An Introduction to Web Analytics

There's no denying that this is the age of the internet. We have online shops to blogs to mails and to dating sites. Web 3.0 promises to enhance this experience by bringing in applications which are embedded into the websites. But as always, everything needs to be measured. Something that is not measurable is not tangible. The process of measuring metrics of websites is called "Web Analytics".
Web Analytics is, in general terms, the study of online activities of people visiting your website. These people are called "visitors" and the activities they perform are called as "Metrics". The most common activities include, how they found the website, what pages they looked at, what they downloaded and so on. Study of these metrics can improve the website and bring in more traffic.

Methods:

Even though there is no standard method for gathering the metrics, the most widely used ones are web server log file analysis and page tagging.

1. Web Server log file analysis:

Web Servers have always maintained logs of all their transactions. Analysis of these logs makes it possible to understand what type of people visited the website, how long they stayed, which was the most interesting page and so on.

Advantages:

1. All web servers maintain logs, so the web page code need not be changed to accommodate the JavaScript.
2. Works for all servers even if JavaScript is disabled on the visitor's browser.
3. The web servers contain data in standard format and not in any third party company's format.

2. Page tagging:

Page tagging was introduced after the web server log file analysis procedure became cumbersome with the expansion of the internet and the search engines. Page tagging uses cookies and JavaScript to tag a page and send the data back to the web analysis company. The JavaScript drops a cookie in the user's computer and the cookie is updated with all the activities the user does. Finally the cookie is sent to the company where it is analyzed.

Advantages:

1. Page tagging can report on many mouse events like onClick etc.
2. There is no need to collect and analyze bulky web logs.
3. Can be done by third party companies and there is no need for companies to invest in web log collection and analysis.

Web Analytics data is compared against some "Key Performance Indicators" to come up with results that can help improve the performance of the website.

Important terms and definitions:

1. Hit: A request for a page from a web server.
2. Page View: The number of times a particular web page is viewed.
3. Visit/Session: The number of visits from a uniquely identified visitor.
4. Unique Visitor: The uniquely identifiable visitor. A unique visitor is usually distinguished by the information in the cookie or by the userid of the visitor.
5. Impressions: An impression is an advertisement that loads on the screen.
6. Bounce Rate: The bounce rate is the rate when the visitor enters and exits the same page without visiting any other page.

There are a lot of companies which are professionally in the business of analyzing website data, the most popular being Omniture and comScore. Google Analytics offers an easy interface for setting up your website for analytics. Piwik is an open source software which can be easily setup to measure analytics on your page.

Due to lack of standards there is no set way for measuring the analytics data. As measuring web analytics gets popular and more and more industries go in for the procedure to get better results from their websites, there seems be a dire need for a standard way to measure the metrics.