One important aspect of web marketing is an understanding of web stats, what they are and how to interpret them.

Web stats in their raw form consist of a computer files, log files, stored on your web server where details about each visit to your website are recorded, every visiting web browser and every file they request is recorded in the logs.

Raw web logs, not terribly informative to the human eye

There are many different software packages, both commercial and open source, that help you mine useful information from the jumble of data that makes up raw web stats.

Properly analysed web logs can tell you: How many visitors your site is getting, what pages they are looking at, what time of day they come, whether they found you through a search engine and what keywords they searched for, how long they stay, what browsing software they’re using and what country they are browsing from.

Processed weblogs, in this case from the analysis program AWStats

This type of information can be hugely useful if you want to dig into how your website is used and who your visitors are. For example, say AcmeWidgets is running a contest to give away one of their top of the line $10,000 products, in return they get the contestants email address for future marketing campaigns. Using information gained from web stats analysis AcmeWidgets can see that of the five thousand people who visit their site every month only ten percent visit the contest page and of those only a small fraction actually fill in the form to enter the contest. Armed with this information AcmeWidgets might make the contest information more prominent on the website and promise to only send a maximum of one email per month to the people who do enter the contest, on the assumption that people generally aren’t aware of the contest and that those who are aware of it are nervous about their email account being spammed.

Acting on the information that you can gain from web stats is still a fine art, there is lots of assumption and conjecture involved in knowing why your visitors do what they do. But web stats do provide a solid, if imperfect, base from which to start this type of analysis and to improve the effectiveness of your website.