GA4

GA4 is the current version of Google Analytics - Google Analytics 4 - which replaced Universal Analytics in July 2023. GA4 is fundamentally different in its measurement model (event-based rather than session-based), its privacy posture (designed around cookieless measurement), its cross-platform tracking (unified web and app), and its reporting interface (more customisable; some familiar reports removed). Every modern marketer has to learn GA4; the migration from Universal Analytics was not optional.

What changed between Universal Analytics and GA4

Five fundamental shifts:

Event-based measurement model. Every interaction in GA4 is an event. Sessions and pageviews are computed from events rather than being primitives.

Cross-platform tracking. Single property for web and app. Unified cross-device reporting.

Privacy-first design. Works without cookies; IP anonymisation by default; first-party data emphasis.

Machine-learning-based reporting. Predictive metrics (churn probability, purchase probability) built in.

Reporting interface overhaul. Different navigation, different reports, different customisation model.

Core GA4 concepts

Six essential concepts:

Events. Every tracked interaction. Pageviews, clicks, scrolls, conversions - all events.

Parameters. Attributes attached to events. A pageview event has page_location, page_title, etc.

User properties. Attributes about the user (signed-in status, plan tier, etc.).

Conversions. Events marked as conversions. Multiple conversions per session allowed.

Audiences. Dynamic user segments based on behaviour.

Explorations. Custom report-building interface replacing Universal Analytics’s custom reports.

Common GA4 setup issues

Five patterns that cause problems:

Missing enhanced measurement. GA4 auto-tracks scrolls, outbound clicks, video plays, downloads, site search. Disabling these by default means missing common events.

Conversion events not marked. Events only become conversions when explicitly marked. Teams often track events but don’t flag them as conversions.

Cross-domain tracking misconfigured. Multi-domain sites need explicit cross-domain configuration. Default GA4 treats cross-domain navigation as session-end.

Content grouping not set up. Universal Analytics’ content groupings must be recreated in GA4. Often skipped during migration.

Referral exclusions incorrect. Default referral exclusions may not cover all relevant domains. Manual review needed.

What GA4 reports well

Four strong areas:

Event-level analysis. Detailed breakdowns of specific user actions.

User journeys. Path analysis across sessions.

Cross-device reporting. Users tracked across web and app.

Predictive metrics. ML-driven predictions about user behaviour.

Where GA4 falls short compared to Universal Analytics

Four gaps:

Simpler questions take more clicks. Reports that were standard in Universal Analytics sometimes require Exploration-builder work in GA4.

Unique pageviews metric is gone. No direct equivalent. ‘Views’ in GA4 equals total pageviews; unique page views require workarounds.

Behaviour-flow report absent. Universal Analytics’s flow reports don’t have direct GA4 equivalents.

Data retention defaults are shorter. GA4 default is 2 months (adjustable up to 14). Universal Analytics stored indefinitely.

GA4 and privacy

Three design features for privacy compliance:

Cookieless measurement support. Built for the cookieless future through Google Signals and Consent Mode.

IP anonymisation by default. IP addresses not stored by default.

Consent Mode integration. GA4 adjusts tracking based on user consent status. Regulated-market-compliant by design.

GA4 and content programmes

Four ways GA4 supports content measurement:

Content-specific events. Scroll depth, time on page, outbound clicks to external sources. GA4 tracks these by default.

Engagement rate metric. Replaces bounce rate; more nuanced measure of engagement.

Path exploration. Shows user journeys through content before converting.

Cohort analysis. Group users by first-visit period; track their behaviour over time.

GA4 attribution changes

Three shifts:

Data-driven attribution is default. ML-based model rather than last-click.

Attribution paths include more touchpoints. Longer attribution windows and more comprehensive path tracking.

Cross-device attribution. Users tracked across devices attributed as single users.

Common GA4 reporting workflows

Four frequent use cases:

Traffic source analysis. ‘Acquisition’ reports → ‘User acquisition’ or ‘Traffic acquisition.’

Content performance. ‘Engagement’ → ‘Pages and screens’ report shows per-URL performance.

Conversion tracking. ‘Advertising’ section plus ‘Events’ reports for conversion analysis.

Custom exploration. For questions not covered by standard reports, Explorations interface.

Penfriend and GA4

Penfriend doesn’t replace GA4; it produces the content GA4 measures. Sites running Penfriend-produced content programmes see them reflected in GA4 as content-driven traffic, engagement, and conversion. The content catalogue shows up in ‘Pages and screens’ reports, organic-search acquisition data, and conversion paths. Clean content with proper schema integrates cleanly with GA4’s event tracking.

Related terms