Up until now, most marketers depended upon data that suggests either the best time of day, or the best hashtags to use (or a combination of both) to get a feel for what their tweet might do in the wild. That might all change, as an MIT professor has now created a computer model that predicts the number of retweets in its entire lifetime after only a couple of minutes of retweet activity.
“A lot of people felt that Twitter usage was totally random and unpredictable, that it was all just noise,” stated Sloan School of Management assistant professor Tauhid Zaman in an article in the LA Times. “But it turns out that there is a systematic, repeatable behavior that you can model.” Who knew?
You can actually see the model in action on his website Twouija: the Retweet Oracle, which is a twist on the old physical Ouija board. The site shows a number of tweets from celebrities and shows, as well as how well they match his predictions.
While on the surface this might seem like a boon to anyone who does promotion via Twitter, I don't think the model by itself adds much in terms of promotion, since you have to actually release the tweet first before it can predict what it will do. By correlating that data over time though, it might provide some better clues as to the parameters that make a tweet more "retweetable."
That said, I can tell you that without any of the fancy computer models or predictions. Just write something worth retweeting.
----------------------------------
Help support this blog. Any purchases made through our Amazon links help support this website with no cost to you.
Interested in the Music 3.0 archives? Buy The Music 3.0 Guide To Social Media for the best of over 800 posts.
You should follow me on Twitter for daily news and updates on production and the music business.
You should follow me on Twitter for daily news and updates on production and the music business.
Check out my Big Picture blog for discussion on common music, engineering and production tips and tricks.
No comments:
Post a Comment