The Complete GA4 Audit Checklist

This is the checklist we use internally when auditing GA4 implementations for enterprise clients. It covers everything from basic configuration to advanced e-commerce and consent setup.

Use this as a self-audit guide. If you find more than 5 issues, your implementation needs professional attention. If you find more than 10, your data is likely unreliable for business decisions.

1. Property & Data Stream Configuration

• GA4 property exists and is the correct type (not a UA property being confused with GA4)
• Data stream is configured for the correct domain(s)
• Enhanced Measurement is enabled with appropriate toggles (page views, scrolls, outbound clicks, site search, file downloads, video engagement)
• Measurement ID is correct and matches what is deployed on the site
• Internal traffic filtering is configured (IP filters for office, agency, developers)
• Referral exclusion list includes payment gateways and third-party checkout domains
• Cross-domain measurement is configured for all domains in the user journey
• Data retention is set to 14 months (not the default 2 months)
• Google Signals is enabled (if demographics and cross-device reporting are needed)
• Reporting identity is set appropriately (blended, observed, or device-based)

2. Tag Implementation (GTM)

• GA4 Configuration Tag fires on all pages (All Pages trigger)
• Only ONE GA4 Configuration Tag exists per measurement ID (no duplicates)
• Configuration Tag uses correct Measurement ID
• Page view is not being sent twice (Configuration Tag + separate page_view event)
• All custom events use proper naming convention (snake_case, no spaces, under 40 characters)
• Event parameters are registered as custom dimensions/metrics in GA4 Admin
• No PII (email, phone, full name) in event parameters or page URLs
• Tags fire in the correct order (GA4 config before custom events)
• Debug mode works (DebugView shows real-time events)
• GTM container is published (not just in preview)

3. Conversion Configuration

• Key events (conversions) are defined and marked in GA4
• Conversion counting method is correct (once per event for leads, every event for purchases)
• Conversion events actually fire when the action happens (test each one)
• No stale/legacy conversions that no longer receive data
• Google Ads conversion import is linked and receiving data (if using GA4 import)
• Conversion values are assigned where applicable
• Primary vs secondary conversion distinction is clear in Google Ads (if linked)
• No duplicate conversion tracking (GA4 event + separate Google Ads tag for the same action)

4. E-Commerce Tracking

• All required e-commerce events are implemented: view_item, view_item_list, add_to_cart, remove_from_cart, begin_checkout, add_payment_info, add_shipping_info, purchase
• Purchase event includes: transaction_id, value, currency, tax, shipping, items array
• Items array includes: item_id, item_name, item_category, price, quantity
• Currency is set correctly (USD, EUR, etc.) and matches across all events
• Revenue in GA4 matches revenue in the e-commerce platform (within 5%)
• Refund events are implemented (if applicable)
• No duplicate purchase events (check for page refresh double-firing)
• view_item_list and view_item have proper item parameters for remarketing audiences

5. User Properties & Audiences

• User-ID is implemented for logged-in users (if applicable)
• User properties are set for key segments (customer_type, membership_tier, etc.)
• Custom audiences are created for key segments (purchasers, cart abandoners, high-value)
• Google Ads audience sharing is enabled (if running ads)
• Remarketing audiences have sufficient membership duration
• No PII stored in user properties

6. Google Ads Integration

• GA4 and Google Ads are linked
• Auto-tagging is enabled in Google Ads
• GA4 audiences are shared with Google Ads
• Key events are imported into Google Ads as conversions
• Imported conversions show data in Google Ads (not zero/inactive)
• Attribution model is consistent between GA4 and Google Ads settings
• Enhanced Conversions is enabled and verified
• If using GA4 import AND native Google Ads tags, no duplicate counting

7. Consent & Privacy

• Consent Mode v2 is implemented (both ad_storage and analytics_storage)
• Default consent state is set correctly (denied for EEA, check for US state requirements)
• CMP integrates with GTM consent mode (consent update fires when user makes choice)
• Tags respect consent signals (no marketing tags fire when consent is denied)
• Consent state is visible in GTM preview mode
• GA4 receives consent mode signals (check in DebugView - consent parameters visible)
• Behavioral modeling is enabled in GA4 (to recover data from non-consented users)
• Privacy Policy and Cookie Policy pages are accessible and current
• Cookie banner appears on first visit in incognito mode
• No tracking scripts fire before the CMP initializes (check Network tab)

8. Data Quality & Validation

• Real-time report shows data when you browse the site
• Page view counts are reasonable (not doubled, not missing pages)
• Session counts align roughly with server logs or other analytics tools
• No "(not set)" values in key dimensions (page title, page path, source/medium)
• Self-referrals are not inflating session counts
• Bot/spam traffic is not inflating numbers (check for suspicious referrers)
• Hostname filter is applied (only your domains should appear)
• UTM parameters are working correctly for campaign tracking
• Site search tracking returns actual search terms (not "(not set)")

9. Reporting & Attribution

• Attribution model is explicitly chosen (not left on default)
• Reporting identity aligns with your data strategy
• Custom reports or explorations are set up for key business questions
• Automated exports are configured if needed (BigQuery, scheduled emails)
• GA4 property is connected to Looker Studio (if using for dashboards)
• BigQuery export is enabled (if raw data analysis is needed)
• Data freshness is understood (GA4 has 24-48 hour processing delay for standard reports)

Scoring your audit

Count the items you flagged as issues. Here is how to read the results:

• 0-5 issues found: Strong implementation. Focus on optimization and advanced features.
• 6-10 issues found: Decent foundation but data reliability is questionable. Fix the gaps before making business decisions from this data.
• 11-20 issues found: Significant problems. Your reports are likely misleading. Prioritize the conversion and consent sections.
• 20+ issues found: The implementation needs to be rebuilt. Consider engaging a specialist before spending more on ads optimizing against this data.