HAD A LONG CONVERSATION WITH A JOURNALIST about, among other things, the value of “unique visitor” stats. I have to confess that I’m not that impressed: to me, having a reader who refreshes 10 times a day is pretty much as valuable as having 10 readers who look once a day. It’s a question of width vs. depth of interest. He more or less agreed but said that unique visitors is just one of those measures that people like, even though tech-types say that “unique visitors” is usually an undercount because of firewalls, etc.

I think that many of the comparisons going on (MetaFilter is apparently claiming more viewers than the WSJ) are pretty bogus — not even apples vs. oranges, but one thing that you hope is some kind of fruit with another thing that you hope is some kind of fruit. And I don’t think that even well-established metrics are worth much — note the discrepancy between the NY Times bestseller list and actual book sales. Or look at newspapers: if a million people buy a newspaper, do a million people read every story in it? Probably not.

People measure stuff because it makes them feel better. But the measures are of limited value, especially in the absence of any agreement on what’s important. Until you agree on what measures are important, and what you want to know, the measures are mostly bogus. But I’m working on a site redesign, and I’m going to include a counter that measures (however inaccurately) “unique visitors” so as to finally be able to answer the question that journalists ask.