Bounce rate is the share of website sessions where a visitor leaves without doing much of anything, a number analytics platforms have tracked in one form or another since the earliest days of web reporting. A high figure on a given page usually signals that whoever landed there didn’t find what they expected, though the metric alone rarely explains why.
The definition shifted meaningfully when Google Analytics 4 (GA4) replaced Universal Analytics. Universal Analytics counted any single-page session as a bounce, full stop, regardless of how long that one page held someone’s attention.
How GA4 Calculates Bounce Rate
GA4 flips the definition around. It measures engaged sessions first and treats bounce rate as whatever’s left over. A session counts as engaged if it runs 10 seconds or longer or includes 2 or more page views, and it counts as engaged the instant it triggers a conversion event too. Bounce rate is simply the percentage of sessions that clear none of those 3 conditions.
That distinction changes what the number actually tells a site owner. A visitor who spends 45 seconds reading a single blog post and then leaves would have counted as a bounce under the old system. Under GA4, that same session reads as engaged, since it comfortably clears the 10-second threshold. The two versions of the metric aren’t directly comparable, which is why a bounce rate that appears to jump or crash right after a GA4 migration is usually a reporting artefact rather than a real behaviour change.
Bounce Rate and SEO
Bounce rate isn’t a confirmed Google ranking factor on its own, and Google representatives have said as much on record more than once. Its SEO relevance is indirect. A page that consistently fails to hold visitor attention is very often a page with a genuine content or intent-match problem, the same underlying issue that also tends to hold rankings back over time.
Reading the number in isolation rarely helps. A high bounce rate on a page built to answer one specific question in a single glance, a store’s opening hours, say, can be a sign the page did its job efficiently rather than a warning sign. Pairing bounce rate with time on page, scroll depth, exit rate, and conversion data gives a much clearer read on whether a page is actually falling short.