‍

Why did the definition change?

Because the old definition punished pages that worked.

Under Universal Analytics, a visitor landing on an article, reading it for six minutes, getting the answer, and leaving satisfied counted identically to someone who landed and left in two seconds. Both were bounces. The metric could not distinguish success from failure, which made it close to useless on content pages.

GA4 inverted it. Engagement rate became the primary metric, and bounce rate became its complement: the share of sessions that were not engaged. A session qualifies as engaged if it lasts longer than 10 seconds, records a key event (what GA4 called a conversion until Google renamed it in 2024), or includes a second pageview. So the six-minute reader is engaged, and the two-second visitor is a bounce.

The 10 second threshold is a default, adjustable up to 60 seconds in the data stream's tag settings. Raising it moves the bounce rate of every page on the site at once, which is worth knowing before someone reports a site-wide regression that was a settings change.

GA4's engaged set is a strict subset of what Universal Analytics called a non-bounce, so GA4 bounce rates come in lower for identical traffic. Figures from either side of a migration cannot be joined, and any trend line spanning that boundary is an artifact.

When is a high bounce rate fine?

More often than the metric's reputation suggests.

• A page that answers the question completely. Someone searching for a definition, finding it, and leaving is a success. This is what a glossary entry is supposed to do.
• Contact and location pages. The visitor got the phone number.
• Pages where the action is off-site. A page whose purpose is a click to an app store or a booking tool.
• Single-page campaign landers where the conversion is a form on that page, if the conversion event is firing correctly.

The question is never whether bounce rate is high. It is whether the page's purpose was served before the visitor left.

When does bounce rate actually matter?

Three situations where it is genuinely diagnostic.

Sudden change. A page whose bounce rate jumps after a release usually has something broken. This is the most reliable use of the metric.

Mismatch between promise and page. High bounce on paid traffic points at message match between the ad and the landing page.

Segment divergence. Mobile bounce far above desktop on the same page usually means a layout, speed, or tap-target problem.

What it does not support is cross-site comparison, ranking pages against each other regardless of purpose, or use as a KPI. Any metric improvable by adding a pointless second click is a poor optimization target.

Why does Webflow inflate bounce rate?

Webflow's in-place form success state produces no second pageview, so a visitor who filled in the form, submitted it, and became a lead can still be counted as a bounce if no conversion event fires on success. Bounce rate on a Webflow landing page is therefore untrustworthy until conversion tracking has been confirmed working on the published site. Event tracking covers the two ways to wire it.

Related terms

Exit rate · Time on page · Scroll depth · Message match · Landing page conversion rate

Service: Conversion Rate Optimization.

FAQ

What is a good bounce rate?

There is no target figure, since the number depends on page type and on which definition the tool applies. The comparison that works is a page against itself before and after a change, with the traffic source held constant.

Is bounce rate a ranking factor?

No. Google has stated that it does not use Google Analytics data as a ranking signal, and bounce rate is an analytics construct rather than something a crawler observes. The behaviors that produce high bounce, such as slow loading and weak relevance, do relate to signals engines measure directly.

What is the difference between bounce rate and exit rate?

Bounce rate counts only sessions that started on the page. Exit rate counts every view that ended a session, whatever route brought the visitor there. Every bounce is an exit, and most exits are not bounces.

Does GA4 still report bounce rate?

Yes, as the inverse of engagement rate rather than as a headline metric. Expect lower numbers than Universal Analytics reported for the same traffic, since a single-page visit that lasts long enough now counts as engaged.