Not long ago, I posted about AdAge including Alexa rankings in the Power150 Index. Long story short, Alexa is about as unreliable of a metric as you can get in Web / blog rankings, because it’s one of the most easily manipulated data sets in existence. Aside from advertisers who truly don’t know any better, no one gives a damn about Alexa rankings.
Todd And and I had a brief back and forth in the comments on that post, and unfortunately I was off a little while last week and am just getting around to responding now (sorry Todd). Todd had a few questions about potentially improving the Power150 Index that I’d like to address:
“So, while our attempt at adding multiple metrics was in good faith, Alexa might not be the best answer. I’ve shared your blog post with the Ad Age team. Again, as background, the Ad Age folks and I were receiving complaints about Bloglines and suggestions for adding Alexa. We thought we were responding favorably to the community by reducing the Bloglines weight and adding Alexa. It seems like Alexa is not welcomed, so let me ask a couple questions…
- In your opinion, should Alexa even be a metric on the Power 150? If so, how would you weight it in proportion to Technorati, Google PR and Bloglines?
- Are there any other metrics that you recommend for ranking blogs that offer an open API?
- We know there is no such thing as a perfectly accurate ranking, but is there anything else you suggest to make the Power 150 stronger – keeping in mind that it is fully automated and requires third-party metrics to have open APIs?”
Here are my thoughts:
- I’d say that (as much as I think Alexa’s worthless in this regard), it certainly has a place in the Power150 Index. When you’re using other metrics that are easily manipulated, subjective, or simply very “non-scientific” (like Google’s arbitrary PageRank punishments to sites using advertising models they don’t like), then why not add Alexa to the mix? I don’t know exactly how it’s weighted, but I can’t comprehend why something benefiting webmasters as a whole would be given more weight than something specifically targeted to blogs. If anything, it certainly was given too much weight, or we wouldn’t have suddenly seen increases in webmaster-oriented blogs as opposed to “true” marketing, advertising, media, and PR blogs. The target audience of the blogger shouldn’t have much weight in where they rank.
- I’d suggest adding a LOT of metrics if you want anything remotely authoritative (in a true sense and not just perception). A handful of metrics is nothing with something like blog rankings. If you really want to improve it, start adding things more relevant to bloggers (such as social bookmarking services). Here’s a list of 300 APIs available with plenty of possibilities for blog rankings – most notably Feedburner (public access may be opt-in, but that’s really no different than Alexa relying only on those opting to install their toolbar, is it?).
I think my answer to number 2 covered number 3. Like Todd said, there are no perfect blog rankings (and I highly doubt there ever will be). The biggest issue with the Power150 is its lack of metrics forcing each one (poor in and of themselves) to carry far too much weight.