Web Statistics

Articles


Syndicate content

Web Statistics

Written by admin
Bookmark and Share

Web statistics track the usage of specific websites, both in terms of overall popularity and how surfers are using the sites. The traffic a website attracts (nowadays in the tens of millions for the most popular destinations) is the net's version of television ratings. Web statistics evaluate the amount of traffic to a site and other estimates of popularity, like time spent surfing the website. They are also a critical tool in analyzing how to improve the design of a website to increase sales or attract more loyal surfers.

Commercial Applications of Web Statistics

As the commercial value of the Internet continues to explode, the need for accurate tracking methods also grows in importance, in particular to gauge the value of online advertising. Few can miss the profusion of pop-ups and flashing banner ads littering the browser window. What, perhaps unfortunately, props this system up is the internet traffic generated by clicking these ads and being redirected to the advertiser's website. Traffic is not the only goal, however, and other means are necessary to see if these visitors are making the advertising worthwhile.

Various approaches have thus cropped up to more precisely give a picture of a site's popularity and commercial value. The oldest methods, typified by web counters, simply counted the total number of visitors or page requests, a statistic which is now mainly useful to gauge the popularity or reach of news and opinion sites. Commercial enterprises are far more interested in detailed statistics about user behavior, especially whether they buy anything after being directed to the site, a subset of web analytics known as conversion tracking.

Methods for Gathering Web Statistics

Surely if you were shopping at a shoe store in a new town, you would be put off if the clerk automatically knew where you had moved from, what type of shoes you like, and how likely it is you'll purchase something. Everyone enjoys their privacy, which is why the more invasive methods of generating web statistics are increasingly annoying internet users. Some web users are even being driven to buy software programs that block data-gathering programs used by websites.

The methods for gathering web statistics are as varied as the types of data being sought, but fall broadly into two types. Logfile analysis is the older and less invasive type, relying on searches through a web server's own record of web requests. The other major strategy, often called page tagging or website tracking, follows a specific user's computer, usually by saving a cookie to the user's hard drive. It is this unwanted injection, which can be used to sniff out a great deal of private information, that raises many objections to the privacy-invading style of collecting web statistics.

Hundreds of software companies worldwide have begun offering web statistics services. Some offer free software, such as simple web counters, in return for the free advertising from showing the company name. Other software is offered as a one-time purchase, usually for simple applications that do not rely on cookies. More complex statistical work can be done for a continuing fee, much like private consulting firms do for traditional business. Another major service related to web statistics is site analysis, which uses web statistics to analyze the effectiveness of a site's design and then seeks to improve upon it.

Spiders

Complications for accurate statistics have arisen over the widening use of spiders. These are software programs, most-often used by major search engines, that crawl through cyberspace recording the variety of pages and links. The information gathered forms the basis of their search results.

Spiders ferreting through a web server can artificially raise the amount of traffic recorded, despite no actual users visiting the site. Several techniques have been developed to recognize spiders and thus fine-tune a website's data. In particular, website trackers and other page tagging methods can usually recognize such spiders and would not record their visits. Sites wanting to know about spider visits can then compare those statistics to their logfile.



Bookmark and Share