Better Know a Web Metric : the Page View

Stephen ColbertWelcome to the first installment in the 10-part series Better Know a Web Metric!

In this feature, I will try and explain Web analytics terms and what they’re all about.

In this week’s episode, we will explore… the Page View!

A quick history

Long before Web 2.0 was a vague idea, when HTML was the bastard child of SGML and other structured document formats, Sir Tim introduced the world to a whole new concept : “Web pages”, electronic documents organised in thematic collections called “Web sites”.

Of course the activity of these so-called Web sites had to be tracked, with new methods to process or ‘parse’ the web server’s access and error logs.

Then came the time of the first Web analytics vendors, providing mostly Web server log analysis. The good thing about log analysis is that it provided more details: what web browser was used, which page element was donloaded, etc… Suddenly Web site owners realized that out of the 10000 ‘hits’ they registered for a given site, 3/4 of those hits really concerned the various elements linked to the site’s pages: images, stylesheets, Javascript files, etc. Page reloads were not included either. So it all boiled down to knowing which pages were seen the most.

The advent of “personal home pages” introduced alternative tracking methods, generally consisting in displaying the number of ‘hits’ the Web site received, usually in the form of a counter image or a text message. These counters were usually based on a CGI script that incremented a value in a text file every time the page was requested from the web server. Nothing fancy but it was efficient… at the time!

Then, around 1995, Javascript came along and changed everything with session cookies, privacy concerns and what have you. Using Javascript-based tracking methods would ensure that the tracking code be executed every time a given Web page is viewed, even if the page had already been loaded into the Web browser’s cache. This is the moment when the notion of page view appears.

Definition

A page view is now considered by the Web analytics industry as the number of times a document was viewed in a Web browser.

Example

Let us take the arbitrary example of my favorite fictional sports site, Foosports.com. I usually start browsing their home page, which has links to the hockey section. The hockey section then takes me to a third page, which displays the roster of the FooVille Badgers (go, Badgers!). I eventually grow bored with the Badgers’ roster and head back to the home page. On second thought, it’s getting late and I switch my computer off.

How do we describe my recent fictional navigation in terms of page views?

  • Foosports.com home page: 1 page view, browser caches it
  • Hockey section: 1 page view, browser caches it
  • FooVille Badgers’ roster page: 1 page view, browser caches it
  • Return to Foosports.com home page: 1 page view but already in browser cache

In this scenario, a log-based Web analytics tool will evaluate this visit as 3 page views. A Javascript-based solution will evaluate it at 4 page views because it ignores the browser’s cache. Both methods are correct but be sure you know which method is used before you come yelling at your favorite Web analytics consultant that numbers dont match 🙂

Considerations

I’d like to make a few adjustments to that statement. Now that the content that is being used on Web sites is generally structured, it generally means that content can be pushed on to platforms other than web browsers, e.g. mobile phones, pocket PCs, Web TV, etc… Do you decide to exclude that content altogether? This is a typical business decision, but excluding page views generated by other media usually results in the multiplication of tracking entities, one for each separate media platform. Alternatively, do you track all page views regardless, but break the total down by media used?

Then there is also the issue of Flash applications, which I would exclude from page views, unless you want to use Flash like a banner and track the number of impressions your Flash app. received. In my humble opinion, Flash has its own paradigm and it is more centered on usability rather than on the number of impressions.
Discuss 🙂

Conclusion

It’s time to wrap up for this first installment of Better Know a Web Metric. I hope you learned something today or at least had as much fun reading it as I had writing it.

In closing, I would like to say that page views are about quantity, not quality.

Remember that page views are about volume, about your Web site’s output, if you will. You will notice that this output fluctuates and follows a plethora of factors that we might cover in later episodes.

In the meantime, happy surfing and feel free to drop me a comment 🙂

Author: Julien Coquet

Expert de la mesure d’audience sur Internet depuis plus de 15 ans, Julien Coquet est consultant senior digital analytics et responsable produit et évangélisation pour Hub’Scan, une solution d’assurance qualité du marquage analytics. > A propos de Julien Coquet

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Better Know a Web Metric : the Page View

Stephen ColbertWelcome to the first installment in the 10-part series Better Know a Web Metric!

In this feature, I will try and explain Web analytics terms and what they’re all about.

In this week’s episode, we will explore… the Page View!

A quick history

Long before Web 2.0 was a vague idea, when HTML was the bastard child of SGML and other structured document formats, Sir Tim introduced the world to a whole new concept : “Web pages”, electronic documents organised in thematic collections called “Web sites”.

Of course the activity of these so-called Web sites had to be tracked, with new methods to process or ‘parse’ the web server’s access and error logs.

Then came the time of the first Web analytics vendors, providing mostly Web server log analysis. The good thing about log analysis is that it provided more details: what web browser was used, which page element was donloaded, etc… Suddenly Web site owners realized that out of the 10000 ‘hits’ they registered for a given site, 3/4 of those hits really concerned the various elements linked to the site’s pages: images, stylesheets, Javascript files, etc. Page reloads were not included either. So it all boiled down to knowing which pages were seen the most.

The advent of “personal home pages” introduced alternative tracking methods, generally consisting in displaying the number of ‘hits’ the Web site received, usually in the form of a counter image or a text message. These counters were usually based on a CGI script that incremented a value in a text file every time the page was requested from the web server. Nothing fancy but it was efficient… at the time!

Then, around 1995, Javascript came along and changed everything with session cookies, privacy concerns and what have you. Using Javascript-based tracking methods would ensure that the tracking code be executed every time a given Web page is viewed, even if the page had already been loaded into the Web browser’s cache. This is the moment when the notion of page view appears.

Definition

A page view is now considered by the Web analytics industry as the number of times a document was viewed in a Web browser.

Example

Let us take the arbitrary example of my favorite fictional sports site, Foosports.com. I usually start browsing their home page, which has links to the hockey section. The hockey section then takes me to a third page, which displays the roster of the FooVille Badgers (go, Badgers!). I eventually grow bored with the Badgers’ roster and head back to the home page. On second thought, it’s getting late and I switch my computer off.

How do we describe my recent fictional navigation in terms of page views?

  • Foosports.com home page: 1 page view, browser caches it
  • Hockey section: 1 page view, browser caches it
  • FooVille Badgers’ roster page: 1 page view, browser caches it
  • Return to Foosports.com home page: 1 page view but already in browser cache

In this scenario, a log-based Web analytics tool will evaluate this visit as 3 page views. A Javascript-based solution will evaluate it at 4 page views because it ignores the browser’s cache. Both methods are correct but be sure you know which method is used before you come yelling at your favorite Web analytics consultant that numbers dont match 🙂

Considerations

I’d like to make a few adjustments to that statement. Now that the content that is being used on Web sites is generally structured, it generally means that content can be pushed on to platforms other than web browsers, e.g. mobile phones, pocket PCs, Web TV, etc… Do you decide to exclude that content altogether? This is a typical business decision, but excluding page views generated by other media usually results in the multiplication of tracking entities, one for each separate media platform. Alternatively, do you track all page views regardless, but break the total down by media used?

Then there is also the issue of Flash applications, which I would exclude from page views, unless you want to use Flash like a banner and track the number of impressions your Flash app. received. In my humble opinion, Flash has its own paradigm and it is more centered on usability rather than on the number of impressions.
Discuss 🙂

Conclusion

It’s time to wrap up for this first installment of Better Know a Web Metric. I hope you learned something today or at least had as much fun reading it as I had writing it.

In closing, I would like to say that page views are about quantity, not quality.

Remember that page views are about volume, about your Web site’s output, if you will. You will notice that this output fluctuates and follows a plethora of factors that we might cover in later episodes.

In the meantime, happy surfing and feel free to drop me a comment 🙂

Author: Julien Coquet

Expert de la mesure d’audience sur Internet depuis plus de 15 ans, Julien Coquet est consultant senior digital analytics et responsable produit et évangélisation pour Hub’Scan, une solution d’assurance qualité du marquage analytics. > A propos de Julien Coquet

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.