What is bounce rate?

What counts as a bounce

A bounce is a session with exactly one pageview and nothing after it: no second page, no click that was tracked, no form submit, no custom event. Bounce rate is those sessions divided by all sessions in the period you are looking at.

The idea is simple; the details are not. Whether a visit "did nothing" depends on what your analytics tool records. A tool that only records pageviews will call every single-page visit a bounce, including someone who read a 3,000-word article to the end. A tool that also records clicks and scroll will not.

A worked example

Say a landing page had 1,000 sessions last week:

• 420 sessions went on to the pricing page.
• 180 stayed on the page but clicked the "watch demo" button, which is tracked as an event.
• 400 loaded the page and left.

Bounce rate = 400 ÷ 1,000 = 40%. If the demo button were not tracked, those 180 sessions would also count as bounces and the rate would jump to 58% — same visitors, same behaviour, different number. This is why you should only compare bounce rates measured with the same setup.

Why tools report different bounce rates

There are two common definitions in use today:

• Single-page, no-interaction sessions. The classic definition, used by Universal Analytics and by most privacy-focused tools. Anything the tool records as an interaction rescues the session from being a bounce.
• The inverse of engagement. Google Analytics 4 defines bounce rate as the share of sessions that were not engaged. A session is engaged if it lasted longer than 10 seconds, had a key event, or had two or more page or screen views. So a visitor who reads one page for 40 seconds and leaves is a bounce in the classic sense but not in GA4.

On top of the definition, three things move the number: which interactions are tracked as events, how sessions are split (a session that is cut in two creates an extra single-page session), and whether bot traffic is filtered. Bots that load one page and leave inflate bounce rate noticeably on small sites.

How TapCub counts bounce rate

TapCub uses the classic definition: a session is a bounce when it has one pageview and no events. Clicks, form submits and page changes recorded by autocapture count as events, so a visitor who clicks a button or opens an accordion is not a bounce, even if they never load a second page.

TapCub also marks a session as engaged when it lasts at least 10 seconds, has two or more pageviews, or has at least one key event. That matches GA4's engagement rule, so if you are migrating you can compare engaged sessions directly and treat bounce rate as a separate, stricter number. Bounce rate appears in the pages report and in every source and channel table, next to sessions, pageviews and duration.

How to read it

Bounce rate is only meaningful against the page's job:

• Blog posts and help articles often have high bounce rates because the reader got the answer. Look at time on page or scroll events before deciding the page failed.
• Landing pages and product pages exist to send people somewhere else. A high bounce rate there usually means the page does not match the promise of the ad, email or search result that brought people in.
• By source, not in total. A site-wide bounce rate mixes every audience together. Split by channel or campaign: if paid social bounces at 80% and organic search at 35% on the same page, the problem is the ad, not the page.

If you are moving off GA4 and want to compare bounce or engagement numbers between the two tools, see how to migrate from Google Analytics 4.

Common mistakes

• Comparing tools. GA4's bounce rate and a classic bounce rate are different metrics with the same name. Compare trends within one tool.
• Chasing a site-wide target. There is no single good bounce rate; it depends on page type and traffic source.
• Lowering it with fake events. Firing an event on page load makes bounce rate drop to near zero and tells you nothing. Only track interactions a person chose to make.
• Ignoring bots. On low-traffic pages a burst of crawler visits can double the rate overnight. Check whether bot filtering is on before investigating.

Related guides: Bounce Rate vs Exit Rate and Analytics Metrics. See also what is a good bounce rate and how to reduce bounce rate.