Why we built the GA4 Auditor, a free Google Analytics debugging tool
GA4 is the standard today – and at the same time one of the most frequent causes of incorrect marketing decisions. Not because Google Analytics 4 is “bad”, but because the measurement chain has become more complex: consent, browser restrictions, event-based tracking, e-commerce implementations, cross-domain flows, payment providers.
At our webmastering firm, Prange , this led to a very simple situation: We constantly had to check the same things during GA4 setups – and the manual approach was time-consuming and annoying. So we built a tool to automate our auditing work. The result is the GA4 Auditor .
The problem: A GA4 audit is mandatory – but it takes time.
A proper GA4 audit doesn’t consist of “just quickly looking at the reports”. It examines the setup and the data reality .
Typical symptoms in projects:
“(not set)” in pages, sources, or campaigns
Direct is suspiciously high
Self-referrals or payment referrals destroy attribution.
Sales figures in GA4 do not match the shop system
Duplicate purchases lead to an “excessively high” ROAS.
A quick quality assurance check is lacking after deployments.
A manual audit by an experienced analyst typically takes 4–8 hours – and after every change, you essentially have to recheck everything. That’s exactly what we wanted to solve.
The idea: Same tests, but in minutes – repeatable at any time
The GA4 Auditor automates precisely these standard checks:
Live configuration via the GA4 Admin API
Real-time check to see if events are actually being well received
Optional (if enabled): In-depth analysis of the last 30 days via BigQuery export
The result is a dashboard with traffic light logic :
Passport : no irregularities
Warning : Risk/setup gap that frequently leads to false reports
Fail : high probability of a genuine tracking or data problem
What the GA4 auditor checks – on three levels
1) Health Checks (quick)
Fast setup and data quality checks based on the GA4 APIs.
2) Traffic light checks (prioritized)
Compressed evaluation (Pass/Warning/Fail) with a focus on BigQuery raw data checks .
3) SST analysis (server-side – pragmatic)
Technical assessment of whether and where server-side tracking makes sense and where measurable problems exist today.
