Reconsidering bounce rate in web analytics

Abstract

In web analytics, ‘bounce rate’ refers to the percentage of sessions that have only one pageview. Bounce sessions are widely considered to be failures by marketers and web analysts. This article reconsiders bounces by looking at single pageview sessions using a different tool, Hotjar, to make virtual video recordings of such sessions. The results show that while some bounces are indeed failures, many are meaningful, including long visits that simply happen not to include any additional pageviews. The main reason for misunderstanding bounces as failures is that, in analytics, there is no marker to denote the end of the session. As a result, long single pageview sessions are improperly measured and look the same as short sessions through the analytics dashboard. Additionally, single pageview sessions are on the rise due to the increase in mobile browsing, as well as the ability of search engines to put the right content in front of searchers, lessening the need to request another page. Marketers are encouraged to use tools other than clickstream analytics to better understand bounce sessions, and to make the first pageview count for the visitor.