I admit it - I’m addicted to Alexa. Through data gathered by their Internet Explorer toolbar, Alexa attempts to rank the popularity of web sites along various metrics. Tech pundits love to point to Alexa stats when drinking the Kool Aid of their web 2.0 site of the week - or calling for a site death watch. Two great recent examples are Digg vs. Slashdot and Flickr vs. Webshots. It’s a bit like sizing up boxers before a fight just by the numbers, genesis of the the famous saying “the tale of the tape”.

While many people argue that Alexa stats are crap, I think they’re valuable given you bear a few points in mind:

1. Don’t take Alexa out of context. Alexa has a narrow but deep window into web usage, i.e. extrapolating Windows IE users with Alexa toolbars out to a general web population. While they’ve got a great volume of data to play with (especially if they leverage Amazon’s usage numbers), don’t assume Alexa’s numbers reflect reality. On the other hand, it is quite reasonable to compare two sites’ Alexa ranks over the medium to long-term if they have a similar focus, audience demographic, international footprint, etc.
2. Until a site reaches a certain traffic level (rank over 40,000?), Alexa just doesn’t have enough data to be accurate. One user can skew stats at this level.
3. Obsessing daily over your own site rank is useless. I’ve seen Wikispaces stats fluctuate wildly on a daily and weekly basis when our internal numbers show nothing of the sort. Watch for long term trends instead.
4. Never confuse a site’s Alexa rank with its chances of business success. Much like the diameter of a heavyweight fighter’s bicep isn’t a great indicator of who will win the fight, Alexa’s tale of the tape says absolutely nothing about a site’s actual value to its users.

With these points in mind, I’ll periodically post Alexa-isms that catch my eye. First up: Squidoo’s first six months.