After migrating from Universal Analytics to GA4, almost every team hits the same wall: the numbers don't match. Sessions are lower. Conversions look different. Bounce rate has disappeared and been replaced by "engagement rate" which seems to show the opposite picture. This is not a bug — it's the result of fundamental architectural differences between the two platforms. Here's exactly what's different and how to make sense of it.
Sessions Are Defined Differently
In UA, a session resets at midnight, after 30 minutes of inactivity, or when campaign parameters change. In GA4, midnight resets are gone — a session can span midnight without resetting. Campaign parameter changes also no longer trigger a new session automatically. The result: GA4 typically reports 10–20% fewer sessions than UA for the same traffic, even when nothing is wrong.
This is expected and correct. Fewer sessions doesn't mean less traffic — it means GA4 is counting sessions more accurately by not artificially splitting them.
Bounce Rate Is Gone — By Design
UA's bounce rate measured sessions with only one pageview. GA4 replaced this with Engagement Rate — the percentage of sessions that lasted longer than 10 seconds, had a conversion event, or had 2+ pageviews. A GA4 "engaged session" is a more meaningful signal than the inverse of UA's bounce rate.
Typical result: GA4 engagement rates of 50–70% often correspond to UA bounce rates of 40–60%. They're measuring related but different things. If you need bounce rate in GA4, you can create a custom metric: sessions minus engaged sessions, divided by sessions.
Conversions Count Differently
UA counted Goals — binary per session (a goal could only fire once per session). GA4 counts conversion events, and each event can fire multiple times in a session. A user who completes two purchases in one session will show 1 Goal completion in UA but 2 conversion events in GA4. For e-commerce properties, this is almost always the more accurate count.
The Underlying Data Model
UA was session-scoped by default. GA4 is event-scoped. Every interaction is an event — pageviews, scrolls, clicks, purchases. This means the same user behaviour produces a different count of "hits" and aggregates differently across reports. Direct metric comparisons between UA and GA4 are almost never valid — what matters is whether the trends in GA4 align with what you know about your business.
Checking GA4 Data Quality
Before assuming discrepancies are architectural, rule out actual data quality issues. Common problems that cause GA4 numbers to look wrong include:
- Duplicate tracking tags firing on the same page (inflates event counts)
- GA4 tag not deployed on all pages that had UA coverage
- Conversion events misconfigured or missing required parameters
- Internal traffic not filtered, polluting session and conversion data
- Data retention left at 2 months, causing historical gaps in comparisons
GA4 Health Check's automated audit checks for all of these across 7 modules and flags any discrepancy between expected and actual event behaviour. Run a 60-second audit to confirm your GA4 data can be trusted on its own terms — get started here.
