Google Ads Audit Checklist: 15 Things Senior Account Managers Check First
A Google Ads audit checklist built from 300 plus account audits. Each check includes the threshold for what bad looks like.
A DTC brand spending $22,000 per month on Google Ads. Overall CPA of $58 against a $65 target. ROAS of 320%. The marketing team reported to leadership that the account was performing well.
A 15 check audit found $9,200 per month in waste. The primary conversion action was counting Add to Cart instead of Purchase, inflating conversion numbers by 4x. Display Network was enabled on two Search campaigns, leaking $1,800 per month to placements with zero purchases. Location targeting was showing ads internationally on a US only brand, sending $2,640 per month to clicks that could never convert. Twelve keywords with quality scores of 1 to 3 were paying a 300% click premium with a combined one purchase per month.
This is the checklist I have refined across 300 plus account audits. It is not a list of best practices. It is a diagnostic sequence where each check builds on the last, and each one includes the threshold that tells you when something is wrong.
The foundation: conversion tracking
Everything downstream depends on whether the data is real.
Check 1: Primary conversion action. Open conversion settings and verify what counts as a conversion. The threshold is simple: if the primary action is anything other than the actual business outcome (purchase, qualified lead, booked demo), the account is optimizing toward fiction. In the DTC account, the primary action was Add to Cart. Google reported 340 “conversions” per month. Actual purchases: 85. Every CPA number in the account was 4x better than reality. Decisions based on those numbers had been compounding the problem for months.
Check 2: Google Ads vs CRM comparison. Pull the conversion count from Google Ads and compare it against the source of truth: the CRM, Shopify, or the back office system. If the gap exceeds 20%, data quality is unreliable and every optimization built on that data is suspect. The DTC account showed 85 purchases in Google Ads, 72 in Shopify. A 15% gap from cross device attribution differences. Acceptable, but worth noting when setting CPA targets.
Check 3: Conversion value assignment. If every conversion carries the same value (or no value at all), the algorithm treats a $30 impulse purchase identically to a $300 high margin order. It optimizes for volume, not revenue. The DTC account had no revenue values assigned. After implementing actual order values, ROAS improved from 280% to 410% within 60 days because the algorithm stopped chasing low AOV purchases that barely covered shipping.
Campaign settings: the silent budget leaks
These are the settings that leak money from the day a campaign is created and rarely get reviewed again. For the full dollar impact analysis of each setting, see the wasted spend diagnostic framework.
Check 4: Location targeting type. The default “Presence or Interest” shows ads to people merely researching a location. Threshold: pull the location report and check if more than 10% of spend comes from regions you do not serve. The DTC brand was US only. 12% of spend ($2,640 per month) came from international clicks. Not a single one converted.
Check 5: Search Network partners. Enabled by default. Threshold: if partner CPA is more than 2x Google Search CPA, disable. The DTC account showed Google Search CPA of $42 and partner CPA of $110. Partners were consuming 9% of budget at nearly 3x the cost per acquisition.
Check 6: Display Network on Search campaigns. Also enabled by default on new Search campaigns. Threshold: any Display spend on a Search campaign is waste unless deliberately intended. The DTC account had two Search campaigns with Display enabled. Combined Display spend: $1,800 per month. Purchases from Display placements: zero. The clicks came from app inventory and content sites where nobody was shopping.
Check 7: Auto applied recommendations. Google automatically implements changes unless you opt out. Threshold: check the change history for any recommendation applied without approval. The DTC account had 14 broad match keywords auto added by Google that duplicated existing exact match terms. The overlap caused the same search queries to match through both, inflating CPCs by 18% on those terms because the account was bidding against itself.
Structure and bidding
Check 8: Match type separation. If broad match and exact match keywords share the same campaign and budget, the algorithm favors broad match because impressions are easier to find. Threshold: if broad match keywords receive more than 60% of total impressions in a shared campaign, exact match is being starved. The DTC account had exact match keywords for “organic cotton t shirt” receiving 12% of impression share while broad match in the same campaign consumed 78% of budget on loosely related terms like “cotton clothing” and “organic fabric.”
Check 9: Smart bidding vs conversion volume. Smart bidding needs data to learn. Threshold: fewer than 15 conversions in the past 30 days means insufficient signal for Target CPA or Target ROAS. The DTC account had a niche accessories campaign with 6 conversions per month running on Target ROAS. The algorithm was erratic, swinging between overbidding and underbidding daily. Switching to Manual CPC stabilized CPA at $38 compared to the $67 average under smart bidding.
Check 10: Target CPA or ROAS vs actuals. Threshold: if the target is more than 2x away from actual performance in either direction, the strategy is fighting reality. The DTC account had one campaign with Target ROAS set at 500% while actual ROAS was 220%. The algorithm became extremely conservative, dropping impression share to 15%. The campaign was barely spending because the target was unachievable, leaving profitable search volume on the table.
Check 11: Campaign consolidation. Each campaign needs enough data to learn. Threshold: any campaign generating fewer than 10 conversions per month is a candidate for consolidation. The DTC account ran 8 product category campaigns. Five of them produced fewer than 10 conversions each. Consolidating into 3 campaigns by grouping related categories improved algorithm learning and reduced CPA by 22% within the first 30 days.
Keywords and search terms
Check 12: Search terms by cost descending. Every search term with spend and zero conversions is potential waste. Threshold: any individual non converting term above 1x target CPA warrants immediate review. For a deeper treatment of how to categorize and act on search term findings, see how to read a search terms report like a senior account manager.
Check 13: Non converting keywords above the spend threshold. Threshold: any keyword that has consumed more than 3x your target CPA without a single conversion has had enough data. In the DTC account with a $65 target CPA, that meant any keyword above $195. “Sustainable clothing” had spent $195 with zero conversions. Its search terms were “sustainable clothing brands list,” “sustainable clothing cheap,” and “sustainable clothing for kids.” None matched the product. Paused.
Check 14: Quality scores for active keywords. Keywords with quality scores of 1 to 3 pay a significant premium per click. A quality score of 1 pays roughly 400% more per click than a quality score of 7 for the same ad position. Threshold: any keyword with a quality score of 1 to 3 that is actively spending deserves scrutiny. The DTC account had 5 keywords at quality scores of 2 to 3 with combined spend of $1,900 per month and 1 purchase. Effective CPA: $1,900.
Check 15: Converting search terms not added as keywords. Threshold: any search term with 2 or more conversions that shows a status of “none” (meaning it matched through a broader keyword rather than being targeted directly). The DTC account had “organic cotton crew neck” converting at 4.2% CTR and $28 CPA, but it was not targeted as a keyword. Adding it as exact match with a dedicated landing page gives you direct control over bids, ad copy, and the user experience.
Prioritizing findings by dollar impact
After running all 15 checks, the question is what to fix first.
Priority one: conversion tracking (checks 1 through 3). If the data is wrong, every other finding is built on a false premise. Fix tracking before touching anything else.
Priority two: active budget leaks (checks 4 through 7). These are dollars leaving the account right now for zero return. Every day they persist, waste compounds.
Priority three: structural inefficiencies (checks 8 through 11). These degrade performance over time but are not immediately burning budget the way a Display Network leak does.
Priority four: keyword and search term optimization (checks 12 through 15). This is where ongoing maintenance lives. These findings represent the recurring work that keeps an account healthy month over month.
For teams that want this prioritization produced automatically on every account, I walk through what a complete audit workflow outputs in a separate article.
How to use this Google Ads audit checklist
Run it in order. The conversion tracking checks come first because every subsequent finding depends on whether the data is trustworthy. If check 1 reveals that conversions are miscounted, the CPA calculations in checks 10, 12, and 13 are meaningless until tracking is fixed.
The first pass takes 60 to 90 minutes. Once you know an account’s patterns, subsequent monthly reviews take 45 to 60 minutes because you know where to focus.
The goal is not to find something wrong with every item. The goal is to confirm each one is right, or to catch the ones that are not before they compound further. The accounts that perform well over time are the ones where someone runs this sequence consistently, not once.
This is the sequence I follow on every account I audit. I run a free audit that applies all 15 checks with prioritized findings. For teams that want this depth applied consistently across every account, AI agents can encode the same diagnostic process.