What is A/B testing?
Learn what A/B testing means, how a controlled marketing experiment works and which mistakes make test results unreliable.
Marketing glossary
Bounce rate can flag a weak page or poor traffic, but only after you understand what your analytics system calls a bounce. Without that definition, it becomes an impressively precise percentage attached to an unclear event.

Historically, many tools described a bounce as a single-page session with no further tracked interaction. Modern platforms may calculate it differently. In Google Analytics 4, for example, bounce rate is the inverse of engaged-session rate, so its logic includes time, page views and key events.
This means you should not compare bounce rates between analytics products, properties or periods without confirming that tracking and definitions stayed consistent. Consent choices, blocked scripts, single-page applications and event configuration can all affect the number.
A solicitor publishes a page answering “How long does probate take in the UK?” Many visitors read the answer and leave without clicking. A traditional analytics setup may count those sessions as bounces, even though the page fulfilled its informational purpose and improved recognition of the firm.
By contrast, a visitor clicking a paid advert for a probate consultation and leaving the landing page immediately may indicate a mismatch, slow page, weak proof or accidental click. The same observed behaviour carries different meaning because the search intent and page objective differ.
Bounce rate is most useful as a diagnostic signal segmented by page, source, campaign, device and audience. A sudden change can reveal:
Combine it with engagement time, scroll or interaction data, conversion rate, qualitative feedback and the business outcome assigned to the page.
The most common mistake is setting an arbitrary goal to reduce bounce rate across the whole site. You could achieve that by triggering a meaningless event after one second, improving the dashboard without improving the experience. Measurement should describe behaviour, not flatter it.
Another mistake is assuming a high rate proves poor content. Reference pages, contact-detail pages and short answers may naturally end a session. Equally, a low rate can hide frustration if people click around because they cannot find what they need.
Before changing a page, verify the tracking, compare like traffic sources, inspect the experience on real devices and decide what useful next action exists. If no additional action is necessary, completion may matter more than another click.
No. It may mean the page answered a question completely. Judge it against the page’s purpose and other evidence, particularly conversions, engagement, source quality and whether visitors had a useful next step.
Platforms can define sessions, engagement and bounces differently. Tracking settings, consent handling and event configuration may also change. Document the measurement method before treating two periods as directly comparable.
Who wrote this
Read next
Learn what A/B testing means, how a controlled marketing experiment works and which mistakes make test results unreliable.
Learn what answer engine optimisation means and how to make useful business content easier for search and AI systems to extract and cite.
Learn what brand positioning means, how to define it and how an SME can occupy a useful, distinctive place in customers' minds.
A short discovery call is usually enough to tell. No deck, no pitch.
Book a discovery call