How is knowledge created? And what's the best way to put that knowledge to use? Those two questions have long been central to the work of Etienne Wenger, an independent researcher, consultant and author. The 49-year-old Wenger, a native of Switzerland, has spent his career spreading the concept of "communities of practice"groups of people within organizations working together to create and apply knowledge.
Wenger started out as a teacher, and then turned to computer science. His study of artificial intelligence and tutoring systems led him to the belief that such systems neglect how knowledge becomes meaningful to people, so he began investigating how people learn on the job. "He helped us begin to see the social side of learning," says John Seely Brown, chief scientist at Xerox Corp. and co-author of The Social Life of Information. "This work absolutely started with Etienne. He is the main figure."
In their new book, Cultivating Communities of Practice: A Guide to Managing Knowledge, Wenger and coauthors Richard McDermott and William Snyder provide a practical guide to creating communities of practice: how to launch and manage them, and how to measure their contribution to the corporation. His message has important implications for CIOs who've been puzzling over how best to exploit knowledge. Such companies as DaimlerChrysler Corp., Shell Oil Co. and McKinsey & Co. have used communities of practice to improve products and processes, explore technologies, and build expertise. By creating organizational capabilities and spotting trends, communities can play a vital strategic role.
"Technology has to make the expert a better expert, rather than replace the expert," Wenger told executive editor Allan Alter in a recent interview. The true task of technology, technologists and CIOs isn't to store and channel information, but to help communities form, be effective and contribute to corporate strategy.
CIO INSIGHT: What's wrong with knowledge management?
WENGER: There is nothing wrong with knowledge management in principle. Knowledge is a critical asset in the knowledge economy, and it has to be managed like any critical asset. More and more firms, such as consulting firms, really are just about knowledge management; that's what they do. What's wrong with knowledge management in practice is that it started in IT.
Knowledge is not just an IT problem. Knowledge is not something that you can simply capture in a database and then manage. To equate knowledge with information is to not see what it takes for knowledge to be knowledge. Science, which may be the most respected way our society creates knowledge, is a very social process of people convincing each other what the facts are. What allows you to claim that you produce knowledge is a process of participating in a scientific community.
People are experts to the extent that they are members of communities. When you go to a doctor, you don't especially care about the personal opinion of the doctor. You want that doctor to represent the best thinking of the community in his area of specialty. If a doctor goes to court for malpractice, the question is not whether the patient lived or died, but whether that doctor applied reasonable medical practice as defined by the community. The best knowledge of the community is what defines you as a professional.
What is the relationship between knowledge management and communities of practice?
Communities of practice are a key to knowledge management, because the practitioners, the people who use the knowledge in their day-to-day work, are in the best position to actually manage that knowledge. They know what knowledge they need, they know how much of that knowledge must be made explicit versus how much they can keep as a tacit understanding among themselves. So the idea of a community-based approach to knowledge management is that it is not the knowledge manager who manages the knowledge, it's the community of practitioners themselves.
The idea of many knowledge management systems was to create a repository that would replace communities by replacing the interaction. Communication was to be mediated by a big Lotus Notes database. I'm not saying technology is irrelevant, but it has to be in the service of community, not a replacement for this community.
So you don't want to push the technology for technology's sake. You want to see how this community functions and what the best technologies to make it function as a community are. Sometimes, it will need a very complex database with very complex search capabilities because it is creating a lot of documents. Simple technologies like teleconferencing and e-mail, and more complex technologies like discussion databases, Web conferencing, collaborative technologies and application sharing can be very useful, particularly if a community is distributed geographically and wants to solve problems together.
We're talking about the same technologies, but put in the service of communities that are the stewards of an area of expertise, rather than replacing the community process with technology.
That's different from having some group in headquarters create and manage knowledge for other people to apply. It's also different from thinking you can just put all of this in a database. It's saying that knowledge is a living thing and it lives in communities. Knowledge is the property of communities.
Knowledge management isn't just about sharing knowledge. People are already dealing with information overload; it's not like people need more to share. This idea of diffused, boundaryless knowledge sharing doesn't work because people simply cannot read everything everybody writes. People need ways to manage their attention. The community is a way to deal with information overload; your interaction with those people has a high chance of being relevant because you know that you share an interest.
Now, once you have a community, you can take it to the next stage, which is creating a public face for the community so that its knowledge can be shared with others. Many communities have a help desk or a technology process by which you can ask a specific question even if you are not a member of the communitya visible place where they can go for information. It provides a point of focus to knowledge within an organization.
Who has used technology well to support a community?
Raytheon Co. has a system that was originally designed for a Six Sigma project. What's interesting is it's an unusual use of project management software. Its project managers use the system to find others who have experience in the kind of project they are starting. It's an alternative to a corporate yellow pages, because instead of finding people through knowledge that they claim they have via a statement of expertise, you find people through experience that they've had that's relevant to your experience. You can see the roles that they've played, and some documents they may have produced. So it's through their work that you find them, not through some statement.
What can the IT organization offer communities?
Most communities don't want to be bothered with all the technological details. To have an IS organization that's really attuned to their needs and ready to enable them to function as communities is a great contribution.
If the IS organization is the steward for knowledge management, you want to offer not just IT support and support for the "cybrarian"the community member in charge of its databasebut also coaching to help community leaders learn how to do their job better. Most organizations that use communities of practice on a systematic and broad scale have what they call a "support team," a group of two to five people who specialize in helping communities launch, sustain themselves, grow, develop their knowledge and be useful to their members.
But the one thing communities really can use is sponsorship: the voice in senior management that says, "Yes, this is important for the business, and we're going to support the work of these communities." Senior-level sponsorship is very important because they have to remove the barriersfor instance, making it legitimate for people to spend time with their communities and speak in their performance reviews about the work they have done in a community. And that takes influence.
The CIO's organization can also provide a community-developing service. That's happening at Johnson & Johnson. At first, the CIO was building communities internally among MIS people around different MIS topics. But now, they are offering the service of developing communities to the rest of the organization. And they are building a whole methodology for building communities, within and across J&J's operating companies.
Is there a risk of overmanaging communities?
Yes. A community of practice is not something you're going to want to micromanage, because if you knew how to micromanage them you would not need them. You need them precisely because they have an area of expertise and are the only ones who know how to maintain it at a world-class level. And there would be resentment if they were micromanaged. People would say, "Wait a minute, aren't we the people who know how to do that?"
Managing a group that organizes itself is tricky. You cannot simply start communities in the way that you could start a team, by saying, "You, you, and you go in and do that." There has to be an interest in what others are doing, and a sense of value from being with one another. Sometimes when I say communities are made out of experts who are the only ones who know how to manage their knowledge, managers say, "Oh, OK, you mean I should leave them alone. If I can't control them then I should leave them alone." But the point is that you should not leave them alone. You should engage them in a conversation about how to run the business. Managers need to learn to engage with groups they don't fully control.
So the value of communities extends beyond the creation of knowledge?
Yes. But if the CIO takes on the sponsorship role, his role isn't just to make it possible for these communities to exist, but to give them a voice in the business. And in doing so, CIOs can become very involved in their company's strategic conversation. You see, communities of practices are not just about technology issues or sharing knowledge. They are also about strategy. They are about building capabilities in the organization that are essential to the business strategy.
An electrical engineer once said to me, "What's neat about a community of practice is that I have a voice as a professional. I'm not just a fungible entity that you can put on one project and then another. In my community, I can talk about what is the right strategy for the organization. And if we have the right strategy, then what do we need to do?"
Communities aren't just infrastructure; they're right there at the strategic front line. You need to decide which communities are really essential and which communities you need to dialogue with in order to understand the future of a technology or market you want to go into, because these people are close to the ground; they are the practitioners.
How can you tap their knowledge to engage them in a strategy conversation? From a CIO standpoint, it requires a different kind of engagement with the organization. If CIOs are to cultivate communities of practice within their own organization and start offering the development of communities as a service for the organization at large, it requires a different relationship between the CIO and the places in the organization where strategy conversations are being held. These conversations ultimately decide what technology capability needs to be built. So I think for CIOs, communities of practice can be an exciting opportunity to become engaged in strategy.
There's more than simply creating a supporting technology infrastructure for communities. It's seeing the potential in these communities, then engaging yourself, their members, and the line leadership in the strategy conversation, that allows you to have a knowledge strategy. And building a knowledge strategy is really what we are talking about.
Please send comments on this story to firstname.lastname@example.org.
This article was originally published on 05-15-2002