What Should a Google Ads Audit Cover?

Jesse Heslinga
Aug 5
14 min read

You received a 12-page PDF. Or maybe you are shopping for an agency and one of them offered a free audit. Either way, you want to know: did it actually cover anything?

A Google Ads audit is useful when it is thorough. It is a waste of time when it is a surface scan dressed up to look like analysis. I have audited accounts that agencies had been running for 12 months and found the conversion tracking broken from day one. No one noticed because nobody went looking.

The good news is that a proper audit is not complicated to define. There are 8 areas that matter. If an audit covers all of them, you get a real picture of the account. If it skips any, the findings are incomplete by definition.

The Short Answer

A real Google Ads audit covers 8 areas: conversion tracking, campaign structure, search terms and negatives, bidding strategy, ad copy and testing, landing pages, geography targeting, and budget allocation.
Conversion tracking is the most important area. If it is broken, nothing else in the account can be trusted.
The most common finding in audits is miscounted or broken tracking, not bad keywords or poor ad copy.
A shallow audit like Google's own Recommendations tab or a 5-page PDF with generic scores is a marketing document, not a diagnostic.
If an audit finds nothing wrong, that is itself a red flag.

What a Google Ads Audit Actually Is

An audit is a structured review of a Google Ads account across every area that affects whether the account is generating real leads at a sensible cost. A performance report tells you what happened. An audit tells you why it happened and what is wrong.

An audit is also not Google's Recommendations tab. That tab is a product feature designed to increase spend. It suggests adding broad match keywords, raising your budget, and turning on more automated campaign types. Those suggestions are built to serve Google's revenue. An audit from your agency or a third party starts from a different question: is this account set up to generate leads efficiently, or is money leaking somewhere?

A real audit produces actionable findings. Each finding names the problem, explains why it matters, and says what to change. "Your Quality Scores could be improved" is not a finding. "Two of your five ad groups have a single ad with no variants, have not been tested in 11 months, and are running to landing pages with a 4-second load time on mobile" is a finding.

Audits are most useful in three situations: when you first take over an account from another agency, when performance has dropped and you want to understand why, and as a periodic health check even when things look fine on the surface.

Verdict: a Google Ads audit is a structured diagnostic. Expect specific findings with specific next steps, not scores and percentages.

Area 1: Conversion Tracking

This is the most important area in any audit, and the most commonly broken.

Conversion tracking is how Google knows which clicks turned into leads. It is also what Smart Bidding uses to decide who to show your ads to. If the tracking is wrong, the bidding is working from bad data, and every decision you make based on reported performance is unreliable.

Four things to check in a tracking audit:

What is being counted as a conversion? Form submissions and phone calls are valid. Clicks to a phone number, page views, and scroll depth are not. Counting those inflates the lead number and misleads Smart Bidding. For a full breakdown of what to count as a lead in Google Ads, that post covers exactly which conversion actions should be set as primary.
Is the tracking firing once per conversion? A double-fired tag records two conversions from one form submission. This is common when both a global site tag and a Google Tag Manager tag are active on the same thank-you page.
Is data actually flowing? Check the Conversion actions section of the account. If the last recorded conversion was three weeks ago and the campaign has been active, something is broken.
Is the primary conversion action set correctly? Secondary actions (micro-conversions like video views or page scrolls) can be tracked without being used for bidding. Many accounts send every tracked action into the primary slot, which gives Smart Bidding mixed signals.

In one clinic account I audited, the conversion action was set to a thank-you page view that was also being hit by direct traffic that never submitted a form. The account was reporting 88 leads per month. Once the tag was switched to a form submission event and the tracking cleaned up, actual lead volume was 27 per month. Smart Bidding had been optimising toward a number that was more than three times reality for the better part of a year.

Common Mistake: Tracking that looks fine because conversions are being reported is not the same as tracking that is accurate. The volume, source, and tag logic all need to be confirmed. I have seen accounts reporting 40 leads a month with a tag firing on every page load, not just on form confirmation.

Verdict: if the tracking is wrong, the audit is already worth it regardless of what the other sections find.

Area 2: Campaign Structure

Campaign structure is how the account is organised: which campaign types are running, what they target, how they relate to each other, and whether that setup makes sense for the account's goals and budget.

Main things to check:

Are the campaign types appropriate? Search campaigns are the right default for lead-gen. Performance Max can work, but it needs enough conversion data to optimise properly. Google recommends 50 conversions per month per campaign; in practice, I see Performance Max struggle below 30. Shopping campaigns are for e-commerce and have no business in a service account. A Display or Video campaign eating budget without a specific remarketing purpose is usually a problem.

Are campaigns segmented logically? A single "All services" campaign with 15 ad groups mixing unrelated services is hard to optimise. Keyword themes competing for the same budget cannot be controlled individually.

Are campaigns fighting each other? If PMax and Search are both targeting the same intent, PMax almost always wins the internal auction. That means your tightly structured Search campaign runs less than you think.

Is the account too complex for its budget? Five campaigns on a €1,500 monthly budget means each campaign has €300. That is often not enough data for any of them to perform reliably.

Verdict: campaign structure problems usually mean budget is spread in a way that prevents any single campaign from performing. Consolidation is almost always the fix.

Area 3: Search Terms and Negative Keywords

Google Search campaigns show your ads in response to searches. The question is: which searches? The search terms report shows the actual queries that triggered your ads. It is one of the most revealing reports in the account.

Common problems in lead-gen accounts:

Informational queries eating budget. People searching "how to fix a leaky tap" are not hiring a plumber today. If that query is triggering your plumbing ads, you are paying for a click that has almost no chance of converting.
Competitor research queries. "Plumber reviews", "best plumber near me", "plumber cost guide" often come from people in research mode, not ready to call.
Job seekers and suppliers. "Plumber jobs", "plumbing apprenticeship", "buy plumbing supplies" are common in accounts that have never added basic negatives.
Wrong geography. If you serve Amsterdam but ads are showing for "plumber Utrecht", that is a structure or geo-targeting problem showing up in the search terms report.

A solid audit pulls 90 days of search terms data and categorises the waste. It will tell you roughly what percentage of spend is going to queries with no realistic chance of converting, and which negative keyword patterns would prevent most of it.

Keep in Mind: Google's move toward broader match types since 2023 has made this worse. Phrase match now behaves more like the old modified broad match, and broad match with Smart Bidding covers a significantly wider range of queries than most advertisers expect. Reviewing the search terms report is no longer optional.

Verdict: the search terms report tells you where your money went. If the audit does not include a structured review and a negative keyword gap analysis, it missed the most actionable cost-saving work in the account.

Area 4: Bidding Strategy

Bidding strategy determines how Google spends your budget at each auction. The wrong strategy can quietly underperform for months without any obvious alarm signal.

The main strategies for lead-gen and when they apply:

Maximize Clicks is appropriate for brand-new accounts with no conversion history. It is not appropriate six months in once you have real data.
Target CPA tells Google what you want to pay per conversion and bids accordingly. It needs at least 30 to 50 conversions in the last 30 days to work properly. Below that, the algorithm does not have enough signal and performance varies widely.
Maximize Conversions with a Target CPA cap is often the right middle ground for accounts with some data but not enough for Target CPA to stabilise.
Target ROAS is for e-commerce. It should almost never appear in a lead-gen account.

Two specific problems to check in an audit:

Is the strategy appropriate for the conversion volume? A Target CPA campaign with 8 conversions in the last 30 days is essentially running without proper signal. Smart Bidding cannot learn effectively from that few data points.

Is the bid strategy mismatched with the conversion tracking? If tracking is overcounting (see Area 1), Target CPA will hit the target on paper while the real cost per lead is much higher. This is the most common way a tracking error hides inside what looks like a bidding problem.

Data Tip: Google's own guidance puts the Target CPA minimum at 50 conversions per month. In accounts I run, 30 is often enough for stable performance, but below that I switch to Maximize Conversions until volume builds.

Verdict: bidding strategy needs to match the actual conversion volume and the accuracy of the tracking feeding it. Both need to be checked together.

Area 5: Ad Copy and Testing

Ad copy is what appears in search results. It determines your click-through rate, your Quality Score, and how well the searcher's intent matches what they see before clicking.

The ad copy review covers:

RSA asset count. Google recommends at least 8 to 10 unique headlines and 3 to 4 descriptions per Responsive Search Ad. Accounts with 3 or 4 headlines give Google almost nothing to test. The ad strength score will say "Poor" or "Average" and the algorithm has no way to find the best combination.

Is there any real testing? Two RSAs in an ad group is the minimum for an A/B structure. But if both ads have nearly identical headlines with no meaningful variation (same offer, same CTA, same format), there is nothing to learn from running them side by side.

Are the headlines specific? Generic headlines like "Professional Plumbers" and "Get a Free Quote" appear in thousands of accounts. They do not differentiate the business, and they do not give Google strong signals about the intent being targeted.

Overuse of pinning. Pinning forces certain headlines into specific positions. Pinning headline 1 and 2 on every ad limits what Google can test and reduces the RSA format's effectiveness.

Verdict: if the account has not tested meaningfully in 6+ months, the ad copy is stale. An audit should flag this and specify what a proper test would look like.

Area 6: Landing Pages

An ad click that lands on the wrong page, a slow page, or a page that makes conversion difficult wastes the full cost of that click. The landing page is technically outside Google Ads, but it belongs inside any competent audit.

Things to check:

Message match. If your ad says "Emergency Plumber Amsterdam, Available Now" and the landing page is a generic homepage, the message breaks. The visitor expected an emergency service, not a company overview. Bounce rate goes up, conversion rate goes down.

Load speed. Google's Core Web Vitals research shows that pages taking over 3 seconds to load lose a significant share of mobile visitors. Check with PageSpeed Insights. If the page takes 5 to 7 seconds on mobile, the ad is paying to show people a loading spinner.

The form. How many fields? Five fields for a basic contact request is too many. Is the phone number visible above the fold? Does the form redirect to a thank-you page (important for tracking)?

Service intent match. A plumber running ads for "boiler repair" should send that traffic to a boiler-specific page, not a general "our services" page.

Pro Tip: Landing page quality feeds your Quality Score, which affects your cost per click. A better-matched, faster landing page does not just convert better. It also makes your ads cheaper to run.

Verdict: the landing page review is where many audits stop being thorough. It is easier to audit the account than the page, but the page is often where the lead actually dies.

Area 7: Geography Targeting

Geography targeting controls where your ads show. For service businesses, this is usually the most specific setting: a city, a radius, or a list of postcodes. Getting it wrong means budget spent on people you cannot serve.

What the audit checks:

Is the geo setting "presence" or "presence or interest"? The default in Google Ads includes people who are interested in your target location, not just present there. For a local service business, that setting can send ads to someone in a different city who recently searched for something about your city. The setting should be "presence only."

Is the radius appropriate? A 50km radius around a single-location business often covers areas the business cannot actually serve. Check whether the geographic spread in the impressions data matches the real service area.

Are location exclusions set? If you serve Amsterdam but not certain surrounding municipalities, those should be excluded explicitly.

Is any location consistently eating budget with no conversions? This shows up in the geographic performance report. Some locations within a broad radius simply do not convert. Excluding them frees budget for the ones that do.

Verdict: geography targeting is often a quick fix with a meaningful impact. Most accounts I audit have either the "presence or interest" setting left on by default, or a radius that is too wide for the actual service area.

Area 8: Budget Allocation

The final area is whether the budget is distributed sensibly across campaigns, and whether the total budget is appropriate for what the account is trying to do.

What to check:

Is any campaign permanently budget-limited? A campaign hitting its daily limit by early afternoon is missing the afternoon and evening searches, which for many service businesses are the highest-intent hours. That may be your best-performing campaign running at half-speed.

Is the budget split logical? If one campaign generates 80% of the leads, does the budget reflect that? Or is it split evenly out of habit?

Does the budget math work? If the target CPL is €100 and the daily budget is €50, Google can only afford half a conversion per day. That is not enough for the algorithm to function. A simple rule: the daily budget should cover at least 3 to 5 target CPLs to give Smart Bidding room to operate.

Keep in Mind: Budget allocation problems are often left out of audits because they are awkward to raise. The finding implies the client is not spending enough, which agencies worry will come across as upselling. A thorough audit raises it anyway, because the maths not working is a structural problem that cannot be optimised around.

Verdict: even a well-structured account underperforms if the budget math does not add up.

The Audit Checklist: Real vs. Shallow

Area	What a real audit checks	What a shallow audit says
Conversion tracking	Tag logic, firing accuracy, what is counted, double-fire check	"Conversion tracking is set up"
Campaign structure	Campaign types, segmentation, internal competition, budget per campaign	"You have 4 campaigns running"
Search terms and negatives	90 days of actual queries, categorised waste, negative gap analysis	"Negative keywords are present"
Bidding strategy	Appropriateness for conversion volume, match to tracking accuracy	"Smart Bidding is enabled"
Ad copy	RSA asset count, meaningful test variants, headline specificity	"Ad strength is Good"
Landing pages	Message match, load speed, form UX, mobile experience	Not covered
Geography targeting	Presence vs. interest setting, radius accuracy, exclusions, geographic CPL	"Targeting is set to your city"
Budget allocation	Campaign-level limits, CPL math, budget vs. volume expectations	"Budget is set to €X/day"

What "the Audit Found Nothing" Actually Means

If an auditor comes back and says your account is fine, one of two things is true: the account is genuinely well-run, or they did not look hard enough.

In six years of auditing accounts, I have seen the first maybe 10% of the time. The other 90%, something was wrong. That is not a critique of every advertiser. It reflects how complex Google Ads has become, how much it has changed since 2022, and how easy it is to overlook something when you are running an account day-to-day.

The most common things missed in shallow audits are subtle. Tracking that records conversions but overcounts them. Geography that silently expanded via the "presence or interest" setting. Performance Max campaigns that look like they are performing but are mostly converting on branded queries that would have come in anyway.

An audit that finds nothing is either incomplete or dishonest. Push back. Ask which specific reports they pulled and what the search terms data showed.

How I Handle This for Lead-Gen Clients

When I audit an account, I work through all 8 areas in this order: tracking first, always, because everything else depends on whether the data can be trusted. Then structure, search terms, bidding, ad copy, landing pages, geography, and budget.

I document each finding with the specific data behind it. "Your account has conversion tracking issues" is not useful. "Your thank-you page fires the conversion tag twice: once from the global site tag and once from a GTM trigger, resulting in double-counted conversions. Your actual conversion volume is approximately half what the dashboard shows" is useful.

The output is a prioritised list of fixes, not a score. If you want to do the work yourself, you can. If you want to hand it to me, the account review is the first step for every new client.

Ready to Find Out What Your Account Is Actually Doing? Most clients I work with come in thinking their account is basically fine and the problem is just the market or the budget. In the majority of cases, the audit finds something structural that has been costing them money for months. The fix is usually not expensive. The delay in finding it is.

Book a free account review

Frequently Asked Questions

How long does a Google Ads audit take?

A thorough audit of a single account typically takes 3 to 5 hours done properly. This includes pulling search terms data, checking tag logic in Google Tag Manager, reviewing landing page speed and match, and going through the geographic performance report. A 30-minute audit is a surface scan.

How much does a Google Ads audit cost?

Independent specialists typically charge €300 to €1,500 depending on account size and depth. Some agencies offer a free audit as a first step, which is fine as long as you verify it covers all 8 areas above, not just account structure and ad strength scores. I offer a free account review as the first step for potential clients.

Can I audit my own Google Ads account?

Yes, if you know where to look. Start with conversion tracking (use Google Tag Assistant or GTM Preview to confirm the tag fires once on the thank-you page), then pull 90 days of search terms, then check the geographic performance report. Those three alone often surface the biggest issues.

What is the most common problem found in a Google Ads audit?

Broken or miscounted conversion tracking. It appears in the majority of accounts I audit, even accounts that have been actively managed. The second most common is wasted spend on irrelevant search queries that have never been excluded. Both problems compound over time because Smart Bidding learns from the bad data.

How often should you audit a Google Ads account?

A full audit makes sense when you take over an account, when performance drops meaningfully, or after any major campaign restructure. Outside of those events, a lighter quarterly check of search terms, bidding, and conversion tracking accuracy catches most drift before it becomes expensive.

What should an audit report include?

Specific findings, not scores. Each finding should name the problem, show the data behind it, explain the impact on performance, and say what to change. An audit report that consists mostly of graphs and percentages without named recommendations is a presentation, not a diagnostic.

Is Google's Recommendations tab an audit?

No. Google's Recommendations tab generates suggestions based on what will increase Google's ad revenue: broad match, budget increases, new automated campaign types. Those may sometimes be valid, but they are generated without understanding your goals, your CPL targets, or your business model. A third-party audit asks different questions.

What is Performance Max and why does it come up in audits?

Performance Max is a campaign type that runs across all of Google's inventory using automation. It comes up in audits frequently because it is often turned on without enough conversion data to run properly, and because it can take traffic away from well-structured Search campaigns. If a PMax campaign is in the account, the audit needs to check what assets are running, what conversion data is feeding it, and whether it is generating leads that would not have come in anyway.

About the author

Jesse Heslinga | Google Partner | 7+ Years Google Ads | Lead-Gen Expert

I run Google Ads for lead-gen service businesses at Groove Media across clinics, home services, and professional services, working with clients directly, no account-manager layer. I build every account around one question: is this spend turning into real customers, not just cheap form fills?