Google Analytics 4 Basics: A Beginner's Orientation
If you opened Google Analytics expecting the dashboard you remember from a few years ago and felt instantly lost, you're not alone. In mid-2023 Google retired the old Universal Analytics and replaced it with Google Analytics 4 (GA4) — a different tool built on a different data model, not a redesign of the old one.
This guide is a beginner's orientation, not an exhaustive manual. The goal is to give you a working mental map: why GA4 looks the way it does, the few concepts that make everything else click, how to get set up, and which reports are actually worth opening in your first month.
- GA4 is event-based. Every interaction — a page view, a click, a purchase — is an event, which is a big shift from the old session-and-pageview model.
- It tracks across web and app using "properties" and "data streams," so one property can hold a site and an app together.
- Setup is two pieces: a property and a tag on your site (often via Google Tag Manager) that sends data to it.
- "Key events" (formerly "conversions") are the events you mark as the actions that matter most to you.
- Explorations let you build custom analyses beyond the standard reports, but start with the standard reports first.
GA4 vs Universal Analytics
The single most useful thing to understand is why GA4 feels unfamiliar. Universal Analytics organised data around sessions and pageviews: a visit was a session, and the page loads inside it were pageviews, with a handful of special "events" bolted on. GA4 throws that hierarchy out. In GA4, everything is an event — including a page view — and a session is simply a group of events that happen close together in time.
That change has consequences you'll notice immediately. Familiar metrics like "bounce rate" were dropped (then partly reintroduced) in favour of engagement-based metrics such as engaged sessions and engagement rate. Reports are fewer and more flexible. And because the model is the same for websites and mobile apps, GA4 can measure both in one place. Google's own introduction to GA4 walks through the rationale.
Stop trying to find the old report. Ask "what event represents this action?" — that question fits GA4's model and almost always leads you to the answer.
The event-based model
An event in GA4 is a named thing that happened, optionally carrying extra detail in the form of parameters. A page_view event carries the page location and title. A purchase event carries the value and currency. This structure is what makes GA4 flexible: you can measure almost any interaction by sending an event with the parameters you care about.
Google groups events into a few types, which is helpful when you're deciding what you need to configure yourself versus what arrives for free.
Automatically collected events (like session_start) and enhanced measurement events (scrolls, outbound clicks, site search, file downloads) need no code — you just toggle enhanced measurement on for the data stream. This is the fastest way to get useful data flowing.
Recommended events use Google's predefined names and parameters (such as purchase or sign_up) so reports work properly; you send them yourself. Custom events are ones you name when nothing fits. Prefer a recommended event over a custom one whenever there's a match.
Setting up GA4
Getting started is two distinct steps that beginners often blur together. First, create an account and a property in Analytics — the property is the container that holds your data. Inside it, you create a data stream for each platform (a web stream for your site, an app stream for an app). The web stream gives you a measurement ID that looks like G-XXXXXXX.
Second, you have to actually put a tag on your site that sends data to that stream. You can paste the Google tag directly into your pages, or — better for the long run — install it through Google Tag Manager, which lets you manage tags without editing code every time. After installing, use the Realtime report or DebugView to confirm data is arriving before you trust anything.
| Term | What it is |
|---|---|
| Account | Top-level container for one organisation |
| Property | The dataset for a website and/or app |
| Data stream | A source of data — one per platform (web, iOS, Android) |
| Measurement ID | The G- ID that links your site tag to the stream |
| Key event | An event you've flagged as a meaningful conversion |
Key reports worth opening
GA4's standard reports live under two collections, Life cycle and User. You don't need all of them. These four answer the questions most beginners actually have.
Reports snapshot / Realtime give you the pulse and let you verify tracking. Acquisition shows where users come from — organic search, paid, referral, direct. Engagement > Pages and screens shows your most-visited content. Engagement > Events lists every event being collected, which is where you confirm your conversions are firing. Spend your first weeks here before touching anything advanced.
Conversions and key events
Analytics is only useful if it measures the actions that matter to you — a purchase, a lead form submission, a newsletter sign-up. In GA4 you do this by marking an event as a key event (this was previously labelled "conversion," and you'll still see that word in older guides and in the linked Google Ads side of things).
The workflow is: make sure the action fires an event, confirm you can see that event in the Events report, then toggle it on as a key event. From then on GA4 counts it and surfaces it across reports. Google's documentation on key events covers the current setup. Choose a small number of genuinely meaningful actions rather than flagging everything — a list of twenty "conversions" tells you nothing.
- Created a property and a web data stream
- Installed the Google tag (directly or via Tag Manager)
- Confirmed data in the Realtime report
- Turned on enhanced measurement
- Identified the 2–4 actions that count as success
- Marked those events as key events
- Set the data retention period to suit your needs
Exploring data
Once the standard reports feel familiar, the Explore section lets you build custom analyses — free-form tables, funnel explorations to see where users drop off in a sequence, and path explorations to see the routes people take through your site. These are powerful, but they reward people who already know what question they're asking. Don't start here.
A good habit: when a standard report can't answer your question, write the question down in plain language first ("at which step do checkout visitors leave?"), then build the smallest exploration that answers exactly that. To turn the answers into decisions, pair GA4 with our guide to the marketing metrics that matter and the conversion rate optimization basics.
Common questions
Is GA4 free?
Yes, the standard version of GA4 is free and is what the vast majority of sites use. There is a paid enterprise tier (Google Analytics 360) with higher limits and extra features, but you almost certainly don't need it to start.
Where did bounce rate go?
GA4 initially replaced bounce rate with engagement metrics, then reintroduced bounce rate as the inverse of engagement rate. The engagement-based view is generally more useful: it counts a session as "engaged" if it lasts long enough, has a key event, or includes multiple page views.
Do I need Google Tag Manager to use GA4?
No — you can paste the Google tag straight into your site. Tag Manager is recommended because it lets you add and edit tags without changing site code each time, which saves real effort once you track more than a couple of things.
Why does GA4's data not match my other tools exactly?
No two analytics tools count identically. Differences in how sessions are defined, how bots are filtered, time zones, and consent settings all cause gaps. Use GA4 to spot trends and compare periods rather than treating any single number as absolute truth.
Where to go next
GA4 is the measurement layer; the value comes from acting on what it shows. Connect it to your advertising with the Google Ads beginners guide, learn which numbers deserve attention in marketing metrics that matter, and use the marketing funnel explainer to understand the journey your reports are describing.