This is our wall of post-its, left after two groups of 20 to 30 staff members had debated, scribbled and stickered in a frantic exercise in speed-brainstorming. I hadn’t tried any IA exercises with such comparatively large groups in my previous life so I was curious (and a little nervous) as to how it would work out. The goal was to gain an understanding of the types of information (and their comparative weighting) that our new intranet wiki will need to support (I should also add that Leisa Reichelt’s posts on disambiguity about guerilla research proved very inspirational in finally tackling this).
I initially split each group into twos and threes, handed out large quantities of post-it pads, markers and biros (ballpoints) and asked people to write down as many topics or subjects or categories of information, documentation, policy and so on that might be found on our organization’s intranet as they possibly could over the course of twenty minutes. As people began to run out of ideas (or steam) I sent them into the next room to cover the walls with post-its.
The stickering brief was to be as intuitive as possible, to group the post-its in whatever way seemed to work best. The next thing was to move post-its around – try them in different locations, discuss any contradictions or differences of opinion and see what happened.
People seemed to have fun. After about fifteen to twenty minutes of shuffling, I began to detect more and more people standing back – not losing interest but contemplating. Before the energy could completely dissipate, we called it a wrap and went back into the next room to see if we’d learned anything in general.
What both groups picked up on included:
- there was more agreement than disagreement about what should be on the site
- they’d discovered that the intranet was going to be much bigger than they remembered the old one being
- navigation isn’t easy!
- too many subject areas seemed to belong in more than one “grouping”
Something I pointed out to them was that they’d just engaged in their first group ’social tagging’ exercise and that that the post-its spread out across the walls actually formed a pretty comprehensive corporate taxonomy.
Next steps? Transferring the tags into a spreadsheet, a lot of iteration and thinking and more consultation (this time through email and the PBwiki we’ve set up as a development tool) until we can arrive at “starter pack” of tags for the new intranet wiki proper. It’s also demonstrated to the teams the importance of basing our navigation on tagging but on tagging which has a carefully pre-defined base. It’s only a base though – it’ll evolve as use of the intranet evolves. It’s there to be constantly extended and edited. Otherwise, it’ll go the way of most rigid taxonomies – a dead snapshot of an organisation that’s thriving else (see also Thomas Vander Wal on folksonomies vs taxonomies – note that my own use of the latter term has more to do with the avowed purpose of a bunch of tags gathered together than formal knowledge management structures).
The other outstanding item is selecting the actual wiki. It needs to (amongst much else):
- support tags(!)
- allow for lockable pages and different levels of user permissions
- allow comments
- run on a Windows server
- have customisable skins
- support RSS feeds (incoming and outgoing)
- include WYSIWYG page editing
- be more or less ready to run “out of the box”
The last refers to a key measurement of the success of the whole operation – the extent to which the ongoing support of the wiki requires support from our tech team. The ideal level of support is zero! The other measurement of success is usage – not so much people looking up the latest contacts list but actually contributing and helping the beast grow.
Suggestions welcome! I still haven’t found the perfect wiki for us yet.
P.S. An article by Bill Ives from last year seems to confirm at least some of the value of the exercise. Whether we’ll eventually graduate to a more sophisticated “enterprise” tagging tool like Dogear or Cogenz, only time will tell.
Filed under: Enterprise 2.0, IA, Web2.0, information management, social bookmarking, social software, tags, wiki
Michael, looking at your wiki requirements I recommend Confluence from Atlassian.
Thanks, Niall. I’ll check it out. So many wikis, so little time…
Michael, THANK YOU for sharing your IA tagging exercise. A lot to think about for next time I do this. My experience is not with tagging exercises for IA brainstorming but rather as part of sharing around a personal network drawing exercise. Always very productive and lots of AAH HAA’s. Giving credit where it is due my first experience using social tagging for knowledge sharing was to encourage dialogue around a keynote presentation at a University of Warwick Knowledge Innovation Network conference. Connecting your IA and tagging post to your Facebook comments on our blog a common thread I see in both social tagging exercises and using Facebook is “FUN”. Both appreciated.
This is a great reminder that sometimes the simplest tools are the best place to start. I have yet to find an easy to use, quick and efficient substitute for the good old pen and paper for some tasks! When I wrote my first novel, I covered the wall in post-it notes so I could structure the narrative and plot!
Hiya – I’m now deep into the distilling and iteration phase so we’ll see just how much actual use I can extract out of this part of the process (a lot, I hope…).
I used index cards in the same way – now, with Scrivener, I find myself back with them in a virtual way. Thanks for the props!
[...] Pre-defined ’starter’ packs of tags are a great idea but try and develop them collectively. [...]
[...] Designing Corporate Intranet Folksonomies/Taxonomies with Posts-Its [...]