Hebrew source: https://mediak.co.il/guide-633
Google Analytics 4, or GA4, is Google's analytics platform for measuring website and app traffic, user behavior and important business actions.
This guide introduces its measurement model, main metrics and reports, and a practical workflow for checking data collection.
1. What is Google Analytics 4?
Google Analytics helps digital property owners understand where visitors come from, which content they view and what they do during a visit.
GA4 replaced Universal Analytics with an event-based measurement model. Rather than using page views and sessions as the main building blocks of collection, GA4 records events and can connect activity across websites, apps and devices where the implementation and available identifiers support it.
2. Events, users and reporting identity
Not every interaction is measured automatically. GA4 collects specified automatic events; other interactions require configuration.
Event types
Automatically collected events include events such as `first_visit` and `session_start` that are collected when the relevant implementation is present.
Enhanced measurement events can be enabled or adjusted in the interface. Examples include page views, scrolls, site searches, eligible video interactions and file downloads. The default `scroll` event is associated with reaching approximately 90% of the page's vertical depth, not every movement of the scrollbar. These features have implementation conditions; video tracking, for example, does not automatically cover every embedded video player.
Recommended events use Google's suggested event names and parameters for particular business activities, such as `generate_lead` or `purchase`.
Custom events are defined for a business need when no suitable recommended event exists.
Reporting identity
Reporting identity determines how GA4 combines activity into users. Its identity spaces include:
- User-ID: IDs supplied by the website or app for signed-in users. This requires implementation and must not contain prohibited personally identifiable information.
- Device ID: A device- or browser-based identifier, such as a web client ID.
- Modeling: Eligible modeled data used to address measurement gaps, including gaps associated with consent choices.
Reporting identity options include Blended, Observed and Device-based. The methods available and used depend on the property's configuration and data eligibility.
Google signals is a separate optional feature with its own activation and consent requirements. It should not be described as a mandatory prerequisite for all cross-device measurement.
Sources:
3. Basic metrics
Users
Active users is a main GA4 user metric. It represents users meeting GA4's activity criteria, including engaged-session and other qualifying signals.
New users represents people whose first visit or first app use is recorded during the selected period.
Sessions
A session is a period of interaction with the website or app. By default, a session times out after 30 minutes of inactivity; that setting can be adjusted.
Engagement rate and bounce rate
An engaged session is a session that lasts longer than the configured engagement threshold, includes a key event, or includes at least two page or screen views. The default time threshold is 10 seconds and can be adjusted.
Engagement rate is the percentage of sessions that are engaged sessions.
Bounce rate is the percentage of sessions that are not engaged sessions. In GA4, it is the inverse of engagement rate: 100% minus engagement rate.
Key events
Key events are events marked as especially important to the business, such as a qualified lead submission or purchase. GA4's standard reporting terminology uses key events for these important actions. Google Ads conversions are a related but distinct concept; not every collected event is automatically a Google Ads conversion.
Sources:
4. Practical example: measure a successful contact form submission with GTM
The following workflow uses Google Tag Manager to send `generate_lead` and then mark it as a key event in GA4.
Step 1: Set up the Google tag
If a base tag is not already configured, create a Google tag in GTM and enter the relevant tag or measurement ID, such as an ID beginning `G-`. Configure the appropriate initialization trigger, commonly Initialization - All Pages for a standard web setup.
Check whether the site already has a Google tag before adding another one, to avoid duplicate collection.
Step 2: Create the event tag
Create a Google Analytics: GA4 Event tag. Select the relevant Google tag or enter the measurement ID, and use the recommended event name:
`generate_lead`
Use the same event name consistently throughout the setup.
Step 3: Choose a trigger that represents success
Option A: Standard HTML form
A Form Submission trigger can be used for a compatible standard HTML form.
Check Validation makes the trigger depend on the browser's built-in form validation. Without that option, it can fire on an attempted submission that does not pass validation.
Browser validation is not proof that a lead was accepted by the server. Server-side processing can still fail. A standard form trigger may also fail to fire for an AJAX form that does not navigate or submit in the expected way.
Option B: AJAX form
A more precise success signal is for the site's developer to push a custom event to the data layer only after the server confirms the form was accepted. For example:
`dataLayer.push({event: 'form_success'});`
In GTM, create a Custom Event trigger listening for the exact deployed event name, in this example `form_success`.
Even this method does not guarantee complete analytics collection. Content blockers, consent settings, network errors or processing problems can still create gaps.
Step 4: Test in Preview and DebugView
1. Open GTM Preview and submit a test form.
2. Confirm that the expected Form Submission or Custom Event fired and triggered the `generate_lead` tag.
3. In GA4, open Admin > DebugView and check that the event arrives in the debug stream.
Also test an invalid or failed submission. It should not be counted as a successful lead simply because someone clicked the submit button.
Step 5: Publish the GTM container
After the test passes, use Submit and the publish/version controls in GTM. Changes tested in Preview are not automatically published to ordinary site visitors.
Step 6: Mark the event as a key event
In GA4, go to Admin > Data display > Events and use Mark as key event for `generate_lead` when the event is available, or configure the key event through the available creation workflow.
Standard reports can take 24 to 48 hours to finish processing. Intraday and daily processing follow different update schedules. Realtime and DebugView are useful checks but do not prove that every standard report has finished processing.
Sources:
5. Main reports
Realtime shows recent activity, commonly covering the last 30 minutes, and helps check whether data is arriving.
User acquisition describes how a user was first acquired.
Traffic acquisition describes the acquisition source of a particular session.
Engagement reports show viewed content, events and pages or screens receiving activity.
Demographics and technology reports provide available breakdowns such as country, city, device and browser. Privacy thresholds, consent and configuration can limit some data.
6. Retention and product editions
Data retention
For standard GA4 properties, user- and event-level retention options commonly include 2 or 14 months. This setting affects data available in tools such as Explorations; it is not a rule that every aggregated standard report disappears after that period.
Choose retention according to the business need and privacy requirements, rather than always selecting the longest period.
Standard versus Analytics 360
The standard version of Google Analytics is available without a product subscription fee. Analytics 360 is an enterprise offering under a paid contract, with higher quotas and service commitments. Implementation, consulting and related cloud services can have separate costs.
Sources:
7. Common beginner mistakes
- Not marking important events as key events. Collected events can still appear in event reports, but unmarked events are not presented as key events in the relevant reporting and conversion workflows.
- Not reviewing internal traffic. Employee or developer visits can distort analysis. Define and test internal traffic filtering before activating a filter that excludes data.
- Relying only on defaults. Review enhanced measurement and use Explorations where they fit the question.
- Forgetting GTM publication. A setup left only in the workspace or Preview is not the same as a published container.
8. Frequently asked questions
What is Google Analytics?
It is Google's measurement platform for understanding website and app traffic, user behavior, acquisition sources and important actions.
How do I view data?
Sign in with a Google account that has access to the property, select the intended property and open Reports. Realtime, acquisition and engagement reports answer different questions.
How does GA4 differ from Universal Analytics?
GA4 uses an event-based collection model. It offers cross-platform measurement options, engagement metrics and privacy controls, including retention settings and consent-related measurement features. Universal Analytics used a different session-based reporting and collection model.
Is Google Analytics free?
The standard product is available without a subscription fee. Analytics 360 is a paid enterprise offering. The costs of implementation and related services are separate.
What is a report?
A report is a structured view of measurement data, usually combining metrics and dimensions around a question such as traffic sources, popular pages or audience characteristics.
9. Topics for further reading
- GA4 key events and conversion measurement.
- Funnel and user-behavior analysis.
- Custom reports and Explorations.
- Troubleshooting data collection.
