Choosing the right web stats package
Web sites stats, also known as web analytics, are the process of studying behavior of visitors of web sites. This study is conducted by collecting data from a web site to determine which parts of the web site work towards the objective of the company or business. Businesses and companies are able to see which of their landing pages have encouraged web surfers to purchase. The data that is collected from the web sites usually include web traffic reports. Email response rates and direct mail campaigning is a popular marketing method for companies and with web site stats, the business is able to see the performance of those read and unread. Sales and lead information is also available in the reports of web analyses. Also available are user performance data (with click heat mapping or other custom metrics as needed). To help improve a web site, the data collected is compared to the key performance indicators.
There are two popular kinds of ways of collecting web site status information. One is with logfile analysis. This method uses the process of reading logfiles that web servers record during all of its transactions. The other way is with page tagging. This uses JavaScript on each page. It then notifies a third party server when pages are accessed by a web browser. Since the beginning, web servers have recorded all transactions made within a logfile. Later on, it came to the attention that a program could read the logfiles and provide data about how popular a web site is. This began the creation of web log analysis software.
By the early 90s, web site stats meant mostly counting the number of hits that were made to the web server. This way of doing things was initially a great idea because every web site usually had a single HTML file. Once images were produced in HTML, web sites began to span multiple HTML files, making the counting of hits not as useful. IPRO was the first true commercial Log Analyzer, created in 1994. To better accurately count the amount of human activity on web servers, two units of measure were produced. In the mid 1990s, page views and visits were created. Page views are when a request is made to the web server; and visits were a sequence of requests from each uniquely identified person; this usually expires after 30 minutes of inactivity. You may still see web sites that have visits and page views features, but are looked upon as unsophisticated measurements.
In the later portion of the 90s, search engine spiders and robots came about. Web proxies and dynamically assigned IP addresses (ISPs) were also developed. These made identifying unique human visitors to a web site much more difficult. Some of the log analyzers used cookies as a means of counting human visitors instead of requests from known spiders.
Problems arose with logfile analysis software when extensive uses of web caches were present. When a web surfer would enter a web site, the second request would be retrieved from the cache of the browser, but the web server wouldn’t receive any requests.