A bounce, in Google Analytics terminology, occurs when a visitor arrives on a page and leaves without triggering any additional interaction. In Universal Analytics, this meant leaving without visiting another page. In GA4, the metric has been refined to engaged sessions: a session is considered engaged if the visitor stays longer than ten seconds, views at least two pages, or completes a conversion event. The change in measurement approach means that bounce rates in GA4 are not directly comparable to those in Universal Analytics.
High bounce rate on a contact page or a checkout page is a clear problem: visitors intended to take an action and left without doing so. High bounce rate on a blog post is ambiguous: the visitor may have read the entire post, gotten exactly what they needed, and left satisfied. That outcome is a success from the reader's perspective even though it registers as a bounce. High bounce rate on a pricing page may indicate that prices are higher than expected or that the pricing presentation is unclear. The metric only becomes actionable when you connect it to the intent behind the visit.
Traffic source affects expected bounce rate significantly. Visitors from paid search ads who were sent to a relevant landing page have high commercial intent and should convert or bounce cleanly. Visitors from social media arrive with lower intent and higher curiosity, producing bounce rates that look worse but may represent appropriate behavior for that traffic type. Visitors from email campaigns are highly engaged and typically produce the lowest bounce rates of any traffic source. Comparing bounce rates across traffic sources reveals which channels are sending the most relevant visitors to your site.
Page load time is the most reliable predictor of bounce rate outside of content relevance. Research from Google shows that pages loading in one to three seconds have average bounce rates of thirty-two to forty percent. Pages loading in five seconds have bounce rates above ninety percent. Every second of additional load time increases the probability of a visitor leaving before the page is fully loaded. If your site has a high bounce rate across all pages and all traffic sources, page speed is the first variable to investigate.
A realistic benchmark for bounce rate depends on your industry and content type. News and media sites typically see bounce rates of sixty-five to ninety percent because most visitors read one article and leave. Service business pages see rates of forty to sixty percent. Landing pages optimized for conversion see rates of thirty to fifty percent. E-commerce sites see rates of twenty to forty-five percent. These benchmarks are general guides, not performance targets, because your specific audience, content quality, and visitor intent will determine what normal looks like for your site.
Bounce rate is a directional indicator worth monitoring but should never be evaluated in isolation from the actual behavior it represents, which requires looking at which pages are bouncing, where the traffic is coming from, and what the visitor was likely trying to accomplish when they arrived.