Google Analytics 4 (GA4) is the current generation of Google's free analytics platform. It replaced Universal Analytics in July 2023, requiring most businesses to migrate or lose their historical data. GA4 represents a fundamental rebuild — not just an update — with significantly different data models, reports, and capabilities.
How GA4 differs from Universal Analytics
- Event-based vs. session-based — everything in GA4 is an event (page view, click, scroll, form submission). Universal Analytics organized data around sessions.
- Cross-platform tracking — GA4 was built to track web + mobile app behavior together; UA was web-only
- Privacy-focused — GA4 uses modeled data (statistical estimates) where consent or tracking is unavailable, reducing reliance on individual user tracking
- BigQuery export free — the raw event data can be exported to BigQuery for advanced analysis, previously a paid feature
- Different default reports — many UA reports don't exist or look different; significant relearning required
- Limited historical data — if you migrated late or fresh-started, you have less historical depth than mature UA properties had
Key GA4 concepts to understand
- Events — everything users do: page_view, scroll, click, file_download, form_submit, purchase. Some auto-collected, some require manual setup.
- Conversions (key events) — events you've marked as business-critical; reporting and bidding can be tied to these
- User properties — attributes describing users (account type, plan tier, persona)
- Audiences — segments of users defined by behavior or properties; can be exported to Google Ads
- Explorations — the analysis interface for custom reporting beyond standard reports
- Attribution models — GA4 uses data-driven attribution by default for paid channel analysis
Common GA4 setup priorities
- Enhanced measurement enabled — auto-tracks scroll, outbound clicks, file downloads, site search, video engagement
- Key events configured — mark form submissions, purchases, demo requests, key page views as conversions
- Google Ads linked — for conversion import and audience export
- Google Tag Manager set up — for any custom event tracking beyond enhanced measurement
- Consent mode v2 implemented — required for EU compliance, increasingly relevant globally
- BigQuery export configured — even if you don't use it now, having historical raw data exported is invaluable later
- Reporting customized — default reports often don't match how you'd want to view your business; custom reports/explorations close that gap
Common GA4 frustrations
- Different metric definitions than UA (e.g., "users" calculated differently)
- Sampled data on large queries
- Thresholding (hiding data when sample sizes are too small)
- 14-month default data retention (extendable to longer with paid tiers)
- The reporting interface feels less intuitive coming from UA
- Attribution windows and data lag — conversions can take 1-3 days to fully appear
When GA4 isn't enough
For mature businesses needing custom funnel analysis, deep cohort work, or attribution beyond Google's models, supplemental tools (Mixpanel, Amplitude, Heap, Northbeam for ecommerce) often layer on top. GA4 is free and necessary; not always sufficient.