Dot Commentary: Statistically Speaking

As the editor of PF Online, some of the questions I hear most often have to do with the importance of various statistics on the Internet: "How many hits should my web site get this year?," "What's the difference between a hit and a page view?," and "How many unique visitors should I have in a given month?"

Your wait is over. At long last, I'm going to put an end to all of the speculation about Internet traffic statistics. My best advice is to...

FUGGEDABOUTIT! (apologies to the producers of "Donnie Brasco.")

The fact of the matter is that most people tend to overvalue Internet traffic statistics. Using hits and page views as a measurement of your site's success tells only a small-and oftentimes erroneous-part of the story. I cringe whenever I hear somebody make a claim like "our web site received over one million hits last year!" Okay, so you got a million hits . . . how nice for you. But what does this actually mean?

Because the web is such a young medium, hits and other units of measurement are still very poorly defined and subject to more interpretation than a hanging chad in Florida. Additionally, many tracking software programs have very different ideas of what exactly constitutes a hit. For this reason, it's nearly impossible to compare your site's traffic to that of a competitor.

A greater issue is that hits and page views are strictly a quantitative analysis of a web site's performance. Sure, they might give you a rough idea of how much traffic your site is getting, but they do not answer many of the really important questions: Are users finding the information they need on your site or are they walking away frustrated? Is your site playing any kind of a role in the user's buying cycle? Are your visitors even legitimate prospects, or just some schmucks who stumbled on to your site while looking for sexycoeds.com?

For me, the single most important indicator of a site's success is user feedback. The handful of e-mails that PF Online receives each month tendering kudos, offering suggestions or describing a usability problem speaks volumes more to me than any set of numbers ever could.

That said, do I think that statistics are inherently bad? Of course not. When used responsibly and in conjunction with other factors such as user feedback, numbers can help tell a powerful story about your site. The point is to make sure that you are citing the right numbers, and to keep in mind that they are just one factor to be considered when gaging a web site's success.

The Numbers Game: So What's What?

Not exactly sure what's what anymore? Join the club. Here are some of the more commonly used (and misused) forms of Internet traffic measurement, as found on Webopedia.com:

hits: The retrieval of any item, like a page or a graphic, from a Web server. For example, when a visitor calls up a Web page with four graphics, that's five hits, one for the page and four for the graphics. For this reason, hits often aren't a good indication of Web traffic.

page views: The accessing of a Web page. Often used by sites to give advertisers a sense of traffic, a page view differs from a hit by counting only the number of times a page has been accessed. Page views, however, have become tougher and tougher to gage, since pages can include frames that divide them into separate parts.

unique visitor: When tracking the amount of traffic on a web site, it refers to a person who visits a web site more than once within a specified period of time. Software that tracks and counts web site traffic can distinguish between visitors who only visit the site once and unique visitors who return to the site. Different from a site's hits or page views-which are measured by the number of files that are requested from a site-unique visitors are measured according to their unique IP addresses, which are like online fingerprints, and unique visitors are counted only once no matter how many times they visit the site. There are some ISPs that use Dynamic Host Configuration Protocol, such as AOL and cable modem providers, which use different IPs for every file requested, making one visitor look like many. In this case, a single IP address does not indicate a unique visitor.