Knowledge Management is a systematic approach to organisational learning; but learning about what? Learning to perform, is the answer. Learning to perform means the identification of better ways to do things, and then sharing and replicating those better ways, and embedding them into the processes and structures of the organisation. We can measure the result of this learning in learning curves.
But "identifying the better ways" requires metrics, and effective performance metrics can be challenging to find.
Let's take the case of a national health service as an example.
We know that metrics has been a very difficult topic in nationalised health services - that many metrics that have historically been applied by government were at best arbitrary and that some may have led to gaming-style behaviours. As a result there has been a suspicion of metrics.
A friend of mine was involved in an attempt to set up a community of practice for hospital head nurses. They had a few meetings, exchanged interesting stories about how they did things in their own hospital, and then went back home and carried on doing things they way they always had.
What was the incentive for them to change their ways of working? Why would they go through all the upheaval of change, unless they were convinced that there was a better way of doing things? Those other stories were interesting, but "we do things differently here".
That's where metrics come in. They help prove that there is a better way to do things that you never thought of, and give you the incentive to learn. And they tell you who to learn from as well.
Imagine that the head nurses, instead of sharing stories, had chosen one or two key metrics. Infection prevention, for example. Pain management. Medication administration. Imagine they had compared numbers, and found the best performers. Imagine they had all committed to improving their own metric, through learning from the others.
That's where knowledge management takes over - in facilitating that learning, knowledge sharing, and knowledge re-use.
Then imagine them applying the knowledge, and seeing the metrics improve. That, for them, would be a proof of concept for knowledge management.
If you find yourself in a context where metrics are not easy, then find something - something at all - that you can measure that is important, and apply knowledge management to that. You will find that the most effective place to start.