140 Twitter Conference Seattle Recap: #140tc Analysis

by Adam Schoenfeld on March 9, 2010

I had the pleasure of attending my second 140 Twitter conference in Seattle yesterday. Keeping up with the LA Event, Parnassus Group put on another great conference. We used 20 Decibels to track, measure, and export the #140tc hashtag data. Our campaign captured tweets for the 2 days leading up to the event, the event itself, and ~12 hours following. We used Steve Broback’s favorite Twitter app, Microsoft Excel, to turn the export into charts. The Parnassus group will be following up with a sentiment analysis of this data shortly – I’ll update here with a link.

140tc Tweet Volume and Velocity

The #140tc hashtag was very active – there were 5,489 tweets from 1,445 people in the period tracked. Ben Parr’s opening keynote generated a high point in activity. Later at ~5pm Mashable tweeted a story about Hootsuite integrating Foursquare. This generated another wave of activity as Mashable Retweets took over the #140tc stream. Chris Pirillo noted that many of these were likely spam bots (we’ll cover that in a future analysis)

140tc Twitter Followers

Not surprisingly, conference attendees are active twitter users. The follower numbers for this group are impressive. Over 1/3 of people have over 1,000 followers, and 7% have over 5,000 followers! The average number of followers is 3,830.

140tc People

The average person tweeted #140tc 3.8 times in the period tracked. However, there some stood out from the crowd. To make the top 20 list, it required 247 tweets! The top 20 were Jaxx_Magee, Kathy E Gill, Alecia Sullivan, Jonathan I. Ezor, Ray Prock Jr., Rich Harris, Brian M. Westbrook, Kathy Gill – Live, Sarah Fosmo, Nansen Malin, John Knight, Craig Sutton, Seattle Wine , Mike Tallent, Jeff Fowle, Eric Earling, nazila, Shona Milne, Damon Cortesi, and dakini_3. Special props to Kathy Gill (@kegill) for having 2 handles in the top 20 list :)

  • I had no idea I was so tweeting that much. There was so much great information I guess I felt I had to share.
  • With regards to the velocity vs. ability for Twapper Keeper to grab tweets:

    At Twapper Keeper we know there are cases that high velocity tagged tweets can be missed. The issue stems from the fact that as we try to keep up with all of the different tags we run into situations where we can't adjust fast enough to topics that begin to accelerate in usage.

    We are finalizing Version 2 of the system (in testing right now) which will tap into the Streaming API which should eliminate these issues in the long run.

    If you have any questions, just let us know.

    @jobrieniii
  • OK, I know I'm looking at a subset of these tweets from TwapperKeeper -- just 8 March -- but that shows me with @kegill having 61 tweets and @kathy_live having only 37. I tweeted as @kathy_live only during the conference. That's a huge difference in data and suggests that TwapperKeeper is managing to grab only a slice of the flow during heavy traffic.

    I just checked the @kathy_live account and I think I counted 74 tweets with the #140tc hashtag (you have 69). I counted 75 from @kegill for Monday -- 22 were @replies and 10 RTs.
  • LOL! I'm not sure that deserves a shout-out. :-)

    The data are interesting because I /think/ that most of the @kegill tweets would be @replies whereas the @kathy_live tweets would be "session notes." I did cross-tweet some, but not that much. I don't think. Now I need to go look at my own stats for that 12 hour period (any post-even tweet was from @kegill).

    I posted a few Wordles today - this is the best one, I think: http://www.wordle.net/show/wrdl/1760899/140TC_-...
  • I tracked this hashtag too - my data differs a bit... http://bit.ly/aoRZet
  • Seems like there was a whole lot of twittering going on. Reading this analysis, one might infer that #140tc was more about the quantity of tweets - and/or Twitter followers - than the quality of the tweets ... or what was being said IRL. Are there or will there by any summaries of what was said, or learned, on a more qualitative level?

    It seems much use of Twitter is about reaction. It would be especially nice to read some reflections ... especially those longer than 140 characters.
  • Thanks Joe. Great point. I think the guys at Parnassus are going to do a follow-up to dive into what was being said. I'll post a link here if and when they do that.
  • Adam, I love this analysis that you push out after every #140tc. Always interesting to look back and see what happened.
  • Thanks Ayush! It's always fun to do.
blog comments powered by Disqus

Previous post: 5 Ways to Manage Twitter at Scale

Next post: Twitter Spam: A Case Study