a starting point.

There’s been a lot of conversation about Novo Nordisk’s recent partnership with Charlie Kimball to tweet over on @racewithinsulin, and many folks questioning whether or not pharmaceutical companies can (or should) get involved in social media at large, and Twitter specifically.

So, Kru Research did a neat study to learn more about how people decide who to follow on Twitter: they came up with a fake drug, designed to treat insomnia, and then they created four Twitter profiles, each following approximately 100 people on Twitter who had tweeted about living with insomnia. Go read the whole thing here.

I think the most interesting part of this study is that someone actually took it on in the first place. It’s an intriguing approach to a problem that doesn’t have a lot of solutions right now.

That being said, I’m not entirely sure how much stock I’d put in these particular results - they’re certainly very interesting as trends, but I’m not sure they’re concrete data points.

The authors make that clear themselves in the final paragraphs, and draw attention to a point which I think is also crucial: what was true of insomnia may not be true of all other health conditions.

While there are many similarities between health communities - they’re ultimately unique ecosystems, living on and off the web.

I’ll certainly be on the look out for the follow up posts to see what they found (and how successful they were), in their attempts to promote an unbranded website, and convert coupons into prescriptions. But on the whole, I think studies like these will be of limited usefulness; in order to conduct a strong study, they’ve had to follow a particular process that isn’t really how people use sites like Twitter (professionally, or personally).

How successful would you be at building a Twitter-following if you only tweeted once?

And how can you tell the number of people who followed you because they were interested in your one tweet, versus the number of people who happened to have auto-follow set up.

That seems to be the weakest point (to my mind) of this information; I’m not sure there’s a way to correct for auto-followers, and that seems pretty important when you have a sample size of approximately 100 people (per account).

As a self proclaimed “social media nut”, and avid Twitter user/defender, I do hope this sort of information will encourage companies, particularly in such highly-regulated industries as pharma, to get involved in social media. But ultimately, I think this is something you learn by doing.

At some point, you just have to accept the risks, look at the benefits, and (hopefully) jump in.