Does this bother you as much as it bothers me? In Google Analytics, when viewing traffic from specific keywords souces, I’m sure the 00:00:00 Avg. Time On Site has bothered you at one time or another. Here is an example of what I mean – the following shows some Organic Google search traffic to a Google Chrome oriented forum that I run:

I constantly wondered why – are they bots? Fake clicks? Search Engine Spiders? Is Google Analytics broken? Well now I know the answer and it’s all in the technical definition of a very important web metric called Bounce Rate.

To know how long a visitor spent on a page it needs two things:

1. The time you arrived on a specific page
2. The time you requested another page on the same site

Read that last element again… the time you requested another page on THE SAME SITE. You have Google Analytics installed on YOUR site but if they leave for another page by entering an URL, closing the browser, clicking on advertisement, etc… Google Analytics does NOT get triggered again and it cannot calculate the time on that page. So… it defaults to 00:00:00.

All your 00:00:00 visits are the visits increasing your Bounce Rate.

By definition, bounce rate is the percentage of visitors who view one page on your site and then, instead of requesting another page on your site, exit and either leave to another website or close their browser altogether. These are the same visits that Google Analytics cannot track the time of and therefore, all those 00:00:00 visits are the bounces.

This totally sucks.

First of all, everyone always talks about needing to lower your bounce rate. Some have even suggested Bounce Rate makes a difference in SERPs but Matt Cutts has pretty much denied any association between the Google Algo and Bounce Rate via a Sphinn article:

Without reading the article, I’ll just say that bounce rates would be not only spammable but noisy. A search industry person recently sent me some questions about how bounce rate is done at Google and I was like “Dude, I have no idea about any things like bounce rate. Why don’t you talk to this nice Google Analytics evangelist who knows about things like bounce rate?” I just don’t even run into people talking about this in my day-to-day life.

Rewarding sites/pages with low bounce rates with better search rankings would not only be COMPLETELY idiotic but incredibly counterproductive and counterintuitive. In my Google Chrome example above, suppose somebody is searching for “Divx plugin for Chrome”. I duplicated that Google Search and here are the results:

As you can see, ChromeSpot.com is ranked #2 – that is the site in question. Supposing someone clicks on it they are brought to the forum thread where a user provides a link to download a file which solves the problem of playing Divx in their Chrome browser. He also explains how to make it work. So the person may spend 20 minutes on the site, have their problem solved and then leave.

The page in question did its job 100% yet is punished with a 00:00:00 Average Time On Site and increased Bounce Rate.

Google Analytics Is Broken

This flaw has HUGE implications on the value I put in Google Analytics for two main metrics:

• Average Time On Site
• Bounce Rate

Think about this: not only the “Bounced” visits not counted towards your Average Time On Site (since it can’t be calculated) but the last page anyone visits before leaving your site can NEVER be calculated and therefore is NEVER included in your overall Average Time On Site Metric.

I understand the logistics of why this isn’t supported but Google should have SOME solution for figuring this problem out. For some sites, I’m sure the bounce rate is HUGE because people find exactly what they’re looking for RIGHT when they land on the site and spend a bunch of time consuming that information and are whisked off. The reason they are leaving isn’t necessarily because they found something better but because they’re satisfied and they’re problem has been solved!

For example, if I need some quick information I often do a search for “keyword wiki” so for example if I want to know some random fact about the country of Switzerland I might search Google for “Switzerland Wiki”. As expected, the Switzerland page on Wikipedia is the first result:

There is a TON of information on that result and I could be consuming it for literally HOURS. Think of a high schooler doing some homework or research paper on Switzerland. They spend 2 hours on the page and then leave. What do they get for their job well done? 00:00:00 Average Time On Site and Increased Bounce Rate. Even if you had entered at Wikipedia.com and searched around for the page, the page you wanted would also likely be the LAST page you visited and it’s also the ONLY ONE THAT DOESN’T COUNT TOWARDS YOUR ANALYTICS STATS.

I’m incredibly happy that I now know the answer behind all those mysterious 00:00:00 sitting on some of my bigger sites. I don’t feel nealy as bad about their existence or at my pretty high bounce rates. But surely this is a problem that Google needs to take a serious look at and come up with a solution for in Analytics. Because right now I’ve lost faith in a few web metrics that some would say are pretty important.

I wonder if there are any other ugly warts that Analytics is hiding? Share your thoughts on Analytics below – I’m eager to hear what type of insight you have!