Do you use Google Analytics on your blog to track your website traffic statistics? If so, have you ever looked at your bounce rate there and wondered "why is this so high?" The problem might have very little to do with your blog and have more to do with the fact that the typical bounce rate metric isn't always relevant to the blog format. Fortunately you can tweak Google Analytics to give you even more useful data. But first....
What is a Bounce Rate?
When it comes to Web traffic, your bounce rate (according to Google Analytics among others) is the percentage of visits that last for only one page. The visitor comes to your site, looks at the page they landed on, and then leaves without clicking deeper into your blog.
That said, bounce rates are frequently debated. Not everyone thinks a single pageview visit should count as bounced traffic. Instead you would look at "true bounces" -- people who visit the site and almost immediately leave, without spending enough time on the site to digest any of the content.
Google Analytics goes with the former definition. I consider the latter to be far more useful, especially for bloggers. Let me explain why.
How Bounce Rates Sometimes Mislead
The point of a bounce rate is to show you which pages on your website are ineffective -- they drive the visitor away rather than keeping them on-site. Of course there are exceptions. For example:
- If someone lands on a product page of a site and immediately buys the product, the bounce rate for the page could be high, but it would be a good thing.
- If someone comes to your site via a search query to have a specific question answered, and your article does answer that question, the bounce rate could be high and it doesn't reflect poorly on the quality of the content on that page.
- You want people to leave your site from a certain page, such as clicking on a specific ad or link to another site or resource (directing traffic from one site to another you own for example). In that case the bounce rate could be high, and that would also be a good thing.
At the same time, a low bounce rate isn't always a sign of a high quality site. For example, some outdated or poorly designed sites still use those annoying intro pages that make you "click here" to finally enter the site. Because every visitor to that home page has to click the link to move on (and many will), the bounce rate could be very low for that page. But it has nothing at all to do with the site offering quality material.
Why Single Pageview Visits Aren't "Bounces" for Many Blogs
The same kind of thing can happen with blogs, where you'll get a low bounce rate for behavior that basically inconveniences readers as opposed to a low bounce rate because your content compels them to dig deeper. For example, you might have a blog where only excerpts are posted on the front page, forcing those visitors to click to view a full blog post.
That artificially affects your bounce rate (as far as Google Analytics considers bounce rates at least). It makes your page look more effective than it might actually be. Sometimes people do this solely to increase pageviews (a bad idea in my opinion). Other times it has to do with aesthetic preferences or design constraints (less of an issue, although it can still annoy readers -- in full disclosure I have sites that use excerpts and others will full posts, and the full post sites actually get significantly more traffic).
Now consider the other side of that issue -- the many blogs that publish not just one full post on their homepage, but several full posts.
Your content can be compelling enough to get people to read through several full articles. But if they read the five or even ten on your homepage and then leave, Google Analytics still considers that bounced traffic. Um, no. It's a faulty metric.
How to Get More Useful Bounce Rate Data
In the case of making full posts viewable on a blog's homepage, the single pageview definition of a bounce is completely irrelevant. What you really need to know is the percentage of visits to the page that last no more than a few seconds -- people who come to the page but are quickly driven away for some reason. That's how you'll know something's wrong.
I looked for some kind of solution to this -- a way to make my Analytics bounce rate information more applicable to the blog format I use, where a single pageview isn't a negative measurement when several full articles are available per page. I came across an article on PadiCode that shows you how to do just that (make sure you look at the "update" section there before doing anything).
In short you would simply open your template file that has your Google Analytics code and go below the last existing _gaq.push declaration. Then add this:
setTimeout('_gaq.push(['_trackEvent', 'NoBounce', 'Over 10 seconds'])',10000);
That will alter your bounce rate calculations and have the bounce rate reflect visitors who leave your site within the first 10 seconds of their visit -- a more realistic definition of a "bounce." You can change that time to something that suits you. You could change 10000 to 30000 if you wanted to track those leaving within 30 seconds for example.
Now you can get a better feel for which pages are driving visitors away, so you can work on improving your content quality, navigation, or other site elements to keep visitors on your blog. And you won't be misled by pageview stats that could really mean someone read several articles in full on your homepage, category pages, or other archive pages.
If you still want to know how many visits were only a single pageview in length, that information is still available too, so you don't lose anything. You'll simply find it in the "depth of visit" section in Google Analytics.