No matter what solution you use, industry standards and Internet interaction have established the three fundamental building blocks of Page Views, Visits, and Visitors. In this article, I take a look at each, comparing and contrasting their subtle and sublime properties (I assume a standard JavaScript tracking methodology).

Page Views

Example: You just loaded this article
Page views are most likely the easiest concept to understand for anyone working in Web Analytics. Most tracking sources use, but are not limited to, a javascript function that generates the "image" request that sends data to your tracking solution. An image request is generally a 1x1 transparent pixel image that contains loads of tracking information in the URL of the image request (the src attribute in the img tag). The tracking server is set to accept any URL given to it and always returns the transparent image that loads on a page. Each page must have an image request for a page view to occur.

With the rise of AJAX in recent times, the concept of a page view becomes somewhat distorted and complicated. The decision, in this case, lies upon the webmaster to determine what should be considered a new page as the content changes in varying degrees and forms.

Caveats: Because the web is filled with all sorts of traffic and an infinite variety of software is used to read that traffic, something as simple as an image request to generate a page view has exceptions. For example, there are many browsers that have JavaScript disabled or other high-security features enabled. This would negate tracking those page views. Text-based browsers that do not load images can also cause the same problem.

Visits

Example: When you came to my site, you opened a session with my web server in order to persist data across multiple page views.

The relationship between Page Views and Visits is many to one (*:1). The individual page views are linked together by a session that is established between you and the tracking server. These sessions generally remain open for only thirty minutes of inactivity, or in other words, the visit will automatically be concluded after 30 minutes of not contacting the server for a new web page or other data.

Caveats: Again in this instance, security measures and differing technologies often cause problems in tracking a Visit and visit-related metrics.

1. Many browsers block third party cookies and sessions from ever being established, and since most tracking solutions utilize a third-party tracking server, the visit is never established, even though the page views are still captured. This would result in several page views being attributed to ZERO visits, and it happens more often than most people think. The solution here is to use first-party cookies, if your solution allows it, so security settings will not block the session.
2. Some browsers simply are not capable of establishing the session. For instance, many browsers found on cheaper cell phones have no cookie handling ability. This excludes most smartphones, of course, but that really makes up only a limited percentage of the population.
3. Overstated visit counts occur when people leave the browser open on a page on your site for longer than 30 minutes. If they come back and then click on a link on the page, then a second visit will have started, even though it may have been in the course of one "visit" to the site.

Because of the browser, tracking could not establish a session (Visit) for this page.

Visitors

Example: No matter how many times you come back to this site, you are still the same visitor as before, thus all of your page views and all of your visits are attributed to you as a Visitor.

The relationship between Visits and Visitors is also many to one (*:1). In an ideal world, persistent cookies can be set on every browser, giving the visitor a unique ID # that helps your tracking solution to distinguish between you and other visitors to the site.

The nice thing about tracking Visitors is that there is really no way to completely mess up this part simply because there are basic requirements for loading a web page, such as an IP address to send the page to. Thus, even if the browser disallows persistent cookies, advanced tracking solutions (i.e. SiteCatalyst) will use a combination of the IP address and user-agent string of the browser in order to create a pseudo-unique value to track the visitor by. That is why the above shown example has one page view and one visitor, but no visits. The backup method to tracking visitors was used in this case. Hopefully this type of traffic to your site is rare, though.

Conclusion

There are exceptions to every rule when it comes to the Internet. As I said in my introductory post on analytics, there are far too many unknowns to expect a clear-cut number. Rather, focus on the trends and understand what you are working with when you look at these three building blocks of analytics. They will become powerful bases for segmentation work and analysis.

Remember:
Page Views tell you how compelling your site is and how well it keeps people there.

Visits reveal the depth of impression on your visitors and tell you how much they desire to come back.

Visitors provide a base metric and segment to compare your other blocks to and provide ratios. For example, a Visits-to-Visitors ratio shows you the average number of visits each visitor makes to your site. A ratio close to one is a bad sign in that you are not receiving return visits.