"Bounce Rate is high, that's a problem, right?" "What's the difference between Bounce Rate and Exit Rate?"

These are the two metrics most often confused on the analytics floor. On top of that, the move from Universal Analytics (UA) to Google Analytics 4 (GA4) redefined Bounce Rate completely and removed Exit Rate from the default reports.

This article walks through the difference between Bounce Rate and Exit Rate, contrasts the UA and GA4 definitions, and ends with the question every EC operator actually cares about: which one should we be looking at when revenue is on the line?

TL;DR

• Different sessions are counted: Exit Rate counts every session that included the page. Bounce Rate counts only sessions that started on that page
• GA4 redefined Bounce Rate: what used to mean "single-page session" in UA now means "session without engagement" (1 minus Engagement Rate) in GA4
• EC operators read 'exit quality' against revenue: a high Bounce Rate or Exit Rate is not automatically bad — combine it with RPS (Revenue Per Session) to decide whether to fix or to leave alone

1. Bounce Rate — The metric GA4 redefined

Bounce Rate means very different things in GA4 and Universal Analytics. The classic UA shortcut "percentage of single-page sessions" no longer holds in GA4.

Bounce Rate in UA (the old definition)

In UA, Bounce Rate was "the percentage of sessions that started with the page where there was only one pageview". The shorthand "people who left after seeing only one page" became the standard mental model.

Bounce Rate in GA4 (the new definition)

GA4 redefined Bounce Rate as "the percentage of sessions that were not engaged" (1 minus Engagement Rate).

A session counts as engaged when any one of the following is true.

• The session lasts longer than 10 seconds
• A Key Event (formerly Conversion) fires
• There are 2 or more page or screen views

A single-page visit that lasts more than 10 seconds is no longer a bounce. Conversely, a 2-page visit under 10 seconds with no Key Event can still be counted as a bounce.

The simple "high Bounce Rate equals bad" reading is even harder to defend in GA4. Whether a sub-10-second exit means "bounced" or "read through" depends entirely on the page's purpose.

2. Exit Rate — Where each page becomes the last stop

Exit Rate is the percentage of sessions including the page where that page was the session's last.

Exit Rate = Sessions that ended on the page / All sessions that included the page

The fundamental difference from Bounce Rate is the session pool. Bounce Rate's denominator is sessions that started on the page. Exit Rate's denominator is every session that passed through the page.

GA4 dropped Exit Rate from default reports

UA had Exit Rate as a default column. GA4 does not list Exit Rate as a default metric.

To see exit behavior in GA4, use one of these instead. Path Exploration visualizes where users came from and where they exited. Aggregating session_end events shows the page right before the session ended. Combining engagement time with exit volume highlights pages with short stay and high exit.

The very habit of looking at "Exit Rate" as a single metric is downstream of UA's session-based model. Under GA4's event-based model the framing has shifted.

3. Bounce Rate vs Exit Rate — At a glance

A side-by-side table makes the difference clearer.

A simpler analogy: Bounce Rate is the share of visitors who turned around at the front door (entry-page lens). Exit Rate is the share of visitors who walked out from each room (per-page lens).

4. From a revenue lens — Are these "bad" metrics?

Both metrics are often labeled "high equals bad", but in practice that shortcut breaks down quickly.

A single-page LP that completes purchase or inquiry on one screen will, by design, see almost everyone "leave after one page". A 90% Bounce Rate is healthy if CVR is good. Article media also: search visitors who read an article and return to results are technically "bounces" but completed the page's actual job.

On the other hand, an EC product detail page where buyers with intent are dropping off has clear improvement levers — button placement, stock indicator, shipping copy. Mid-funnel checkout steps showing a spike on one step usually point to form design issues.

In other words, where the bounce or exit happens, and in what context, is the actual signal. A standalone rate number rarely points to an action.

5. Read "exit quality" with RPS

Both metrics ultimately need to be judged against revenue. Use RPS (Revenue Per Session) — Revenue / Sessions — to read "exit quality".

• High exit, low RPS → fix priority ★★★ (revenue leakage is largest)
• High exit, high RPS → fix priority ★ (post-conversion natural exit)
• Low exit, low RPS → traffic flows but revenue does not (rethink CTA)
• Low exit, high RPS → keep as is and replicate elsewhere

Plotting Exit Rate × RPS as a 2 × 2 matrix makes it instantly clear which page deserves the next hour of work.

Question for you

How often do you check Bounce Rate or Exit Rate in isolation versus alongside revenue? In your team, has the GA4 redefinition of Bounce Rate changed how you read those reports — or is the old "single-page session" mental model still in play?

For the full guide with all charts and the revenue-side breakdown:
Bounce Rate vs Exit Rate — The GA4 Basics You Were Afraid to Ask