Exploring social business strategy

As part of my ongoing research, I’ve been paying close attention to developments around Twitter’s “list” feature (launched in October) – “A great way to organize the people you follow and discover new and interesting accounts.” Yesterday I stumbled upon an interesting finding (that has me scratching my ahead a bit) in relation to follower vs. list counts as a measure of popularity.

One would think that there would be a very high correlation between the two – if a lot of people follow you, you are likely to make a lot of lists. While I haven’t actually run a regression to prove that, as I’ve looked around it generally seems like a fairly safe assumption. But there’s one interesting anomaly I’ve found recently – @Marvel vs. @Agent_M.

@Marvel is “the official Twitter for Marvel Comics, Movies, Games and More.” Agent_M is the “editor for Marvel.com. Writer, blogger, loves tacos, tattoos, comics…” I’ve been watching these two accounts with interest for some time, because the former has about 43 thousand followers, and the latter has about 1.4 million. This would seem to say something important about relative popularity, and it’s interesting when the editor is more widely followed than the content.

But the “list” count tells a slightly different story. @Marvel has been added to 1,467 lists, while @Agent_M has been added to 1,234. So even though Agent_M has 0ver 30 times more followers, his account has been added to fewer lists. Divide lists Marvel is on by total followers you get 3.4%; for Agent_M you get 0.1%. ; Why is that? And what does it mean?

I have a few ideas (that basically amount to pure speculation), but there’s one other wrinkle I’d like to throw in here from my previous research. Regular readers have heard me talk about the Twitter brand engagement matrix before (see here, here, here, here, and here). The basic idea is to differentiate between brand and personality centric accounts, and broadcasting vs. conversational strategies, by using a 2 by 2 matrix. It looks like this (with a few examples thrown in):

The general message (as each of the examples represents) is you can have success from each of the quadrants. In this case, @Agent_M is squarely in the personality / conversational quadrant (upper right). @Marvel is squarely in brand /broadcast quadrant (lower left). Two diametrically opposed approaches, with two very different results.

So in this case, the personality / conversational approach has proven far more effective at attracting followers – a sign of popularity. But the brand / broadcast approach has proven far more effective (on a relative basis) at making it onto lists – a different sign of popularity. So why? Is it a different type of person that follows the different type of accounts? Does it indicate that a higher proportion of @Marvel’s followers are actually interested – and perhaps a significant chunk of Agent_M’s aren’t paying attention? What exactly is it about the type of content @Marvel posts that makes it more likely to crack lists?

Can’t claim to have the answer (at least not yet ). But I think there are some interesting questions here that I will be paying attention to…