GA4
Data & TrackingAlso: Google Analytics 4 · Google Analytics
Quick definition
GA4 stands for Google Analytics 4. It is Google's current web and app analytics platform, released as the replacement for Universal Analytics in 2023. GA4 tracks user behaviour as a stream of events rather than sessions and page views, which changes how reports are structured and how conversions are defined.
How it varies across Australia
Across Australian businesses we have reviewed, GA4 is installed on most sites but configured correctly on far fewer. The gap between having GA4 and getting reliable data from GA4 is wide. Most common failures are missing conversion events, no consent mode, and default sampling thresholds hiding real traffic patterns.
See data and tracking maturity across Australian industries →How GA4 thinks differently
Every interaction is an event. Page views, scrolls, clicks, form submissions. The session model is secondary.
Events carry metadata called parameters. A purchase event carries value, currency, and items. Parameters are what make reports useful.
You mark specific events as conversions. Nothing is a conversion by default except purchases on ecommerce setups.
The custom report builder inside GA4. More flexible than the standard reports and where most useful analysis happens.
GA4 can export raw, unsampled event data to Google BigQuery. The only way to get reliable data at scale.
What it actually means
GA4 is Google's answer to a tracking world that changed: mobile apps, privacy regulations, cookies disappearing, and users jumping between devices. Universal Analytics was built for a simpler world where a session on a desktop browser told most of the story. GA4 tries to stitch a fuller picture together.
The core shift is from sessions to events. In Universal Analytics, everything was framed around sessions and page views. In GA4, every user interaction is an event with its own parameters. A scroll is an event. A video play is an event. A purchase is an event. This makes the data model more flexible but also harder to read out of the box.
The practical consequence is that GA4's default reports look sparse compared to Universal Analytics. The useful analysis happens in Explorations, in the Advertising workspace, and for serious use cases, in a BigQuery export where the raw data is unsampled and queryable.
GA4 also introduced consent mode, which uses modelling to estimate conversions from users who declined tracking. The model is useful but imperfect. It makes the 'total conversions' figure in GA4 a blend of observed and modelled data, which needs to be understood before you act on it.
Having GA4 installed and having GA4 configured are two different things. Most businesses have the former and think they have the latter.
How it shows up
GA4 shows up as the base layer of your marketing measurement stack. It tells you which pages get traffic, where users drop off, which events fire, and which channels drive conversions as GA4 defines them.
It connects to Google Ads for conversion import, to Search Console for organic performance, and to BigQuery for raw data export. Most of the other measurement tools in your stack either push data into GA4 or pull from it.
When GA4 is configured well, it gives you a reliable first-party view of user behaviour that doesn't depend on ad-platform reporting. When it is configured poorly or not at all, every channel report you rely on is built on a gap.
The Australian context
Australian businesses operating under the Privacy Act 1988 and, for those in relevant sectors, the Australian Privacy Principles (APPs), need to consider consent before firing GA4 tracking. Google's consent mode is the standard implementation path, but it requires explicit integration with a consent management platform to function correctly.
The Australian Privacy Act is under reform. The proposed amendments would tighten consent requirements and increase penalties for breaches. Australian businesses running GA4 without consent mode are likely ahead of a compliance deadline, not behind it.
Australian English localisation in GA4 is automatic but currency and timezone configuration is manual. Both need to be set correctly at property creation. Resetting them later is possible but resets historical data interpretation.
Where people get this wrong
Related terms
Common questions
Do I still need Universal Analytics?
No. Google shut down standard Universal Analytics properties in July 2023 and stopped processing new hits. GA4 is the current platform. If you are still referencing Universal Analytics data, you are looking at historical exports only.
Is GA4 free?
The standard GA4 property is free. GA4 360, which is the enterprise tier with higher data limits, no sampling, and SLA-backed support, is a paid product available through Google Marketing Platform resellers including some Australian agencies.
Why do my GA4 conversion numbers not match my ad platform numbers?
Attribution models differ. GA4 uses data-driven attribution by default and counts conversions across all channels. Google Ads counts conversions it can attribute to itself within its own window. The gap is normal. Neither number is wrong but they are answering different questions.
Should I connect GA4 to BigQuery?
If you have meaningful traffic volume, yes. The BigQuery export gives you raw, unsampled event data you can query directly. Standard GA4 reports sample above certain thresholds and the Explorations interface has row limits. BigQuery removes both constraints.
Keep exploring
About New Rebellion
New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.
How we think →