Why page counters are unreliable

Page counters count, at best, accesses to pages. Here access means that some piece of software, such as a Web browser or a search robot or a cacheing Web server, refers to a page. No human being needs to be involved. Or, alternatively, through a single such access, a page might be made available to thousands of people hundreds of times. In most cases the situation is between these extremes. The point is that you never know.

For a good discussion of the unreliability of Web page access statistics in general (including both statistics from server logs and counter statistics), see the document Why web usage statistics are (worse than) meaningless.

Darin McGrew has written a hit counter rant from a different perspective, making very good points.

Jukka Korpela
This document is essentially a footnote to the discussion of counters in my document So you want to create a home page?.