The 44-Point Google Ads Audit Checklist
The same checklist we run on every new Google Ads account before quoting work. Each item has a pass/fail criterion and a brief explanation of why it matters. Print this page or save it as PDF from your browser.
1Campaign Architecture
Campaigns are organized by goal (Search / Shopping / Display / PMax / Demand Gen), not mixed into one container. Why it matters: Mixed-network campaigns disable bid-strategy control and dilute optimization signals.
Each ad group contains ≤ 20 tightly themed keywords (single intent per ad group, ideally ≤ 10). Why it matters: Quality Score is calculated at the ad-group level. Loose themes drag CTR and CPC across all keywords in the group.
Match types are segmented (broad / phrase / exact in separate ad groups or campaigns). Why it matters: Without separation, you cannot adjust bids or pause underperforming match types independently.
Campaign budgets reflect strategic priority — not equal-spread or historical inertia. Why it matters: Equal budgets across campaigns of vastly different ROI guarantees you over-spend on weak campaigns and under-spend on strong ones.
Shared budgets are NOT used unless explicitly intentional (most accounts misuse them). Why it matters: Shared budgets prevent per-campaign optimization and obscure performance attribution.
Performance Max campaigns are split by margin tier / catalog segment, not one monolithic PMax for the whole catalog. Why it matters: A single PMax campaign averages performance across product tiers and optimizes for whatever sells easiest, regardless of profitability.
2Keyword & Search Term Hygiene
Search Terms report has been audited in the last 7 days. Why it matters: Smart Bidding will spend on weird long-tail queries if they happen to convert once. Weekly auditing catches drift.
Negative keyword lists are shared at the account level for universal exclusions (jobs, free, salary, "how to", competitor staff names). Why it matters: Per-campaign negative lists are duplicate-effort; shared account-level lists scale with new campaigns.
Brand terms are intentionally targeted with a dedicated branded campaign — not accidental cross-spend from generic campaigns. Why it matters: Brand traffic should have its own ROAS target (typically 10–20× higher) and protect against competitor poaching.
Competitor brand terms are either explicitly targeted (with dedicated copy and LP) or fully negative. Why it matters: Accidental competitor-brand traffic burns budget without intent. Intentional competitor-conquest is high-ROI inventory in B2B.
Broad match keywords have ≥ 30 conversions/month OR are paused. Why it matters: Broad match without enough conversion signal to feed Smart Bidding becomes random-traffic acquisition.
Underperforming keywords (high cost, zero conversions in last 60 days, ≥ 50 clicks) are paused. Why it matters: These are pure budget drag — Smart Bidding will keep spending on them while it "explores".
3Conversion Tracking
Google Ads conversion data matches GA4 conversion data within 5% variance. Why it matters: Larger gaps indicate broken cross-domain tracking, GTM misfires, or double-counting. Smart Bidding optimizes off whichever signal is wired.
For lead-gen accounts: offline conversions are imported from CRM (closed/qualified leads). Why it matters: Without CRM imports, Smart Bidding optimizes for form fills — including the 80% that never become customers.
For e-commerce accounts: enhanced conversions with hashed email/phone are enabled. Why it matters: Enhanced conversions recover 20–40% of attribution lost to ITP, ad blockers, and cookieless browsers.
Conversion actions are categorized correctly (primary for bidding, secondary for observation). Why it matters: Smart Bidding optimizes only toward primary conversions. Miscategorized actions either dilute the target or fail to count.
Conversion value reflects actual revenue/LTV, not flat dummy values (e.g., $1 per lead). Why it matters: Flat values make Target ROAS strategies meaningless. The algorithm cannot prioritize high-value conversions.
GA4 is set as the conversion source where appropriate (vs. Google Ads native pixel). Why it matters: GA4 attribution model + offline imports = better cross-channel signal than the native pixel alone.
4Smart Bidding & Budget
Each Smart Bidding campaign has accumulated ≥ 30 conversions in the last 30 days. Why it matters: Below 30 conversions/month, Smart Bidding cannot stabilize — performance becomes erratic.
Bidding strategy matches business goal: Target CPA for lead-gen, Target ROAS for e-commerce, Max Conversions only for launch/learning phase. Why it matters: Mismatched strategies optimize for the wrong outcome (e.g., Max Conversions on an e-commerce account ignores AOV variance).
Target CPA / Target ROAS values are realistic — within 20% of historical 90-day average. Why it matters: Targets set too aggressively choke spend. Targets set too loose waste budget on low-quality conversions.
Budget pacing alerts are configured (Google Ads → Tools → Notifications or Looker Studio). Why it matters: Budget overspend by 2× is common when daily bids increase and no one notices until end-of-month.
Day-parting and location-bid adjustments are applied where data justifies them (≥ 60 days of segmented data). Why it matters: Smart Bidding handles most adjustments, but certain patterns (off-hours / specific geos) still benefit from manual overrides.
5Ad Copy & Assets (RSAs)
Each Responsive Search Ad has 11+ headlines (Google's ideal threshold) and 4 descriptions. Why it matters: RSAs with full asset diversity produce more headline combinations and higher CTR. Sparse RSAs underperform consistently.
Ad Strength shows "Good" or "Excellent" on all live RSAs. Why it matters: "Average" or "Poor" Ad Strength capping impression share regardless of bid.
Sitelink, Callout, Structured Snippet, Call, and (if relevant) Location and Price assets are all active. Why it matters: Each enabled asset increases ad visual footprint and CTR by typically 5–15%.
Asset reporting is reviewed monthly — underperforming headlines/descriptions paused, new variants tested. Why it matters: Set-and-forget RSAs decay as user search behavior shifts. Quarterly refresh is the baseline; monthly is best practice.
No ad copy claims trigger Google policy violations (guarantees without disclaimers, undisclosed APR for finance, "cure" claims for health). Why it matters: Policy violations cause silent disapprovals that quietly drop your delivery without alerting you.
6Landing Page & Quality Score
Each ad group has a landing page that matches the keyword intent (no generic homepage targeting for specific keywords). Why it matters: Landing page relevance is a Quality Score component. Generic LP for specific keywords forces higher CPC.
Landing pages pass Core Web Vitals (LCP < 2.5s, INP < 200ms, CLS < 0.1). Why it matters: Page experience signal affects both Quality Score and direct conversion rate.
Mobile responsive layout — 60%+ of paid traffic is mobile. Why it matters: Mobile-unfriendly landing pages convert at 30–50% lower rate than desktop, but you pay the same CPC.
Single primary CTA above the fold on each LP — no decision paralysis. Why it matters: Multiple competing CTAs cut conversion rate by 15–30% per added CTA.
Form fields are minimal (3–5 max for lead-gen, optimized for mobile keyboard types). Why it matters: Each additional form field drops conversion ~7%. Mobile keyboard mismatch (text keyboard on phone field) adds another 5–10% drop.
Conversion tracking fires correctly on every LP (test with GA4 DebugView and Google Tag Assistant). Why it matters: Silent tracking failures on a single LP can hide 20%+ of your conversion data.
7Audiences & Remarketing
Customer Match lists from CRM are uploaded and refreshed at least monthly. Why it matters: Customer Match enables exclusion of existing customers from acquisition campaigns and inclusion in upsell/retention campaigns.
Remarketing audiences are layered onto Search/Shopping campaigns as "observation" (with bid adjustments) where appropriate. Why it matters: Returning visitors convert at 2–4× higher rate. Bid adjustments on these audiences improve CPA dramatically.
Excluded audiences are set on prospecting campaigns (e.g., existing customers excluded from new-customer acquisition). Why it matters: Without exclusions, you re-acquire your own customers at full CPC.
In-Market and Affinity audiences are layered for relevant categories. Why it matters: These Google-built audiences provide intent signal even when the user is not actively searching.
Lookalike / Similar Audiences from past customers feed PMax audience signals. Why it matters: PMax campaigns use audience signals as bidding hints. Lookalike audiences accelerate the learning phase.
8Reporting & Account Hygiene
A live dashboard (Looker Studio recommended) connects Google Ads → GA4 → CRM with primary KPIs. Why it matters: Monthly PDF reports lag reality. Live dashboards let you and stakeholders react within hours, not weeks.
Account access is properly scoped — admin access reserved for the agency lead, standard access for the team. Why it matters: Over-granted admin access creates audit trail gaps and security exposure.
Account history shows weekly optimization activity (changes log in Tools → Change History). Why it matters: Empty change history = abandoned account. Smart Bidding cannot replace ongoing hygiene.
Conversion goals match the current business priority (e.g., trial signup vs. paid conversion has shifted with your model). Why it matters: Old optimization goals waste budget if your business model evolved (very common in SaaS year-over-year).
A documented monthly optimization checklist exists and is followed. Why it matters: Account hygiene depends on consistent process, not heroics from a single account manager.
Found 5+ items failing on your account?
Most accounts we audit have 15+ items failing. We can run the full audit and deliver a written remediation plan in 5 business days.
Book a free 30-min call →