Google Ads Audit: How I Review an Account, in the Order That Matters

Last updated: 7 November 2026

A Google Ads audit is only useful if it checks things in the right order. I have worked in paid search since 2013 (my founder profile), and the most common mistake I see is fixing bids and ads while the tracking underneath is wrong. This guide shows the order I follow, what I look for at each step, and what one audit led to in a real account.

Quick Answer: What is a Google Ads audit, and what should it check first?

A Google Ads audit is a structured review of an account to find wasted spend and whatever blocks growth. I check it in a fixed order: conversion tracking first, then structure and settings, search terms, negative keywords, bidding, and landing pages. Each step depends on the one before it, so the order matters.

This table shows the order I follow and the question each step answers.

What does a Google Ads audit actually involve?

An audit is a read-only review. I look at the account, the tracking and the landing pages, write down what is wrong and rank it, but I change nothing while auditing. Read-only access is enough, since it lets a user view campaigns and run performance reports without editing (Google Ads access levels).

The scope depends on the account. A local lead generation account may have a few Search campaigns and one call conversion. An online store may run Shopping, Performance Max, Search and remarketing, with a Merchant Center feed behind them. The order of checks stays the same in both, because each step depends on the one before it.

I read the change history early. It shows who changed what and when across the past two years, including paused campaigns, added keywords and old budgets (Google on change history). A sudden drop often lines up with a change someone forgot to mention.

Policy limits belong in the audit too. One Judaica store I worked on sells religious products, and Google does not allow advertiser-curated audiences in that sensitive category, such as Customer Match, your data segments and lookalike segments (Google personalized ads policy). An audit that recommends remarketing lists there would be wrong from the start.

Why do I check conversion tracking before anything else?

Because every automated bid strategy learns from what the account counts as a conversion. If tracking is wrong, every later finding rests on bad data. So I check tracking before I change bids or structure, every time.

Google splits conversion actions into primary and secondary. Primary actions show in the Conversions column and are used for bidding, while secondary actions only show in All conversions and do not affect bids (primary and secondary actions). The first thing I look at is which actions are primary, and whether each one is a real sale or lead.

What do I test in the tracking setup?

These are the tracking checks I run on every account:

• Which conversion actions are primary, and whether any are soft actions such as page views or add to cart.
• Whether the same sale is counted twice, for example by a Google Ads tag and an imported GA4 event.
• Whether conversion values are real order values or one flat number.
• Whether Google Tag Manager fires the tags on the right pages and forms.
• Whether enhanced conversions are set up and working.

Enhanced conversions use first-party data, such as email addresses, to improve the accuracy of conversion measurement. The data is hashed with SHA256 before it is sent to Google, and there is a version for website sales and one for leads that close offline (Google on enhanced conversions). My guide to Google Ads conversion tracking covers setup in more detail.

What does weak tracking look like in a real account?

In a commercial cleaning equipment account I audited, the conversion setup gave Google’s bidding very little clean data to work with. Forms and calls were not sending accurate signals (cleaning equipment case study). I cleaned the tracking first, and moved to conversion-focused bidding only once there was enough data to support it.

How do I review account structure and settings?

I check whether each campaign has one clear job. Brand and non-brand searches should sit in separate campaigns, and so should Shopping, Performance Max and Search. When intents are mixed, budget drifts to whatever is cheapest to win, not to what matters most.

The cleaning equipment account showed this well. Buy, hire and service searches shared the same campaigns, so someone wanting short-term hire saw the same ad as someone wanting a maintenance plan (case study). I split the account by product line and intent: scrubbers and sweepers, robotic cleaners, machinery hire and fleet maintenance, each with its own budget, bids and ads.

Location settings come next. Google’s default, Presence or interest, can show ads to people who have shown interest in your area, not only people in it. The Presence option limits ads to people who are in or regularly in your locations (advanced location options). For a local service business, that setting decides whether you pay for out-of-area clicks.

I also check auto-apply recommendations and treat optimization score with care. Google describes the score as an estimate of how well the account is set up to perform, and says it is not used by Quality Score (Google on optimization score). A high score is not proof of profit, so I never use it as an audit verdict.

What does the search terms report tell me in an audit?

It tells me what people actually typed before they clicked, which is often very different from the keyword list. Google defines a search term as the words a person enters when searching, while a keyword is what you add to an ad group to reach them (Google on the search terms report). The gap between the two is where most wasted spend hides.

Read it knowing its limits. Google leaves out search terms without enough query activity, to protect privacy, so the report is a strong sample rather than a full list. I sort by spend, then flag terms with clicks and no conversions, terms from the wrong audience, and terms that belong in a different campaign.

Match types explain a lot of what I find. On new Search campaigns, I start on exact match for roughly the first 15 days, because the traffic matches even though each click costs more. Broad match only works once an account has conversion data. My search terms report guide shows how I read it week to week.

What happened when the problem was the traffic, not the price?

In the Judaica store account, the earlier setup was paying for the wrong searches, not overpaying for the right ones. Spend was moved behind campaigns that matched searches for named products. Over 120 days, Google Ads cost fell about 37%, while ROAS went from about 4.90 to 17.15 against the previous 120 days (Judaica store case study).

The case study is open about its limits, and an audit should be too. The platform values are not deduplicated, the earlier period is not a control group, and ROAS there means revenue, not profit. That is why I judge results against break-even ROAS rather than a headline number.

How should negative keywords be set up?

Negatives should be built up front and held at account level wherever possible. I add them in phrase and exact match before launch, rather than waiting for waste to show up in reports. I prefer account-level negatives to campaign lists alone, so a miss at campaign level does not cost anything.

Google applies account-level negative keywords to search and shopping inventory in relevant campaign types, and they work for Performance Max as well. There is a limit of 1,000 negatives at account level (Google on negative keywords). That limit is enough for most accounts, but it rewards tidy lists over thousands of one-off terms.

In the cleaning equipment account, the negative list blocked residential and home cleaning searches, job and career searches, training courses and DIY repair terms (case study). Those searches had been spending a budget meant for facility managers. In an audit, I look for exactly this kind of leak: whole groups of searchers who will never buy.

Is the bid strategy right for the data in the account?

A bid strategy is only right if the account has the data to support it. For Search and Shopping, Google’s target ROAS needs at least 15 conversions in the past 30 days, with values assigned to them (Google on target ROAS). Below that, a target setting chases noise.

On low-volume accounts that run exact match only, CPCs run high, so I test manual CPC to control cost per click. Accounts with enough conversion data go on Maximize conversion value. Once ROAS has been stable for about 30 days with no target, I switch to target ROAS and push the target up by 15% to 20%. Google lets you enter your own target or take its suggestion, which is based on your actual ROAS over the last few weeks (target ROAS settings).

I judge every campaign against break-even ROAS, which is 1 divided by profit margin. As an example, a store with a 25% margin breaks even at a ROAS of 4, because 1 divided by 0.25 is 4. Anything above that earns more gross profit than it costs in ads, as I explain in my break-even ROAS guide.

I never quote a client a ROAS figure I will hit, especially for relaunched campaigns still in learning. And when the goal is new customers, I do not push live campaigns back into learning. I leave them as they are and test new-customer acquisition in a separate campaign that excludes the existing customer list.

Do the landing pages and product feed match the ads?

A click only pays off if the page delivers what the ad promised. I check the landing page for each main campaign on mobile and desktop, and confirm that the headline, offer and form match the search. In the cleaning equipment account, each campaign was aligned to a matching landing page and the enquiry paths were fixed (case study).

For Performance Max, I check which pages Google is allowed to send traffic to. Blog posts and other non-selling pages should be excluded. Google’s URL exclusion setting can do this, but the campaign’s final URL cannot be excluded, and in feed-based campaigns the homepage may still serve on some inventory (URL exclusion in Performance Max).

For online stores, the feed is part of the landing experience. I check product titles and descriptions against real search demand from Keyword Planner, Semrush volume and the search terms report, not AI guesses. In the Judaica store, the feed became the main lever once remarketing was off the table (case study).

What happened after I audited a B2B lead generation account?

After the audit and fixes, a commercial cleaning equipment supplier in Australia went from 29 to 154 Google Ads leads across two comparable 90-day periods, and cost per lead fell 73% (case study). The audit had found three problems. Searches from homeowners, job seekers and DIY repairers were eating a budget meant for facility managers, and buy, hire and service intent shared the same campaigns.

The third problem was the conversion setup, which gave bidding very little clean data. Each problem had a fix, and each fix depended on the one before it.

The fixes followed the audit order. Tracking was cleaned so forms and calls sent accurate signals. The account was split by product line and intent, negatives blocked the consumer searches, targeting was narrowed to Southeast Queensland service areas, and ads were rewritten for facility managers. Bidding moved to conversion-focused strategies once data volume allowed.

This table compares the roughly 90 days before the changes with the 90 days after, using the Google Ads figures published in the case study.

Source: Commercial cleaning equipment Google Ads case study, Google Ads dashboard figures for each period.

Spend went up and cost per conversion still fell, because the extra budget went to searches that could buy. These are leads counted by Google Ads, not signed contracts, and your account will not behave the same way. What carries over is the order: tracking, structure, search terms, negatives, bidding, then pages.

How do you turn audit findings into an action plan?

Rank findings by what they put at risk, and fix them in dependency order. A long list of equal-weight issues helps nobody. The owner needs to know what to fix this week and what can wait.

This is how I rank what I find:

1. Anything that corrupts data, such as duplicate or wrong primary conversions.
2. Anything that spends money on searches that cannot buy, such as missing negatives or loose match types.
3. Structure and settings that mix intents or send traffic to the wrong place.
4. Bid strategy changes, once the data underneath is clean.
5. Ad copy, assets and landing page tests.

I change one layer at a time where I can. If tracking and bidding change in the same week, nobody can tell which change moved the numbers. Changes go in as small, dated steps, so the change history tells the story later.

FAQs

How long does a Google Ads audit take?

It depends on the size of the account. A small lead generation account with a few Search campaigns takes far less time than a store running Shopping, Performance Max and Search together. Tracking checks usually take the longest, because I test conversions on the website rather than trusting the columns inside Google Ads.

Can I audit my own Google Ads account?

Yes, and the order in this guide works for a self-audit. Start with conversion tracking and check which actions are primary. Then read the search terms report sorted by spend, add negatives in phrase and exact match, and confirm your bid strategy fits your data. An outside audit mostly adds a second pair of eyes.

What access does a Google Ads audit need?

Read-only access is enough. Google’s read-only access level lets a user view campaigns, planning tools and billing information, and run campaign performance reports, without editing anything. For the tracking checks, I also ask for view access to GA4 and Google Tag Manager, so I can see how conversions are recorded.

Is optimization score a good way to audit an account?

No, not on its own. Google describes optimization score as an estimate of how well an account is set up to perform, scored from 0 to 100%, and says it is not used by Quality Score. A high score can sit alongside broken tracking or wasted spend, so treat it as a list of ideas.

How often should you audit Google Ads?

Run a full audit when you take over an account, before you raise the budget, and after any big change to the website or tracking. Between audits, review the search terms report and conversion tracking regularly, because waste and broken tags build up quietly. A full audit then becomes a check rather than a rescue.

Will an audit change anything in my account?

No. An audit is a review, and with read-only access nothing can be edited. I write down what I find and rank it. Changes only happen after you agree to them, and then I make them in small, dated steps, so the change history shows exactly what moved and when.

What does my free Google Ads audit cover, and what should you do next?

My free audit follows the same order as this guide. I check conversion tracking, account structure and settings, search terms, negative keywords, bid strategy fit, and the landing pages behind your main campaigns. For online stores, I add the Merchant Center feed and the Performance Max setup.

I only need read-only access, so nothing in your account can change while I look. You get the findings ranked by impact, with what I would fix first and why. I will not promise you a ROAS figure, because nobody can before the tracking and the data have been checked.

Business owners describe working with me better than I can. Here is how one ecommerce brand owner puts it.

My public record is easy to check. I am Top Rated Plus on Upwork, with $400K+ earned and 144 client reviews as of November 2026 (my Upwork profile). If you would rather hand over the whole account, you can hire a freelance Google Ads specialist directly through this site.

If your account has never had its tracking checked, start there, before you read another report. If you want a second pair of eyes, send the account over and I will run the checks in this guide and rank what I find. You can also browse more guides on my Google Ads blog.

How this post was written: drawn from Ishant Sharma’s own account work and the published case studies linked above, researched against current Google Ads documentation, drafted with AI assistance and checked against the sources before publishing.