A Google Ads audit isn't a quarterly ritual — it's a diagnostic tool. Run one when performance drops, before taking over a new account, or when spend is climbing but results aren't moving. The goal is straightforward: find where money is being wasted, identify what's actually working, and build a clear picture of what needs to change.
This guide walks you through the exact process senior paid media managers use to audit an account systematically. Not just skimming surface metrics. You'll cover conversion tracking, campaign settings, keyword strategy, ad quality, audience targeting, and performance analysis. Each step is designed to surface a specific category of problems.
Work through them in order. By the end, you'll have a prioritized list of fixes, not a vague sense that something's off.
One note before you start: document everything as you go. Screenshots, notes, a simple spreadsheet — whatever works. An audit without a record is just browsing.
Every other metric in the account is unreliable until you know conversions are tracking correctly. This is the step most people skip or rush through. Don't.
Go to Tools > Conversions in the Google Ads interface. Every active conversion action should show a "Recording conversions" status. If you see "No recent conversions" or "Inactive," that's a red flag — either the tag isn't firing or the action hasn't received traffic recently. Dig into which scenario applies before moving on.
Next, check for duplicate conversion tracking. This is one of the most common issues in accounts that have been managed by multiple people or agencies. It typically happens when both a Google Ads native tag and an imported Google Analytics goal are counting the same event — a form submission, a purchase, a phone call. The result is inflated conversion numbers and Smart Bidding algorithms optimizing toward phantom data.
Confirm that conversion values are set correctly. For eCommerce accounts, dynamic revenue values should be passing through. For lead-gen accounts, assigned values need to reflect actual business logic — not just placeholder numbers someone entered years ago. Missing or incorrect values make ROAS bidding meaningless.
Look at the "All conversions" versus "Conversions" column discrepancy. Google includes cross-device conversions and store visit conversions in "All conversions" that may not appear in the "Conversions" column depending on account settings. Understand which column your automated bidding strategies are optimizing toward — that's what actually matters.
Finally, confirm that the right conversion actions are included in the "Conversions" column. Soft actions like page views or video plays should never be in there influencing bid decisions.
Success indicator: Every active conversion action shows a recent ping, no duplicates exist, and only meaningful conversion actions are included in the column that drives bidding.
Campaign settings are where well-intentioned defaults quietly drain budgets. Most of these issues were set at launch and never revisited.
Start with network settings. Search campaigns in Google Ads default to include both Search Partners and the Display Network unless you manually uncheck them. These should almost always be separated. Display traffic mixed into a Search campaign makes performance data harder to read and often pulls budget toward lower-intent placements. Check every Search campaign and note which ones are running on mixed networks.
Location targeting is another common problem. The default setting in some campaign types is "Presence or interest," which serves ads to users who have shown interest in your target geography — not just users physically located there. For most advertisers, you want "Presence" only. If you're running a local service business and your ads are showing in other states, this is likely why.
Review bidding strategies against campaign goals and data volume. Google recommends roughly 30 to 50 conversions per month for Target CPA to function reliably, per their own documentation. An account running Target CPA on a campaign generating five conversions a month is essentially running on guesswork. Manual CPC or Maximize Clicks may be more appropriate until data accumulates.
Check ad scheduling. Many accounts have schedules set from an early assumption that was never validated. Cross-reference the schedule against actual conversion data by hour and day. You may find you're limiting reach during hours that consistently convert.
Look at device bid adjustments. Are mobile bids set intentionally based on performance data, or were they left at default? Pull device performance segmented by conversions and CPA. The numbers usually tell a clear story about where adjustments are warranted.
Flag any campaigns showing low impression share due to budget. These campaigns need a budget conversation, not optimization work.
Success indicator: Every setting has a documented reason behind it, not just a default that was never changed.
Pull the Search Terms report for the last 90 days. Go to Keywords > Search Terms. This is where wasted spend hides. Filter immediately for terms with significant spend and zero conversions. In most accounts, this list is longer than it should be.
Evaluate match type distribution. Accounts running exclusively broad match without strong negative keyword lists tend to bleed budget on irrelevant queries. Broad match has its place, but it requires active management. If you see broad match keywords pulling in search terms that have nothing to do with the product or service, the negative list hasn't kept pace.
Review negative keywords at both the campaign and account level. Many accounts have thin or outdated negatives that haven't been updated in months. Look for obvious gaps: competitor terms you don't want to show for, informational queries that signal research rather than purchase intent, and geographic terms that conflict with your targeting.
Check for keyword cannibalization. Multiple ad groups or campaigns bidding on overlapping terms splits Quality Score data and inflates CPCs. If the same keyword appears in three different ad groups, Google is essentially competing against itself on your behalf.
Look at Quality Scores at the keyword level. Scores of 3 or below indicate a mismatch between the keyword, the ad copy, and the landing page. These keywords cost more per click and drag down ad rank across the ad group. Either fix the alignment or pause them.
One structural note: single-keyword ad groups (SKAGs) were once a widely recommended practice. They create significant management overhead without proportional benefit in most accounts today, particularly given how RSAs and Smart Bidding have changed the optimization dynamic. If you see an account built entirely on SKAGs, flag it for consolidation.
Success indicator: Search terms are relevant to the actual offer, negatives are current and substantive, and no meaningful budget is going to queries outside the target intent.
Google phased out Expanded Text Ads — they stopped serving on June 30, 2022. If you still see ETAs in an account, they're not delivering. Responsive Search Ads are the standard format now, and every active ad group needs at least one.
Pull up RSA asset strength ratings. Google rates them as Poor, Average, Good, or Excellent. Poor or Average ratings typically mean the headlines and descriptions aren't diverse enough, aren't covering different themes, or are too similar to each other. RSAs work by testing combinations — if all your headlines say essentially the same thing, there's nothing to test.
Check for over-pinned assets. Pinning is useful when specific messaging must appear in a specific position — a legal disclaimer, a brand name, a key differentiator. But pinning every headline defeats the purpose of RSAs entirely and limits Google's ability to find winning combinations. If you see an RSA where five or six assets are pinned, it's functioning more like a static ad.
Look at whether multiple RSAs are running per ad group. Testing ad variations requires having more than one. If only one RSA exists per ad group, you're leaving performance data on the table.
Review ad assets (formerly called extensions): sitelinks, callouts, structured snippets, and call assets. These affect ad rank per Google's documented formula and improve click-through rate. Missing assets are one of the quickest wins in an audit — they're often overlooked and take minimal time to add.
Check that ad copy reflects the actual landing page messaging. Misalignment between what the ad promises and what the landing page delivers hurts Quality Score and conversion rate simultaneously.
Success indicator: Each ad group has active RSAs with strong asset ratings, relevant assets are in place, and ad copy matches landing page messaging.
Start with Search campaigns. Check which audiences are applied and whether they're set to observation or targeting mode. Observation mode collects performance data without restricting reach — that's appropriate for most Search campaigns. Targeting mode limits your ads to only those audiences. Make sure the mode is intentional, not accidental.
Pull audience performance data. If in-market audiences or remarketing lists show dramatically different conversion rates compared to the baseline, bid adjustments should reflect that. An audience converting at twice the average rate with no bid adjustment is money left on the table.
For Display or Performance Max campaigns, look at audience signals. Signals based on actual customer data — Customer Match lists, website visitors — typically outperform generic interest categories. If the account has first-party data available but isn't using Customer Match, that's worth flagging.
Check demographic performance: age, gender, and household income segments often convert at very different rates but receive identical bids by default. Pull the data. The patterns are usually clear, and adjustments are straightforward once you see them.
Verify that Customer Match lists are uploaded and active. For accounts with existing customer data, this is frequently an underused advantage — both for exclusions (suppressing current customers from acquisition campaigns) and for prioritizing high-value prospects.
Success indicator: Bid adjustments are grounded in actual performance data, not guesswork, and high-value audiences are actively being prioritized.
Compare current period performance against the prior period and year-over-year. Look specifically at CPC trends, conversion rate shifts, and impression share changes. A rising CPC with flat conversion rates points to an auction problem. A declining conversion rate with stable traffic points to a landing page or offer problem. The direction of the metrics tells you where to look.
Use the Recommendations tab with skepticism. Google's automated suggestions are real, but they consistently favor spend increases and broader targeting. Evaluate each recommendation against actual account goals, not Google's optimization score. A high optimization score doesn't mean the account is performing well.
Segment performance by device, time of day, and geography. Campaign-level data masks pockets of strong and weak performance. A campaign averaging a $50 CPA might be running at $25 on desktop and $90 on mobile — that's a meaningful finding that disappears in aggregate numbers.
If the account includes Performance Max campaigns, pull asset group performance and search category insights. PMax offers less transparency than standard Search campaigns — that's a documented limitation. Work with what's available: asset group performance data and the search categories report give you enough to identify what's driving results and what isn't.
Build your prioritized fix list. Sort issues into three buckets: tracking and structural problems that need to be fixed immediately, optimization opportunities to address within two weeks, and longer-term tests to schedule and monitor. This keeps the work manageable and ensures the highest-impact issues get addressed first.
Document the baseline before making any changes: current CPA, ROAS, CTR, and Quality Score averages. You need these numbers to measure whether your fixes actually worked.
Success indicator: You have a written list of issues ranked by estimated impact, with clear owners and timelines for each fix.
Work through these six steps and you'll have a clear, prioritized picture of what's wrong and what's worth keeping. Most accounts have three to five high-impact issues, not twenty. Conversion tracking, campaign settings, and keyword waste tend to be where the biggest dollars are recovered. Ad copy and audience targeting usually come next.
The audit itself isn't the end goal. It's the starting point for a better-performing account.
If you've run through this process and the issues are bigger than your team can address, or you inherited an account that needs a full rebuild, that's a signal worth acting on. Triad Media Lab offers senior-level Google Ads management with no account handoffs and no black-box reporting — experienced practitioners who can take what your audit uncovered and turn it into a working plan. Learn more about our services to see how we approach paid search management.
