A great deal of money is allocated to the setting up and maintenance of websites and to advertising campaigns run over the net so, obviously, advertisers want to know that their investment is being rewarded. As a result there has been a growing demand for sophisticated analytic software.

Web analytics refers to the analysis and interpretation of data collected from a website to provide important insights into the demographics of the visitors and the effectiveness of the website itself.

This data comprises reports on traffic to the website, including returning visitors; time spent on the site; amount spent by individual customers; email response rates and geographical information about the visitors. Web analytic software can also track the ‘click through’ behaviour of users, pinpointing the sites from which they access the website.

Interpretation of the data depends upon the requirements of the user but includes predicting which customers will repurchase a product, which products customers are most likely to buy in the future and which target market specific products should be aimed at.

In other words, the data is compared against the key performance indicators for the site and used as part of the marketing strategy to attract more visitors or increase the amount spent by existing customers.

Typically data is collected using one of two main technological approaches: logfile analysis or page tagging.

The advantage of logfile analysis is that the web server automatically records all transactions in log files which means that the data is readily available, on the company’s own servers and in a standard format. Because this information is easy to access, logfile analysis requires only an initial purchase of software and is usually carried out internally.

Although logfile analysis has become increasingly sophisticated, there are still problems associated with this method of data collection. One of these concerns is the use of web caches which means that very often a second visit to a webpage by an individual will go through the browser’s cache instead of the request being received, and logged, by the web server.

Because caching compromises the accuracy of logfile analysis page tagging was developed as a data collection method.

Page tagging enables the web server to provide a reliable record of each transaction but the website has to be changed to accommodate it. This method also relies on the cooperation of the visitors’ browsers. Cookies are used to identify individual visitors and JavaScript, which runs automatically, provides information about visits to a web page without the problem of caching. JavaScript can also be made to acquire additional information about the visitor to the webpage if this is required by a client.

The data can then be collected by a remote server. This means that, although data collection and analysis can be carried out in-house, it is more commonly outsourced and, for a monthly fee, web analytic companies are able to process the data and generate statistics for their clients even if they do not have their own web servers.

The approach chosen would be determined by economic considerations and the advertiser’s requirements but, given the expenditure involved, few companies can afford to remain ignorant about their return on investment where internet advertising campaigns are concerned.