Early stage startup marketing & community management + the great outdoors.

Twitter Analysis, Part I

Generally I feel that “Twitter analysis” is a bit of an oxymoron – it’s missing the qualitative benefits of the social connections that are made offline and/or long term.  I am, however, interested in sharing content that my core audience will find interesting.  Short of polling 1,600 people, the easiest way to do that is to look at the numbers and here are a few concepts I’m thinking about for improving my feed.

I’ll do this <sarcasm>earth-shattering </sarcasm> analysis  in segments as I have ideas or suggestions and time allows.  Here’s the first pass…

I do “at” mentions far too often.  Formulists pulled a list for me of the top 50 people who have @replied to me in the past year and includes the number of times I have mentioned them.  The numbers were wildly differential.  Here are a few quick stats;

  • The total times I mentioned people was 382 v. was mentioned 720 (53%)
  • Person who mentioned me most: @JimStorer with 42 mentions.
  • Runner up: @m750 with 40 mentions.
  • Person I mentioned the most: @JimStorer with 97 mentions.
  • Runner up: @m750 with 56 mentions
  • The best ratio was @shakti672, who replied 200% of the time.
  • The worst was @Devon, who replied 14% of the time.

I, seemingly, have used Twitter as a conversational tool far more often than other people.  Another option is that I simply haven’t been active enough to warrant others mentioning me – either in online or offline methods, or I may just be sharing too much noise / junk / unoriginal content… Or I’m just yelling at clouds.

My takeaway is that people seemingly care less about social mentions and more about the content delivered, so I’ll make a conscious effort to decrease the conversational mentions.  My secondary takeaway is that I need to hang out with Jim & Aaron more offline.

10/13 Update – I’m having doubts in the integrity of the data pulled from Formulists, namely since @PeteKuhn is excluded from the list.  Pete is a close friend and after I grew suspicious that he wasn’t completely excluded from the Formulists data I manually went through his feed and counted at least 10 instances of mentions where we’ve had conversations via Twitter – enough to put him in the top 10, let alone top 50.  This erodes a lot of my confidence in the initial data but I still think it is, directionally, correct.  I don’t have any reason to believe Formulists is favoring inbound mentions over outbound mentions, or vice-versa.  …yet, anyway.

Leave a comment

Your email address will not be published. Required fields are marked *