Bounce Rate
Bounce rate is the percentage of visitors who land on a page and leave without taking any further action or visiting another page. It is a behavior metric that hints at whether a page satisfied the people who arrived, though it needs careful interpretation.
Bounce rate is the percentage of visitors who arrive on a page and then leave without doing anything else, no second page view, no further interaction that the analytics tool counts. If 100 people land on your article and 70 of them read it and then close the tab without clicking anywhere, you have a 70 percent bounce rate. People love to treat bounce rate as a simple grade where lower is always better, and that is where most of the trouble starts. A bounce is not automatically a failure. Sometimes a high bounce rate is exactly what a successful page should produce, and reading it without context is one of the most common analytics mistakes I see.
A high bounce rate is not a verdict. It is a question. The question is: did this visitor get what they came for and leave happy, or get nothing and flee? The number alone cannot tell you which.
When a bounce is good and when it is bad
The single most important thing to understand about bounce rate is that its meaning flips depending on the page's job. The same 80 percent bounce rate can be a triumph on one page and a disaster on another. What separates them is intent. You have to ask what you actually wanted the visitor to do, and only then can the bounce rate tell you anything useful.
- Good bounce: someone searches "what time zone is Denver," lands on your page, gets the answer in two seconds, and leaves satisfied. They bounced, and they got exactly what they wanted. Mission accomplished.
- Bad bounce: someone lands on a product page, sees a slow, confusing layout, and leaves immediately without browsing or buying. That bounce is a lost customer and a real problem.
- Ambiguous bounce: someone reads your full 2,000-word guide top to bottom and then leaves. Older tools may count that as a bounce even though the visit was a complete success.
targetA note on how bounce rate is measured today
How bounce gets counted has shifted over the years. Classic analytics counted any single-page session as a bounce, no matter how long or engaged. Newer measurement leans on engagement instead, looking at whether the visitor stayed, scrolled, or interacted before leaving. Whichever definition your tool uses, the rule holds: never read the raw bounce number without knowing what the page was supposed to achieve and how the metric was actually counted.
Example
Picture a recipe site. The page for "how many tablespoons in a cup" might have a 90 percent bounce rate, and that is perfect, because people want one number and they got it. Meanwhile the homepage having a 90 percent bounce rate would be alarming, because the homepage exists to send people deeper into the site. Same metric, opposite meanings. This is why I always tie bounce rate back to the specific job of the page before I draw any conclusion, and why I weigh it against signals like on-page content quality rather than reading it in isolation.
Context is everything
Bounce rate without context is noise. Define what success looks like for each page first, then interpret the bounce rate against that goal. An informational page and a checkout page should never be judged by the same bounce standard.
If you do find a page with a bounce rate that genuinely worries you, the fixes are usually about meeting expectations and removing friction. Make sure the page delivers what the search result promised, load it fast, make it easy to read on a phone, and give people a clear next step if you want them to take one. A bounce often means a broken promise between the search listing and the page, and closing that gap is some of the highest-leverage work you can do.
lightbulbPRO TIP
Do not chase a lower bounce rate as a goal in itself. Chase a satisfied visitor. If you obsess over the number, you risk adding pointless extra clicks just to game it, which makes the actual experience worse. Optimize the experience and let the metric follow.
RELATED TERMS
Want this handled by someone who has measured search for 20 years?
Work with me