Analyzing Twitter

I've been scratching at developing some methodology that newsrooms can use for measuring the success of their Twitter strategies for some time now.  Unfortunately, I haven't had much time to really focus on it.  However, a couple of nuggets I saw today reinforce my hypotheses that simple counts of followers and tweets mean very little.  You really need to dig deep and spend the time (sigh) analyzing follower profiles and tweet content. 

Thanks to Mashable today, I've finally found the site that keeps a running count of how many users use Twitter via its web site vs. third-party clients or applications like TweetDeck.  Twitstat currently shows that only 20 percent (!) of Twitter usage comes from the web site.  Now I'm really going to scrutinize the methodology of any reports on Twitter usage.  I suspect most of them – especially the panel researchers like comScore or Nielsen – only capture site traffic.

This stat from iMedia Connection is from September but it's probably still useful.  "Roughly one quarter (24 percent) of Twitter users have never tweeted or
have ceased doing so, according to data from audience measurement firm Crowd Science. That number is very close to the percentage of users who tweet on a daily basis (27 percent)." 

140conf I'm really looking forward to the 140 Characters Conference in Los Angeles next week.  After that I'm sure I'll be inspired to really buckle down and figure this out!