How do you measure the effectiveness of your web site? Many people look at their statistics to see numbers of hits, unique visitors, geographical spread and so on. Whilst these are useful indicators, they do not provide a full indicator of effectiveness. Many companies provide a range of statistical analyses for your web site, but two reports just published suggest that the current way of measuring web sites is ineffective and needs to change. One of the problems is that much web site counting depends on "cookies" - little pieces of text that are automatically downloaded to a reader's computer. The statistical programs can count the cookies and collect information which helps build up the data you look at. The problem is that many people reject cookies or clear out their cookie collection - indeed modern browsers allow you to do this automatically; plus there are several "washing" programs that delete such "Internet junk" from your PC. The result is that web sites are over-counting cookies. If you cleared out the cookie from my web site and then re-visited, you would get another cookie and be counted as "unique" when in fact you are not. Consequently, many web statistics that people look at are actually misleading. The industry is discussing new ways of measuring web site data - including the way TV and radio assesses audiences by using a panel of people who record everything they watch and listen to. However, I have a much simpler measure - is my web site selling more today than it did yesterday? If so, it's effective.