I was discussing SEO with a colleague the other day, and an interesting thought occurred to me related to a few of Google's Analytics metrics - specifically bounce rate, time on page, and time on site. Before I talk about that thought however, I feel that I should offer a bit of context. See, I've a bit of a problem with how I use the Internet.
I'm sure at least a few of you can relate to it. I've this tendency - especially when I'm compiling research for a project or article - to rather absent-mindedly open every single page that catches my eye all at once. Thanks to Chrome's tabbed browsing, I'm able to leave all of them open, browsing each one at my leisure.
It doesn't stop there, either. Occasionally, while reading one article, I may see another on the same site that catches my eye. If I'm not finished reading the first one, I'll simply pop the new one into yet another tab.
This probably isn't the best habit, especially considering how cluttered a browser full of tabs tends to look, but I'm fairly certain I'm not alone in falling prey to it. I'd imagine that there exists a ton of users who do precisely the same thing that I do. There are probably plenty of users who leave a multitude of tabs open in their browser, potentially ignoring them for hours at a time while they occupy themselves with other matters.
Right. You've got the context. Let's get straight to the point, shall we?
My thought was a fairly simple one: in light of tabbed browsing, can time on page and bounce rate really be looked upon as wholly accurate metrics?
I mean, think about it. You've got people leaving a particular page of a website open for hours at a time in their browser, with little to no interaction. Even if they close the page without clicking on anything, that's bound to mess with your results.
And what about when they navigate to a new page in a separate tab?
A Complicated Answer For A Complicated Question
I've got good news and bad news. The good news is that tabbed browsing doesn't entirely invalidate time on page, time on site, or bounce rate. The bad news is that it makes calculating the three immensely more complicated, as it's far more difficult to get an accurate picture of how your visitors are using your site - or what path they're taking as they navigate through it.
"Google analytics has absolutely no clue about the fact that you are using multiple tabs," writes Cardinal Path's Stephane Hamel. "As long as the session is alive - by default 30 minutes between two web analytics "hits" - the session is still alive."
Analytics tools deal with this issue in a few different ways, according to digital marketing veteran Avinash Kaushik. Either they create a different session for each tab in the browser, or they lump all the tabs and hits together to construct a single session. As you've probably surmised, neither method is terribly ideal; each one skews the results in a different way.
In the case of the first, you're getting a ton of false positives, with potentially multiple hits on your site from the same visitor. In the case of the second, it's nigh impossible to tell how much time they spent on each specific page. Either way, getting an accurate measurement is an exercise in futility.
Bounce rate is similarly skewed by tabs, explains the Lost Saloon Blog. Since a large portion of bounce rate (and the closely-related exit rate) is tied to a user's navigation through a site, the use of browser tabs can make it extremely difficult to determine a user's path. Again, this means that any attempts to use the metric as a measurement of your site's success will likely hit a brick wall.
"When using tabbed browsing over multiple web pages from multiple websites at the same time, it is challenging to correctly measure the times or order of navigation," the blog reads, adding that "there may be variations on the implementation of session management logic among the browsers while using tabbed browsing."
So...bounce rate, time on page, and time on site aren't rendered entirely useless by browser tabs, but they're definitely muddied a whole lot. Taken by themselves, they won't give you a complete picture of how people are engaging with your site.
You'll need to combine them with a number of other metrics in order to use them effectively. (Of course, that's something you should be doing anyway).