When it comes to Internet audience share, a prominent statistician examining www traffic responded, "What you are trying to do is like trying to count the stars in the sky."
Traditional Methods
Traditional audience metrics were fairly simple. For instance, a newspaper publisher would count the number of subscriptions to get a basic number of readers. But, what about a newspaper that is delivered to a library? Or, how about a magazine that is delivered to a dentist's office? Because of these types of variables, algorithms were developed to estimate viewership.
Modern Methods
In the realm of traditional media, calculating an ads "reach" became fairly standardized. However, as media became more interactive, calculating audience metrics became much more complex. In fact, it may be fraudulant for a business to represent that it can provide audience share numbers, advertising metrics, Internet traffic statistics or any other "eyeball" counts.
The Hit List
The following is an abbreviated list of terms that need to be understood before trying to measure a website's traffic:
- Hits
- Re-distribution
- Robots, crawlers and spiders
- Search engine results
- Proxy servers
- Caching
- Mirrors
- "Snapshots" of the web (such as, archive.org)
- Cookies
- Java and scripting
- Unique viewers
- IP used as Intellectual Property
- IP used as Internet Protocol address
- Domain Name Service (DNS)
- Decentralized domain names
- Hypertext Markup Language (html)
- File extensions, such as, .html, .htm, .mp3, .ogg, .mpg, .avi, .jpg and .gif
- Uniform Resource Locater (URL) [also known as Universal Resource Locater]
- Hypertext Transfer Protocol (http)
Does Counting Hits Count?
Given the complexity of counting hits, is there any value in attempting to measure traffic? Yes. There can be some merit to audience metrics. A highly trained team can draw some conclusions from a traffic analysis. However, the value has less to do with "sales" then it has to do with "savings". The mad robot is an example. By studying the traffic created by the mad robot, it could be shut down. By shutting down the mad robot, more real "eyeballs" could reach the site. In the process, a huge cost savings was realized through reduced bandwidth and hardware.
On the other hand, the best method for determining hits isn't counting the number of files pulled from a server. The best method for measuring hits is an advertisement's effectiveness.
That is the beauty of the world wide web. Unlike traditional media, the web is an interactive medium. That interactiveness is much easier to measure and results in increased business.
Conclusion
If the interest is in sales, the audience share that a website posses is much less important than the effectiveness of the site. If you want to increase the effectiveness of a website, forget about the number of hits and find a knowledge based team to help you improve sales.