A GA4 property left unchecked will degrade. Tags break silently when sites are updated. Consent banners get reconfigured. UTM campaigns go live with typos. Bot traffic spikes. Revenue attribution shifts. None of these announce themselves in the GA4 interface. The only reliable way to catch them before they distort weeks of data is a consistent monitoring routine that you run on a fixed schedule. A healthy property should be quick to review. If the checklist keeps expanding, that is usually a sign that the implementation needs a deeper audit.

How to work through a GA4 audit systematically

This checklist is best used as a recurring operational review, not as a claim that every property should fit one rigid timing or threshold model. Work from upstream configuration and collection checks toward downstream reporting symptoms so you do not spend time investigating a report issue that was really caused by a broken tag or filter.

Property configuration

Verify the property time zone, currency, data retention period, and Reporting Identity. Confirm Google Signals status and whether it aligns with your consent framework. Check that the Internal Traffic data filter is Active, not Testing.

Data streams

Confirm each data stream is receiving hits by checking the last 48 hours of activity. Review Enhanced Measurement settings and disable any auto-collected events that are not deliberately used. Verify the measurement ID matches the one deployed on site.

Filters and data quality controls

Check all active data filters. Confirm the IP ranges defined in the Internal Traffic filter still cover your team. Review whether a bot filtering exception is in place for known high-volume internal testing tools.

Events

Open the Events report. Check for unexpected event name casing, duplicate events, and any event with a suspiciously high count that was not present in the previous period. Verify that Enhanced Measurement events are not firing unexpectedly after a site change.

Conversions

Confirm that only meaningful business outcomes are marked as key events. Check that page_view, scroll, session_start, and first_visit are not toggled as conversions. Verify conversion counts are consistent with your backend or CRM over the same period.

Attribution

Review Traffic Acquisition for self-referrals (your own domain appearing as a referral source indicates broken cross-domain tracking). Confirm payment processor domains are not appearing as session sources. Check the Referral Exclusion List is current.

Reports

Verify that standard reports return data and that no dimension shows an unusually high (not set) or (other) row. Confirm custom reports and collections are loading correctly. Check for thresholding indicators in demographic reports if Google Signals is enabled.

Integrations

In Admin > Product Links, confirm Google Ads, Search Console, and BigQuery connections are active. Check the BigQuery daily export has completed for the previous day. Verify that Google Ads conversion imports are receiving data from GA4.

Configuration

Start with property, stream, filter, and consent checks before reading the reports

Collection

Validate tags, key events, and attribution continuity before trusting output metrics

Reporting

Use report symptoms as evidence to investigate, not as proof by themselves

Escalation

Escalate when the change is unexplained and materially affects business decisions

Traffic check

Compare this week's sessions to a recent baseline and to known business context. The point is not to use one universal percentage threshold. The point is to spot a movement that is large enough to be surprising for this property. Pair this with the broaderproperty configuration checkson a recurring deeper-dive cycle so structural problems are caught before they show up as traffic anomalies.

Open Traffic Acquisition report. Set date range to last 7 days vs previous 7 days
Check total sessions for unexplained step-changes relative to recent history
Check channel split for unusual redistribution that may indicate attribution breakage
If Direct, Referral, or Unassigned moves sharply without a business reason, review tagging, consent timing, and domain handoff behavior

Conversion and revenue check

Verifying that priority business outcomes are still being collected correctly is the core of the routine. The deeper checks behind these signals are covered in theconversion tracking setup guide.

Open Monetisation > Ecommerce Purchases. Compare this week to last week
Check purchase count for unexplained movement against business context
Revenue per transaction: sudden change indicates currency parameter errors or tax/shipping being included or excluded inconsistently
Check purchase event count in the Events report: if purchase events fire but revenue is zero, the value or currency parameter has broken

Data quality check

Use this pass to catch silent collection problems before they distort reporting. The fulldata hygiene auditcovers the deeper review pattern, and(not set) valuesare usually the first symptom of upstream collection drift.

Check Unassigned traffic share in Traffic Acquisition and investigate if it rises materially above the property's normal pattern
Check the Events report for any event with an unusually high count that was not present last week
Open DebugView and verify no unexpected real user traffic is appearing (which would indicate debug_mode=true in production)
Check the Internal Traffic filter is still Active in Admin > Data Settings > Data Filters

Want to automate this checklist and get alerted when something breaks?

Review audit findings

Attribution check

Attribution checks should also include verifying that the underlyingdata streamsare still routing traffic correctly, since stream-level issues often surface first as referral noise.

Check Referral sources in Traffic Acquisition for any new referral domains appearing this week
If your own domain appears as a referral source, cross-domain tracking or Unwanted Referrals configuration has broken
Check that payment processor domains (PayPal, Stripe, Klarna) are not appearing as session sources
For active paid campaigns: verify the campaign appears in the Campaigns report with correct source and medium

Monthly add-ons

Run a 30-day revenue reconciliation against ecommerce platform backend
Review GTM container version history for any unexpected publishes
Check BigQuery last export date if active
Review custom dimensions cardinality and quota usage
Check data retention setting has not been changed

When to escalate immediately

Certain signals should trigger immediate escalation regardless of where they appear in the weekly routine:

Sessions drop sharply day-over-day with no deployment or seasonal explanation
Revenue in GA4 drops to zero despite sales continuing in the ecommerce platform
The consent banner is not loading (often caught through a zero-session gap for EEA traffic)
A new referral source appears that is your own domain (cross-domain tracking broken)
DebugView shows continuous real-user traffic (debug mode in production)

Weekly monitoring checklist summary

Sessions: week-over-week change within expected range
Channel split: no unexplained large shifts in channel share
Conversions and revenue: counts and amounts consistent with business context
Unassigned traffic share: within the expected range for this property
No production debug traffic in DebugView
No own-domain self-referrals in Traffic Acquisition
Internal Traffic filter is Active
Core audit areas are revisited on a recurring deep-dive cycle
Escalation process documented for critical signals (for example zero revenue or a major unexplained traffic drop)

GA4 audit checklist: a practical monitoring routine for analysts