Fallacy of Twitter Authority Based on Followers


Michael Arrington backing up Loic Le Meur’s call for something akin to a Twitter PageRank algorithm with authority based on the number of followers:

Should Twitter Add Authority-based Search?: “I’m with him on this. Most of the time I just want to read everything people are writing about a topic to more or less take the temperature of the masses on whatever I’m researching. But sometimes it would be nice to hear what just the top users are saying on a particular topic, too, since so many more people hear their message.”

I have 3,000 or something followers but I think this is a terrible idea with the following logic:

1) Pagerank sucks (now) for blogs and isn’t a true measure of a blog’s worth, value or credibility.

2) Even then, Twitter is not blogging. Ranking people according to something as transient and flimsy as the number of followers is a worse idea than ranking blogs according to their number of inbound links. Oh, and imagine the gamers.

3) Twitter is a not only a micro-presence platform, it’s a micro-community platform. What purpose would such a “follower algorithm” serve?

Some (most) of my favorite and most “valuable” people I follow on Twitter have under 1k followers. Calling them less credible or their tweets less substantive based solely on the number of followers is silly.

4) I agree with Arrington that it is nice to hear what “top users are saying on a particular topic” rather than crowdsurfing. However, there are already great tools for that. It’s called the follow function combined with RSS or Summize or Yahoo Pipes or Google Alerts, etc. The “top user” on a particular topic such as Hebrew Bible or some niche realm that I’m interested in is not necessarily going to have thousands of followers.

The best metric here is individual intuition and discernment.

5) This isn’t an argument for “wisdom of the crowds” or the “power of the conversation” etc. I’m not a big fan of that mentality, either. Those types of 2006-esque arguments are annoying at best.

Instead, my point is that it would incredibly difficult to institute something like a “worth quotient” on all users of Twitter (even more so than blogging). Putting something like a rank or worth based on the (easily gamed) number of followers a person has makes it even worse.

There Has to Be a Better Way

Don’t get me wrong, If Arrington or Le Meur or Twitter could come up with a ranking or worth algorithm based on something inventive and truly reflective of value, I’d be all for it. If Twitter could put together something revolutionary for determining authority akin to PageRank back in the ’90’s, I’d be the person yelling the loudest from the mountaintop for adoption.

However, this ain’t it.

This seems more like A-Listers grasping at straws to me.


14 thoughts on “Fallacy of Twitter Authority Based on Followers”

  1. Take Guy Kawaski: by all measures he is a very influential person, and has 30K+ followers. Yet, in my Twitter stream, he is one of the least read … his constant posting of AllTop topics is too time consuming for me to engage. I'll glance at his tweets, as he still has a lot of great things to say, but he is far less influential in my day-to-day conversations than the Ux professional with 100 followers.I think a slightly better measure would be the number of @replies received. When people are very specifically seeking out your advice, it shows greater influence than simply listening to you talk.

  2. Good point on Guy. I didn't think about it when writing this post, but Arrington's tweets were of such low value (since they are just links back to each TC post and I have an RSS reader for that), I unfollowed. Same with Guy, etc. Not sure about @replies, either. Everytime Kevin Rose tweets that he's having pizza, I'm sure he gets 1,000 @ replies. However, that's of considerably less “value” (to most of us at least) than someone posting about something relevant or interesting with much less in the follower column.Gaurav suggests here (http://bit.ly/26ejwv) that perhaps retweets are the answer. I think not (http://bit.ly/PGqr)Basically, I don't think you can apply these sorts of quantitative metrics to Twitter. The model just doesn't support the “more is better” mentality as with in-bound links. I think the right approach is to leave value or worth or authority judgements up to the discernment of individuals.

  3. I agree; the original purpose of Twitter was to create “digital intimacy” with the people you know or the online acquaintances you can trust. Doing this would be akin to limiting yourself on Facebook or LinkedIn to the people with the most “friends”, which not only is completely arbitrary, but would do more harm than good as the intimacy is replaced by anonymity.

Leave a Reply