What is bounce rate in GA4?

Google defines an engaged session as one that lasts longer than 10 seconds, has a key event, or has two or more page or screen views. These are alternative criteria; a session does not need to meet all of them. Google’s engagement and bounce definitions.

For a web data stream, the engaged-session timer can be changed in the tag settings. Record the configured timer when comparing reports; the documented 10-second criterion is not a universal setting for every property. Session and timer settings.

Read the metric against the page’s job

A visitor who finds your opening hours and leaves may have completed their task. A visitor who cannot submit a booking form may have failed. The rate alone does not distinguish those situations. Look at the intended action and the page experience before deciding what needs changing.

The GA4 bounce-rate formula

Bounce rate = non-engaged sessions ÷ total sessions × 100. Equivalently, subtract engagement rate, expressed as a percentage, from 100%. Use the same date range, filters and session population for both metrics.

Compare like page types and traffic

Build comparisons around the task: a contact page, a product page and a definition page need different next steps. Then separate traffic sources and devices rather than assigning every page an industry target. A changed timer or key-event configuration also changes what the rate means.

A hypothetical session report

Assume a report contains 200 sessions, of which 160 are engaged and 40 are not. Bounce rate is (40 ÷ 200) × 100 = 20%; engagement rate is 80%. These are illustrative inputs, not a benchmark or a customer result.

An exit describes leaving the site; it does not, by itself, classify a session as engaged or non-engaged. Pogo-sticking describes returning to search results and selecting another result. Neither an exit nor a high bounce rate tells you which search result a visitor selected next.

Investigate before trying to lower the rate

Google recommends checking whether low engagement is concentrated in particular channels, pages or screens, and checking the measurement code when the problem is widespread. Engagement troubleshooting.

Open the relevant page on mobile, try the intended action and compare the content with the referring message. Fix the observed problem, then monitor the same segment with the same measurement settings.

Measurement changes can look like improvements

Do not add a meaningless key event just to reduce bounce rate. Check unexpected changes for duplicate page views or altered key-event settings. Keep historical reports labeled by analytics version; the older single-page explanation is not the GA4 definition.

What to ask from an engagement audit

Ask for the affected landing page, traffic segment, configured timer and event checks behind each finding. A useful recommendation identifies an observed obstacle, such as an unusable form, rather than promising a ranking or revenue lift from lowering a percentage.

Bounce-rate questions

Can a single-page session avoid being a bounce?

Yes. In GA4, a single-page session can be engaged through its duration or a qualifying key event, even without another page view.

Does a lower bounce rate prove better Google rankings?

No. Google describes aggregated and anonymized interaction data in its relevance systems, but that description does not establish GA4 bounce rate as a ranking metric. A change in your analytics report is not evidence of a ranking effect.

Engagement rate · GA4 · Conversion rate

Sources