@ the friends within followers

Sometime back, I’d written about micro ambassadors, where I’d also touched upon the long tail of twitter influence that is made up by individual users, and marketing opportunities therein. I read a few posts recently that made me think on the subject a bit more.

Some of the posts referred to a research paper on Twitter, by HP, which reveals that

the driver of usage is a sparse and hidden network of connections underlying the “declared” set of friends and followers.

A few numbers on the respondents (from the study) – 309740 users (this sample is 6% of the twitter universe, info courtesy Jeremiah’s post, the comments on the post are also very interesting), who on an average posted 255 posts, had 85 followers, and followed 80 other users. Among the 309740 users, only 211024 posted at least twice. The average @ replies ( conversations between 2 or more users, specifically mentioned in the tweets) were 25.4% .

A few findings I thought were interesting. The number of posts increases with the number of followers but saturates after a certain point. However, this saturation does not occur if we consider the number of ‘friends’ (followers with whom a user has had 2 or more @ conversations). The study also shows that on a number of ‘friends’ vs number of followers chart, the number of friends saturate after a certain number of followers is crossed. Understandable, since in a day, one can only have so many conversations with so many friends. My twitter statistics (though not the same as the average user in the study) corroborate these friends and usage findings more or less. It is thus debatable whether there’s any sense in just increasing the follower count. A certain Guy definitely wouldn’t agree, and it does finally depend on the intent. 🙂

It also questions the follower-influence- WOM marketing model, and its scalability. I’m actually quite happy with this since I have never been comfortable with this line of marketing strategy on a trust based network. While its not scalable generally, there are exceptions – Guy’s Alltop is one easy example.  The relationship he has with his ‘followers’ obviously doesn’t fit into the followers-friends definition discussed earlier.

The echoing powers of RT (re tweeting) is another thing to be kept in mind. If I follow someone, and i find some content interesting/informative, there is a good chance I’ll RT that, even if I do not have regular conversations with that person.

Lastly, this equation might change if Twitter implements groups. Even though its limited to Japan now, there is a good chance that the rest of the world could get it soon. Meanwhile, you can always use Twittgroups. Groups would allow consolidation of crowds interested in certain things. Marketing would definitely be easier then.

And finally, to wrap up, the favourite twitter pastime- revenue models. If such marketing is not a scalable option, and will not excite brands to use twitter a lot, what will? I read two very good posts on the subject of revenues. RWW has a post on the search of Twitter being used as a revenue tool, since it gives live results from all of Twitter, and doesn’t wait for any indexing like Google does. In fact, the idea of companies using Twitter as an early alarm signal is something I have come across before. Nick Bilton has an interesting idea on Twitter throwing up the kind of immediately relevant ads we would like to see with “some really intelligent data mining and cross pollination”. I quite like that idea too. 🙂

until next time, tweet and ye shall find 🙂