Mar 192009
 
Shares

Complexity iPhone Camerakit App 22Ever since I joined Twitter (GaryPHayes) I have been fascinated by the subtle ‘etiquette’ of being followed, following and timely updates (as well as the enormous growth and creative potential twitter now affords). It is also interesting watching those traditional media brands and celebrities with a non-twitter and web 2.0 online reputation enter into the fray. What effect do they have? Do they corrupt this young new channel before it has found it’s own feet or is the invasion of old brands and celebs part of its maturation?

Laurel Papworth has far more in-depth coverage of this movement and etiquette across many and various posts on her main blog here but one thing became evident to me as traditional media and celebrities started to ‘infiltate’ Twitter – the instant emergence of old world, short head, long tail distribution. Those brands (individual and companies) already popular in other media on setting up in twitterville started to gain followers like magnets, they swarmed to them – in many cases regardless of what they were tweeting (film and pop stars particularly). We also see old form media channels such as news updates, emerging as useful ‘feeds’ and gaining instant popularity too. Merging with all these are the new stars, traditional bloggers find the transition to micro-blogging easy and so on and so on…

As Twitter has an open API the stats are relatively easy to pull out and there are quite a few sites that do much better analysis than mine below such as TwitterFacts blog, Damon Cortesi and TweetStats. For my little effort below thanks to Twitterholic and its dynamically updated top 1000 (based on followers), I was able to do a quick big picture overview – data taken on the 17 March 2009 !. Before we dig down into the charts themselves a quick high level stat on the Top 1000 tweeters

The top 1000 tweeters have generated 3.45 million tweets and are following 12 million but being followed by 35 million. (note: followers and followings are of course not unique, but the updates/tweets are)

The first chart is what I simply call the  Twitter Long Tail. Starting at the far left with top tweeters CNN Breaking News and Barack Obama at 543k and 486k respectively we move across to the 1000th top tweeter in the world Brad Will with just under 8k followers. I have highlighted a few random tweeters in-between for reference – key thing to note of course is the obvious almost perfect Long Tail shape (I would imagine over time this would smoothe even more – we are still early days)

twit_lt_06

The highlighted selection here include world renowned bloggers Robert Scoble and Darren Rowse (problogger), passionate artistes Imogen Heap and Stephen Fry, TV getting in on the act Ellen Show and Letterman plus trad media and social media folk. It is interesting for example that The Ellen Show Twitter ID appeared on the 16 March and generated around 200 000 followers off the back of one show – sadly there were only a handful of updates and virtually no following back, a poor user experience – traditional media really needs to make sure it doesn’t corrupt these ‘delicate’ new media channels as it so often does and then tells everyone they don’t really work!

While we are on the global view worth noting that adding all the followers up (thats means each persons follower amount) we end up with 35 million (remember that will contain many duplicates). The point though is to demonstrate the short head’ness here where followers are effectively a ‘rating’ (abstract) of popularity.

Of that 35 million totalled followers

  • 55% are in the top 100
  • 67% are in the top 200 and
  • 85% are in the top 500

To demonstrate this rather spookily smoothe long tail curve I removed the top 50 (that have rather exponentially big figures) and looked at the top 50-500. I started to think also here about the number of updates – do updates bring in followers or is it all about pre-twitter trust and reputation – of course its a to be calculated mix of the two of them – but look below at updates and position…
twit_lt_01
I went further down this road and looked at the top 100 and their update distribution – the spikes are named. Fascinating again to see that updates do not equal popularity (OK that’s obvious and I will stop labouring that one) but there is a significant high amount of updates going on the in 13-30 areas – remember though we are looking at the creme-de-la-creme of tweeters here and might be too ‘zoomed in’ for meaningful insight?

twit_lt_02

If your still with me, for reference, here is a quick snapshot of the top 50 World tweeps based purely on following (now you can go and follow them all!). As I keep saying this is not the whole story as we can see – for example CNN following 1 person (is pure broadcast) and Al Gore with only 14 updates (is pure pre-twitter reputation – or 14 amazing world shattering tweets?! – I will go with the former). Of course automated tweeting is rife and there are many in the top thousand who have or are resorting to bots to send messages in their ‘down time’. More after the list…
twit_lt_04

Some time ago I thought a twitter quotient that took into account updates/followings too is important and the chart below is the same top 1000 tweeters now ordered by a Gary algorithm (made famous on Twitter Agency and Laurel’s post of Australian Journalists on twitter), which changes the landscape significantly. Reproduced from my little contribution to twitter agency here.

Here is a little formula I just cooked up called the Tweet-GQ (Tweet Gary Quotient) that works out a Twitter rating. To be considered as a valuable system to be used on top 100s etc. Before I go into explanation, here is the secret formula

( ((Following/3)+Followers) x (Followers/Updates) ) / 10

This takes into account the raw numbers of followers weighted over following. More importantly it then has an critical multiplier – that of how many updates you do in relation to the followers you generate. So simply, it rewards high numbers of followers but also takes into account how many tweets or updates it took you to get that many followers.

To do this yourself without needing a degree in pure math (or an online calculator – to be done by someone). Here is a simple 3 step DIY version.

  1. Divide followings by 3 and then add this to followers – write the number down
  2. Divide followers by updates – write the number down
  3. Multiply the two numbers above and divide by ten – et voila. Your very own TweetGQ

twit_lt_05

Finally and while I am on this twitter topic heres a lovely mosaic of 360 out of my current 1300 followers…seems so insignificant now 🙂 But this shows off the power of open API – each of the faces are clickable and therefore followable – is that a word. Bye for now, see you in the twitterverse.

Get your twitter mosaic here.

Get your twitter mosaic here.

Shares

  34 Responses to “Twitter Long Tail – Broadcastization & Pre-Twitter Reputation”

  1. […] Posted in Blog, English, Galleries, Media, Netstuff | No Comments » I just found this in an article by Gary P Hayes, introducing the calculation of the Tweet-GQ (Twitter Gary Quotient): ( […]

  2. […] Twitter Long Tail – Broadcastization & Pre-Twitter Reputation | PERSONALIZE MEDIA all kinds of twitter stats, maybe mainly showing that everyone uses twitter differently and is successfuk for different reasons. (tags: twitter stats statistics longtail) […]

  3. […] Twitter Long Tail – Broadcastization & Pre-Twitter Reputation – Personalize Media – Oct ‘09 […]

  4. The story of the long tail, staring the Web2.0…

    After my blog last week concerning my dislike for Twitter, I would like to discuss how Twitter is an important Web.2.0 social media tool and my candidate for leveraging the long tail. Firstly, I would like to go over why the “long tail” is so important…

  5. Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/458d8F9 vía @garyphayes

  6. Amazed my 2-6 yr old posts still being rt'd RT @EdsonRshow Twitter Long Tail, Broadcastization & Pre-Twitter Reputation http://t.co/458d8F9

  7. Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @garyphayes

  8. Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @garyphayes

  9. RT @maelson81: Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @garyphayes

  10. RT @maelson81: Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @garyphayes

  11. RT @maelson81: Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @GaryPHayes

  12. RT @maelson81: Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @GaryPHayes

  13. Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/JfVJCXc via @garyphayes

  14. Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/JfVJCXc via @garyphayes

  15. RT @maelson81: Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @garyphayes

  16. RT @maelson81: Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/q9YHQai vía @garyphayes

  17. Broadcastization & Twitter Reputation http://t.co/tj9AWLl

  18. Broadcastization & Twitter Reputation http://t.co/tj9AWLl

  19. Broadcastization & Twitter Reputation http://t.co/tj9AWLl

  20. Broadcastization & Twitter Reputation http://t.co/tj9AWLl

  21. Broadcastization & Twitter Reputation http://t.co/tj9AWLl

  22. Broadcastization & Twitter Reputation http://t.co/tj9AWLl

  23. Archive: : Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/TdWH6Aji

  24. Archive: : Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/TdWH6Aji

  25. Archive: : Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/TdWH6Aji

  26. Archive: : Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/TdWH6Aji

  27. Archive: : Twitter Long Tail – Broadcastization & Pre-Twitter Reputation http://t.co/TdWH6Aji

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

(required)

(required)

CommentLuv badge