Bounce Rate Has No Place In Content Marketing | Outbrain Blog

CONTENT MARKETING & DISCOVERY

Bounce Rate As You Understand It Has No Place In Content Marketing

| Will Fleiss

Bounce rate and content marketing

Google Analytics defines a website’s bounce rate as follows: “Bounce Rate is the percentage of single-page sessions (i.e. sessions in which the person left your site from the entrance page without interacting with the page).”

In other words, a page’s bounce rate is the percentage of visits that land on that page and leave without visiting a second page or interacting with the page they landed on.

The following visitor actions will result in a bounce according to Google Analytics

  1. Visitor clicks the back button
  2. Visitor closes the browser
  3. Visitor types a new URL in the browser
  4. Visitor clicks an external link
  5. Visitor does not click to another page for 30 minutes

So, basically it’s only NOT a bounce if they visit a second page.

OK, What’s Your Point?

The point is this. If a visitor lands on your blog post and reads the entire thing, that content has achieved it’s goal. If you’re looking at the bounce rate as a success metric for that content, 100% of the visitors could have read the entire article, scrolled to the very end, and then hit “back” on their browser, and Google Analytics will still say that page has a bounce rate of 100%.

Google Analytics is telling you that all those readers who just read every word of your carefully crafted article bounced… bounced?! Are you kidding me? They just hung on your every word!

How to make bounce rate meaningful for content marketing

There is a solution to make bounce rate a more accurate measurement. When someone lands on a piece of content let’s say it takes 5 to 15 seconds (maybe 20 secs, depending on your site) for her to understand if this is something she wants to read. If the reader decides to stay, you don’t want to count her visit as a bounce, right?

By adding a line of code to your Google Analytics tracking script, you can automatically trigger an event when readers stay for a minimum amount of time. You decide what you think the appropriate time should be.

If you use the regular Google Analytics code add this line:

setTimeout(“_gaq.push([‘_trackEvent’, ’15_seconds’, ‘read’])”,15000);

If you are using the new Universal code add this line:

setTimeout(“ga(‘send’, ‘event’, ‘unbounce’, ’15_sec’)”, 15000);

By doing this, the bounce rate becomes a significant metric. It is giving you an accurate number to rely on, and answering a very relevant question. Is the audience staying and reading the content or not?

For a full guide, read our post on using Google Tag Manager to increase engagement on your content.

Testing the Adjusted Bounce Rate

Here’s what we saw after adding an engagement event that fires 30 seconds after someone landed on the page. Over a period of 7 days our bounce rate went from an average of 82% to 29%.

Google Analytics adjusted bounce rate

Now we have a much better understanding of what content actually gets read, instead of falsely assuming this visitors bounced without engaging with our content.

Key Takeaways

  1. Bounce rate as most people understand it is not a good KPI for content marketing
  2. If you’re using bounce rate as a content marketing KPI, you MUST implement a timing engagement event into your Google Analytics tracking script.

Effectively measuring your content marketing is not easy, but those who measure and optimize the right metrics will be the ones who ultimately see the greatest return on investment. According to a poll conducted by SmartInsights the majority of marketers agree content marketing will have the single biggest commercial impact in 2015 among all digital marketing techniques.

2015 content marketing top digital marketing technique

If that will be true for your business or brand understanding bounce rate and other performance content marketing metrics are critical to your success.

How are you measuring your content marketing?

Featured image courtesy of frankieleon via Flickr

Hitting "Publish" is only The Beginning!

Download our free A-Z Guide to Brand Publishing to find out how to make the most of your content strategy.

Will Fleiss

Will Fleiss

Will leads content marketing at Outbrain. He's been doing online marketing for 10+ years, spending time on the agency-side at... Read more

Add A Comment

* Your email address will not be published. All fields are required.

  • Aditya Verma| April 5, 2015 at 2:02AM

    Agree with this assessment. Bounce rate needs to be seen in context of the time spent by the user in consuming that content. If the user is spending over 20-30 seconds on the blog or content page, it is reasonable to assume that he would have consumed the content. He could exit the page for various reason like

    1. There was no other relevant content for him to consume
    2. His requirement was met with the post
    3. The architecture of the site was poor, so multiple content discovery was not possible

    Overall a good post.

    Cheers

    Aditya

    • Will Fleiss
      Will Fleiss| April 6, 2015 at 8:08AM

      Yep, thanks for that recap Aditya! Really helpful!

  • Brian| April 9, 2015 at 12:12PM

    Valid points but there are certainly valid counter points as well. Such as:

    1. Publishers need multiple page views for the ad rotation in order to make paid media such as content marketing platforms successful. That is why there are so many top 10 lists that require 10 page views to read, otherwise it would be one page.
    2. Most advertisers objectives, not all, will not be met if one page is visited. Almost, no one is buying or signing up for anything with one page view. Even a landing page with a form has a thank you page (not a bounce).
    3. There typically is a direct correlation to time on site and bounce rate. If I’m tracking my traffic sources I already know the one page I’m sending that traffic to and can clearly see the 85% bounce rate and 5 seconds average time on site. Yes, there are exceptions.
    4. Bounce rate in relation to other sources is valuable provided proper bench marks are in place. Knowing that upper funnel tactics (paid content marketing) will have a higher bounce rate compared to lower funnel tactics (SEO).

    This isn’t an endorsement for using Bounce Rate as a sole KPI but it can have its place in content marketing with respective to a site owner’s objectives and other KPIs.

    • Will Fleiss
      Will Fleiss| April 9, 2015 at 10:22PM

      Hey Brian! Thanks for taking the time to comment. Some some really interesting points here. Here are my thoughts.

      1. Publishers are definitely a different story as their business model is driven by page views.
      2. I have to disagree here. In order for content to receive the credit it’s due in the customer journey, we need to pay way more attention to user actions after the web session. If your content speaks to your customer’s needs, and the visitor consumes that entire piece of content, it’s ok that they don’t immediately buy your product. The sooner businesses and brands become more comfortable with attribution tracking, the sooner great content will emerge as a major driver of growth.
      3. Time on site and time on page metrics can also be problematic for measuring content marketing as Google Analytics only records the time on page if the user visits a second page. This will be the topic of another post, but similar to bounce rate, the visitor could spend 10 seconds or 10 minutes on a page, and Google Analytics will record the time as zero seconds if a second page is not visited.
      4. Yep, it’s critical to compare metrics by source, but also by content type. A blog post shouldn’t be looked at in the same way as a landing page.

      The “As You Understand It” part of the title is important to note. Even if it’s still used as one of several KPIs, it’s important for people to fully understand what they are measuring. For me the word “bounce” evokes a very clear image of a ball touching a surface for only a split second, which in reality makes the term a bit of a misnomer.

      Thanks again Brian!

  • Kevin| April 10, 2015 at 11:11AM

    Interesting article. Thanks for posting. We use SiteCatalyst, rather than GA and I’m wondering if there is a similar script that can be implemented and whether SiteCatalyst defines bounces the same as GA does.

    • Will Fleiss
      Will Fleiss| April 10, 2015 at 12:12PM

      Hey Kevin,

      Great question. I did a bit of digging without any luck. Stay tuned… we’ll get you an answer!

    • Adam Greco| April 13, 2015 at 1:13PM

      You could certainly use Custom Links to set a success event or interim page view after a specific amount of time in Adobe Analytics (SiteCatalyst) as well.

  • scott krause| April 10, 2015 at 1:13PM

    As a reader, I always scan my newsfeeds (feedly, twitter, facebook, nuzzel) and cherry pick 5-10 stories I want to read either in that session or throughout the day. I open all of these in new tabs and imagine this practice is very common.

    I may not get to any of these stories until a good half hour or several hours after opening them. How would this affect this adjusted bounce rate metric?

  • Scott Lee| April 14, 2015 at 7:19PM

    Great points. I have been testing Outbrain for my company (Spoken English Practice website http://www.spokenenglishpractice.com) and the high bounce rate has been a point of concern for the team. You make some valid points on why bounce rate should not be looked at as a single factor of success. That said, the bounce rate for Outbrain traffic is around 80% while normal organic traffic visiting the same content on our site shows a bounce rate of around 50%. Any thoughts why there is such a significant gap?

    • Will Fleiss
      Will Fleiss| May 13, 2015 at 10:10AM

      Hey Scott,

      My apologies for the delayed response! I’m glad you asked that question.

      Visitor behavior of someone in content consumption mode vs search mode is definitely going to be different. Someone coming to that piece of content from organic search has landed there as a result of a specific search term, while someone who found your content to be interesting while browsing the web is in a different mindset.

      In many cases content discovery is what seeds the desire for someone to search for something specific, so it’s important to try to connect the dots in the customer journey for each individual that visits your site.

      Are you using any kind of attribution measurement to understand how people first visited your website?

  • Ben| April 19, 2015 at 10:22PM

    We added this code to our site but our site visits tanked almost immediately in GA, from an average of 1000 hits a day to around 100 overnight!. What did we do wrong? Any advice? We just added the code as per below. As you can see we only changed the timing. Help?

    (function(i,s,o,g,r,a,m){i[‘GoogleAnalyticsObject’]=r;i[r]=i[r]||function(){
    (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
    m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
    })(window,document,’script’,’//www.google-analytics.com/analytics.js’,’ga’);

    ga(‘create’, ‘UA-XXXXXXXXX’, ‘auto’);
    ga(‘send’, ‘pageview’);
    setTimeout(“ga(‘send’, ‘event’, ‘unbounce’, ’30_sec’)”, 30000);

  • Mann Chawla| April 23, 2015 at 12:00AM

    Hi, this is a really nice read.

    I have one question- why is it important to add the additional code? GA provides avg time spent as well. If the avg time spent is 15secs and bounce rate is 80%, isn’t it safe to assume that most of the traffic is bouncing off after 15sec?

    Let me know your thoughts.

    Cheers, Mann

    • Sean| June 13, 2015 at 2:02AM

      I was thinking the same thing. They need to discuss the time spent on the site, which wasn’te discussed.

  • iqbal| April 23, 2015 at 8:08AM

    well detailed and informative article. Running and online eCommerce store, eliminating or reducing higher bounce rate is a challenge for web marketeers. This article will help them understanding the idea in details.

  • preet| April 29, 2015 at 12:12PM

    Nice Thanks For Posting.

  • cyril bedat| May 12, 2015 at 6:06AM

    You are forgetting a very big point that is making your analysis wrong. A bounce rate is what you describe + time spent less than 1 second on the page. Hence, bounce rate is still a good metric to use if you are judging on your content performance.

    • Will Fleiss
      Will Fleiss| July 27, 2015 at 11:11AM

      Hey Cyril,

      Thanks for commenting! Unfortunately that’s not how Google Analytics calculates it, and that’s where most people get their bounce rate metric.

  • Christine| June 4, 2015 at 7:19PM

    If the user clicks on an ad, like a Google Adsense link, and nothing else, will it count as a bounce?

  • Jeff| June 19, 2015 at 10:10AM

    I believe that Bounce Rate is one Variable/KPI that cannot be over looked especially in Content Marketing. So to say that “Bounce Rate has no Place is Content Marketing” is a very bold statement. Bounce Rate @ a High Level can describe the overall Health of your Channel’s Engagement and whether or not your Content is Consistent with your Audience. Overtime if you decide to side step on the Bounce Rate you’ll start to see your Paid Media Channel become less relevant.

    If your Paid Media Channel has an Avg. Bounce Rate of 90% in the last 30 days, why would you disregard that metric?

    • Will Fleiss
      Will Fleiss| June 19, 2015 at 4:16PM

      Thanks for the comment Jeff! I couldn’t agree more. That’s why the title is “Bounce rate as you understand it has not place in content marketing.”

      Most people don’t understand exactly what it’s measuring. If a channel that’s being used for direct response marketing has a 90% avg. bounce rate you absolutely need to pay attention to that metric, but the goal of content marketing is to build a relationship with great content. If I’m paying to promote a blog post I want to know if they read the full post, because that could signal the beginning of a beautiful relationship 🙂

  • Raj Roy| June 23, 2015 at 5:05AM

    I think this is an interesting article, and a very valid point, but doesn’t quite dig deep enough. As a content publisher, it’s a big problem we face.

    Once you start to fiddle your bounce rate, it makes comparisons between competitors, other sites you run, or between other marketing channels harder to understand. As one person points out, you can look at average visit time alongside the bounce rate; however, this buckets up all users into an average.

    There’s a nice post by Justin Cutroni (http://cutroni.com/blog/2012/02/21/advanced-content-tracking-with-google-analytics-part-1/) which looks at how you can categorise Readers and Scanners; and far your visitors scroll down the page. Although this fiddles the bounce rate too.

    And Upworthy have devised an Attention Minutes metric (http://blog.upworthy.com/post/75795679502/what-uniques-and-pageviews-leave-out-and-why).

    We don’t have any traditional ‘conversions’ such as a purchase, sign-up or offline enquiry. I know we’re awash in metrics and KPIs, but I feel that it would be helpful for the industry to develop another metric to measure user interactions with content in a more meaningful way.

  • Marcin R.| July 3, 2015 at 10:10AM

    Hey 🙂 That’s a nice article. Informative and quite well-written. If you have some spare time and want to find out more about bounce rate, check out this article:

    http://www.mavenec.com/blog/bounce-rate-google-analytics/

  • Merit Lansing| February 1, 2016 at 6:18PM

    This article is great. However, I implemented the tracking in Tag Manager directly as per the instructions about 6 weeks ago, and now our bounce rate seems to be completely broken. That is, it’s showing as less than 1% site wide. Our time on site event tracking (30 seconds trigger) is showing about an 85% completion rate (which is awesome) – so shouldn’t our bounce rate be reflective of that rate, and be about 15% (i.e. the people who aren’t staying on the site page for 30 seconds). Doesn’t zero percent seem too low?

  • Simen| August 24, 2016 at 3:15PM

    Some good points and the dialogue between you and Brian (9th of April 2015) is also interesting.
    My question is this: Is there some sort of consensus on what the edited bounce rate metric should be?
    It is useful to be able to compare performance across multiple sights and this does not work for this metric if everybody has their own twist.

  • Betty| September 1, 2016 at 4:04AM

    Good point! Google Analytics Counter Tracker plugin on WordPress is Very simple and easy to use.If you Connect your Google Analytics account, you get Google Analytics stats; Group stats by hour, day, month and year; you can See data for different time periods in your reports; it is Compatible with almost every version of WordPress; and Mobile-Friendly Design; it is Multilingual.

Loading

Get Weekly Content Marketing Tips

Subscribe to get weekly tips and insights directly to your inbox