GA4 vs Universal Analytics: Key Differences and Migration Guide (2026)

Quick answer: Universal Analytics (UA) measured sessions; GA4 measures events. That single change ripples through everything else — fewer reported users, redefined metrics (bounce rate is gone, engagement rate replaces it), a flatter account structure, and far more customization. UA stopped collecting data on July 1, 2023, and Google began deleting UA properties in July 2024, so for almost every business this is no longer "should I migrate?" but "is my GA4 set up correctly?"

This guide covers the real differences that matter when you're reading reports or finishing a migration — not just the marketing bullet points. If your numbers look lower than they did in UA, that's expected, and we explain exactly why below.

The state of play in 2026

Here's the timeline that matters:

• July 1, 2023 — standard UA properties stopped processing new hits.
• July 1, 2024 — Google began permanently deleting UA properties and their historical data.
• Today — UA is gone. If you didn't export your historical data, it's likely no longer recoverable. (See how to check what's left and preserve it.)

Industry data shows roughly 87% of former UA users have migrated to GA4 — but the average property uses only about a dozen of GA4's 40+ available event types. In other words, most teams have GA4 installed but not configured. That gap is where bad data comes from.

Sessions vs events: the core difference

Universal Analytics was built around the session — a container for a visit, with pageviews and events nested inside it. Almost every UA report rolled up to sessions.

GA4 throws that model out. In GA4, everything is an event: a page view is an event, a scroll is an event, a click is an event, a purchase is an event. There is no privileged "session" wrapper — a session is just a derived grouping of events that share a session_id. Each event can carry up to 25 custom parameters, which is why GA4 is far more flexible but also far easier to misconfigure.

Practical consequence: in UA you asked "how many sessions did this page get?" In GA4 you ask "how many page_view events had this page_location?" Same question, different mental model — and reports that don't translate one-to-one.

GA4 vs UA: side-by-side comparison

For a field-by-field translation of every metric, see the GA4 vs UA metrics translation table.

How the metrics changed

The most confusing part of the move isn't missing features — it's that familiar words now mean different things.

Users

UA headlined Total Users. GA4 headlines Active Users (users with an engaged session). Because the definition is stricter, the same audience produces a smaller number in GA4.

Bounce rate → engagement rate

UA's bounce rate counted single-page sessions with no interaction. GA4 removed it for a long time, then brought a redefined version back — but the metric you should actually read is engagement rate, which is roughly its inverse and based on a "10-seconds-or-an-event-or-two-pageviews" engagement threshold. A 60% UA bounce rate does not equal a 40% GA4 engagement rate. We break this down in GA4 bounce rate vs engagement rate explained.

Sessions

GA4 sessions are modeled (estimated) and don't restart at midnight or on a new campaign source, so they trend lower than UA sessions — typically 10% fewer, though anywhere from −10% to +20% is considered normal variance.

Why your numbers look lower

This is the single most common question after migration, and it's almost never a tracking bug. GA4 reports fewer users and sessions than UA for structural reasons:

• Active vs Total Users — different, stricter definition.
• Cross-device unification — UA counted the same person on phone + laptop as two users; GA4 often resolves them to one.
• Session reset rules — UA spun up a new session at midnight and on each new campaign source; GA4 doesn't.
• Modeled data & thresholding — GA4 estimates and withholds small-cohort data for privacy.

If the drop is larger than ~20%, then it's worth investigating. Read the full breakdown in Why GA4 shows fewer users and sessions than Universal Analytics, and if a big slice of your traffic landed in "(unassigned)," see how to fix unassigned traffic.

Account structure & conversions

Views are gone. UA's three-tier Account → Property → View hierarchy let you maintain a raw view, a test view, and a filtered "master" view. GA4 has no Views. Instead you use data streams (one per web/app source), plus data filters, audiences, and report comparisons to slice data. Teams that relied on multiple Views for governance need to rethink that workflow.

Goals became Key Events. UA "Goals" (destination, duration, pages/session, event) are replaced by marking any event as a Key Event (formerly "Conversion"). There's no 20-goal limit, and the setup is event-first. Mapping your old goals correctly is the highest-risk step of a migration — covered in UA goals to GA4 key events & conversions.

What happened to your UA data

GA4 did not import your UA history. The two systems have different schemas and definitions, so there was never a clean migration path for historical data. After July 2024 deletions, UA data only survives if someone exported it — via the UA interface, the Reporting API, or a BigQuery 360 export — before the property was removed. If you need year-over-year comparisons, check what you preserved now: exporting and preserving Universal Analytics data.

Migration checklist

If you're finishing (or auditing) a migration, work through this:

1. Confirm the GA4 tag fires on every page via GTM or gtag, with no duplicate configuration tags.
2. Re-create your conversions as Key Events — don't assume enhanced measurement covers them.
3. Rebuild custom dimensions/metrics as event or user-scoped parameters and register them.
4. Set data retention to 14 months (the max on the free tier) so Explorations have history. See GA4 data retention.
5. Connect Google Ads & Search Console, and re-import conversions on the Ads side.
6. Turn on the free BigQuery export for raw, unsampled, owned data.
7. Validate against expectations — purchases, lead forms, and key pages should report plausible volumes.
8. Run a full audit before you trust the data for decisions. Use our 25-point GA4 audit checklist or step-by-step audit walkthrough.

Not sure your setup is clean? Snifflytics runs an automated GA4 audit that flags misconfigured key events, missing parameters, retention problems, and attribution leaks in minutes.

FAQ

Is Universal Analytics still available in 2026?

No. UA stopped collecting data on July 1, 2023, and Google began deleting UA properties in July 2024. There is no way to resume UA tracking.

Why does GA4 show fewer users than Universal Analytics?

Mainly because GA4 reports Active Users (not Total Users), unifies the same person across devices, and counts sessions differently. A drop of around 10% is normal; see the full explanation.

Did GA4 import my old Universal Analytics data?

No. Historical UA data was never migrated into GA4. It only survives if you exported it before the property was deleted.

What replaced bounce rate in GA4?

Engagement rate is the metric to use. GA4 later re-added a redefined bounce rate, but it's simply the inverse of engagement rate and isn't comparable to UA's bounce rate.

What replaced UA Goals in GA4?

Key Events (previously called Conversions). You mark any event as a Key Event rather than configuring goal types.