Bounce rate in GA4: what it means now, and when to worry
GA4 defines bounce rate as the share of sessions that were not engaged. The rule, the formula, how it differs from the old one, and when a high rate matters.
On this page
Bounce rate used to mean "people who saw one page and left". In GA4 it means something else: the share of sessions that were not engaged. Same name, different rule, and the numbers are not comparable with the old ones.
The GA4 definition
GA4 calls a session engaged if any one of these is true (Google):
- it lasted longer than 10 seconds,
- it had a key event,
- it had 2 or more page or screen views.
Then:
- Engagement rate = engaged sessions ÷ sessions.
- Bounce rate = sessions that were not engaged ÷ sessions, which is 100 percent minus engagement rate.
So a visitor who lands on a blog post, reads for two minutes and leaves is not a bounce in GA4. In Universal Analytics they were, because they saw one page.
The 10 second threshold is the default. You can change it per web data stream, under Configure tag settings, Adjust session timeout, Adjust timer for engaged sessions (Google). Change it and every bounce rate after that date is on a new scale, so note the date.
How it differs from Universal Analytics
Universal Analytics counted a bounce as a session with a single interaction, usually one pageview. Reading time did not matter. A long, useful article with no second click had a bounce rate near 100 percent.
| Universal Analytics | GA4 | |
|---|---|---|
| A bounce is | A single-interaction session | A session that is not engaged |
| Two minutes on one page | Bounce | Not a bounce |
| Five seconds on one page | Bounce | Bounce |
| One page, then a form sent | Bounce, unless the form sent an interaction event | Not a bounce, if the form is a key event |
Universal Analytics stopped processing data for standard properties on July 1, 2023, so any comparison across that date compares two different metrics.
How other tools count it
There is no standard. Each tool picks its own rule:
- Many privacy-focused tools count a bounce as a visit with a single pageview, like Universal Analytics did.
- Sporetag counts a bounce as a visit with one pageview and less than 10 seconds of time on the page, so a one-page visit that was read is not a bounce.
A bounce rate only means something next to its definition. Never compare one tool's bounce rate with another's.
When a high bounce rate is a problem
A high bounce rate is not bad in itself. Look at it per page and per source, not for the whole site.
Usually fine:
- A contact page where people copy the phone number and call.
- A blog post that answers the question in the first screen.
- A page people land on from a search for your opening hours.
Worth fixing:
- A paid landing page with a high bounce rate from one ad but not others: the ad promises something the page does not show.
- A product page that bounces most on phones: likely slow, or the key information sits below the fold.
- A sudden jump on one day: often a tracking change, such as a broken tag on the second page, a consent banner change, or a new source of bot traffic.
Why bounce rate suddenly changes
- A second copy of the tag. Two tags send two pageviews per page, so almost no session looks like a bounce. A bounce rate falling to near zero overnight is the classic sign.
- The tag missing on inner pages. Visitors who go deeper look like they left.
- A consent banner. If the banner holds the tag until consent, visitors who leave before answering are not counted at all, and the remaining sessions skew toward engaged ones.
- Bot traffic. Bots often load one page and leave in under a second. A wave of them raises bounce rate.
- A change to the engaged session timer.
Ways to lower bounce rate that actually help
- Match the page to the source. A visitor from an ad about a discount should see the discount at once.
- Load fast on mobile data. A page that is blank for several seconds loses people before your tag even counts them.
- Put the answer, price or next step in the first screen.
- Give one clear next step, not ten.
Do not lower it with tricks such as firing fake events on scroll. The number improves and nothing else does.
Common questions
What is a good bounce rate in GA4?
It depends on the page and the source. A landing page from a paid ad and a blog post from search will never look alike. Compare a page with itself over time, and one source with another for the same page.
Why is my GA4 bounce rate so much lower than in Universal Analytics?
GA4 does not count a session as a bounce if it lasted more than 10 seconds, had a key event, or had two or more pages. Many sessions that bounced in Universal Analytics are engaged in GA4.
Where do I find bounce rate in GA4?
It is not in every default report. Add it to a report with Customize report, Metrics, Add metric, Bounce rate, or use it in Explore.
Is bounce rate the same as exit rate?
No. Exit rate is the share of pageviews of a page that were the last in their session, whatever happened before. Bounce rate is about whole sessions.
Sources
- Google Analytics Help: Engagement rate and bounce rate. Checked October 2026.
- Tag Manager Help: Configure your Google tag settings. Checked October 2026.
- Sporetag's definition is in its own code: a visit with one pageview and under 10 seconds of time on the page.
Spotted something wrong or out of date? Email hello@sporetag.com and we will fix it.