Breaking down the value of all the tweeting on Twitter
This probably isn’t terribly surprising, given what we know about Twitter and the stream of consciousness platform it offers users. But a recent study of Twitter traffic found that 40 percent of the content is pretty much senseless babble.
A story at BBC.com, looks at the study by Pear Analytics, a U.S. market research firm that analyzed some 2,000 messages. They were grouped into six categories:
- Pointless babble: 40.5 percent
- Conversational: 37.5 percent
- Pass along value: 8.7 percent
- Spam: 3.75 percent
- Self promotion: 5.85 percent
- News: 3 percent
I’m sure there’s some arguments to be made here regarding the way the analysis classified these tweets. After all, what is useless to one person might be of interest to another. Still, the survey - and the company says it will do others in the future - does seem to demonstrate one of the challenges facing users of Twitter - getting through the noise to the good stuff.
Tim has covered a wide range of topics, including tourism, crime, aviation and gambling, since becoming a reporter in 1990. The Oklahoma native joined the Post-Dispatch in 2007 after spending nine years in Orlando. In his spare time, he's often exploring one virtual world or another. He can be reached at tbarker@post-dispatch.com. 
Wow. A whopping 2000 messages. That’s a big enough sample to report on.
/sarcasm
While I don’t doubt there are loads of useless tweets, I find that I get the information I want just fine. Of course, I don’t follow SuzyQXXX. My list includes: @phandroid, @criticalpath, @SPConf, @Microsoft, @Ackerman1120, @dgoold, @jprutherford, @Axceler, @bobandtom, @tommabe, @StLouisGameTime, @NormSanders, @andrewconnell, @miklasz, and many more. Some are comedy (bobandtom, tommabe, jimgafigan), some are sports (dgoold, jprutherford), and some are tech(Microsoft, Axceler, phandroid).
Classifying tweets is purely subjective. I would expect Apple loyalists would consider @Microsoft to be useless just as I would find something about COBOL to be pointless.
It would be interesting to do a similar study of what we talk about in the “real world.”
That’s a pretty good point by Kurt. Most of the talk at work is about anything non-work related.