Web Analytics (How It Works) Explained in 4 Minutes

By on in with 7 Comments

I was tinkering around a few weeks ago trying to figure out the best way to communicate an idea out to a group of people and hit on using Snagit to record me talking my way through a few PowerPoint slides that had some basic diagrams on them and then uploading the resulting video to YouTube (in that case, as a private video). It worked great — perfectly okay audio quality (I used a USB headset) and perfectly okay graphics. Lo-fi, but using the tools I already had at hand.

Below is an audio slideshow that uses the same approach to provide a very basic overview of how page tag-based web analytics tools work. If you’re a web analyst, I sincerely hope there is nothing new to you here. But, if you’re a web analyst who has repeatedly beaten your head against a brick wall when trying to explain to some marketers you work with that they need to put campaign tracking parameters on the links they use…maybe it’s a video you can send their way! It’s right at 4 minutes long, with a subtle-but-shameless suck-up to my favorite Irish web analyst at the 1:30 mark (it really never hurts to suck up to an Irish(wo)man, now, does it?).

The video is a much simplified overview of what I went into in greater detail in an earlier blog post.

If you’d like to download the slides (.pptx) for your own use (attribution appreciated but not required, and edit at will), you can do so here.

I’d love to hear what you think (of the format and/or of the content)!

Similar Posts:

7 Comments


  1. Pingback The Anatomy of a URL: Protocol, Hostname, Path, and Parameters | Gilligan on Data by Tim Wilson

  2. Pingback Do Omniture SiteCatalyst metrics correlate more with comScore than Webtrends or vice versa? - Quora

  3. Pingback » The Anatomy of a URL: Protocol, Hostname, Path, Parameters | Tim Wilson's Blog at Web Analytics Demystified

  4. great intro!
    why do we need jscript to grab data? why is this job not done on the server side? e.g. reading the client request header and the query string from the url? (e.g cmp=xyz)

    many thx
    mike

  5. That’s a great question, Mike! And, initially, that was how mainstream web analytics tools worked — they parsed the server log files and grabbed all of that data. Client-side (Javascript) solutions came along for two main reasons: 1) to capture browser-cached page views, and 2) to capture additional information that is *not* included in the http:// header.

    I actually wrote a post comparing the two methods…yikes!… 5 years ago. I’m sure there are more recent write-ups comparing them as well.

    http://www.gilliganondata.com/index.php/2008/01/02/capturing-web-traffic-data-two-methods-that-suck/

Leave your Comment


Notify me of followup comments via e-mail. You can also subscribe without commenting.

« »