The Science of the Perfect Tweet

If you’ve ever tried to promote something or someone on Twitter, you may have pondered the art of the perfect tweet (i.e. one that will garner the most attention and retweets). Turns out, it may actually be more of a science. And surprisingly, you don’t necessarily need to feature the words “Justin” and “Bieber.”

According to The Atlantic, a paper (PDF) published by UCLA and Hewlett-Packard’s HP Labs features an algorithm that can predict the popularity or “spreadability” of news-related tweets with an incredible 84 percent accuracy, even before they are actually sent out into the Twitterverse. This means that with the tools created by the researchers, news organizations (and individuals), can actually tailor and structure those 140 characters for maximum attention.

The short of it: technology is the most popular type of news story, followed by health-related stories and “fun stuff.”

Name recognition was also fairly important (okay, so the words “Justin” and “Bieber” might have some clout). But what most greatly affected the shareability of a news tweet was the organization or person doing the tweeting. People are more likely to share a story with their followers that they know comes from a reputable source, like The New York Times.

And interestingly, the objectivity of the language used in the tweet made very little difference. In other words, you don’t get more readers or retweets by saying something like, “OMG I’M CRYING! BEIBER’S ALBUM IS OUT!” rather than simply saying, “Justin Beiber’s new album released today.” So please, stop the hyperbolic screaming.

You can read the whole article in The Atlantic here, and the research paper here.