Parting the Fog of Visitors

Imagine this. You're the publisher of a metropolitan newspaper, tasked with gathering statistics on how many people read your publication on a weekly basis. Of course you count your subscriber base and your newsstand copies sold, but do you include those folks who only glanced at the newspaper headline as they passed by the newspaper box? If you're a web designer or marketer who's counting every visitor to your site, even those who hit the home page then retreat, you're basically doing just that.

Your newspapers advertisers, not to mention the newspaper's auditors, would take exception if you tried to tell them that those people who simply glance at the headline 'count.' Certainly they matter if the concern is trying to figure out why a certain demographic isn't purchasing the paper, or if you're trying out a new design to see if it's more inviting and appealing. But what matters more in this case are the people who have the opportunity to delve deeper into the paper, reading the op-ed section and the entertainment pages, looking at the advertisements.

This notion of 'readers who matter' can easily carry over to the web, translating into 'visitors that matter.' Face it. Regarding your website, at any given time measuring any given component, there are simply visitors who create noise and distraction from the visitors that are actually doing something. Don't get me wrong; I'm not saying that you should take your log files and simply dump information on anyone who didn't buy a product or sign up for a newsletter. What I am saying is that depending on what you're looking for, and no matter what you're looking for, it's easier to see what you're doing if you don't have a fog of visitors unrelated to that task clouding your vision.

Grouping Visitors is Key
How do we get past this noise pollution? If you look at the overall behavior of all of your web visitors, you'll be inundated with a ton of information from your log files. Trying to discern and identify patterns amongst such a large group is extremely difficult, if not impossible. If you segment or divide your users into logical groups and then compare and contrast how they behave and what they do, you'll have more accurate, actionable information that actually means something.

Segmenting your site visitors allows you to develop a clearer overall picture of your site's performance. If the metric you're trying to measure is how many people buy a particular product, it makes sense to measure that within the context of how many people interacted with the site in a meaningful way, period. A real world example of this would be if you were a supermarket manager wanting to determine the effectiveness of a particular product promotion: doesn't it make more sense to compare and contrast the number of people who bought the product versus the number of people who put the product in their shopping cart, then left that shopping cart in the aisle? What value would looking at shoppers who simply walked in the front door, turned around and walked back out provide? None. Why include those shoppers in your promotion's overall picture? Don't.

Segmenting Visitors Pinpoints Successes-and Weaknesses
Another example of the benefit of segmenting and grouping visitors can be found in analyzing traffic from PPC campaigns versus traffic from search engine results. Both may come from Google, but they may both have different goals. The PPC campaign goal, for example, may be to get visitors to sign up for a free weekly newsletter, while the search engine goals may be something more vague like keeping visitors on the site for a certain amount of time. By segmenting these visitors into two specific groups, and temporarily discarding information about 'all site visitors,' web designers, marketers and SEO folks can more accurately pinpoint the success or shortcomings in a particular offering. Are your PPC visitors more or less likely to sign up for that weekly newsletter than the 'average' Google visitor? Dividing these groups can help you find out.

Grouping visitors also allows you to better test campaigns' effectiveness. Let's use the example of the PPC versus search engine result visitor. Say that during the first week, 15% of PPC traffic signed up for the newsletter, and only 3% of search engine visitors did. Is the copy on your PPC placement a direct call to action, leading visitors to your desired behavior? Change the copy in your ad, and wait a week. Regroup your visitors and reanalyze your results. Did your PPC traffic newsletter signups increase? Did your campaign deliver the information/action promised to visitors in your ad copy? Fantastic. Did they decrease? Did visitors come into the site via the PPC campaign, then turn right back around and head out? Try again.

The bottom line is that, at any given point in time, not every visitor matters. In fact, I can't think of one occasion where measuring traffic as a whole would provide useful, actionable results. By simultaneously eliminating the people who don't matter and looking more closely at the ones who do AS A PERCENTAGE OF THE REMAINDER, you'll be better able to make site and campaign changes that result in a larger return on investment and a better user experience.