Twitter Users Say Only 36% of Tweets Worth Reading

February 3, 2012

This article is included in these additional categories:

Analytics, Automated & MarTech | Data-driven | Social Media

cmu-tweet-attitudes.jpgTwitter users rated only 36% of the tweets they received as worth reading, while they expressed ambivalence about 39% and said 25% were not worth reading, according to [pdf] research released in January 2012 by Carnegie Mellon University (CMU), which looked at data gleaned from December 2010 to January 2011. “Question to followers” tweets were the least disliked of the various categories studied, with an 18% probability of being rated not worth reading. By contrast, “me now” tweets regarding current mood or activity were among the most disliked, with just a 22% chance of being rated worth reading.

Boring Tweets Most to Blame

Data from “Who Gives a Tweet” indicates that being boring was a prevalent problem on Twitter, as this reason accounted for 82% of all explanations for followers rating a tweet not worth reading. The Twitter audience appeared to value learning about new content, though: the leading reason for rating a tweet worth reading was that it was informative (48%), while being funny (24%) was also a popular attribute.

Longer Tweets Get More Clicks

Although the CMU study found followers to appreciate conciseness, link-containing tweets have a better chance of getting clicks if they are longer, according to recent analysis from Dan Zarrella. In fact, among the 200,000 link-containing tweets he analyzed, those between 120 and 130 characters in length had the highest click-through rates (CTR).

Other Findings:

  • According to Zarrella’s analysis, the highest CTRs are on links that appear approximately one-quarter of the way through the tweet.
  • Tweets containing “via,” “@,” “RT,” “please,” and “check” had higher CTRs than tweets without those words. Users beware, though: according to the CMU study, Twitter-specific syntax was a common source of complaint, particular the overuse of hashtags and @mentions.
  • Zarrella also found that tweets posted on Friday, Saturday, and Sunday had higher CTRs than those posted during the week. Tweets posted in the afternoon hours (ET) also had higher CTRs than those posted in the morning.

About the Data: The CMU analysis is based on data from December 30, 2010 to January 17, 2011. The dataset includes 43,738 tweet ratings from 1,443 users. These users rated the accounts they followed, an even broader population of 21,014 Twitter users. All analysis is drawn only from follower ratings.

45th Parallel Design Ad

Explore More Charts.

Pin It on Pinterest

Share This