How to Audit a Google Ads Account — Step by Step (The Framework That Exposes What’s Actually Wrong)

The average Google Ads account we audit for the first time has at least three significant structural problems — and in most cases, at least one of them has been silently draining budget for over a year.

It’s rarely incompetence. It’s usually an account that was set up reasonably well at launch, then managed on autopilot while the business changed around it. Match types drifted. Negative keyword lists stopped getting updated. Conversion tracking was “good enough” until it wasn’t. And at some point the numbers stopped making sense and nobody could explain why.

This guide is the exact Google Ads audit framework we use when we take on a new account — every section, every check, every benchmark. If you’re evaluating your own account before calling an agency, use this to know exactly what you’re walking into. If you’re an in-house marketer defending your spend to leadership, this gives you a defensible diagnosis you can actually act on.

Step 1: Start With Conversion Tracking — Because Everything Else Is Built on This

Before you look at a single keyword or bid strategy, you need to know whether your conversion data is trustworthy. If it isn’t, every other insight in your PPC audit is built on sand.

Go to Tools → Conversions and audit every single conversion action in the account. Look for four things.

First, are you tracking the right events? We regularly see accounts where the primary conversion action is “Page View” or “Session Start” — which means Smart Bidding is optimizing to get people to land on a page, not to take any meaningful action. That’s a budget incinerator dressed up as a strategy.

Second, are you double-counting? A single form submission that fires both a Google Ads conversion tag and a Google Analytics goal imported into Ads counts twice. For a lead gen account spending $20,000/month, double-counting can make your CPL look 40% better than it actually is. Your boss loves those numbers. They’re lying to you.

Third, are you using data-driven attribution? Last-click is still the default for many accounts and it systematically undervalues upper-funnel keywords that assist conversions. If your account has more than 300 conversions per month, switch to data-driven. Below that threshold, use linear. Don’t use last-click in 2025 unless you have a very specific reason.

Fourth — and this one stings — check whether Enhanced Conversions is set up. If it isn’t, your conversion measurement is already degraded due to iOS privacy changes, cookie restrictions, and browser-level tracking prevention. We wrote a full breakdown of what Enhanced Conversions actually does and why your measurement is broken without it — it’s worth reading before you touch your bid strategies.

Step 2: Account Structure — Where Chaos Goes to Hide

A messy account structure doesn’t just look bad in screenshots. It directly causes performance problems — misaligned Quality Scores, budget cannibalization between campaigns, and Smart Bidding algorithms that are starved of signal because conversions are spread too thin across too many campaigns.

In this step of your Google Ads account review, you’re looking for four structural red flags.

Too many campaigns, too little data per campaign. If you have 15 campaigns and a $5,000/month budget, most of them are getting fewer than 30 conversions per month — which is the minimum Smart Bidding needs to function. You’re not getting strategic segmentation; you’re getting data fragmentation. Consolidate aggressively.

Ad groups with 20+ keywords. This is a relic of the Quality Score obsession from 2015. Single keyword ad groups (SKAGs) are largely obsolete in the era of RSAs and broad match, but ad groups that try to cover wildly different intents under one roof are still a problem. Theme tightly.

Branded and non-branded keywords in the same campaign. Your branded terms will always have astronomically higher CTRs and conversion rates, which distorts the campaign-level data your bidding algorithm is learning from. Branded campaigns should always be isolated. Always.

Missing negative keyword lists at the account level. If you don’t have a shared negative keyword list applied to all campaigns that blocks obvious irrelevant terms (competitors’ branded terms you don’t want to bid on, industry-adjacent but irrelevant searches, job-seeker terms), you’re paying for those clicks everywhere, every day. Our full guide to Google Ads account structure best practices covers the exact framework we use to keep campaigns profitable at any budget level.

Step 3: The Search Terms Report — The Most Honest Document in Your Account

The Search Terms Report is where accounts either prove they’re being managed or prove they’re being neglected. Pull it for the last 90 days and filter for cost. What you find will tell you more about account health than any metric in the dashboard.

You’re looking for three things specifically.

Irrelevant queries eating meaningful budget. Sort by cost, descending. Go through the top 50 search terms and ask: would I pay for this click? If the answer is no for more than 20% of your top-spend queries, you have a negative keyword problem, a match type problem, or both.

High-spend terms that aren’t keywords. If a search term has generated $500+ in spend and strong conversion data, why isn’t it an exact match keyword? Too many accounts let broad or phrase match silently run on valuable queries without ever promoting those queries to their own keywords with dedicated bids and ad copy.

The ratio of “unknown” terms. Google increasingly hides search term data, labeling it as “Other search terms” due to privacy thresholds. In heavily broad-match accounts, we’ve seen 60–70% of spend hidden under this label. That’s not a reporting nuisance — that’s the majority of your budget spent on queries you cannot review, cannot add as negatives, and cannot optimize against. If you’re running broad match without the control structure to support it, a rigorous negative keyword strategy isn’t optional — it’s the only thing keeping broad match from becoming a budget fire.

Speaking of which: your choice of match type has massive downstream effects on everything this audit is surfacing. If you haven’t worked through the broad match vs. exact match decision for 2026 with fresh eyes, do that before you restructure your keyword list.

Step 4: Bidding Strategy Audit — Is the Algorithm Working For You or Against You?

Smart Bidding is not a “set it and forget it” system. It’s a machine that performs exactly as well as the data you feed it — and in most accounts we audit, the data quality is the problem, not the algorithm.

Check the following in your Google Ads audit checklist for bidding.

Are you using tCPA or tROAS with targets that match reality? A tCPA target set at $40 when your actual historical CPL is $120 isn’t aggressive — it’s broken. The algorithm will either refuse to spend (because it can’t find conversions at that cost) or it’ll start finding “cheaper” conversions that are actually lower quality. Your targets should be informed by your 90-day historical data, then tightened gradually — no more than 10–15% at a time.

Are campaigns in a learning period right now? Smart Bidding re-enters learning whenever you change a target, budget, or conversion action. An account that’s constantly being tweaked is an account that never exits learning — and never performs. Check the “Status” column on your campaigns. If you see “Learning” on more than one or two campaigns, you’ve been over-managing the algorithm instead of letting it work.

Is your bid strategy aligned with your actual business goal? A lead gen account using Maximize Clicks is optimizing for traffic volume, not conversions. An ecommerce account using Manual CPC at scale is leaving conversion value optimization on the table. These misalignments are incredibly common and genuinely expensive. For a deeper look at how the algorithm makes decisions — and when to override it — our breakdown of how Google Ads Smart Bidding actually works is the best place to start.

Step 5: Ad Copy and Asset Audit — Where Good Accounts Quietly Get Mediocre

Responsive Search Ads give Google enormous latitude to mix and match your headlines and descriptions. That’s powerful when your copy is strong. It’s a disaster when your copy is weak, repetitive, or stuffed with generic filler.

For each RSA in your audit, check three things.

Ad Strength rating. “Poor” or “Average” rated ads are a problem — not because Ad Strength directly determines performance, but because ads that get low ratings are usually ads with repetitive or non-differentiated assets. Google will deprioritize them in the auction. Aim for “Good” or “Excellent” on your top-spend ad groups.

Are headlines doing different jobs? A strong RSA has headlines that cover at least four distinct categories: keyword inclusion, value proposition, differentiator, and call to action. If six of your fifteen headlines are variations of “Get a Free Quote Today,” Google can’t learn which message resonates — it’s just serving random iterations of the same phrase.

Ad extensions (now “assets”) — are they actually being used? Check the Combinations report to see which asset combinations are actually serving. Then check your Assets tab to confirm you have sitelinks, callouts, structured snippets, and at minimum one image asset per campaign. Accounts that skip assets are paying the same CPCs for less real estate. That’s a self-imposed penalty.

Do your ads match your landing pages? This is the most commonly ignored audit item and the one that costs the most. If your headline says “Custom Steel Fabrication — Ships in 48 Hours” and your landing page is a generic home page with no mention of lead time, your Quality Score suffers, your CPCs go up, and your conversion rate tanks — all at once. One message, start to finish.

Step 6: Audience, Geographic, and Device Targeting — The Bid Modifier Graveyard

Most accounts have never revisited their bid modifiers after initial setup. This step of the PPC audit almost always surfaces quick wins.

Geographic performance. Go to Locations → Geographic Report and sort by cost. Look at CPL or ROAS by state or city, not just nationally. We’ve seen B2B accounts where two states drove 80% of closed revenue but were receiving the same budget allocation as 48 others. Shift budget toward your highest-converting geographies, add negative bid adjustments (or full exclusions) for your worst performers.

Device performance. Pull performance by device and look at conversion rate, not just click volume. In B2B lead gen accounts, mobile typically converts at 30–50% the rate of desktop — but without a device bid adjustment, you’re paying the same for that traffic. Unless your landing page is genuinely mobile-optimized and your audience researches on phones, you should be bidding down on mobile by at least 20–30%.

Audience observation layers. Are you using audiences in observation mode to gather data on how different segments perform? If not, you’re flying blind. Add your customer match lists, remarketing audiences, and in-market segments in observation mode to every campaign. After 30 days you’ll have the data to make informed bid adjustments. This isn’t optional — it’s table stakes for a well-managed account.

Are you excluding audiences that will never convert? Past converters who you don’t want to retarget, competitors’ employees if you’re running LinkedIn-matched lists, or audiences that have historically shown zero conversion rate — if they’re not excluded, you’re paying to reach them. The distinction between remarketing and retargeting matters here too, especially when you’re deciding which audiences belong in separate campaigns versus bid-adjusted layers.

Your Google Ads Audit Checklist — The Quick Reference

Use this as your running checklist when conducting a full Google Ads account review. Each item should get a pass/fail and a priority rating (P1 = fix immediately, P2 = fix this month, P3 = monitor).

Conversion Tracking

• All conversion actions reviewed — only meaningful actions set as “Primary”
• No duplicate conversion counting confirmed
• Attribution model set to data-driven (or linear if volume is low)
• Enhanced Conversions configured
• Offline conversion import in place (if applicable)

Account Structure

• Branded campaigns isolated from non-branded
• Campaign-level budget sufficient to generate ≥30 conversions/month each
• Shared negative keyword list applied to all campaigns
• Ad groups themed tightly around intent, not just keyword cluster

Search Terms & Keywords

• Last 90 days of search term data reviewed; irrelevant terms added as negatives
• Top-spend queries promoted to exact match keywords where appropriate
• Match type strategy deliberate and documented
• “Other search terms” percentage below 40% of spend

Bidding

• Bid strategy aligned with primary business objective
• tCPA/tROAS targets within 20% of 90-day actual
• No campaigns stuck in learning period longer than 2 weeks

Ads & Assets

• All RSAs rated “Good” or “Excellent”
• Sitelinks, callouts, structured snippets active on all campaigns
• Ad-to-landing page message consistency confirmed

Targeting & Modifiers

• Geographic bid adjustments informed by actual conversion data
• Device bid adjustments applied based on CVR differential
• Observation audiences active on all campaigns
• Exclusion lists applied where relevant

Frequently Asked Questions

How long does a proper Google Ads audit take?

A thorough audit of a single account — covering tracking, structure, keywords, bidding, ads, and targeting — takes 3 to 5 hours for an experienced PPC manager. Anything faster is a surface-level review, not a real audit. If an agency offers you a “free audit” that comes back in 20 minutes as a PDF with bar charts, they ran a tool report, not an audit.

How often should I audit my Google Ads account?

Full structural audits should happen quarterly. Mini-audits covering search terms, negative keywords, and conversion tracking integrity should happen monthly. Accounts that only get reviewed when something breaks will always be two steps behind where they could be.

What’s the most common issue found in a Google Ads account review?

Broken or misconfigured conversion tracking, by a significant margin. It’s not glamorous, but it’s the root cause of more poor decisions — wrong bid strategy targets, misallocated budgets, campaigns that look great but aren’t — than any other single issue.

Can I audit my own account, or do I need an agency?

You can absolutely audit your own account using this framework. The challenge is that you may not know what “good” looks like for your specific industry, spend level, and campaign type. Benchmarks matter — a 5% conversion rate might be excellent in one vertical and mediocre in another. That context is where an experienced agency adds real value. If you’re unsure whether your numbers are good, our ROAS benchmarks by industry give you a starting point for calibration.

What should I do after I complete the audit?

Prioritize your findings into three buckets: things that are actively costing you money now (fix this week), structural problems that are capping performance (fix this month), and optimizations that would improve an already-functional account (ongoing). Don’t try to fix everything at once — especially anything that will trigger a Smart Bidding learning period, since stacking multiple changes resets the learning window each time.

My account looks fine on the surface but leads are low quality. What am I missing?

This is a conversion tracking problem disguised as a targeting problem. If you’re optimizing for lead form submissions but those submissions aren’t qualified buyers, you’ve taught the algorithm to find more form submitters — not more customers. The fix is offline conversion tracking, which feeds actual sales data back into Google Ads so the algorithm optimizes for people who buy, not just people who click a button.

Should I be worried about Performance Max cannibalization during an audit?

Yes — and this is one of the most underdiagnosed issues we find. If you’re running PMax alongside Search campaigns, PMax will often claim credit for conversions that Search campaigns drove, particularly on branded queries. During your audit, check whether you have brand exclusions applied to your PMax campaigns. If you don’t, your Search campaign data is being artificially suppressed. This gets complicated fast — our complete Performance Max guide for 2026 covers exactly how to structure the two campaign types so they don’t fight each other.