The Independent Twitter 100: reflections & actions

Today marks the second annual Independent Twitter 100, where the national paper provides a curated guide to some of the titan’s of the British twittersphere. We’ve been lucky to collaborate with the Independent as a data provider for the past two years.

It’s a great thing for the Independent to do this.

Too many ‘lists’ and guides are dominated by Americans. And bridging the grassroots movement of twitter with the traditional authority of the newspaper can spark interesting conversations. Here we have a list of Britons across the spectrum.

And Twitter is huge – and confusing and full of undiscovered gems. The Independent’s work today will uncover those gems.

The project is also challenging. Twitter recently passed 500m registered users. Most major celebrities are on twitter. Brands are on twitter. Many people have more than one account. Many people don’t identify their geography accurately.

A list of 100 people strictly based on a PeerIndex would be a tedious list of hyper-celebrities. Pop stars, footballers and X-Factor judges, the type of people whose Twitter impact drives a PeerIndex of 80 or more. In essence a list of the people on your telly on a Saturday night.

Why? Because twitter is big. And 100 people isn’t a lot out of millions. Think of the tallest person you actually know. The absolute tallest, the person who towers of your friends at Christmas. Chances are, even though they are the tallest person you know, they would be several inches shorter than the 101st or even 250th tallest person in the UK. (Unless you are one of the few hundred people who actually knows the 250th tallest person inthe UK).

So for any collection of names to make sense, to bring colour to the diversity of voices on Twitter – curation is necessary. Editorialisation is needed, not because the process is innacurate, but because the editors and judges need to apply their judgement to make lists functionally valuable and meaningful.

There are other times where the algorithms do their job but cant account for taste. For example, the infrequently tweeting celebrity or the press-releasing corporate leader. These people might score extremely highly because they generate interactions when they do talk – but do they follow the spirit of what makes twitter interesting? Probably not.

And of course, you might disagree with inclusions or exclusions on the list.

The solution? Simples, as that TV meerkat would say.

Create your own lists. 

You can build your own list of people on PeerIndex. We call it the groups function and it is available for free.

A group can have up to 2,000 people and be on any subject. You can invite collaborators to help you build the group. And you can invite group members to join the group to improve the data.

Groups can be assigned to specific topics (like ‘Hedge funds’ or ‘Premier League’) and you can see how group members score in topics as well as more generally by PeerIndex.

Build a group for fans of your football team.

Build one for your profession.

Build one for your town.

This is about social media after all.

We provide the tools.

You are in control.


On the reputation graph

We recently had the opportunity to present at Laurent Huang’s Lift Conference on Geneva where we covered the subject of ‘The reputation graph – and why the Web needs one’.

The summary: the web is about people not pages. We don’t have a good people-reputation signal (call it PeopleRank if you will). Such a people-reputation signal would vastly reduce the cost of creating meaningful and deep connections with other people. And create untold amounts of new value for all of us.

The event was captured on video and also on this great single drawing:



Watch live streaming video from liftconference at