Chris Collison: Making KM Practical. Chris Collison, Centrica
A key to successful KM is leadership behaviours. These will determine whether organisational culture can change and whether it can be sustained without the centre.
Chris' wife is a school teacher working in a primary school which has adapted accelerated learning techniques. This involves a morning “brain gym” where kids do exercises (such as rubbing your head and tummy at the same time) which requires the left & right sides of the brain to be work together. Neuroscience says that this creates new connections in the brain and leads to faster learning.
We can make the metaphor of creating connections across the organisation to make it more effective. KM becomes less about capturing bodies of knowledge and is instead about capturing the relationships, identify the people who know the recipe, getting them talking. You don't use packaged knowledge without asking questions, so who do you ask?
You can consider a continuum ranging from capturing to connections. BP (of which Chris used to be a member) is very much at the connecting end of, whereas the US Army is on the capturing end. There is no right and wrong only a question of finding the right balance.
Chris describes the situation in the aftermath of a hurrican in the southern USA in a state with strong connections to the president. The 82nd airbone (an elite combat unit) is, rather unusually, asked to handle the situation instead of the reserves. The colonel in charge is told: “No screw-ups!” This is a combat “hot-action” officer who has never been involved in a civilian operation before in his life.
He goes to the website of the Centre for Army Lessons Learned and asks: What does the army know about hurricane support clean-up? Four hours later he has:
- a profile for deployment including types & numbers of troops, skills, budget, etc…
- the 10 Q's he will be asked by CNN in the first 30 minutes on the scene
- a list of state & federal agencies to be liased with including local army contacts for each
- volunteers for an advisory group including 2 generals & a colonel
A blend of resources and relationships. A distillation of knowledge with case histories. A committment to a relationship. A total knowledge bundle. This kind of knowledge bank captures the knowledge that can be made explicit and the relationships which can deliver that which cannot.
Learning after doing: The retrospect. Sometimes, after a project is complete, people do not understand what it's successes and failures actually were. The retrospect is held afterwards to find the lessons learned and highlite what has been achieved.
Learning whilst doing: The After-Action-Review (AAR). An army technique which is performed immediately an action has been performed. It may only last 10 minutes. It's very direct, asking 4 simple questions:
- What was the objective
- What was supposed to happen
- What actually happened
- What can we learn from this
AAR's are a good way to figure out: “How are we doing as a team?”
Peer Assistants: It is helpful to be able to identify groups of people who can help us on a project. The idea is that what I know + what you know (i.e. what we know) leads to a greater sense of what is possible. The key is in producing a diverse enough blend of people to avoid group think whilst not making it so big that it becomes impossible to achieve action.
A key to peer assists is a yellow pages service that allows you to identify people who you would like to assist you. Chris disagreed with Dave Snowden's assertion that YP's always fall into disuse because they are too hard to maintain. The idea is to create a personal home page that helps people identify with you (and your perspective) as an individual.
Chris mentioned Tacit which is a package that can scan data sources (such as email) looking for connections between people (probably based upon keyword analysis). This is an alternative approach which may or may not be as effective. Probably both together would be the best way to identify good peers.
What the tacit approach lacks is any sense of who makes a good peer since analysis of email cannot determine who is likely to volunteer information willingly.
What is clear that it is essential to make the system as informal as possible to help establish relationships even before people have actually met.
Lessons learned: Humans have a limitation on absorbing knowledge, repositories are not enough. You need to create bundles of knowledge, distilled for a specific context and with the appropriate relationships to support the use of this knowledge asset. This requires narrative, experience and links to the individuals involved.