In my last post, I made some comments and observations around Avinash Kaushik's recent series of posts regarding a search for the single most useful web analytics report. Today, I'd like to bring it down a level and talk about one of the most useful raw metrics that most analytics programs track - Average Time On Site.

At first glance, average time on site doesn't seem like the most interesting piece of user behavior to be tracking. ROI or ROAS has got to be more important for making decisions and tracking the effectiveness of campaigns, right? Well, maybe. For any site that generates revenue, tracking the ROI of campaigns is a great way to focus in on which campaigns are effective. The thing is, though, ROI is almost too focused.

Only visitors who actually result in a conversion make it into the ROI calculation (thanks to John Marshall for many interesting talks on this subject). Any visitor who comes to your site because of a keyword search or email campaign who doesn't convert, gets lost. So, by paying too much heed to this metric, you are passing up on useful data about how your Web site's customers interact with your site and your advertising campaigns.

Here's where average time on site comes in. It's an egalitarian metric that applies to all visitors. Even if a set of keywords does not directly lead to conversions (and therefore not showing any positive ROI), it's possible some of those keywords may be attracting users who tend to spend more time on the site than others. These are important keywords. They are building brand awareness and pulling in visitors who are likely to return to your site. Try looking at the segment of new versus returning visitors on campaigns that have high average time on site but no conversions. That leads to another advantage of average time on site. Since it is a simple and easy to understand metric, it is also easy to understand when comparing several different segments.

Finally, when paired with ROI metrics, it acts as a second-opinion when comparing campaign performance. It is interesting to compare trends in ROI for sets of campaigns to trends in average time on site. Again, add a little segmentation into the mix and you begin to get a grasp on how visitors are interacting with your site and then how you can leverage this insight to make improvements.

So there you go. Average time on site. If you have any examples of insights gleaned from average time on site, or any metric for that matter, drop a line in the comments section.

Comments (0)

Subscribe to this comment's feed

Name

Email

Title

Comment

smaller | bigger

I have read and agree to the Terms of Usage.

Add Comment