# Articles in The 'Google Analytics' Tag

February 7 2008

## The difference between Bounce Rate and Exit Percentage

Greetings, and welcome to the new MoreVisibility Analytics and Site Intelligence Blog! My name is Joe Teixeira, and I’m the Manager of Web Intelligence here at MoreVisibility. I’ll be doing most of the posting here, but from time to time, some of my co-workers and colleagues will join in on the fun.

I was thinking of what my first blog post would be about, and I’ve decided to talk about one of the most common items that is brought up in Web Analytics discussions, especially for newcomers to Web Analytics. Most analytics packages, like Google Analytics, show a metric named “Bounce Rate”, and also show a metric named “Exit %”, or “Exit Percentage”. At first glance, these metrics may look very similar, and you may even interpret them to mean the same thing. However, they are two COMPLETELY separate metrics, calculated two entirely different ways.

First, lets define a “Bounce”. A “Bounce” is a single-page visit to your website. For example, John lands on your homepage, www.xyz.com, and leaves your site without visiting any other pages on your website – that’s a “Bounce”. The “Bounce Rate” is calculated by taking the total number of Bounces (to your website or a set of pages, depending on what you’re looking at), and dividing it by the total number of Visits (to your website or a set of pages, depending on what you’re looking at).

The Exit Percentage is calculated by taking the total number of Exits, and dividing them by the total number of Pageviews (Not Visits – Pageviews). The Exit Percentage doesn’t care whether or not any of the Pageviews were from visitors who viewed 1 page, or viewed 1,000 pages – it simply does the math, and prints it in your Web Analytics interface.

Usually after explaining this difference, the follow-up questions that I usually get are “So, which one do I look at? / Which one do I use?”

I have a pretty simple rule: “Never make any analysis based off of one metric or one statistic”. So, you should never think of “looking” or “using” one individual metric to make any kind of decisions – you should always look at the complete picture of your website’s data, and then go from there. However, every rule has an exception, and this one is no different. If there was any metric in Web Analytics that you could make a very strong argument for using by itself, without the support of any other metrics, it would be the Bounce Rate. Think about what the Bounce Rate is calculating – it’s calculating the percentage of visits to your website who viewed one page on your website, and then left. If the visitors to your website were engaged and reacted positively to your website when they first landed on it, wouldn’t you think they would at least visit another page on your site, instead of leaving it entirely? Especially if you are looking for people to interact and visit the other pages on your site?

With Exit Percentage, all you can say is “This Percentage of Exits happened from this page / this set of pages”, without separating visits that Bounced from visits that did not Bounce. It’s impossible to draw any conclusions or formulate any hypotheses from this, as you can from the Bounce Rate metric. Also, keep in mind that, at some point in time, a visitor to your website ultimately has to leave your website (unless they are some sort of android that can stay awake and on a website infinitely).

So, I would recommend for you to pay close attention to your Bounce Rate – especially if you have a website featuring multiple pages (not counting blogs or single informational pages). Of course, the lower the Bounce Rate, the better, and the more engaged visitors are with your website. I can’t give you any kind of solid figure or benchmark on what your Bounce Rate should look like, but chances are that if 2 out of every 3 visits to your website are Bouncing, you may have a big problem that requires your immediate attention. If 1 out of three (or less) are Bouncing, chances are probably pretty good that you’re doing something right.

October 23 2007

## Google Analytics Cranks Out Yet Another Update

After five months of getting used to the new analytics interface, Google figured it was time to shake things up again. The following three changes are on the horizon:

Site Search – Expanding upon current Site Search capabilities, we will now be able to get data on keywords, categories, and products across time and user segments.

Event Tracking — With this addition, you will now be able to tag and track Flash and Ajax events. I can tell you from experience how difficult it is to really track traffic from Flash sites.

Tagless Outbound Link Tacking — This feature will allow for users to track their exit links (links that visitors clicked on that take them to another website).

Although these features will start in beta, many in the industry see this as a pioneer move for Google to set themselves apart from their competitors, making their’s the metric tool of choice. This just goes to show how vital it is to incorporate Analytics into your Search Engine Marketing (SEM) Campaigns. With the data this program can provide, there is no end to the value it can bring with your SEM efforts!

July 30 2007

## Microsoft Sprung a Leak

The secret is out: Microsoft is developing an analytics tool for Adcenter that will compete directly with Google Analytics. The information was leaked a few days ago by Dave Naylor, who learned of the analytics tool (codenamed “Gatineau”) at a Microsoft briefing in London. Somehow he got a hold of some screen shots and published them on his blog. This prompted a blog response from Ian Thomas, a Microsoft rep involved in Gatineau’s development and transition to the market. According to Thomas, Microsoft was about to make a public announcement about the tool when the leak occurred.