When Predicting Twitter Success, the Source is the Thing
Unfortunately, all we have to go on when crafting these Tweets is our gut instinct (and maybe some previous data) about what will take off and what won’t. You write, you send, you hope. I’d say it’s not rocket science but it kinda is.
Researchers from UCLA and Hewlett-Packard’s HP Labs conducted a study to determine the formula for the perfect Tweet. They started with over 40,000 recently published online articles, then matched that data with corresponding Tweets, then let it all flow through for 4 days to get the full lifecycle of any given Tweet. (With me so far?)
Then, they sorted all the data based on four options: the source of the news, the news category, the emotional connection and name-dropping.
What they ended up with was an algorithm that allowed them to predict the popularity of a Tweet, prior to sending it out, with 84% accuracy.
The math looks like this:
For those of you who don’t know your coefficient from your t-density score, here it is in plain English.
Emotional pleas didn’t do any better than straight facts. Technology news in general did well, but just being “news” doesn’t make a Tweet more Tweetable. They found that some celebrity names and brands (Justin Bieber, Lady Gaga) brought out the best in Twitter users but those were more exceptions than rules.
What did make a difference? Source. In the end, all this data and all these numbers point to the fact that a trusted source is the key to Re-Tweet success. Not the most well-known source, or the biggest source, or the one with the most power, but the one that people trust to give them the straight scoop.
That’s good news for you. What that means is that you don’t have to have Lady Gaga on the payroll in order to make a splash on Twitter. But you do have to provide solid, trustworthy, interesting, useable content on a regular basis. Do that and the social media system will take care of the rest.