Quick Answer
A Google Ads audit is a structured check of whether your account records real sales and spends money where buyers are. A real audit starts with conversion tracking, then follows the money into your search terms and product spend. It should show you the exact wasted search terms and products, plus the report where each one was found.
Most Google Ads accounts I open look healthy on the surface. The campaigns hit their targets and the dashboard shows a strong return, so nobody digs deeper. But Google Ads can only report what its tracking tells it, and in my experience about one in three accounts has conversion tracking that is broken or misleading. On one roofing supplies account, Google Ads was claiming more revenue than the entire store took in.
That is why tracking comes first in every audit I run, and the checks below follow the money in order of how much is at stake, with steps to run the core ones yourself.
What a Google Ads audit is (and what it is not)
A Google Ads audit checks two things: whether your account is set up to measure real sales, and whether your budget is going where buyers are. My first pass has nothing to do with performance, and it happens before I look at a single number. I check how the account is set up, starting with which conversion actions are counted and what is excluded. Most of what is wrong with an account is visible at this stage.
An audit is also different from the optimization score Google shows in your account. Google describes that score as “an estimate of how well your Google Ads account is set to perform,” on a scale from 0 to 100% (Google Ads Help). It is calculated partly from Google’s own recommendations, and applying or dismissing them moves the score. It cannot tell you whether the conversions behind it are real, which is why I never open it first.
Why an audit matters: where the money goes missing
In the WordStream 2026 Google Ads benchmarks, the median cost per click across 13,474 US search campaigns was $5.42. The median cost per lead was $66.69, so a few hundred irrelevant clicks a month cost well over $1,000.
When I audit a new account, I typically find around 20 percent of the budget is wasted, and on some accounts more. On ecommerce accounts, a lot of that goes to display and other networks when it could be going to Shopping listings. Someone typing a product name into Google is much closer to buying than someone who was shown an image while reading something else, so the majority of an ecommerce budget should sit in Shopping.
Broken tracking is the bigger risk, and when I find it, the problem is rarely that the tags are missing altogether. The usual cause is double counting, or two or three conversion actions set as primary, so the bidding is fed the same sale more than once and spends against it.
Check 1: Is your conversion tracking telling the truth?
Conversion tracking is the first thing I open in every audit, because more money rides on it than on anything else. If the conversion data is wrong, every decision built on it is wrong too, including the ones Smart Bidding (Google’s automated bidding) makes for you. Google states plainly that Smart Bidding requires conversion tracking, so bad tracking means the automation is bidding toward the wrong target.
Where to look first
In Google Ads, go to Goals, then Conversions, then Summary (in the older layout, the same page sat under Tools and settings). It lists every conversion action in the account, each marked primary or secondary. According to Google, primary conversion actions are reported in the Conversions column and used for bidding, while secondary actions are for observation only.
On that page, I check three things before I move on to search terms:
- whether two or three primary actions are counting the same sale
- whether the account is properly linked to Google Merchant Center and Google Analytics 4 (GA4)
- whether the purchase action fires once, after payment, on the order confirmation
The purchase event is the most common break I find that the owner had no idea about. It fires in the wrong place or it fires twice, and nothing on the screen tells you. A plugin update or a new payment step is enough to make the account start counting people who reached the payment page instead of people who paid. It looks like it is working the whole time, and the bidding keeps buying more of whatever produced those numbers.
The self-test: put Google’s revenue next to your store’s
Take last month’s conversion value in Google Ads and put it next to the total revenue your store or sales system recorded for the same month, across every channel. If Google Ads is claiming more than your total, the tracking is overstating results. If it claims a share that looks too high next to what your other channels bring in, go back to the conversion actions and look for a sale being counted twice.
Click date versus conversion date
By default, Google Ads reports conversions against the date of the click, not the date of the sale. Google offers separate “by conv. time” columns, such as Conversions (by conv. time), which report each conversion on the day it happened. If your customers take a week to decide, the default view matches revenue against spend from a different week.
The gap between those two views is often wide enough to change the decision you were about to make. Add the “by conv. time” columns beside the default ones before you judge any recent change to bids or budgets.
A real example: when Google Ads claimed more than the whole store
One account I worked on was a fiberglass roofing supplies retailer, Shopping-led, spending roughly £6k to £19k a month on Google Ads. Google Ads was reporting a return of 9 to 10 times spend, and it looked like one of the best-performing accounts I had seen.
The problem was visible the moment we put it next to the store’s own numbers. In July 2025, Google Ads claimed £175,648 in revenue while the whole store, across every channel, took £130,116, and in August it claimed £169,718 against £123,155. Google Ads was reporting more revenue than the entire business was making, and nobody had noticed because nobody had put the two numbers side by side.
On January 13, 2026, we moved the account onto profit-based conversion tracking, so the revenue signal came from the store rather than from the ad platform’s own attribution. Then we rebuilt budget allocation from the corrected data, because for the first time we could see which campaigns were producing sales and which were only claiming credit for them.
In the second half of January, blended return across the business came in at 8.40, against 6.08 in the first half, on nearly three times the spend. February, the first full clean month, ran at 4.89 reported in Google Ads against 6.5 blended. The headline number in the platform got worse while the business got better, which is the whole point.
I want to be straight about two things, and the first is that early January is seasonally slow, so the before and after inside that month is not a like-for-like comparison. The second is that the gain came from reallocating budget once the data was trustworthy, which the tracking change made possible but did not produce on its own.
Check 2: Where the money leaks
The search terms report
The search terms report shows what people typed before your ad appeared. You will find it under Campaigns, then Insights and reports, then Search terms, according to Google’s search terms guide. This is usually where much of the wasted spend is sitting in plain sight.
Sort the report by cost and read down from the most expensive term. Any term with spend and no sales that has nothing to do with what you sell is a candidate for a negative keyword. Google leaves out low-volume terms for privacy reasons, so treat the report as a sample of what people searched.
Negative keywords and match types
Negative keywords tell Google which searches to keep your ads away from. On nearly every account I open, negatives and placement exclusions were never put in. Campaigns still show for search terms that should have been shut off long ago, because nobody has opened that report since the account was built. Google also lets you build an account-level negative keyword list that applies across eligible Search and Shopping inventory.
Match types decide how loosely Google matches your keywords to what people search. Google’s match type guide says broad match can show ads on searches “that don’t contain the direct meaning of your keywords,” and recommends pairing it with Smart Bidding. That pairing depends on Smart Bidding being fed real conversions, which is why I check match types after tracking.
Products eating spend inside a campaign that looks fine
This is the leak owners never catch, because at campaign level there is nothing to react to. The campaign hits its target, so nobody looks any deeper. On a roofing supplier we worked with, we went through the product data year by year and found:
- a casting resin aimed at the crafts and hobby market, taking between 800 and 1,700 clicks a year and returning about £2,300
- a large refurb kit spending steadily and returning less than the smaller sizes of the same kit
- a garden sofa set, in a roofing feed, pulled in by a bad feed mapping that nobody had noticed
None of that was a decision anyone made, so we paused the products with no case for staying. We held budget on the core products that were driving the account, and the money moved to products people were buying. To run this check yourself, look at cost and conversion value for each individual product over at least a full year, since a single month can hide slow, steady losers.
Where Performance Max spends your budget
Performance Max runs across several Google channels from one budget. Google’s channel performance report breaks its results out by channel, including Search, Display, and YouTube. You will find it under Campaigns, then Insights and reports, then Channel performance.
Sometimes a chunk of the spend is going to display or search while all the revenue is coming from the Shopping listings. When that happens, we move it to a feed-only setup or to a standard Shopping campaign so the budget goes back to where the buying intent is.
Check 3: Structure, bidding, and budget
Next I check whether the account is built to spend well, starting with search impression share: the impressions your ads received as a percentage of all the impressions they were eligible for. Two columns in the same report tell you why your ads missed the rest of those impressions:
- Search lost IS (budget): the share of time your ads did not show on the Search Network because the budget ran out
- Search lost IS (rank): the share of time your ads did not show because of poor Ad Rank in the auction
Add them through the columns menu under Competitive metrics, at campaign level (Google Ads Help).
Lost share to budget and lost share to rank are two completely different problems with two different fixes. If a profitable campaign is losing share to budget, you are leaving sales on the table and more budget may help. If it is losing share to rank, more budget will not fix it, and the work is in improving Ad Rank through your ads and bids.
Your bidding strategy also has to match how much conversion data you have. Google recommends measuring Smart Bidding over periods with at least 30 conversions, such as a month or longer, or 50 for Target ROAS (Google Ads Help). A campaign that records a handful of conversions a month gives the automation little to learn from, and double-counted conversions make that worse.
For structure, I ask whether each campaign has a clear job and whether you can tell where the money goes. Google sets out its own recommended approach in its guide to account structure.
Check 4: Ads, assets, and landing pages
Assets are the extra parts of an ad, such as sitelinks and callouts, which Google called extensions until 2022. I check whether they are actually in place, or whether nobody has touched them since the account was built. That matters, because Google reports that advertisers who improve Ad Strength for responsive search ads and sitelinks from “Poor” to “Excellent” see 15% more conversions on average.
Google calls Quality Score “a helpful diagnostic tool, not a key performance indicator.” According to the same help page, the auction uses real-time checks of expected clickthrough rate, ad relevance, and landing page experience, while the score itself is not used (Google Ads Help). Treat a low score as a pointer to the part that needs work, and do not chase the number.
Last, click through your own ads on a phone, the way a customer would. Check that the landing page matches what the ad promised and that the checkout or contact form works from start to finish.
Check 5: Settings that quietly drain budget
Some costly problems are single settings nobody revisited after launch, and I check these three on every account:
- Search partners. Google includes search partners by default when you create a Search campaign (Google Ads Help). Compare their results with Google Search and switch them off under Networks if they are not paying.
- Location options. “Presence or interest,” which Google recommends as a best practice on Search, can reach people who have shown interest in your area as well as people in it, while “Presence” sticks to people in your targeted locations (Google Ads Help). A business that serves only a local area should check which one is set.
- Account-level exclusions. I look at the negative keyword lists and the placement and brand exclusions set for the whole account, because a gap here affects every campaign.
The Google Ads audit checklist, ranked by money at stake
Work through this in order, since the early checks decide whether every later number can be trusted.
- Conversion actions (Goals, then Conversions, then Summary): one primary action per sale, and a purchase that fires once, after payment.
- Revenue self-test: Google Ads conversion value against total store revenue for the same month.
- Conversion timing: the “by conv. time” columns against the default click-date view.
- Search terms report: irrelevant terms with spend, sorted by cost.
- Negative keywords: gaps at campaign and account level.
- Match types: broad match spend, and whether it runs on clean conversion data.
- Product-level spend: cost and conversion value per product over a full year.
- Performance Max channels: where the spend goes versus where the revenue comes from.
- Impression share: Search lost IS (budget) and Search lost IS (rank) per campaign.
- Bidding versus volume: at least 30 conversions over the period you judge (a month or longer), or 50 for Target ROAS.
- Ads and assets: assets in place, and Ad Strength.
- Landing pages: your own ad clicked on a phone, through to purchase or contact.
- Settings: search partners, location options, and account-level exclusions.
Do it yourself or hire it out: a real audit versus a sales pitch
You can run the first few checks yourself, and the revenue self-test alone will tell you whether you have a tracking problem. Outside help earns its fee on the harder work behind the checks, such as rebuilding conversion tracking so the revenue signal comes from your store. It also pays off when someone has to go through years of product data to find the items quietly losing money.
If someone offers you a free audit, judge it by what it is built on. In my view, a free audit is worthless when it grades your account using conversion data nobody checked was real. Before you trust one, ask whoever is offering it to show you:
- the wasted spend shown as the actual search terms and products, rather than a percentage
- the name of the report where each problem was found, so you can open it and see the same thing yourself
- evidence that they have compared your conversion data with your own sales before judging anything
An audit that leads with the optimization score, or with a percentage of “wasted budget” and no list behind it, is a pitch. 94n offers a free account audit to select accounts, and whoever you ask, you should hold them to the standard above.
Frequently asked questions
What does a Google Ads audit include?
A real Google Ads audit starts with conversion tracking, checking which actions are primary and whether Google’s reported revenue matches your store’s. It then follows the money through the search terms report and spend by product, including Performance Max. Impression share and bidding come next, with ads and account settings such as search partners last.
Is a free Google Ads audit worth it?
A free audit is worth it only if it checks your conversion data against your real sales first. If it grades your account on conversion numbers nobody verified, the findings may be wrong. Ask to see the wasted spend listed as actual search terms and products, with the report each problem came from, so you can check it yourself.
Is the Google Ads optimization score an audit?
No, because Google describes optimization score as an estimate of how well your account is set to perform, on a scale from 0 to 100%, and applying or dismissing Google’s recommendations changes it. It does not check whether your conversion tracking is accurate, so an account can score well while bidding on double-counted sales.
Where to start
Start with the tracking, because every other number in the account depends on it. Run the revenue self-test and check your primary conversion actions this week, then move on to search terms and product-level spend.
If the self-test shows a gap you cannot explain, or you want a second pair of eyes on the product data, the Google Ads team at 94n is happy to look. We will show you what we find, and where we found it.