9

I was tracking data for a site and realize that there is a sudden drop in bounce rate in Google Analytics. The sudden drop happened across 2 days and remained consistent after.

The bounce rate in January 2019 is about 70% whereas after 2 days, it became 12-15%. I am unaware of what my colleague did so I am trying to figure it out now.

Is a bounce rate of 12% normal?

enter image description here

16

I'd be willing to bet that your colleague implemented on-page events.

By default Google Anaytics only measures page views. Bounces are defined as users who only view one page; even if that user spends hours on one page.

Events allow you to track what happens within the page. You can send GA information about items such as:

  • User clicking on buttons
  • User scrolling the page
  • User spending X minutes on the page
  • User typing on the page
  • User watching a video on the page
  • Additional AJAX content gets loaded

All of those events can be marked as "interactive" which causes GA to say "this user did not bounce." They also could be marked as "non-interactive" which does not affect the bounce rate.

If I'm right, you will see data in the "Behavior" -> "Events" -> "Overview" report starting right when your bounce rate went way down.

A 12% bounce rate when events are implemented is quite normal. It is also very normal for some types of sites to have a 75% bounce rate before events:

  • Single page applications (SPAs) only have multiple page views if users refresh or click to the site a second time.
  • Many sites provide all the information that most users need right in the landing page. That is common for articles and calculators. I feel that it shouldn't be a "bounce" if the user reads the entire article.

Events are the way to let GA know that you are satisfying your users even when they are only viewing one page.

You may want to review the events that are being sent and make sure that each is being marked as "interactive" or "non-interactive" appropriately.

  • 2
    Well this is a real eye-opener. You would think that a person landing on the exact page they needed from a Google result would be calculated more favorably. By adding these events does it also improve the SEO algorithm by showing Google that your user did not "bounce" per the redefined definition? – MonkeyZeus Mar 4 at 18:15
  • 3
    Google Analytics statistics have no direct impact on SEO. Improving your bounce rate by implementing events doesn't change user behavior and can't affect your rankings. For ranking purposes, Google cares that users are satisfied. Rankings could get worse if lots of users head back to the search results after visiting your site and click on another site instead. For a detailed explanation of why, see my answer to "Does a site's bounce rate influence Google rankings?" – Stephen Ostermiller Mar 4 at 18:25
  • This is a new area for me. Thank you. After a quick bit of reading it appears that Google Tag Manager is loaded Asynchronously. Still, is there a performance hit on pages using GTM? – Trebor Mar 5 at 15:38
  • You don't need to use Google tag manager for events. It can be done that way, but I've never used it for that purpose. I've always implemented events with JavaScript loaded in the normal js way. – Stephen Ostermiller Mar 5 at 17:52
2

The bounce rate, generally becomes low when any of the following changes are done:

  1. Better Landing Page Experience
  2. Content served to the user is highly valuable compared to others
  3. No changes on-page are made but traffic from a low quality source, stopped coming.

To understand why, check the following with your Analytics and colleague:

  • Any changes to landing page/content?
  • Any changes to the sources of visitors to your webpages?

Is a bounce rate of 12% is not normal?

This totally depends on your website category and traffic sources. For my website, I'm maintaining less than your 12% bounce rate, which not a lot of people could achieve. (The % is not relevant as it also has to be accounted with the no. of users that visit the webpage in a particular period.)

1

Is that an average bounce rate over different pages/sources?

Usually if it drops to zero it is due to double pageview events firing but since you are showing a consistent 12% I imagine that the tracking issue is only affecting some pages.

I like to use Tag Assistant Chrome add-on for debugging

1

Those all are good things, and of course all of us want data to look like that, no doubt :) , but do not forget that there is "other" side of medal. This can be result based on intercepted script/wrongly setup or double tracking, if someone accidentally implemented code twice, or if you have Both GTM and GA scripts on website code. Check those as well. I hope i am wrong but...

  • I would expect the bounce rate to be lower (almost zero) if GA is being called twice. If that were true there should also be a corresponding big jump in the page views. Certainly worth checking. – Stephen Ostermiller Mar 5 at 13:57
  • 1
    IF it is GTM over GA, not quite, i know odd thing, but that actually happened, bounce rate dropped on 5,6% avg. – Filozof666 Mar 6 at 6:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.