A bounce rate is the percentage of website or app visitors who view one page or one piece of content and then leave the site.
A website’s bounce rate is generally calculated by dividing single-page visits by entries.
Bounce rate is most beneficial when used as a starting point to evaluate how well a company’s site or app is connecting with its audiences.
Companies should view bounce rate alongside other information and data, and within the context of what success looks like for their business.
Despite some drawbacks, bounce rate still offers value, especially when used alongside customer segmentation and other metrics.
Ben Gaines is a director of product management for Adobe Analytics. In this role, he works closely with Adobe customers to understand their data and insight needs and advise them. He also guides Adobe product strategy and mentors a team of product managers and technical writers. Prior to joining the product team at Adobe, he has been a manager of digital analytics at ESPN, a community manager and technical writer at Adobe, and a technical support engineer at Omniture, prior to its acquisition by Adobe. Along with Blair Reeves, he is the co-author of Building Products for the Enterprise: Product Management in Enterprise Software, available now through O'Reilly Media.
Q: How do you calculate bounce rate?
A: Every digital analytics platform and product analytics solution has some form of built-in bounce rate metric, but the calculations to arrive at that metric vary slightly from solution to solution. In general, bounce rate is calculated as visits to a single page only — also known as bounces — divided by entries, or the total number of people who have come to the website through that landing page. But what counts as a single-page visit differs between sites, companies, and solutions.
One of the great debates of digital analytics history is whether a user clicking or interacting with something on a landing page still constitutes a bounce even if the user never actually makes it to the next page. Let's say there's a form on a landing page that the user starts filling out. But they don't finish the form or submit it. Is that a bounce? They clearly were engaged with it enough to interact with the form and provide information, but they never actually got to the next page.
A video also presents complications for calculating bounce rate. If there’s a video on the landing page and the visitor watches it before leaving, does that constitute a bounce or not?
In Adobe Analytics, for example, a bounce is defined as a session that is made up of a single hit, or server call, from the website to Adobe Analytics. When the user clicks “play” on the video, another call gets sent to Adobe Analytics, so the visit is no longer considered a single-page session. However, other tools define the actions differently — such as because the user only interacted with a single page during their visit, it is considered a bounce. A company can choose which interpretation of bounce rate most closely matches their definition of success.
Q: How can companies connect bounce rate to a larger strategy?
A: The significance of bounce rate needs to be carefully considered on a case-by-case basis. Where it provides the most value is as a tool for customer segmentation. An average bounce rate can be a good jumping off point to understand how effective a landing page is at reaching an intended target audience.
The best way to use bounce rate is by breaking down the site to see which pages appeal to which audiences. Then an organization can determine what content is and isn’t effective at engaging visitors and converting them to the next action. If it’s found that a certain page has a high bounce rate—meaning it doesn’t resonate with a specific audience, the organization can start to formulate a content or design strategy that will appeal to that audience segment.
Bounce rates can be used alongside methods like A/B testing to figure out why customers are leaving or staying, and to optimize sites to better serve customers.
Q: What is the difference between a bounce rate and an exit rate?
A: Bounce rate is specifically focused on landing pages or landing content. It measures the percentage of people who land on a landing page and then leave immediately. Exit rate, on the other hand, is not specific to a landing page. Someone could come to the site, browse 10 pages, and then leave. The last page before the visitor leaves the site or the app is the exit page.
When we talk about exit rate, we're discussing the percentage of all page views where the page was the last one viewed during a session. Bounce rate, however, can only be used to refer to a single page, a single piece of content, or a single screen view.
Companies tend to look more closely at bounce rate when making decisions, because landing pages are so important. No one ever stays on a website forever, so exits are unavoidable. However, a high bounce rate can be a signal of potential issues with the webpage or app. If a company is trying to understand the effectiveness of a landing page, bounce rate is the better place to start.
Both exit rate and bounce rate are only starting points for analysis. They're never the end point. A person can’t look at a bounce rate, whether low or high, and make an immediate judgement about the value and effectiveness of a page. They always have to dig deeper to find out why that percentage of people leave, and sometimes bounce rate can be misleading.
Q: What are some weaknesses of analyzing bounce rate?
A: Bounce rate can be an interesting starting point for analysis, but many analysts have used it as too much of a proxy for success. It can be tempting to look at a landing page or look at a higher bounce rate and make a judgment about the content or the design of that page based on this one number. But it's only one of five or six metrics that should be used, even for a surface level analysis of a piece of content, let alone a deeper dive.
One weakness of using bounce rate as an indicator for success is that it doesn’t offer a reason why people leave your site from the landing page, so you might draw the wrong conclusions. There are many reasons why someone might end up on a landing page. For example, an outdoor gear company might create a special landing page for an email campaign. The page focuses on rock climbing equipment and is targeted toward the portion of its audience that scale rocks, but the marketing team sends the campaign email to the entire audience.
Out of the company’s whole audience, only five percent of customers are interested in rock climbing, but other people click through the email as well because they like the brand or saw the discount and want to know more. When the segment of the audience who doesn’t care about rock climbing realizes the page is only offering or promoting climbing gear, they will bounce.
The bounce rate is high, because a lot of people are leaving the landing page rather than clicking through, but that doesn’t mean the landing page was ineffective. The email segmentation was unfocused. If the marketing team had only sent the email to the five percent of the audience interested in rock climbing, they would have seen a low bounce rate.
In addition, the bounce rate doesn’t matter as much as the conversion rate, or the actual revenue generated. If 40 percent of those who clicked through the landing page convert, and spend several hundred dollars each on equipment, that is overall more successful than another page that had a lower bounce rate, but also a lower conversion rate and less revenue. The actual value of the second landing page is much lower, despite having a lower bounce rate on the landing page.
Bounce rate is often taken at face value, without the necessary context. How the data is collected also affects the effectiveness of bounce rate. For instance, when companies build a landing page, they are rarely considering how having a video that autoplays after five seconds will ultimately affect the bounce rate. There's a lot happening behind the scenes that can lead to a misinterpretation of the metric, much more so than factors like revenue or conversion rate, which are pretty standard and rely on metrics that are universally defined across companies and vendors in the digital analytics space.
Q: How can companies avoid issues with bounce rate?
A: There's rarely ever one metric that rules them all, and that is true for bounce rate as well. The way to put bounce rate in perspective is to consider it in relation to how a company defines success.
For most industries, the goal of an app or website is not to have people click around and continue looking at different pages, so a good bounce rate doesn’t necessarily lead to success with business goals. There will usually be some conversion metric or lifetime value of a customer that a company is trying to optimize for.
If success for your company is related to revenue, you can look at bounce rate next to conversion rate to determine the effectiveness of your landing pages. Start with the metrics that matter most, and then view bounce rate in the context of those factors, instead of starting from bounce rate and possibly never determining if the landing page drove success.
Q: Does bounce rate take into account the amount of time spent on a site?
A: Generally speaking, bounce rate does not factor in time, but a company can customize their calculations if the time period for the session is something they are interested in. A media company, for example, might be interested in the length of time spent on a site, because it could indicate whether a consumer finished an entire article or watched an entire video before bouncing. That might be part of how that company defines success.
For most industries, however, the goal of a landing page for an e-commerce site is to drive the visitor to a downstream conversion event, so, if the visitor doesn’t proceed, it doesn’t matter how much time they spent on the landing page.
Q: How will the approach to bounce rate change in the future?
A: The move to single-page applications in many ways reduces the effectiveness of bounce rate because a visitor essentially stays on the same page even when they click through to the next experience. The user experience is so different on many sites now than it was even five years ago, let alone 10 years ago. Across the digital analytics industry, bounce rate is referred to much less frequently.
The industry understands now that user experiences are so different and user behavior is so nuanced, that bounce rate is a challenging metric to get actionable insights out of. As more and more behavior shifts to mobile devices, the way companies look at and approach bounce rate will change. People hop between apps in a way that keeps a session alive for a while, from an analytics perspective, even when the app is in only the background. Generally speaking, as user behavior and the way customers interact with a company changes, the usefulness of bounce rate decreases. But bounce rate can still offer value when used alongside customer segmentation.