Overtaken By Events

"If only we could get the right information to the right people in the right palce at the right time"

How many times have you heard someone say that? And it's true. It would be good if we could. However, this is a question that needs to be answered backwards. If you try to answer it forwards you will come unstuck.

The traditional approach to start with the "right information" piece. We collect information. Lots of it. Then we look at who are the "right people" - who are our target audiences to bombard with this stuff? If we are smart, we ask them if we have got the right information. And generally they say "Huh, yeah, I think so. I'm busy with something else right now". And then we set about building the "right place" - a database with a taxonomy & probably mobile access, etc. Generally,we get half way through doing this and then the money runs out. Or we finish it & no one uses it.

We never get as far as the "right time" but this is really crucial question. Remember the answer that the users gave you 6 months ago? That - like so much in life - was all about timing. The critical question to ask is: "How predictable are the information needs of this user group?" And what that really boils down to is: "How predictable are people's jobs?"

If they are very predictable then everything is easy. You can simply push information to users when they need it. However a characteristic of most knowledge workers' jobs (and I include knowledge workers in factories as much as in offices) is that they are highly unpredictable. The order in which tasks occur may vary. Methods provide a useful guide for the intelligent / experienced but not a repeatable process. So you are dependent on them pulling the information they need at the point that they need it. That assumes that they know your system sufficiently well to use it, that they trust what is in there and that your system is usable for them. Those are big assumptions. Esp. if they hardly ever use it because it sits outside their normal working activities.

So here are two hypotheses:
1. Push approaches to information provision can work well when they are linked to predictable events.
2. You need to work on pull approaches for unpredictable events. And pull approaches, by their very nature, must be user-driven.

KM strategies started off as document-centric (& some still are). They have become people-centric (CoPs, PKM). Do they also need to become event-centric?

I feel a 2 x 2 matrix coming on that plots events by frequency vs. impact. You might identify events using narrative techniques (such as future backwards), incident logs, process maps or participant observation. You would then identify events that are handled successfully vs. those that are not. There is much to learn by studying both...