Universal Analytics vs GA4: What Actually Changed

By Rahul Saini, Search Counsel Co. • Updated August 2026

1. Universal Analytics vs GA4 at a glance

The biggest mistake is assuming a familiar metric name means the underlying calculation stayed the same.

2. What can you safely compare between Universal Analytics and GA4?

This is the most important question if your organization still has exported UA history.

3. The core change: Universal Analytics hits vs GA4 events

The event model is the root of most other differences.

Universal Analytics had several hit types and organized reporting heavily around sessions. Pageviews, events, transactions and social interactions were treated as distinct hit categories.

GA4 collects interactions as events. An event can carry parameters that provide additional context.

A page view, for example, is collected through the page_view event rather than existing as an entirely separate reporting architecture.

This is why recreating an old UA dashboard in GA4 column-for-column is often the wrong goal.

If you are building your reporting stack around GA4 now, use our GA4 for SEO guide for current implementation.

4. GA4 sessions vs Universal Analytics sessions

In Universal Analytics, a session could end and a new one begin because of rules such as midnight or new campaign information.

GA4 instead calculates sessions using session identifiers created when the session_start event occurs, with a default inactivity timeout of 30 minutes.

A simple illustrative example

Under old UA rules, midnight and campaign changes could create additional session boundaries. GA4 does not apply those legacy resets in the same way.

That is one reason the same real-world visit can produce different session totals.

5. GA4 users vs Universal Analytics users

User counts are another common source of false comparisons.

Google says GA4 and Universal Analytics can differ because they may use different identity methods, different property scope and different user metrics.

If your GA4 user count is lower than an old UA report, first confirm whether you are comparing Active Users with old UA Users.

Also check:

• User-ID configuration;
• reporting identity;
• which data streams are included;
• whether old UA Views excluded traffic with filters;
• tagging differences between the properties.

6. GA4 bounce rate vs Universal Analytics bounce rate

Google currently defines an engaged session as one that meets at least one of these conditions:

• lasts longer than 10 seconds;
• contains a key event; or
• has two or more pageviews or screenviews.

Engagement rate = engaged sessions ÷ total sessions.

GA4 bounce rate = sessions that were not engaged.

Also remember that engagement rate can be affected by your key-event configuration. If a very low-value action is marked as a key event, it can make more sessions qualify as engaged.

7. Universal Analytics Goals vs GA4 key events

“Goals became key events” is directionally correct, but it hides an important measurement difference.

Universal Analytics Goals were built around the UA goal framework. In GA4, important actions are represented by events and can be marked as key events.

GA4 currently supports two key-event counting methods:

That distinction matters because old UA Goals were generally session-limited in ways that may not match a GA4 key event configured to count once per event.

For implementation, use our GA4 key events and organic conversion tracking guide.

8. Attribution changed too

A difference that looks like a conversion problem can actually be an attribution problem.

GA4 uses different scopes for traffic-source dimensions:

• First user dimensions describe how the user was originally acquired.
• Session dimensions describe how a specific session began.
• Event-scoped dimensions can be used in key-event attribution reporting.

Google says user- and session-scoped traffic dimensions use paid-and-organic last-click logic, while event-scoped key-event reporting can use the property’s selected reporting attribution model. Data-driven attribution is commonly the default for event-scoped reporting.

9. Universal Analytics Views are not GA4 data streams

Universal Analytics commonly used an Account → Property → View structure. Views could apply filters and create separate reporting perspectives.

GA4 does not have UA-style Views.

Instead, a GA4 property can collect information from data streams, such as:

• a website;
• an iOS app;
• an Android app.

A data stream is a source of collected data. It is not a direct replacement for an old UA View.

This difference can materially change user and session totals if your GA4 property includes more streams or a broader scope than the old UA View you used for reporting.

10. Data retention and BigQuery: do not repeat the UA history problem

Universal Analytics disappearing in 2024 was a painful reminder that analytics history should not exist in only one interface.

GA4 data retention

Standard GA4 properties currently support event-data retention settings of up to 14 months for affected user/event-level analysis, with 2 months as a common default in Explorations.

This does not mean every aggregate report disappears after 14 months. The retention setting is particularly important for user-level and event-level data used in exploratory analysis.

GA4 BigQuery export

GA4 standard properties can export raw event data to BigQuery. Google’s current daily export limit for standard properties is up to 1 million events per day.

Once data is exported into your BigQuery project, it becomes part of your own warehouse rather than relying only on the Analytics interface.

For organizations that care about long-term raw analytics history, that is one of the most important architectural differences between the UA era and GA4.

11. Why your GA4 and Universal Analytics numbers do not match

A mismatch does not automatically mean the traffic changed, but it also does not automatically mean “GA4 just counts differently.”

Use the pattern to decide what to investigate.

12. How to report Universal Analytics and GA4 history together

If you already preserved UA data, you can still use it for long-term business context. Just do not pretend the measurement system stayed unchanged.

Bad reporting

Better reporting

Add a note explaining:

• when GA4 became the reporting source;
• which GA4 metric replaced the old KPI;
• whether the definition changed;
• whether an overlap period was used for validation.

If you have a period where UA and GA4 ran in parallel, use it to understand your own property’s difference—but do not turn that one property’s discrepancy into a universal industry benchmark.

For broader reporting design, see our SEO analytics and reporting guide and SEO KPIs that matter.

13. A six-check reconciliation framework

Before deciding that UA and GA4 disagree, check these six things.

FAQ: Universal Analytics vs GA4

Primary sources

This guide prioritizes current Google Analytics documentation because many older UA-vs-GA4 articles still contain migration-era advice that is no longer accurate in 2026.

Final takeaway

Universal Analytics and GA4 measure many of the same business questions, but they do not measure them with the same underlying system.

That is why the right question in 2026 is no longer:

“How do I migrate to GA4?”

The useful question is:

Keep old UA exports for historical context if you have them. Mark the platform transition clearly. Validate sessions, users, key events, attribution and property scope before comparing numbers. And build your future analytics history around GA4 rather than trying to reconstruct an old measurement model that no longer exists.