Google Ads Agency Pre-Launch Audit Checklist: 30 Checks Before the First Optimization

Why a Pre-Launch Audit Is Different From an Ongoing Performance Review

An ongoing performance review answers “how is the account doing?” A quarterly audit answers “what needs to change?” A pre-launch audit answers a different question: “can we trust the data this account produces?” The distinction matters because the first two weeks of agency management set the trajectory for the next 90 days. If you optimize on broken conversion data, over-paced budgets, or conflicting campaign settings, you don't just waste spend — you poison the conversion lag window that every subsequent bid decision depends on.

This pre-launch audit is a verification gate. It runs before the first optimization, before any bid change, before any budget increase. Its output is a baseline snapshot and a list of verified findings, each staged as a proposed mutation for human review and approval. The checks below are organized into six phases that map to the order in which you should verify the account: access, measurement, structure, budgets, constraints, and reporting.

Phase 1 — Access, Ownership, and Billing Verification (Checks 1–4)

Check 1: Verify MCC vs Single-Account Access Level

Determine whether you're operating through a manager account (MCC) or direct account access. Document the access level for every user: admin, standard, or read-only. Admin access is required for billing changes, user management, and full settings visibility. Standard access blocks several audit items, so escalate before proceeding.

Check 2: Confirm Admin Access on the Target Account

A common agency mistake is assuming MCC access grants admin rights on child accounts. Verify the actual role on each account you'll manage. If you only have standard access, you cannot change billing profiles, manage users, or access all settings. Document the access level in your handoff notes and escalate any gaps before the audit continues.

Check 3: Billing Profile and Payment Method

Verify the billing country, currency, payment method, and auto-payment threshold. A payment failure during a campaign launch causes immediate delivery loss. A currency mismatch can silently inflate costs by 3–5% on FX conversion. Confirm the billing profile matches the client's legal entity, and flag any account where the payment method is expired or the threshold is set below the expected monthly spend.

Check 4: Account-Level vs Manager-Level Settings

Check for account-level negative keywords, account-level exclusions, and manager-level settings that override campaign-level configurations. These hidden settings are the most common source of “why is my campaign not serving?” tickets. A manager-level exclusion for a placement or audience can silently block delivery across all child accounts, and account-level negatives can block keywords the client explicitly wants to target.

Phase 2 — Conversion Tracking and Measurement Integrity (Checks 5–10)

This is the highest-priority phase. Every optimization decision in the next 90 days depends on conversion data being complete, deduplicated, and correctly attributed. If conversion tracking is broken, every other check is theater.

Check 5: Conversion Action Inventory

List every conversion action in the account. For each, document the source (website tag, GA4 import, offline upload, app), the category (purchase, lead, signup), and the counting method (every vs one). This inventory is the foundation for every subsequent check. A typical mid-size account has 8–15 conversion actions; a messy account has 30+ with overlapping definitions.

Check 6: Duplicate Conversion Actions

The classic error: the same purchase event tracked via the Google tag, a GA4 import, and a third-party tag like a form tool. This inflates conversion counts by 2–3x and causes automated bidding to over-optimize. Check the “Include in Conversions” column to identify which actions actually feed bidding. Deduplicate before any optimization. A purchase counted three times will make a target ROAS campaign believe it's performing at 300% of reality.

Check 7: Conversion Window Settings

Verify the conversion window matches the client's sales cycle. A B2B SaaS account with a 60-day sales cycle using a 7-day window is undercounting by 60–70%. The table below shows recommended windows by industry.

Check 8: Cross-Account Conversion Tracking

If the client runs multiple accounts (search, PMax, shopping, YouTube), verify the same conversion actions are shared or consistently defined across all. Mismatched definitions cause PMax to optimize for a different event than search. For example, if search counts a lead as a form submission and PMax counts it as a page view, the two accounts will compete for different outcomes and the client's blended CPA will be meaningless.

Check 9: Offline Conversion Imports

Check whether offline conversion uploads are configured and whether the upload key (gclid or wbraid) is captured correctly. A broken upload key means all offline conversions silently fail, and automated bidding optimizes on a fraction of true conversions. Verify the upload frequency, the data freshness, and whether the client's CRM is passing the correct click ID. This is the most commonly broken measurement component in B2B accounts.

Check 10: GA4 Property Linking

Verify the GA4 property is linked at the correct level and that imported conversion actions are the ones you actually want to bid on. GA4 imports often include events like scroll depth or page_view that should never feed a bid strategy. Also check for duplicate imports: the same event imported twice under different names will split conversion credit and confuse reporting.

Phase 3 — Account Structure and Campaign Settings (Checks 11–17)

Check 11: Campaign Hierarchy and Naming Conventions

Verify campaigns are organized by business line, geography, or funnel stage — not by match type or device. A campaign named “Brand – Exact – Desktop” is a structural red flag that will fight automated bidding. Document the campaign hierarchy and flag any campaign that overlaps with another on the same keywords, audience, or geography.

Check 12: Keyword Match Type Distribution

For mature accounts, 60–70% of spend should sit on exact and phrase match. If broad match dominates without a corresponding negative keyword strategy, the account is bleeding spend on irrelevant queries. Document the distribution by match type and flag any match type that exceeds 50% of spend. Also check for duplicate keywords across match types in the same ad group, which causes internal auction competition.

Check 13: Negative Keyword Lists

Verify shared negative lists are applied at the campaign or account level. Check for conflicts: a negative that blocks a keyword the client explicitly wants to target. Also check for missing negatives on high-spend, low-converting queries from the last 30 days. Pull the search terms report and identify the top 20 spend-without-conversion queries; each one should either become a negative or have a landing page reason for existing.

Check 14: Location and Language Settings

Check the location targeting option: “People in or regularly in your targeted locations” vs “People searching for your targeted locations.” The former includes travelers who may not convert. Verify location exclusions are set for regions the client cannot serve, and confirm language settings match the client's service languages. A common error is targeting “all languages” when the client only serves English speakers.

Check 15: Ad Schedule and Bid Adjustments

Check for stale dayparting bid adjustments that conflict with the new bidding strategy. Automated bidding ignores some adjustments and respects others — a legacy +50% Monday adjustment can silently distort delivery. Document all ad schedule adjustments and flag any that were set more than 90 days ago or that exceed +20% in either direction.

Check 16: Device Bid Adjustments

Check for legacy device modifiers. With automated bidding, device adjustments are largely ignored, but they create confusion in reporting and can conflict with portfolio bid strategies. A campaign with a -50% tablet adjustment and a target CPA strategy will produce confusing data because the bid strategy is optimizing across devices while the adjustment tries to suppress one device.

Check 17: Asset Group and Ad Extension Coverage

Verify each campaign has sufficient assets: 8–10 headlines, 4–5 descriptions, sitelinks, callouts, and structured snippets. PMax asset groups need at least 5 images and 3 videos. Accounts with thin assets lose 20–40% of available impression share in responsive ad auctions. Document asset coverage per campaign and flag any campaign with fewer than 6 headlines or 3 descriptions.

Phase 4 — Budgets, Bidding, and Pacing (Checks 18–23)

Check 18: Daily Budget vs Recommended

Compare current daily budget to Google's recommended budget. A campaign capped at 30% of recommended loses impression share and skews auction insights. Document the gap and flag any campaign where the recommended budget is more than 2x the current budget. These campaigns are delivery-constrained, and no bid optimization will fix them until the budget is addressed.

Check 19: Shared Budgets

Verify shared budgets are allocated correctly across campaigns. A shared budget with one campaign consuming 80% of daily spend is a common silent failure. Check the shared budget history to see which campaigns are actually getting delivery. If a high-priority campaign is being starved by a low-priority campaign in the same shared budget, split the budgets before optimizing bids.

Check 20: Bid Strategy Alignment

Verify the bid strategy matches the campaign goal. A campaign with a target ROAS of 400% that is actually a lead-gen campaign is misconfigured. Document the bid strategy for each campaign and flag mismatches. Common mismatches: using maximize conversions on a brand campaign where target CPA is more appropriate, or using target ROAS on a campaign with fewer than 15 conversions in the last 30 days.

Check 21: Target CPA/ROAS Sanity Check

Compare targets to historical performance. If the account's average CPA is $80 and the target is $40, the campaign will throttle delivery to near zero. If the target is $120, the campaign will overspend. Use the 30/60/90-day historical baseline to validate targets. A target CPA that is more than 30% below the trailing 90-day average CPA is a red flag that the target was set aspirationally, not operationally.

Check 22: Conversion Lag Consideration

For accounts with a 30-day conversion lag, verify that bid strategy targets account for the lag. A target CPA set on 7-day data will be too aggressive because it misses 60–70% of conversions that will eventually attribute. This is the single most common cause of “the account was fine before you touched it” complaints. Document the conversion lag distribution and adjust targets to reflect the full attribution window.

Check 23: Pacing Equation

Compute month-to-date pacing for every campaign with a monthly budget. The math is straightforward: expected spend by day X equals the monthly budget divided by days in the month, multiplied by days elapsed. The pacing ratio is actual month-to-date spend divided by expected spend. The required daily spend for the remainder is the remaining budget divided by remaining days. A pacing ratio above 130% means the account is over-paced and will exhaust the budget before month-end. Fix pacing before changing bids — otherwise you're optimizing a budget that's already broken.

Phase 5 — Client Constraints, Compliance, and Brand Safety (Checks 24–27)

Check 24: Brand Terms Policy

Verify whether the client allows bidding on their own brand terms and whether competitor brand terms are permitted. Document this in the account notes. A client who explicitly prohibits competitor bidding will terminate the relationship if you launch competitor keywords. Also document whether the client wants brand campaigns separated from non-brand campaigns, which affects structure and budget allocation.

Check 25: Competitor Terms

Check existing campaigns for competitor keywords that violate the client's policy. Also check for trademark issues in ad copy — a competitor's trademark in headlines is a legal risk. Document any competitor terms found and flag them for client approval before the first optimization. This is a relationship-protection check as much as a performance check.

Check 26: Restricted Categories

Verify the account doesn't target restricted categories (healthcare, financial services, alcohol, gambling) without proper certification. Google's restricted category policies vary by country, and a violation can suspend the entire account. Check the account for certifications, and verify that any restricted-category campaigns have the required disclaimers and landing page compliance.

Check 27: Landing Page Compliance

Verify destination URLs match the client's approved landing pages and that tracking parameters are appended correctly. A broken UTM parameter structure means the client's analytics will show untagged traffic, and you'll spend the first month defending your data. Check for redirect chains, broken pages, and pages that violate Google's landing page experience guidelines. A landing page with a 5-second load time will destroy conversion rate regardless of how well the account is optimized.

Phase 6 — Baseline, Reporting, and Human-in-the-Loop Handoff (Checks 28–30)

Check 28: Historical Baseline Snapshot

Capture 30/60/90-day performance metrics before any changes: impressions, clicks, CTR, CPC, conversions, CPA, ROAS, impression share, and average position. This baseline is the reference point for every future optimization decision. Without it, you cannot prove the impact of your changes. Store the baseline in a shared document that the client can access, and reference it in every weekly report.

Check 29: Reporting Cadence and KPIs

Align on the reporting schedule (weekly, bi-weekly, monthly) and the exact KPI definitions with the client before making changes. A client who defines a lead as a form submission and an agency that defines it as a phone call will disagree on every report. Document the KPI definitions, the reporting template, and the distribution list. This check prevents the most common source of agency-client friction.

Check 30: Change Log and Approval Workflow

Establish how changes will be proposed, reviewed, and approved. This is the operational backbone of the engagement. Every optimization should be a documented mutation with a proposed impact, a review step, and an approval gate. The approval workflow should specify who approves each change tier: minor bid adjustments by the agency lead, budget changes by the client, and structural changes by both.

Budget Tier Matrix: How Audit Depth Scales

The depth of the audit should scale with the account's spend. A $5k/mo account doesn't need a 5-day audit; a $200k/mo account can't survive a 2-hour one. The table below shows how audit depth should scale across budget tiers.

The key principle: audit depth scales with the cost of being wrong. A $200k/mo account with broken conversion tracking wastes $6,600 per day. A $5k/mo account wastes $165 per day. The audit investment should match the risk.

Turning Audit Findings Into Staged Mutations With PPC Tuner

Once the 30 checks are complete, the audit output is a list of verified findings. Each finding becomes a proposed mutation: a specific, reversible change staged for review and approval. This is where PPC Tuner fits. PPC Tuner is the Gemini 3.8 AI human-in-the-loop alternative to tools that auto-apply changes. Instead of letting an algorithm push changes directly to a live account, PPC Tuner stages every mutation inside its secure web application workspace. A human reviews the proposed change, sees the expected impact, and approves or rejects it before anything goes live.

This workflow matters most during the pre-launch phase. The first week of agency management is when the account is most fragile. A staged approval workflow prevents the avoidable first-week mistakes this audit is designed to catch. Every finding from the 30 checks — a duplicate conversion action, an over-paced budget, a conflicting bid adjustment — becomes a proposed mutation with a clear description, expected impact, and rollback plan.

Tools like Optmyzr, Adalysis, and Opteo offer automation, but they differ in how changes are applied and reviewed. See how PPC Tuner compares to Optmyzr, Adalysis, and Opteo. For agencies coming from a managed-service model, the PPC Tuner vs WordStream comparison covers the differences in control and transparency.

The audit findings themselves can be quantified with free diagnostic tools. The Google Ads Waste Calculator estimates the spend lost to broken tracking and misconfigured settings. The Lost IS Calculator quantifies the impression share lost to budget and rank constraints. The PMax Cannibalization Checker identifies where PMax is stealing conversions from search campaigns.

The pre-launch audit is not a one-time ritual. It's the first iteration of a continuous verification loop: audit, stage, review, approve, measure. The 30 checks establish the baseline. The staged mutation workflow protects it. Every subsequent optimization references the baseline, and every change goes through the same human-in-the-loop approval gate.