What are good bounce rate benchmarks

How to draw the right conclusions from your bounce rate

For website operators, the bounce rate is an indicator of the functioning of the landing page. It is often ignored that this rate alone cannot define the value of a page visit. Nevertheless, a high bounce rate may have a negative impact on Google's perception.

The bounce rate: different depending on the type of page

Google itself defines a jump as follows:

With Google Analytics, a jump is calculated as a session in which only a single request to the Analytics server is triggered, for example if a user visits a single page on your website and then leaves the page without further requests to Analytics for this session -Server are triggered.

The bounce rate is determined by dividing all visits for a page by the number of all sessions.

Kissmetrics illustrated this formula. The blog also states that Google can use very high bounce rates as a factor for pageranking. So if your bounce rate is 80 percent, is your page quality at risk and with it your ranking on Google?

First of all, it should be mentioned that such high bounce rates are not uncommon for certain sites. In addition to blogs, which sometimes only have one page, simple landing pages fall into this category. A bounce rate of 70 to 90 percent is normal here. The situation is different with retail websites, for example, where only 20 to 40 percent bounce rate is the standard. This is also stated in the Kissmetrics infographic; however, the analysis by customedialabs also agrees with these figures.

Unsurprisingly, the rates for service sites or portals (such as Facebook and Co.) are low, while they are extremely high for dictionaries, etc.

What can the bounce rates actually say?

But are the bounce rates always directly related to the quality of the website and the page visits? On the one hand, this can at least be assumed to the extent that comparable websites that have lower bounce rates may not just get them by chance, i.e. maybe do something better. On the other hand, high bounce rates can be considered less serious when the number of visitors is already extremely high, i.e. the traffic is already there.

In addition, Google cannot classify all interactions; the bounce rate is relatively rigid. For example, it can happen that a page visitor reads the start page for more than ten minutes - but without performing any interaction, such as clicking on a CTA. If the user then leaves the page (even for a short time, maybe to come back afterwards), this is counted as a bounce; but this hardly looks like an actual jump.

Furthermore, Google cannot differentiate when users browse different tabs on a page. Even if a conversion happens in a new tab, there can be a bounce in the first tab, says Neil Patel. Another aspect that Patel cites is fuzzy traffic classifications. If redirections from HTTP to sites with SSL certificates take place and these are not classified as referral but as direct traffic, this can also have a negative effect on the bounce rate; after all, there is no possibility of interaction here.

So what can you do to help you see what's going on behind the bounce rate? Neil Patel has a few suggestions.

  • Tracking in-page events
  • The segmentation of inbound traffic
  • Eliminate phantom bounces and test page speed
  • Delay pop-ups
  • Note exit rates

Tracking in-page events says more about visits, which are counted as bounce

In addition to the page views, in-page events are actual indicators of engagement one Page. If a user just stays on the landing page without loading another page and leaves the page after ten minutes, that's a jump. But if he watched an implemented video for five minutes during this time, and then maybe clicked a PDF download, he was anything but uninterested on your side.

In order to be able to track such events, specific events can be created in the admin dashboard in Google Analytics. To do this, certain code must be added to the website (or app). Google explains how this works.

If you get an overview of such events, then you can better assess the level of commitment the users are showing on the site, even if they do not load any further pages. But more importantly, Google recognizes that interactions are taking place, so that such a visit is no longer classified as a bounce.

Segmenting inbound traffic can provide clarity

The segmentation of inbound traffic goes back to an idea of ​​unbounce. This is intended to create separate landing pages for different traffic sources, so that clarity arises as to which of these sources are more valuable for the website and how to work with them.

Also, since Google hardly distinguishes direct traffic from social traffic, these different landing pages can be useful. Neil Patel gives an example of how jumps can be minimized in this way. When a user clicks on an ad on Facebook, they are taken directly to a conversion landing page because they have already shown interest - the possibility of a bounce seems low.

Another aspect to be considered when analyzing inbound traffic is the user's device. Mobile bounce rates are particularly high. They are up to 16 percent higher than with the desktop, according to customedialabs. This may also be related to the fact that many pages have not yet been optimally adapted for mobile use. Website operators should pay special attention to this. If the mobile bounce rates are significantly higher, then it is important to start here, because the mobile page accesses will only increase further.

Test your page speed

An elementary aspect of the bounce rates is the loading speed of your page, also mobile. If your site loads too slowly, high bounce rates are almost inevitable. So-called phantom bounces in particular, in which the user has not even seen a page before leaving due to the long loading time, pose a problem in this case. Especially in the mobile age, the loading speed must always be optimized. Also because Google will make page speed a ranking factor from July 2018.

A study by Google shows the influence of mobile charging speed on the bounce rate.

Page speed optimization should be the first step when bounce rates are too high; and it could lead to a significant reduction.

Delay pop-ups so as not to jeopardize your ranking

Another tip from Neil Patel is about pop-ups. On a landing page they are of course a good way to lead the user directly into an interaction that can then also be registered by Google. But: if pages with pop-ups from Google - especially in the mobile area - are classified as intrusive, this may damage the ranking. Especially when there are a lot of interstitials on the page.

Product Manager explains on the Webmasters Blog Doantam Phan Examples of techniques that make content harder to access for users:

Showing a popup that covers the main content, either immediately after the user navigates to a page from the search results, or while they are looking through the page.

In this case the ranking is at risk. In order to continue to use the pop-up option so that users don't tend to jump off, but at the same time keep the Google ranking intact, there is an option: the pop-ups are only used with a delay. Patel believes that Google will not “punish” them if they only appear after three to five seconds, meaning the user had time to look at the page beforehand.

In addition, so-called exit pop-ups can be used, which also do not result in any penalties from Google. Pop-ups appear when the user is about to leave the page. They can display offers and sometimes entice the user to click so that there is no bounce.

An important point that Patel draws attention to in the course of this strategy is that such pop-ups can also influence the loading speed. So you always have to consider these effects.

Don't forget the exit rates

In addition to the bounce rates, exit rates can also be determined. The specific differentiation to the bounce rate is specified by Google:

The difference betweenExit rate andBounce rate for a specific page can be summarized as follows:

  1. For all calls there is theExit rate what percentage of the page views thelast were at the meeting.
  2. For all sessions that start with the page is theBounce rate the percentage of calls that theonly onePageview of the session were.
  3. TheBounce rate a page is only based on sessions starting on that page.

Patel's visual example confirms his statement that high bounce rates usually also increase exit rates. What is of interest, however, are the differences between these rates, which you can analyze. If there are large discrepancies here, then the actual user engagement must be investigated.

The exit rates could be an indicator of lack of engagement that is even more specific than the bounce rates.

Ultimately, these aspects can be considered when the bounce rate hovers over your site like a dark cloud. Small, contextual interventions can help you see what the bounce rate is really telling you. Nonetheless, the goal remains to reduce them. This doesn't get any easier in times of mobile access. For this, however, the awareness of the reasons for a jump and its background becomes all the more important.