Four of each is the baseline.
Is there a video asset on any campaign? PMAX and Demand Gen need them and most accounts don't have them.
Are there image assets sized for every placement, or are some campaigns running with one square logo and nothing else?

Asset thinness is the most common silent issue. Account looks "fine" because the ads are showing. They're just not showing in the formats and placements that produce the best CTR.

7. The Change History Nobody Reads

This is the one most audits skip entirely and it's where the surprises live. The change history log shows who made what changes when. For accounts that have had multiple managers, agencies, or freelancers touch them, it's where you find:

Bid changes made yesterday by an automated rule nobody knew was active
Conversion goals quietly switched in 2024 that nobody flagged
Budget changes that explain the unexplained spend spike last month
Logins from accounts that shouldn't have access

My take: every audit should start with 90 days of change history before touching anything else. It saves hours of "why is this account behaving like this" diagnosis. It also occasionally surfaces something serious - I had a client this month whose change history showed login activity from an account that shouldn't have had access, and an MCC-wide review caught a credential compromise across multiple accounts. That's a rare one, but the only reason I caught it was the change log.

Red Flags Your Google Ads Audit Is Rubbish

You can usually tell within five pages:

It's a 60-page PDF and you can't find a single dollar figure for impact
Every finding ends with "consider testing" or "review and optimise"
It recommends "switching to Smart Bidding" without checking whether the data supports it
It doesn't reference your actual conversion data, just account settings
The recommendations are generic enough that they'd apply to any account in any industry
There's no prioritised order, just a list

A good audit looks more like a punch list than a report. Ten findings, ranked, each with the dollar impact and the order to fix them in. If you can't act on it Monday morning, it's a deliverable not an audit.

Key Takeaways

Conversion tracking is the first thing to check. Everything else depends on it being right.
Performance Max almost always eats brand traffic unless explicitly excluded. Check this every time.
Look at search term waste as an n-gram pattern, not a list of individual queries.
The bidding strategy has to fit the conversion volume. Most accounts have at least one campaign where it doesn't.
Schedule, geo, and creative depth get skipped in most audits and usually have real money in them.
90 days of change history before any other diagnosis. The surprises live there.
If the audit's a 60-page PDF without a single dollar figure of impact, it's a checklist, not an audit.

If your account is overdue an audit and you're tired of templated PDFs, that's the kind of work I do. The output is a short prioritised punch list with dollar impact next to each finding, not a slide deck. Get in touch if you'd like one.

You can also read Google's own quality score documentation for the official line on what Google considers a healthy account. The practical version is what's above.

What to Look for in a Google Ads Audit