Bounce Rate
A high bounce rate doesn't always mean the page isn't working. Sometimes it means it answered the visitor's question so well they didn't need to click elsewhere.
“An 80% bounce rate on an article that exhaustively answers a specific question isn't a failure. It's the expected behavior of someone who found what they were looking for.”
Otto GTM ObservatoryBounce rate needs context to mean anything
Bounce rate measures the percentage of sessions ending after a single page view, but this behavior has opposite meanings depending on context: on a blog article fully answering a specific question, a high bounce paired with long reading time is a positive signal. On the same metric applied to a landing page designed to drive conversion, a high bounce almost always signals a problem with messaging, targeting, or user experience.
Reading bounce rate without segmenting by page type and traffic source leads to wrong interpretations in both directions: false alarms about content that's actually working well, and false reassurance about conversion pages that are actually losing qualified visitors.
Anti-patterns
Common mistake: treating a high bounce rate as a universal problem to solve, without distinguishing between informational pages (where it can be normal) and conversion pages (where it's a critical signal).
Second mistake: comparing the site's aggregate bounce rate over time without considering changes in traffic mix between sources, attributing to a content problem what's actually a change in the composition of the audience arriving at the site.
Practical Application
A company notices a 76% bounce rate on its pricing page and mistakenly interprets it as positive, by analogy with blog articles. Analyzing average time on page (11 seconds, versus 3 minutes for articles) it discovers visitors leave quickly without reading: after a pricing clarity revision, bounce rate drops to 41% and demo requests from the page rise 65%.