What They Never Told You About Bounce Rate

First of all, what is bounce rate?

Bounce rate is defined as the percentage of visitors who land on a page of your site and then leave your site (not clicking on any other pages).

They only view a single page and then ‘bounce off’.

There’s even an official bounce rate equation.  I’ll spare you (and me) from that, but there are other similar terms too, for example…

What is exit rate compared to bounce rate?

Exit rate is the percentage of visitors who leave your site on a certain page compared to how many total visits that page gets.

You might have a page that gets about 100 visits a month where 90 of those 100 visitors leave. If so, that page has a 90% exit rate.

The thing is everyone’s going to leave your site at some time or another. You can’t avoid it.

But at first glance you might think because a page has a high exit rate the page sucks.

On the flip side, if you’re running a lot of advertising (like Adsense) then you may have a page where a lot of people click off your site and onto one of your ads. In that case it might be a good thing. Your visitors may be reading your content and then clicking off on a super targeted ad, allowing you to earn money. So it’s not always bad.

You can’t just assume because a page has a high exit rate it needs to be fixed. You’ve got to take a look at the exact page in question. You might want to look at the average time a visitor spends on that page and what you’ve got there. Ask people you know to visit it and have them give you their impression.

The same goes for bounce rate. A high bounce rate doesn’t necessarily mean your site sucks.

What if people land on one of your pages and spend a ton of time reading it (maybe even sharing it or bookmarking it for later)?

If that’s the only page they land on it will still count as a bounce, but as you can see, the visitor had a positive experience with your content.

Here’s a few things that cause a bounce:

  • The visitor clicks the back button and leaves the site.
  • They close the window or tab.
  • They type in a new URL.
  • The session times out (usually after 30 minutes).
  • They click on a link leading to a different site.

As you can see, not all of these actions mean a visitor is dissatisfied with your site (yet they all count as a bounce).

What if people fill out your opt in form?  Often the visitor is re-routed through a page hosted on aweber or Infusionsoft, or whatever autoresponder company you use so it looks like they clicked off.

Or what if you’ve got a phone number on your site and they call it? While that will likely count as a bounce they were interested enough to give your company a call.

What if they go to your social media page? That will count as a bounce too if they only visit one page on your site first and then click on a link to your Facebook page.

So there are lots of actions resulting in a bounce.  And as you can see, just because a visitor might be counted as a bounce, it doesn’t mean they had a negative experience with your web page (although that’s what many webmasters believe).

Adjusted Bounce Rate

The thing is, time on site and bounce rate together can give you a more accurate image of what people think of your site. Google

Analytics actually has an adjusted bounce rate feature. This ‘adjusted bounce rate’ feature is rarely ever mentioned, so I want to bring it to your attention.

Bounce rate by itself is really kind of useless. But with this adjusted bounce rate you can also factor in the length of time visitors spend on your site.

With an adjusted bounce rate, a bounce is only counted when a visitor lands on a page, they were there for a really short amount of time, and they click off your site (not visiting any other pages).

In order to get an adjusted bounce rate to show up if you’re using Analytics, you need to add the bolded line of code to your tracking code:

<script type=”text/javascript”>
var _gaq = _gaq || [];
_gaq.push([‘_setAccount’, ‘UA-XXXXXXX-1’]);
_gaq.push([‘_trackPageview’]);
setTimeout(“_gaq.push([‘_trackEvent’, ’15_seconds’, ‘read’])”,15000);
(function() {
var ga = document.createElement(‘script’); ga.type = ‘text/javascript’; ga.async = true;
ga.src = (‘https:’ == document.location.protocol ? ‘https://ssl’ : ‘http://www’) + ‘.google
analytics.com/ga.js’;
var s = document.getElementsByTagName(‘script’)[0]; s.parentNode.insertBefore(ga, s);
})();
</script>

You can change the amount of time to whatever you’d like. Sixty seconds might make more sense than the 15 second length Google talks about using.

If your high bounce rate is freaking you out, you can look at this and see if it goes down. Because really, if people are spending 60 seconds on your pages before they leave, they probably got some value out of being there.

If you use this adjusted bounce rate, now only the people who visited one page and who were there for less than the time you specify count as a bounce. This is a more accurate measure to what Google looks at when considering how users interact with your site.

The Freak Out About Bounce Rate

The bounce rate reported in Google Analytics (whether you’re using the regular or adjusted bounce rate), or even another tracking package, and what Google actually factors into their algorithm about your site are different.

The bounce rate you see is not really what they’re using to assess your site. Here’s why…

What if you don’t have Analytics installed?

Not everyone uses Analytics. Some people use another tracking software (that Google does not have access to) and many don’t use anything at all.

What if you don’t have Analytics installed properly?

Not everyone puts Analytics on every page of their site, or copies and pastes the code properly. Or even puts the right code on the right page.

Analytics is just another thing we can potentially manipulate.

We’re able to copy and paste the analytics code on the pages of our site and we can mess it up (intentionally or accidentally), so Google isn’t going to use that to score your bounce rate.

What Can Google Really See?

Every time a visitor uses Google to search, it gives Google more data.  Google can see what visitors are searching for, what they clicked on, how long they were on those web pages, and whether they come back to their search engine results page and click on another listing or perform another search.

Here’s what someone who has a bad experience on your site might do:

They click on an organic listing, go through to your site, then come back to the same search results page in 2 seconds.  I think we can all agree that probably reflects badly on the web page the visitor clicked on.

Contrast that with someone who only clicks on one page of your site, but spent 10 minutes there.

That’s really all Google can see. They have access to information about their search engine results pages, but not necessarily about your site (I don’t believe even if you install Analytics on your site that they’re using that data for your rankings).

If the visitor is using the Chrome browser or a Google toolbar, I’d assume Google can see a little more about their interaction with the sites they visit.

And if the visitor is browsing while they’re logged into a Google product (like Gmail) then that may allow Google to see more interactions too.

But many people don’t search using Chrome or while logged into a Google product.  So Google has very limited data about the nature of their search, to the point where all they can really see is what visitors are clicking on from a search engine results page and when they come back to Google for another search.

The main point is, a bounce (like what you’re thinking about and seeing if you use Analytics or another tracking tool) is not the same thing Google factors into their algorithm.

In most cases, Google can’t even see if a visitor clicks around on your site.

They’re pretty much in the dark about what went on there.

That’s why the time on site really has to be a metric that’s involved with bounce rate because just a straight bounce doesn’t mean a whole lot (and will be difficult for them to measure for all sites). What Google’s really trying to figure out is whether people have a good or bad experience with your site.

Does it suck? If so they don’t want it ranking high in their results pages.

Adjusted bounce rate is probably a closer metric, but even that isn’t really the whole story.

I’m not saying bounce rate is worthless. It’s just not used in Google’s algorithm for ranking sites like many might expect.  So you can’t say just because your bounce rate’s high that Google will not rank you as high. Again, that’s because they don’t look at bounce rate the same way you do.

But it’s still useful to look at your bounce rate along with the average time a user spends on your site. If your bounce rate is high (80% or above) and you’ve got a really low average time on site (like 30 seconds or less), that’s an indication you’ve got a problem.

How to Reduce Bounce Rate

If you’re thinking your bounce rate could be lowered and that your site doesn’t make visitors as happy as it could, here are some things you can look into:

1. Browser performance. 

It’s possible your site’s design doesn’t work well in every browser. You can check this by looking at your bounce rate by browser. If it goes way up for visitors using Firefox, that may mean it doesn’t look right in the Firefox web browser.

2. Slow loading pages. 

Look at types of connections your users have and see if slower connections have higher bounce rates. If so it may be that slow loading pages are turning visitors off (if it takes 10 seconds to load a page I know I go somewhere else and don’t look back).

3. Consider your design.

Back when the web was newer, you could get away with really ugly design (bordering on no design, just pages with text thrown up on them). Now poor design just isn’t tolerated.

4. Mobile device friendly.

How does your site look on a mobile device or tablet?  Check your bounce rate against the different devices and look at your site from all the different devices you can get your hands on.

5. See if specific keywords go with high bounce rates.

Does the content on the page match the word? Make sure your content is relevant for the search term you’re ranking high for (or are trying to rank high for).

6. Make sure the Analytics tracking code (or whatever tracking package you’re using) is on every page of your site.

If you’re using WordPress for most of your site and have also created a few static HTML pages make sure you add the tracking code manually onto those HTML pages.

7. Link to other pages on your site.

You can include links throughout your content and present your reader with related articles at the end of your content so they can read more about the topic they’re interested in.

8. Consider getting rid of pop-ups.

Pop-ups can turn a lot of people off so if you’re running them on your site and you notice a high bounce rate you may want to consider turning them off.

9. Check your navigation.

It may be that visitors can’t figure out how to find stuff on your site. If they get frustrated, then they may leave.  You might want to check out a tool like Crazy Egg that allows you to see a heat map of your site showing you where users click and what they do on your pages.

10. Consider breaking up your information.

If visitors are hit with a wall of text then many of them will exit quickly.  It’s best to break up text with paragraphs, subheadings, bullet points and include images, tables, video, or whatever else you can.  Readability is an important factor to keep in mind.

11. Consider moving your ads.

If visitors are hit with ads right up front many will turn right around. You may want to re-think your ad placement, especially if you’ve got a lot of ads above the fold.

What next?

  • If you’ve got tracking software installed on your website, go check out your current bounce rate.
  • If you’re using Analytics, consider adding the line of code I mentioned earlier in the article to the tracking code on your web pages so you can see the Adjusted Bounce Rate.
  • If you’ve got a higher bounce rate than you’d like, try a few of the tips above and then check your statistics in a few days, see if it improved.

Leave a comment below. Let me and other readers know how you’ve been able to lower your bounce rate. If you’ve got a really high rate and would like some suggestions, leave your site URL below (along with the current bounce rate) and we may be able to check out your site, giving you a few ideas.

19 replies
  1. Louise
    Louise says:

    Makes for very interesting reading Lisa. Thank you for sharing with us. I wonder to what degree Google will make comparisons to sights like Facebook whereby no doubt the bounce rate would be very low, not to mention other news media sites where people read articles? My point being is does Google factor in what type of site it is and whether or not high bounce rates are justifiable? For example some sites will have a home page with the word ‘enter’ which no doubt makes the bounce rate extremely high?

    Reply
    • Lisa Parmley
      Lisa Parmley says:

      Hi Louise,
      I can only make a guess on how they factor it in, but it would make sense if they go by the search term. So they may calculate it for sites that show up on any given search results page. If your site has a low bounce rate for any given search term and someone else has a higher bounce then yours comes across as better. That would be my guess.
      I don’t know how they could figure out the type of site. I do know people think it varies by the industry, but the way the information would be the most valuable for Google is if they could use it to compare your site to others showing up on the same search page.

      Reply
    • Ernest
      Ernest says:

      I am a little confused here! If Google does not factor the bounce rate into their algorithm why speculate about how Google sees the bounce rate on different types of sites? Bounce rate, and especially the adjusted bounce rate that Lisa so kindly shared with us, is only used for optimization of your own website(s). Or am I missing something?

      Reply
      • Lisa Parmley
        Lisa Parmley says:

        No, that’s all it’s used for. But many people don’t realize that. Some studies break down bounce rates by industries and people use it to see how their site compares (whether they realize it’s not the same as the metrice Google uses I don’t know).

        Reply
  2. Linda
    Linda says:

    Thanks, I have been concerned over bounce rate. People can vary on what rate is good or not so good. I use Google Analytics but didn’t know about the line of code for the ‘adjusted bounce rate feature’.

    Reply
  3. Mara Dall
    Mara Dall says:

    Useful info, Lisa – thanks again for another in-depth analysis. I’ve only just started to look at G Analytics for my sites, and I know bounce rate is something that – yes, causes freak-out!
    It’s good to have an informed perspective – I didn’t realise that bounce rate doesn’t take any account of time on site. I’m going to follow your advice & use the adjusted bounce rate, which is much more useful information to have.

    Reply
  4. Comporta
    Comporta says:

    Hi Lisa, I’ve been able to reduce my bounce rate to around 30% and then when I started targeting all countries, i.e. those in foreign languages, my bounce rate increased again. I have the translator on the site but it doesn’t seem to help. Do you have any suggestions or ideas on how I may target foreign countries but not increase my bounce rate.? Kind regards, Comporta.

    Reply
    • Lisa Parmley
      Lisa Parmley says:

      That’s a good question. I bet if you have a logo and other graphics in English that will turn non-English speaking people off because they’ll immediately see the site’s not in their native language. Not sure if that’s the case with your site or not, but it’s an idea.
      Also, using the translator is not the same as writing articles in another language. I only speak English so I don’t know this first-hand, but that’s what I’ve heard. So it may not be possible unless you target other languages with a completely separate site, with content written in that language.

      Reply
    • Mighil Mohan
      Mighil Mohan says:

      I think using translator is not gonna help you much @Comporta (The whole content will be of less quality! ). You should target the global audience with seperate sites with manually written contents.
      (like en.wikipedia.org, de.wikipedia.org etc.)
      Please dont ask me “how will google make the difference between automated translation and a manual one” ?
      Have a good day!

      Reply
  5. Inspiring Citizen Rafi
    Inspiring Citizen Rafi says:

    Hi,
    My bounce rate from search engines is on a higher side whereas from other sources it is comparitively low..!
    And I see that US visitors leave the page sooner compared to other geographical locations..Is there any specific reason for this?

    Reply
  6. Ron
    Ron says:

    Lisa,
    Millions of people have either the Google Toolbar or Bing Toolbar installed. Although many people don’t know this, they might be passing information back to Google or Microsoft about the sites and pages they visit on the web. Most SEO Consultants understand this, but I find most people outside of the industry have no idea. This is an incredible way for the engines to understand actual bounce rate. Imagine you visited a page from Search and spent 12 minutes on the page. Then you either type in another URL directly in the browser or via the toolbar (Google or Bing). The toolbar can pick this up and understand how much time you spent. Again, that’s much different than seeing a “bounce” in your analytics package.
    Thanks
    Ron D

    Reply
  7. Willem
    Willem says:

    Thanks very much for the tip on the Adjusted Bounce Rate. With my website I also had the problem that I didn’t know whether the visitors just bounced off or at least took some interest in what I am selling.
    I implemented the Adjusted Bounce Rate and set it for 90 seconds. Now I know who has done some investigating. Thanks again and may your website prosper.
    Best regards,
    Willem

    Reply
  8. Nova York
    Nova York says:

    Really good information about Bounce rate. Even though I have already read about it, your information is really detailed.
    Is there any way the bounce rate may affect the Trustrank of a blog?

    Reply
    • Lisa Parmley
      Lisa Parmley says:

      If Google measures a site’s bounce rate as high enough then yes, it will be lowered in their search results pages.

      Reply
  9. faceforwardla
    faceforwardla says:

    that is good information in google measures a site because google bonuse rate is high and enough then .it is lowers search any type information search result pages. nice sharing useful information is our website benefit information.

    Reply
  10. Alex Zorach
    Alex Zorach says:

    I agree with you about bounce rate being over-emphasized as a resource. Having almost four year’s worth of GA data now for my main project, RateTea, I’ve found that bounce rate simply isn’t useful. It lumps too many things, positive negative, into a single measure, and the positive and negative factors cancel out in unpredictable ways.
    One example that I want to add to what you’ve written here, which I’ve seen in my work, is that redesigning your site in a way that is more logical and easy-to-navigate can actually increase bounce rate. For example, if someone types something into Google and lands on a page on your site, but doesn’t find exactly what they’re looking for, they might click around a few times, only to close out of the site feeling frustrated. That won’t register as a bounce, and it will even boost your pages/visit, but it is not a good sign.
    Restructuring the site to improve usability, SEO, and make more logical landing pages could actually increase bounce rate–because users are more likely to land on a page where they find exactly what they’re looking for. But the higher bounce rate might reflect more satisfied, happy users.
    I’ve actually seen that as I improve my sites’ navigability, the bounce rate often increases and the number of pages/visit decrease. In conventional logic this is viewed as a “bad thing” but I think that’s a backwards way of looking at it. Often, these bounces are reflecting satisfied users, and the same users are more likely to share the page, upvote on reddit, or return to the site in the future. So I’ve seen traffic and revenue increase even as bounce rate increases.

    Reply

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *