A few months ago, in an article about on page SEO we briefly quoted the bounce rate, a.k.a the rebound frequency, calling this value “ambiguous” because a high rate does not necessarely mean that the site is not working, since it really depends on the nature of the project itself. So let’s go back on the topic and let’s try to deepen the meaning of bounce rate, what does it measure and why it is so important to always keep an eye out on these metrics.
What is the bounce rate
The expression bounce rate literally indicates the rebound frequency, meaning the percentage of users that leave the website after visiting only one page, bouncing back toward the search results page or the website that first originated the outbound link. It is a pivotal metric in order to analyze the site’s traffic and consequently have an idea of the users’ behaviour regarding the offered contents.
Bounce rate values
According to Wikipedia, “a lot of statistical systems establish bounce rate at 30 seconds“: if the user leaves the page within 30 seconds we can safely say he is uninterested in the displayed contents. Over time, this referring value has greatly dropped, and some commercial softwares fix it at 5 seconds tops!
As a matter of fact, there are no such things as fixed values to establish the ideal rebound frequency, even because the term itself is strongly relative and changes – as we were saying – based on the site’s type; generally, a low dropout rate means that users are appreciating the site’s organization of contents and graphic look, that convince him to keep explore other pages.
How to measure the rebound frequency
The bounce is then a session that only regards a single page of the site; in Google Analytics the rebound is measured “as a session that only activates a single request to the Analytics server”, as sometimes happen when the user opens a specific page and leaves the site without further request to the Analytics server.
The bounce rate in Google’s language measures the “correlation between the sessions of a single page divided by all of the others or the percentage of all the sessions on your website in which the users viewed only one page and activated only one request to the Analytics server”.
Clarifications to understand the bounce rate
W e should clarify at this point that the bounce rate does not measure how much time a user spends on a site’s page nor the user engagement: this is a factor that confuses a lot, but it is actually possible that a site is providing a very engaging quality page but, at the same time, has a high bounce rate, because of the very fact that this metric does not calculate the duration of the onsite sessions.
Therefore, to the previously provided definition we could add to be more specific: the bounce happens every time a user enters the website, reads a page and the system does not register any other useful signals, because the person clicks on the browser’s back arrow, closes the browser’s window itself, clicks on an external link or uses the browser’s search bar to move toward other sites and URLs.
The average bounce rate of the sites
Based on these concepts, the RocketFuel ‘s team took into account a small sample of sites to study the average bounce rate and to try and estimate the “good” values. According to this study the majority of websites presents a rebound frequency between 26 and 70%, and more specifically it is possible to see a sort of ranking system of bounce rates:
- 25% or less: something is probably broken (like a wrong Google Analytics installation, for example).
- 26-40%: excellent. It is an indicative value for a well built and professionally designed website, that matches the users’ need.
- 41-55%: average.
- 56-70%: higher than normal, but it could make sense depending on the type of website
- 70% or higher: very high value, that might point out site’s problems (but also be linked to the uniqueness of the pages).
Generally, it is believed that a very high bounce rate means that the majority of visitors that lands on the site are not truly interested to the offered (and placed in SERP) content, that they did not appreciated the design or yet again that they did not find what they were looking for. But also having a bounce rate equal or below 25% could be a problem: these values depends as previously stated on an error inside the analytics implementation, or on a site’s building mode that forces users to do at least one action before leaving (such as gateway or landing pages he has to go through before actually reaching the main site), thereby undermining the users experience itself.
On average, then, a high value of rebound frequency (correctly computed) is symptom of several deeper problems of the site, and more specifically of a not ideal user experience or poor targeting of the pages, from which could result consequential SEO difficulties.
How to figure out if a site’s bounce rate is too high
As previously stated, a lot depends on the site because individually taken and out of context metrics are not truly useful to a project or strategy.
For instance, the users of a site centered on events often only care about day, time and location of the event of their interest: once acquired the info (perhaps in the shortest possible amount of time), they just leave the page and site. So, those site’s bounce rate will inevitably be high, but it does not stand out as critical factor because it perfectly meets its real users’ needs.
In an ecommerce ‘s case, though, clients will probably linger a bit more and for this the ideal bounce rate should be a bit lower; an increasing rebound trend is a kind of clue that could alarm online shopping sites because points out that users are leaving the website very quickly and almost surely this will translate in less sellings (but one always have to analyze the actual conversions to be absolutely certain of it).
The average rebound frequency per site’s category
And so, another study – performed by Clicktale – tried to determine the average levels of bounce rate per site’s category, revealing that for each blog page the rebound frequency is generally established between 70 and 90%, for informational sites it drops down to between 40-60 % while for those sites providing services it should stay between 10-30 %. For an ecommerce, more specifically, the average bounce rate should be at 33,9%.
In conclusion, we go back to the quoted Analytics guide: Google clarifies that a high rebound frequency is not always a problem or straight “bad”, because closely depends on the analyzed site. If the success of the site “depends on the fact that the users displays more than one page, then yes, a high bounce rate is bad news”, they explains from Mountain View, because – for instance – if the “homepage is the access door to the rest of the site (i.e for new articles, products pages, checkout elaboration) and a large percentage of users only view your homepage, a high bounce rate is not a target to aim to”. On the other hand, “if you have a single-page site like a blog or if you offer other kind of contents for which is normal to expect single-page sessions, then a high rebound frequency is perfectly understandable”, they reassure us in the end.