conversation matters

Creating a Culture in Which Teams and Workgroups Can Engage in Collective Sensemaking

For teams or workgroups to be effective and competitive they must, 1) understand their customers’ requirements and the frequent changes to those requirements, 2) take action
and be fully cognizant of the consequences of those actions, both intended and unintended, 3) detect changes both in the internal and external environment, and 4) develop their collective understanding of the complexities these many factors create.

(In the rest of this paper I will use the term “team”, to refer to both workgroups and teams.)

To accomplish those tasks team members need to engage in a number of learning behaviors, for example, seeking feedback from each other, customers, and other parts of the
organization; sharing information with each other; asking each other for help; talking about errors or problems; challenging the interpretation of others, experimenting to gain insight; and reflecting together.

However, many of those learning behaviors are perceived by team members as risky. For example, members may fear that admitting an error will make them appear incompetent to others, likewise, asking for help. They may be concerned that if people in positions of power notice such actions it could reduce their chances for promotion or job assignments. Individuals who offer opinions that differ from the rest of the team risk being seen as “not a team player” or worse, as being obstructive. At a minimum the individual risks damaging his or her own self-image. Teams may choose not to risk conducting experiments that would gain them insight or trying out new processes that could potentially improve their performance, fearing that they would be blamed for failures. Argyris (1982) has shown that when team members perceive the possibility of embarrassment or threat, they act in ways that inhibit the team from learning; in short they remain silent or resort to meaningless generalities rather than risk negative consequences.

To overcome the reluctance to speak up, team members need to feel “psychologically safe,” a term Amy Edmondson (1999) applies when team members have a sense of confidence that the team will not embarrass, reject, or punish members for speaking up. Psychological safety is a shared belief, held at the team level, that the team is safe for interpersonal risk-taking. For the most part, this belief is tacit, that is, “taken for granted and not given direct attention by either individuals or by the team as a whole.”

Psychological safety is, however, not the same as group cohesiveness, which, as Janis’ (1982) research has shown, can reduce the willingness to disagree with others’ views – the phenomenon labeled “groupthink.” Nor is psychological safety a matter of team members getting along well together or having no conflict. In fact, psychological safety makes it possible for conflicts to be openly raised and discussed. Edmondson notes that psychological safety is the “presence of a blend of trust, respect for each other's competence, and caring about each other as people.”

If as Edmondson, Argyris, Janis, and others point out, talking about the need for members to trust each other or the leader will not engender psychological safety, how then can psychological safety be achieved in a team?

Edmondson Study
Edmondson conducted a study in 2002 that provides hope that psychological safety can be achieved without lengthy training, special facilitation, or a large-scale change effort. The study was conducted within a large organization where Edmondson held extensive interviews and observations with twelve teams, who were spread across a variety of team types and varying levels. Remarkably, Edmondson found that within the same organization some teams were able to engage in learning behaviors while other teams were not. The findings from this in-depth study indicate that:
• The learning process itself occurs at the team level and is focused on a bounded task(s) or opportunity (e.g. development of a product, strategic planning, delivery of service) and occurs through reflective conversations within the group.
• Culture is localized. It is not the culture of the organization that encourages or discourages the learning behaviors needed for effective reflection on complex issues, it is the culture developed within each specific team.
• “When the group’s belief is that the team is not psychologically safe, individual team members are unwilling to actively and honestly contribute their ideas, evaluations, or suggestions. As a result groups are less able to make sound decisions and implement timely decisions in response to changes in the environment.” (Edmondson 2002)
• Team leadership impacts whether a team is able to develop a sense of psychological safety. In the study, teams that reflected effectively and implemented resulting changes had minimal power differences. The leaders of those teams encouraged input and debate. Where power differences between leader and members were high, little learning occurred.
• Teams at the highest level were as likely to be impacted by power differentials as were frontline teams and were therefore unable to seek feedback from each other, customers, or other parts of the organization; share information among themselves; ask each other for help; talk about errors or problems; challenge each other; experiment to gain insight; or reflect together.

Implications of Edmondson’s Study for Collective Sensemaking
A very hopeful implication of Edmondson’s study is that it is possible to create a culture that supports collective sensemaking within a team or unit, even if other parts of the same organization do not have a compatible culture. This means that it is not necessary to wait on top management support in order for change to occur locally. Nor is it necessary for the whole organization to change in order for any one team to make use of effective learning behaviors.

Another positive implication is that it does not take special skills or facilitation to engage a team in collective sensemaking. The successful teams in this study had no special training or help, yet were able to develop psychological safety within the team.

Power Differential
The study found that the manager/leader role was markedly different in those teams that had less power differential between manager and team members. Thus to develop a culture that supports collective sensemaking may require a role shift for managers from the more traditional “boss” role to one that is coach and convener. Edmondson reports on an interview held with a design consultant who worked with many different teams across the organization. The consultant contrasted the leadership of “Beanstalk,” a team that reflected effectively together and was able to implement resulting changes, with “Radar,” a team that did not reflect together and was unable to make needed changes or course correction. The consultant explains, “’[For Beanstalk] I pick up the phone and call anyone…but for Radar, I have to go through Jan.’ In Beanstalk, any of the members spoke for the team. In Radar, Jan held onto the role of spokesperson. In team meetings, Jan was a ‘boss’ who took on the role of making final decisions. In Beanstalk, Martha was a facilitator who encouraged input and consensus.”

Meeting to Make Sense
Collective sensemaking is the very human action of jointly creating meaning out of the incredible amount of data and input that continuously surrounds an organizational unit or team. Collective sensemaking is a creative act, rather than an act of discovering or uncovering what is already there. The meaning created through collective sensemaking does not exist before the conversation that creates it. In conversation the meaning that is created is continually revised as new data and new patterns in the data emerge.

Teams cannot learn unless they take the time to bring the whole team together to collectively make sense of what is happening. A leader, who prefers to talk with team members one at a time, can not produce this level of learning. Nor does bringing a team together to hear announcements or presentations produce understanding of complex situations. For learning to occur meetings need to be convened in a conversational format where team members address their comments and questions to other team members rather than focusing primarily on the leader.

Learning occurs when team members take action and then meet together to reflect on the intended and unintended consequence of that action. It requires the multiple perspectives of all the team members, each of whom has experienced the consequences in a different way and each of whom interprets the meaning of those consequences differently, in order to provide the rich medium needed for the team to make sense of what happened and what needs to happen next. Weick (1995) notes that “the understanding that results from sensemaking is not a definitive answer, rather it is a configuration that is adequate for the organization to plan and take its next action.”

The Quality of the Conversation
Edmondson’s study indicates that taking the time to reflect together was necessary but not sufficient for members to engage in the risky behavior necessary for learning to occur. Some teams in the study took the time to discuss issues but the discussion was of such a low quality that neither new understanding nor action resulted. As an example, the task of a high level team, labeled “Strategy,” was to create a new business strategy for the organization. Although the team met frequently and held lengthy conversations, during the six months of the study, no plan was developed. Edmondson provides this example of “Strategy’s” dialogue in which, George, a senior manufacturing executive, responds to an earlier metaphor about directing the “ship” of the company by turning the rudder.

George: Listening to Bob talk about the ship, I’d like to explore the difference between the metaphor of the ship and how the rudder gets turned and when, in contrast to a flotilla, where there’s lots of little rudders and we’re trying to orchestrate the flotilla. I think this contrast is important. At one level, we talk about this ship and all the complexities of trying to determine not only its direction but how to operationalize the ship in total to get to a certain place, vs. allowing a certain degree of freedom that the flotilla analogy evokes.

CEO interrupting: There’s a question of doing what you want to do and doing it how you want to do it. But you can’t have people just going off and doing what they want to do. You know, some of them may be playing baseball all day long. But, we have to have some alignment with the corporate directives.”

Reading this dialogue, we can assume that the CEO believes George is advocating an absurd amount of that freedom, yet neither says what he means nor offers concrete suggestions. By speaking abstractly both protect themselves from criticism, and in so doing prevent the group from reaching an understanding that would allow them to move forward. Although this group meets, and even has a spirited dialogue, members do not feel safe enough to say what they mean nor to challenge each other openly. As a result no new knowledge or action is created.

Psychological Safety as a Product of Experience
The power differential is not the only factor that impacts psychological safety. Psychological safety is also a result of the interaction experience members have in the group, both with each other and with the leader. Psychological safety begins to grow when a member of a group admits an error or challenges another and then experiences the response of other team members as curiosity rather than blame. Other team members, witnessing a non-blaming interaction are more likely, perhaps at a later time, to offer an idea of their own. When enough safe interactions have occurred, the group as a whole begins to feel the environment is safe. Edmondson gives an example of a dialogue in Beanstalk that represents how conversation might function when psychological safety exists.

After one member, Angela, described, ‘printer problems with those labels’ and asked, ‘Who can we ask for help?’ another member, Rob, responded, ‘How about asking the vendors who make the labels? They probably know how to fix it.’ And Ken offered to make a phone call – closing the loop. Rob also reported on his use of new, trial equipment for conducting these tests, ‘I used the “color analyzer” [he paused} I know it’s not the right word' and looked to Ken for help. Ken responded supportively, ‘photospectrometer.’ Rob continued, ‘It’s worth the $12,000 because we will save $25,000.’ Ken agreed with Rob’s assessment and promised to follow through on acquiring the machine.

As this dialogue illustrates, although a courageous team member is needed to initiate the dialogue, it is the way others respond that is significant in creating psychological safety.

Silence and Support
The silence of other team members is as discouraging to a member who raises a different opinion or challenge, as is a blaming response. Other team members may agree with a person that speaks up, and after the meeting may even tell the challenger that what he said was right. But the fact that the listener does not chance agreeing during the meeting reinforces the belief that speaking up is too risky.

However, an interesting phenomena identified by Ashe (1952) is that if, during the meeting, even one other person on the team responds in a supportive manner, the member who raised the issue is encouraged to speak up again. Thus team members themselves can begin to alter the culture of a group by supporting each other in meeting settings.

Conversations That Build Relationships
A more deliberate technique to begin to change the culture of the group is structured socialization – time set aside during a meeting for small groups to talk about issues that allow them to learn about each other’s competence and allow them to establish relationships at a deeper level. The issues for discussion should be work related, but also topics that provide the opportunity for each person to disclose values and beliefs. For example:
• “What gives meaning to your work?”
• “Tell a story about the best team you’ve ever been a part of.”
• “Talk about a highlight experience you’ve had in working for this company, a time when you felt you made a real contribution.”

These appreciative topics are best discussed in groups of two or three that are small enough for the listeners to express their positive reception to what is being said. That expression might be verbal, a nod, or a smile - any of which indicates to the speaker that what he is saying is accepted. As mentioned earlier, Edmondson found that psychological trust is the “presence of a blend of trust, respect for each other's competence, and caring about each other as people.” Respect for competence to some extent grows out of daily interaction while working on a project or task, but structured socialization can provide the opportunity to learn about past projects and successes, extending that knowledge in ways that may not occur in daily exchanges. Likewise to build “caring about each other as people” there must be opportunities for team members to reveal aspects of self that are not part of normal workplace discourse and to reveal it in a way that listeners recognize themselves in the telling.

Shared Experience
A final suggestion comes from the work of Karl Weick (1995) who acknowledges that many theorists see shared meaning as a way to produce effective teams. However, Weick suggests that it may not be possible for team members to create shared meaning since any meaning an individual creates is a product of their past experiences, which is necessarily unique to each. He explains, “…so if people share anything, what they share are actions, activities, moments of conversations, and joint tasks, each of which they then make sense of using categories that are more idiosyncratic. ....if people want to share meaning, then they need to talk about their shared experience in close proximity to its occurrence and hammer out a common way to encode it and talk about it.” Some shared experiences naturally occur because of the joint work teams engage in, (for example, "that difficult client we worked for" or "the well we drilled in the North Sea"). He notes that “in those situations the critical element is time to process (reflect) what we learned from the experience. Not everyone will learn the same thing from the same experience but that is not critical, what is critical is that they are able to reference the same event that others recognize. The question always is ‘What does it mean?’”

In situations where the work itself does not provide shared experience it is helpful to design events. An example of a designed shared experience comes from Kaiser Foundation Health plan (KFHP) and Kaiser Foundation Hospitals (KFH). The health care system practices a form of Process Improvement that requires employees from a unit, for example the emergency room, to work together as a team. Alide Chase, Senior Vice President of Quality and Service, explains the joint experience each team has before beginning their Process Improvement training.

Before each new team begins their PI training, they engage in Borrow Forward. A team of 4-5 persons from, for example the neonatal unit, makes a visit to a neonatal unit in another location that has already successfully implemented adaptive design. The visiting team goes in a state of inquiry. Each visiting team member shadows his/her counterpart for 2 to 3 days. Before leaving, the visiting team holds a meeting with those they shadowed to talk with them about the observations and insights they experienced. This is a reflective meeting in which both parties learn. The visiting team articulates their insights (see we learn when we talk) which helps them clarify what they learned for themselves and the host team gains new understanding of their own processes by seeing those processes from a different perspective.

After their return, the visiting team begins their PI training and implementation. They hold weekly meetings to reflect on the actions they have taken and the results achieved. Borrow Forward provides a shared experience from which they derive shared meaning, and provides a lens through which they are able to think about their work.

Summary
For a team to be effective and competitive it must be engaged in learning behaviors that are too often perceived as risky by members of the team. To take that risk, team members need to feel psychologically safe, that is, “have a sense of confidence that the team will not embarrass, reject, or punish members for speaking up.” The actions that help to bring about collective sensemaking are:
• reducing the power differential between leaders and members
• teams taking the time to reflect together on a regular basis about their actions, results, concerns, and innovative new ideas
• members actively providing support for each other in meetings
• holding small group discussions about appreciative topics to build relationships and enhance the knowledge of others’ competence
• engaging in shared experiences that serve as a reference point for meaning.

References
Amy Edmonson, Psychological safety and learning behavior in work teams. Administrative Science Quarterly, June, 1999

Amy C. Edmondson, The local and Variegated Nature of Learning in Organizations: A Group-Level Perspective, Organization Science, March/April, 2002

Chris Argyris, Reasoning, Learning and Action: Individual and Organizational, San Francisco, CA: Jossey-Bass, 1982

Solomon Asch, Social Psychology, New York: Prentice-Hall, 1952

Irving Janis, Groupthink: Psychological Studies of Policy Decisions and Fiascoes 2nd Ed, Cengage Learning, 1982

Karl Weick, Sensemaking in Organizations, Thousand Oaks, CA: Sage, 1995

Marissa is right: Yahoo just needs the Oscillation Principle

Marissa is right! In order for Yahoo to have a culture of collaboration employees need to be face-to-face with each other to talk about the critical issues they collectively find themselves facing. But it is equally true that we can’t turn back the clock on virtual work nor on the growing distributed workforce that is serving us so well around the world. Virtual is here to stay.

Forcing people back to the workplace is not the solution because too often when they are in the workplace they are either sitting in a meeting listening to endless presentations, or in a cubicle sending emails to each other. Neither of those activities is worth the cost in time or travel. The only reason to come together face-to-face is for people to be in conversation with each other! And real conversation happens all too infrequently in workplaces, as my own research has shown.

The Oscillation Principle
What a distributed workforce needs, in order to work effectively, is a regularly scheduled oscillation between virtual work and collective sensemaking. Collective Sensemaking is an organized conversation, intentionally held to make sense of the circumstances in which organizational members collectively find themselves. Taylor and Van Every (2000) explain, “Sensemaking is a way station on the road to a consensually constructed, coordinated system of action.”

The deep rhythm of oscillation between face-to-face meetings for collective sensemaking and virtual work, addresses one of the greatest deficits of a virtual work force, that is, one part of the organization taking action without reference to how that action may impact other parts of the organization or impact the whole. Through collective sensemaking all perspectives on a topic are given voice so that an understanding of the whole emerges as well as clarity about the relationship between the parts. In such conversations organizational members often discover assets of which they were unaware.

As this diagram (Maznevski & Chudoba 2000) illustrates there are high intensity periods of collective sensemaking oscillating with periods of virtual work in which the group interacts through less rich media, such as, email, phone calls, SMS or teleconference. The periods of collective sensemaking renew the trust and relationships which are a precondition for the collaboration and information exchange that are limited to mediated interaction when organizational members are again at a distance. And understandably, those virtual interactions are more productive because critical relationships have been renewed during collective sensemaking.

The frequency of the oscillation and duration of the periods of collective sensemaking depend on two factors:
1. the interdependencies in the group's task, and
2. the complexity of the issues.
For example, a virtual team that is designing a product for an emerging market might need to come together for 2 days every two months. A team with less complex issues might come together every 3 months for one day. And a division engaged in a change initiative – one that requires employees to take into account the whole system so that changes planned for one part do not conflict with other parts - might bring everyone together to initiate the change and then come together as a whole every six months, with individual departments meeting for a day once a month.

Collective Sensemaking
If you were to observe collective sensemaking in a daylong meeting, you would see posted, a list of work issues that team members have identified for discussion. All day long people would be working in a series of small groups, some times homogenous and some times mixed. And periodically during the day you would see the whole group gathering to exchange the ideas they had been working on. If you listened in on the conversations you would hear people sharing information, asking for help from others, discussing unexpected outcomes, and working through differences. What you would not see would be motivational speeches or PowerPoint presentations.

Collective sensemaking allows a group to build shared understanding of what has happened since the last time they were together and what needs to happen going forward. “Organization members interpret their environment in and through interactions with others, constructing accounts that allow them to comprehend the world and act collectively.” (Maitlis 2005) That is what Marissa wants for Yahoo - employees interacting enough that they can act collectively. Oscillation between virtual work and collective sensemaking would achieve that end without having to eliminate the virtual work that is so productive.

How Knowledge Workers Learn Judgment

In today’s world employees often must deal with tasks and problems that require much more that simply following a predetermined step by step procedure – the problems require the exercise of judgment. Judgment is needed when we are faced with thorny questions about which there are no right or wrong answers. When asked one of those thorny questions we often say, “Well, it’s a matter of judgment.” Soldiers, for example, face many situations in which one correct course in not clear, as in this example from one of the US Army on-line communities, PlatoonLeader.

“We were headed out to a pretty easy mission. My front truck reports that there’s a dead body under a car in the middle of the road. He was on the ground. He had been in his car and he had been shot. At that point, the gunner from my lead truck noticed a double-decker bus that had stopped. And there was a guy up on the top deck who appeared to have a blue video camera, and he was just hanging out the window video-taping us. Our rules of engagement permitted us to engage anybody video-taping an attack. I looked down my sight and noticed the same thing. The gunner looked through binos and noticed the same thing. He asked me, “Hey sir, can I go ahead and shoot him?”

In the above example the Platoon sergeant must figure out if the video-taping implies an ambush or if it is just some guy video taping a strange scene.

Weick provides an example in which a nurse in a neonatal unit must make two judgment calls.

“I took care of a 900-gram baby who was about 26 or 27 weeks many years ago who had been doing well for about two weeks. He had an open ductus that day. The difference between the way he looked at 9 a.m. and the way he looked at 11 a.m. was very dramatic. I was at that point really concerned about what was going to happen next. There are a lot of complications of the patent ductus, not just in itself, but the fact that it causes a lot of other things. I was really concerned that the baby was starting to show symptoms of all of them.”

You look at this kid because you know this kid, and you know what he looked like two hours ago. It is a dramatic difference to you, but it’s hard to describe that to someone in words. You go to the resident and say: “Look, I’m really worried about X, Y, Z,” and they go: “OK.” Then you wait one half hour to 40 minutes, then you go to the Fellow (the teaching physician supervising the resident) and say: “You know, I am really worried about X, Y, Z.” They say: “We’ll talk about it on rounds.”

The nurse faces two situations that require judgment, the first is to recognize that the baby is in trouble and the second is how to convey that information to a physician in a way will cause him to turn his attention to the infant.

Such situations occur more and more frequently, not only in the military and hospitals but in much of the work we think of as knowledge work, everything from an architect thinking about a design for a steeply sloping site, to an intelligence analyst determining whether the troop build up on a border signals a threat, to a manager faced with how to implement a change initiative.

In order to respond to these kinds of situations, employees need more than the skills they learn in training, more than what is provided in a manual of regulations, and even more than what can be picked up through reading the best practice of others. There is no way to anticipate all the possible situations a knowledge worker could face during their daily work, and therefore no way to provide sufficient procedures or directions. Rather knowledge workers need to continuously read the situation in front of them and then, based on that interpretation, determine the appropriate next action to take - in other words, to use their judgment.

In Talking About Machines, Julian Orr describes the lunch meetings that copy repair technicians hold to solve the difficult problems they face fixing copy machines. He notes that much of their talk is telling stories about the machines they have been working on. The stories serve to make sense of diagnoses confronting them, but also to reaffirm what it means to be a “competent” repair technician in terms of work habits, what clothing is worn to a client site, how a repair technician approaches a problem, how one works with colleagues and a host of other values and norms that turns a person into an effective repair technician.

In contrast to these informal lunch meetings, Orr describes the view of the corporation about how repair problems are solved, that is, that problems are solved by using the documentation issued to the repair technicians. “The diagnostic procedures prescribe a series of tests, with each action defined in considerable detail, and each branching condition presented as a simple Yes/No choice. Such documentation is based on the scientific management principle that if instructions are detailed and complete, the organization can hire cheaper employees, with less skill, who can do the job by following the instructions contained in the documentation. However, to the technicians the actual question to which they are asked to respond Yes or No is often extremely convoluted… The technicians are quite philosophical about the shortcomings of the documentation, saying that, ‘the machine is far too complex to anticipate correctly all of the possible failures.’ ... They view the documentation as a useful resource to consult when their own expertise cannot solve the machine’s problem.” But it is in the conversations where they learn to deal with the tricky problems that never show up in the documentation.

The corporate view, as reported by the copy repair technicians, is that knowledge and skill are a matter of individual competence, which is gained by attending training, reading journals, and/or listening to lectures. The underlying assumptions of that view are that, 1) there are individuals with expertise who can provide the knowledge required to be effective, through documents or lecture, and 2) that the required knowledge is relatively stable, it changes little over time.

As more and more of the workforce is populated by knowledge workers our premise about of how people develop the judgment to be effective is changing. The newer view holds that:
1. complex knowledge and skills are distributed across the practitioners who use that skill, with no one individual knowing all that the group knows, and
2. knowledge is continually changing as the group of practitioners learn from the act of practicing their craft. Ideas are not fixed and elements of thought are formed and reformed through experience. Knowledge then is not stable, but is ever changing.

The newer view takes into account the difference between “know what” and “know how.” For example, it is possible, to learn the “know what” of negotiation strategies by reading the books about negotiation. But books cannot make a person a skilled negotiator. To be an effective negotiator, requires “know how” or what I have earlier called judgment.

How then do people learn to make judgments? There are three elements involved in developing judgment:
• The most fundamental element is the individual taking action and observing the results – in other words experience. Experience is gained, not through one occurrence, but over enough occurrences that a pattern begins to emerge across all of the actions taken. This implies the liberty to experiment, to try new things and, of course, to fail. Edmondson calls this “intelligent failures at the frontier.” Judgment cannot be learned in the absence of failure because the breadth of experience would be too narrow. If an organization punishes failure or if employees feel the need to hide failure, there is little opportunity to develop judgment.
• The second element is reflection on those actions and patterns. Reflection requires thinking back on the actions taken, the reasoning behind the choice of action, the context in which the action took place, and the outcomes that resulted from the actions, many of which are unintentional. While experience is inevitable, learning is not. To learn from requires deliberate reflection. Reflection is most effective when it is done in the company of others who are involved in the same kind of work. When we summarize what happened for someone else, we begin to understand it better ourselves. We learn when we talk. Each practitioner functions within a somewhat different context, responding a bit differently. Consequently the learning of each practitioner is unique. When those practitioners reflect together on their actions, “creative abrasion” occurs, which leads to greater understanding in those engaged in the conversation. There are a number of processes to choose from that are both conversational and systematic in nature, for example Action Learning, Knowledge Café, Anecdote circles collective sensemaking, coaching ourselves.
• The third element is associating oneself with a community of practitioners and thereby understanding the work and it’s talk from the inside.” One learns to be an effective nurse, copy repair technician, or soldier by talking with others about their work. Such talk develops not only the necessary understanding of technique, but also of values held, norms of behavior and how to think like a practitioner in that field. In other words, how “to be.”

These elements, to a great or lesser degree, are present in the informal interaction among people doing the same work, as we saw in the copy repair example. But in a world where the community of practitioners is spread across the globe, and many practitioners work at locations out of the office, the development of judgment cannot not be left to chance and proximity. It must be designed and supported by the organization including: 1) experimentation that leads to learning, 2) treating failure as an opportunity for learning, 3) establishing a systematic process through which reflective conversation occurs about both team and individual actions, 4) and promoting communities of practice.

The Power of Action Learning: A Process for Building a Collaborative Culture

One of the most powerful tools for moving an organization’s culture toward collaboration and knowledge sharing is a process called Action Learning. It is not a new process, but one that was used both in Europe and the US, long before the term “Knowledge Management” was first heard in organizations.

I find that many KM professionals, particularly those whose background is more based in technology, often remain unaware of the wealth of useful knowledge management processes available to them from disciplines such as organizational development, change management, and organizational learning. Action Learning is one of those, as are many of the other processes I have referenced in blog posts for example, Knowledge Cafe, Open Space Technology, Participatory Action Research, and Appreciative Inquiry.

Action Learning is based on taking one or more crucial organizational problems or opportunities and having small groups of managers, analyze their dynamics, implement proposed solutions derived from the insightful questions of their colleagues; monitor
the results; and through being held responsible for these actions, learn from the results so that they develop greater competence in future problem solving and opportunity taking.

There are four elements that make Action Learning effective:
1. A crucial organizational problem/opportunity
2. Organizational members willing to take risks to develop themselves and their organization
3. Authority to take action on the problem/opportunity
4. A system for learning reflectively – sets, which are small groups that meet regularly over several months

There are many formats for Action Learning. Two of the most frequently used formats I describe here, 1) the members of the set each work on a different problem/opportunity, and 2) all members of the set work on the same problem/opportunity.

Format 1 - Each Set Member Works on a Different Problem/Opportunity

• Each set is composed of 4-6 participants.
• Set members are drawn from different parts of the organization in order to bring a variety of perspectives to the issues they deal with.
• Each set member, with their supervisor, identifies a significant problem/opportunity within the supervisor’s sphere of influence. The set member develops a solution and then implements that solution. For example:
⁃ Design and implement a collaborative process for long-term strategic vision development and deployment
⁃ Design and implement a materials-flow process for a new body shop
⁃ Develop and implement a new billing system
• Set members, with their supervisor, also identify a developmental challenge for the set member to work on.
• A set advisor is appointed to help the set reflect on what they are learning from the actions they take.
• Typically, sets hold a day long meeting every two weeks over a period of four to six months.
• In between meetings set members take action on their issues and then report back on the results at the following meeting.
• During set meetings each member has an hour to focus on his/her problem/opportunity. To help the set member uncover assumptions, other members adopt a collaborative, reflective, questioning approach.
• Set members meet regularly with their supervisor to report on their progress with their problem/opportunity.

Format 2 - Set Members Work as a Group on the Same Problem/Opportunity

• The Sponsor of Action Learning, who is typically a high level leader, identifies several difficult, real problems/opportunities that the organization faces. They are problems/opportunities in which:
o No clear expertise already exists
o There are no known answers or there have been many failed attempts to resolve the issue
• Each set is composed of 4-6 employees who come from different disciplines across the organization.
• From the Sponsor’s list of problems/opportunities each set selects an issue in which they feel a vested interest. The set members 1) study the issue, 2) develop a solution, and 3) work with the appropriate client to implement the solution. For example:
o Design and implement a solution to improve the company’s performance in responding to customers’ requests
o Create and implement a process for attracting and retaining technology partners
• A set advisor is appointed to help the set with their reflections.
• Set members are expected to devoted 20-30% of their time to the problem for a period of 6 months. In some organizations set members spend full time on the problem/opportunity for a period of 3-4 months.
• Each set member identifies a developmental goal to pursue while working on the group problem. For example:
o Learning to express empathy for others
o Developing skills in handling conflict
o Developing skills of influence without authority
• The set holds a daylong meeting every two weeks.
• In the set meetings, set members have their own airtime to report on their progress and to respond to insightful questions from other set members.
• The set invites into set meetings any expertise or resource they need in dealing with the problem/opportunity they are working on.
• Each set has a high level manager who monitors their progress and who can open doors as needed for the set members to study the problem/opportunity and/or to implement the solution.
• The set reports to the sponsor on the outcomes it has achieved at the final meeting.

Most action learning programs have 5-6 sets working at the same time. A kick-off meeting is held with all the sets coming together as a large group in a workshop. The kick-off is attended by set members, their direct supervisors, set advisors, clients and the sponsor. A mid-term workshop is held to provide the opportunity for sets to learn from each other’s experience and a final workshop to continue that learning and to report on outcomes. The following diagram illustrates the set and workshop meetings.

The power of Action Learn comes from releasing and reinterpreting the accumulated experiences of members of the organization. The combination of this released energy and the act of moving the authority for problem solving to those people who must live with the consequences, changes the culture to one that is more collaborative, more willing to share knowledge, and more able to initiate change. As one participant noted,

“The result of working together on Action Learning is that now people are more willing to co-operate with one another, to share their expertise with one another… and to use the phone! Morale has definitely improved. There’s more understanding that we survive collectively and not individually…They’re also coming up with ideas for what else we could be doing differently – that wouldn’t have happened two years ago.”

I think of Action Learning as the earliest Knowledge Management process because of the principles upon which it is based:
• The format designed for peers to learn from each other
• The value placed on learning from experience
• The emphasis on "reflection on action"
• The importance of asking questions to pull rather than push knowledge
• The small group viewed as the unit of learning
• Conversation seen as the most effective means of learning between peers

Building a culture of collaboration
Action learning builds a culture of team work and of collaboration. As set members meet together over time, a community of learning develops. Set members come to realize that employees in other parts of the organization are much like themselves, rather than thinking of them as the unknown “others.” As Action Learning continues with more sets meeting, a networked organization begins to develop. Set members, who have come to know and trust colleagues in other departments, find themselves willing to reach out to them to solve issues that might take months to solve without those relationships.

Many organizations have used Action Learning over the years to build collaborative and networked organizations including: Boeing, Caterpillar, Conoco, Dupont, GE, IBM, NASA, Novartis, Nokia, Samsung, Siemens, Sodexho, US Post Office, and Unisys.

I’ve had the opportunity to set up Action Learning programs in a number of those organizations and I consistently find it one of the best ways to foster the cultural side of knowledge management.

Improving Knowledge Worker Productivity

As Knowledge Management professionals our job is the help organizations leverage their knowledge. Our attention is focused on the knowledge worker and our major task is to devise ways for those knowledge workers to share the knowledge they have gained with their peers. In other blog posts I have suggested a second task KM professionals should be engaged in, that is, helping organizations address the complex issues they face - those that are not amenable to technical or improvement solutions. In this post I propose a third task for knowledge managers, helping organizations improve knowledge worker productivity. I have drawn heavily on the ideas of Peter Drucker, who invented the term knowledge worker and who has been the most prolific and insightful voice in describing how knowledge work should be managed.

Knowledge workers now comprise 40% of the American workforce. According to Morgan Stanley economist, Stephen Roach, knowledge workers are the most rapidly growing segment of white-collar employment. Within the last seven years knowledge worker employment growth has averaged 3.5% per year making their productivity vital to the competiveness of both organizations and the country.

To discuss knowledge worker productivity, it is necessary to first define knowledge workers and to differentiate how they do their work from other kinds of workers. Although there are many possible definitions, I will use Drucker’s simple, but profound definition, “A knowledge worker knows more about how to do the task he has responsibility for than does his boss.” Drucker provides several easily recognized examples of this phenomenon:
 The meteorologist on an air base, who is vastly inferior in rank to the airbase commander, knows more about weather forecasting than the air base commander does.
 An engineer servicing a customer does not know more about the product than the engineering manager does, but she does know more about the customer and the customer’s needs.
 The hospital administrator does not know how to do clinical testing so cannot tell the pathologist in the medical lab what good testing is or how it should be done.

These examples illustrate the first striking difference between knowledge workers and other types of workers. Before knowledge workers became such a critical part of the workforce, bosses knew how to do the tasks of those they supervised, having done the same tasks, often little changed, only a few years before. Knowledge workers, however, need to acquire new knowledge every 4-5 years or else they become obsolete. In the knowledge age, bosses, even if they have done the same task in the past, quickly fall behind.

The second difference is in the nature of work itself. Some work can be clearly laid out in a step-by-step procedure. The skill may be difficult to learn, but there is no question about what to do next. Prescribed steps work very well in an environment where work is largely visible, standalone, and unchanging. Knowledge work, however, is invisible, interdependent and constantly changing. Knowledge workers, whether they are scientists, engineers, marketers, accountants or administrators, must continuously read the situation in front of them and then, based on that interpretation, determine the appropriate next action to take. For example, a physical therapist must ascertain what steps to take with a patient depending upon his assessment of the patient’s condition as well as how the patient is responding to therapy. By necessity a physical therapist must be responsible for his own contribution to a patient’s health. Likewise an intelligence analyst must use her own judgment about the veracity of a source as well as her judgment about the importance of the source’s information to her analysis. The quality of her analytic report is based on that judgment. The second difference then, is the nature of the work, that is, using one’s own judgment versus a clearly defined procedure.

The third difference is that knowledge workers “own the means of production,” to use Drucker’s phrase. That is, the knowledge they possess is in their minds so when they leave the organization the means of production leaves with them. They view that knowledge as their personal possession. According to a study by Constant, Kiesler and Sproull, knowledge workers make a major distinction between tangible information such as written documents or computer programs and intangible information that is learned through experience. They would share a computer program or a document with others because they view it as the property of the organization. But knowledge gained from experience, which reflects on their identity and self worth, is shared only when they receive some personal benefit in return. That benefit could be the admiration of peers, the appreciation of a friend they have helped out or the satisfaction felt from paying back a colleague who once helped them. Knowledge workers view it as their discretion to share the knowledge gained through the hard work of learning from their own experience, or not.

Knowledge workers are not dependent on the organization as are industrial workers whose experience is useful only in the place where it was acquired - it is not portable. Conversely knowledge workers’ knowledge is portable. A petroleum engineer at Chevron is equally valuable to Exxon, perhaps even more so because he or she brings new perspective to difficult issues. If knowledge workers are not dependent on the organization they work for, neither are they independent because, whether full time employees, contract workers, or consultants, they require an organization to do their work. Their status then is interdependent. Knowledge workers function more like an associate than a subordinate which makes their interaction with supervisors closer to a conversation than receiving orders. In other words, each must fully share what they know and listen with a willingness to learn from the other. As Boris Groysberg and Michael Slind say in their new book, Talk, Inc. “Traditional corporate communication must give way to a process that is more dynamic and more sophisticated. Most important, that process must be conversational.”

It is only within the last 10 years that much of the growth in knowledge workers has occurred. This has left most managers unprepared for interacting with this new kind of worker. The contrasts outlined above, show that managing a knowledge worker needs to be very different from the command and control method of management that most organizations currently employ. Not only does command and control not improve knowledge worker productivity, it too often does the opposite - it makes knowledge workers less productive.

My own research on organizational conversations has provided numerous examples of the way command and control management reduces knowledge worker productivity. For example, in the intelligence industry, reports written by analysts typically are reviewed and revised by several supervisors who know much less about the topic than does the analyst who wrote the report – yet by the time analytic reports are published they have, too often, been modified in a way that no longer represents the analysts’ findings. In sales organizations I see situations where sales representatives are told by their bosses to modify customers’ requirements even though the sales people know full well that the changes will result in dissatisfied customers. In healthcare, nurses, who have a great deal of knowledge about how their work could be more effective and productive, have no way to communicate that knowledge to hospital administrators. In the telecommunications industry software analysts follow processes, defined by their bosses, which add little value and serve to unnecessarily delay their work. To be productive knowledge workers have to be responsible for their own work processes.

Taking responsibility for work processes is one of three issues that help to make knowledge workers more productive. The others are, 2) knowledge workers require continuous learning as well as the on-going opportunity to teach what they know to others, and 3) knowledge workers need to be treated as an asset rather than a cost. In this post I address on the first issue and will write about issues two and three in an up-coming post.

Taking Responsibility for Knowledge Worker Productivity
Knowledge workers themselves must take the responsibility for their productivity because they are the only ones who know what productive work should look like. But that does not mean they should have carte blanche to do what ever they want. Rather, to increase their productivity, knowledge workers must design for themselves a system that includes clarifying, 1) what the task of their work is, and 2) what quality looks like in accomplishing that task and how to determine whether, both as individuals and as a group they are meeting that level of quality. Further, they must take responsibility for monitoring the quality of each individual and of the group. Although managers cannot do this task for the knowledge workers, managers can make use of the results in terms of allocating resources and working to remove organizational obstacles that reduce productivity.

Identifying the Task
The first requirement to increase knowledge worker productivity is to bring together knowledge workers that are doing a specific type of job to ask, “What is your task? What should it be? What should you be expected to contribute? and What hampers you in doing your task that should be eliminated?” (Drucker, 1999) The question requires an extended conversation among those who do the work. It is a mistake to leaving it to HR to administer a survey or conduct interviews in order to find the answers. The answer must be worked out through the give and take of conversation among the knowledge workers themselves. Knowledge workers possess both the ideas and the analytic skills to do their own sensemaking. Moreover, if those who will have to live with the results do the sensemaking themselves, they are more likely to be satisfied and supportive of the results than if someone else hands them an answer. By participating in the conversation, they will know, not only what the result was, but also the reasoning that led to that result. Even if some members of the group are not in complete agreement with the outcome, they will understand why that outcome was reached.

The question, “What is your task?” may, at first glance, seem axiomatic, but it is not. The question is not, “What does the organization expect of us?” rather “As the only people who understand our work, what do we think our task should be?”
Once understanding is reached about the task, management can assign the non-value added tasks to workers who have less skill and therefore lower salaries. Drucker describes one such situation. “Nurses in a major hospital were asked the above questions [about their tasks]. They were sharply divided as to what their task was, with one group saying "patient care" and another saying "satisfying the physicians.” However, they were in complete agreement on the things that made them unproductive. They called them "chores" - paperwork, arranging flowers, answering the phone calls of patients' relatives, answering the patients' bells, and so on. All, or nearly all, of these could be turned over to a non-nurse floor clerk, paid a fraction of a nurse's pay. This accomplished, the productivity of the nurses on the floor immediately more than doubled, as measured by the time nurses spent at the patients' beds. Patient satisfaction more than doubled and turnover of nurses (which had been catastrophically high) almost disappeared-all within four months.” (Drucker 1999)

What quality looks like in accomplishing that task, and how to determine whether they are meeting that level of quality
Clarifying the knowledge workers task is the first step, but with responsibility comes accountability. Since bosses do not have the knowledge to hold knowledge workers accountable, the workers must devise a system to monitor and hold each other accountable for quality.

The first accountability discussion is to define what quality looks like. This is again a conversation that knowledge workers themselves must have. It may take several tries to define quality in a way that is accurate and measurable. “Surgeons, for example, are routinely measured, especially by their colleagues, by their success rates in difficult and dangerous procedures (e.g., by the survival rates of their open-heart surgical patients or the full recovery rates of their orthopedic-surgery patients).”

Brand Velocity, a consulting company, provides a useful example of workers holding each other accountable. Brand Velocity developed a points system instead of a hierarchical based compensation system. Points are awarded for selling great work, delivering great work, and recruiting and developing a diverse group of people who can do the same. This system has been in place for five years. “[The points system] means that anyone can earn more money than their boss and earn equity linked directly to his or her individual and collective contributions.”

Points are calculated per engagement. The gross profit is translated into points (gross profit = net revenue –cost related to sales and delivery) and awarded to those who contributed to generating this gross profit. 25% of the points go to the sales lead, a 15% pool is distributed by the sales lead for sales assistance, 30% goes to the delivery lead, and a 30% pool is distributed by the delivery lead for delivery assistance. Points can also be tailored to nonfinancial measures, such as employee recruiting and development. The Brand Velocity team collaboratively decided how many points they wanted to award to valuable contributions, although it took several tries to produce a system they were satisfied with.

This points system is one way workers can hold each other accountable for doing productive work. Another example is provided by Drucker describing a telephone company. “The technologist had to work by himself. He could not be supervised. He, therefore, had to define quality, and he had to deliver it. It took… several years before that was answered. At first the telephone company thought that this meant a sample check, which had supervisors go out and look at a sample (maybe every 20th or 30th job done by an individual service person) and check it for quality. This very soon turned out to be the wrong way of doing the job, annoying servicemen and customers alike. Then the telephone company defined quality as "no complaints"-and they soon found out that only extremely unhappy customers complained. It then had to redefine quality as "positive customer satisfaction.' In the end, this then meant that the serviceman himself controlled quality (e.g., by calling up a week or ten days after he had done a job and asking the customer whether the work was satisfactory and whether there was anything more the technician could possibly do to give the customer the best possible and most satisfactory service.”

Knowledge Management Professionals
There are several ways that knowledge management professionals can assist managers in their efforts to increase knowledge workers' productivity. The first is to educate ourselves about knowledge workers. Secondly, to help managers become aware of the differences between knowledge workers and more traditional workers. In part this can be accomplished by providing articles and research to management to familiarize them with the differences. Some sources are the writings of Drucker, Mintzberg, and Hamel , and the websites of Coaching Ourselves and Management Innovation eXchange. Finally, for mangers who are interested in exploring these ideas, knowledge management professionals can help them set up experiments and then track pre and post levels of productivity

The Three Eras of Knowledge Management

In this video I describe the Three Eras of knowledge management that I have previously written about on this blog, Where Knowledge Management has Been and Where it is Going – Part One, Part Two, and Part Three. My understanding about the third era continues to grow so I have elaborated the third era in this post.

 

In the earlier blog post I called the third era, “collective knowledge” and I remain convinced that collective knowledge is at the heart of this change. It is the "means," but what is being managed are ideas. So I am using “Idea Management" as the label for the third era both in this video and in the accompanying chart. The first two eras, Information Management and Experience Management dealt with existing knowledge, that is, knowledge that an individual or a group has gained and is available to be shared with others. The third era is about the creation or development of ideas that have not existed before. It is not the management of anything organizational members have learned through their work experience, but what they create jointly when they are brought together in an environment that supports the use of collective knowledge. That support includes convening, cognitive diversity and transparency.

In the last couple of years I have had the opportunity to work with organizations that are using third era practices and that experience has sharpened my thinking about what is occurring and why. I’m still not sure I have all the words right, but the big change I see is that management has begun to value the sensemaking capabilities of employees, that is, the ability of employees to jointly make sense of complex situations. Over the three eras management’s awareness of the kinds of knowledge that are valuable continues to expand. Initially management only valued explicit knowledge as evidenced by the resources put against building repositories, then they developed an appreciation for the experiential knowledge of employees and allocated resources to build communities, and now there is a growing acknowledgement of the sensemaking capabilities of employees and an accompanying willingness to provide the resources to convene groups for that purpose.

Of course, management being interested in the opinions, ideas, and knowledge of employees is not new. Organizations have long made use of employee surveys, fireside chats, suggestions boxes, and town hall meetings to collect ideas from employees. But in the past management has reserved for itself the right to make sense of what was collected from employees. The subtext of such practices was, “Tell us [management] your concerns and suggestions and we [management] will figure out a way to fix it.” Now working with leading edge organizations that are, by their actions, saying something quite different, I hear, “Let's convene the people who do this work and have them think together about how to make sense of this issue.”

A recent example of convening is a meeting I facilitated for NASA. Over the years each of the ten NASA Centers has grown its own KM strategy, quite independent of the other Centers and Offices. NASA administration, in recognition of the variability in the effectiveness of the various KM strategies, and spurred by a somewhat critical government report, appointed a CKO, Ed Hoffman, and asked him to develop a NASA-wide KM strategy. Ed is a seasoned KM professional, fully capable of developing such a strategy. But rather than doing that, he chose to convene a three-day meeting that brought together fifty KM professionals from across the ten NASA Centers, to think together about what the knowledge strategy of NASA should be. The meeting made use of the collective knowledge and analytic capabilities of the KM professionals. Outside perspective was provided by Christos Kostopoulos, Senior Economist at Knowledge and Learning Department- Corporate, from the World Bank and Grigg Gurvais, CKO from the FBI. All available information about NASA’s KM programs, internal NASA documents, and the government report was provided to the group. By the end of the meeting the group had identified the elements of a system-wide strategy. The following illustration shows the commitments participants made at the end of the meeting.

The NASA example illustrates the three enablers of the third era, 1) convening, 2) cognitive diversity and 3) transparency.

1. Convening
Convening is the skill and practice of bringing groups together to develop understanding of complex issues, create new knowledge and spur innovation. It is about:
• designing meetings as conversations rather than presentations
• identifying who needs to be in the conversation, including those who do the work and are impacted by it
• framing the question in a way that opens thinking
• arranging the space to facilitate conversation
• using small groups as the unit of learning
I have written about convening and the role of the leader in The Power of the Conversation Architect to Address Complex, Adaptive Challenges

Cognitive Diversity
Cognitive diversity is the deliberate use of difference to bring new understanding to an issue. When faced with complex issues our inclination is to collect more data, survey, or assign a task force to conduct interviews; when what is needed is a new way to frame the issue. Cognitive diversity brings people trained in different heuristics, problem solving strategies, interpretations, and perspectives into the room. Cognitive diversity can be found in different parts of the organization (e.g. marketing, finance, engineering), in different disciplines (e.g. biology, neuroscience, archeology), or outside the organization (e.g. suppliers, customers, consultants, academicians, alliances).

Transparency
Transparency includes the willingness of management to say, “I don’t know” and therefore to employ the organization’s collective knowledge. It is also about management providing all the available information and data on an issue so that those convened have what they need to do the work of sensemaking. Organizational members also have a role in transparency, that is, to be open about what is happening at their level, rather than hiding or discounting bad news to appease management – to bring the best available knowledge to bear on organizational issues

I would welcome other’s thinking on the third era and the supporting elements of convening, transparency and cognitive diversity, as well as useful references and examples.

My thanks to Xuehui Liu, a KM Professional in the Strategy and Planning Department of Huawei Technologies Ltd., Zhenshen, China. In addition to being a skilled knowledge management professional, Mr. Liu proved to have great skills as a videographer.

The Hallways of Learning

A number of years ago I wrote an article, “The Hallways of Learning” published in Organizational Dynamics, in which I suggested that we might look at the creative, open ended conversations we have in the hallways of our organizations as a metaphor for the kind of conversations we need in the many organizational meetings we hold each day. Our hallway conversations have several useful attributes; they remove some of the sense of hierarchy making participants in the conversation more equal; they invite multiple perspectives because anyone who wanders by is invited in; and likewise, people are free to move on if they find the conversation uninteresting.

The article was read by some members of a research group at the National Institute for Working Life (Arbetslivsinstitutet) in Stockholm, Sweden and they saw the ideas as a way to enhance their own collective learning. This was a group of about 20 researchers and, as the name implies, they studied workplace issues, including, environment, safety, labor, and management. The normal practice of this group was for each researcher to work on a different project, although infrequently 2 or 3 would collaborate on a project. The group was in temporary quarters while waiting to move into a new building.

In the new building the research group was to occupy one whole floor of individual offices.

They began to contemplate what a design would look like that would facilitate greater collaboration and learning among them and decided they needed a “Hallway of Learning.” One of the researchers was an architect who was very interested in how space impacts the way people work and learn together. The small group met with her to explain what they were thinking about and asked her help to design a “Hallway” in the new space which they wanted to call Researcher’s Square.

I was doing some consulting work with the Institute and while I was there they told me about their plans to build a Researcher’s Square. Together we constructed a small study to test whether having a “Hallway” would make a difference in their ability to learn from each other and to collaborate. We agreed that I would interview all the researchers before they moved into their new quarters and then return in six months after they moved in to conduct a second set of interviews to see what impact the “Hallway” had on the way they worked together.

In the initial interviews I asked each person five questions. Following are the questions and the general response I got to each:
1. “Tell me about your own project”
Everyone was able to explain their own project in great detail. The researchers were doing some very interesting work and as the interviews proceeded, I began to see connections between the projects that were being described, although the interviewees themselves did not suggest any connections between projects.
2. “Describe the projects of others in the research group.”
The responses I got about others’ projects were vague. Each person knew something about one or two other projects, but when I probed, I quickly reached the limits of their understanding. No one mentioned any successes or difficulties other projects were having.
“Jan does something with municipalities.”
“Tommy studies union issues.”
“Marianne and Lena are working on the Ericsson project. But I’m not sure what it is about.”
3. “How would you explain the overall mission of the research group?”
On this question the responses were quite varied. There seemed little agreement on what the mission of the research group was.
4. “During any given week, how many people would you casually stop by their office to chat for at least 5 minutes, maybe just standing in the doorway?”
Each person was able to name 2-3 individuals they chatted with for at least 5 minutes once a week.
5. ”How do you anticipate using the Researcher’s Square?”
Except for the group that had invented “Researchers’ Square,” responses were primarily about food rather than conversation. Most saw it as a glorified coffee bar.
“I don’t think I will use it much, I have too much work to do to spend time there”
“I might go down to get a newspaper and read it for a few minutes while I drink my coffee.”
“I think I’ll probably eat my lunch there and chat with others.”
“It will be good to be able to get a quick cup of coffee.”

A few months later the research group moved into the new building. Researcher’s Square was in the middle of a long hallway and took up the space of 4 offices. The real hallway proceeded through Researcher’s Square, so to get from one end of the hall to the other people had to walk through Researcher’s Square.

On the Northwest side of the space there was a coffee bar with a number of bar stools and opposite it on the Northeast, were several café tables and chairs. On the South side of the hallway was a large conference table with chairs around it. At one end of the conference table was a white board and on the other end a rack that held the recent publications of the researchers. For illustration purposes (I have only shown three office on either side of the Researcher’s Square, but their were actually 10 per side for a total of 20 offices.)

This conference table was the only one on the floor. If a group wanted to have a meeting they conducted it in Researcher’s Square, which allowed passersby to overhear the conversation. Passersby could also read whatever was on the board and if it was a topic that interested them could stop and listen for a few minutes or even add their own ideas to the conversation going on around the table.

Six months after the researchers moved into the newly designed office space I returned to continue working on the project I had started earlier and I took the opportunity to interview the researchers again using the same questions.

The difference in their responses were remarkable.
• Question 2 - there was a tremendous increase in knowledge of what others were working on. Some interviewees knew some level of detail about every project and many knew the issues, both successes and frustrations, of at least half of the projects. Moreover, the researchers were able to say how one project connected to another and to find intersections between the projects. They talked about how they had made suggestions to others or asked others for help on their own project. Questions were asked and answered on methodology, statistics, content, proposal writing, references to articles and information about client organizations.
• Question 3 about mission revealed that a much greater alignment had developed among the researchers about the organizational vision and purpose. The responses were broader and more inclusive of other research efforts than they had been initially.
• Question 4 about the number of doorway conversations showed an increase of 300%. Increasing from 2-3 doorway conversations a week to 9-10 a week after Researcher’s Square. Although the conversations were more likely to happen in Researcher’s Square than in the doorway, or to move from the doorway to the Square.

• Question 5 originally was, ”How do you anticipate using the Researcher’s Square?” Since researchers had now been using the Square for six months, I changed that question to, “How are you now using Researcher’s Square?” As opposed to the earlier responses, very few of the comments were about food. Rather interviewees told me about learning from others and building relationships.
“Sometimes when I pass and see a group sitting around the table I can see what others are doing.”
“Maybe that’s the most important thing: since its a living room it makes living here easy!”
“When you go to someone’s room you have a purpose, but in the Square you just bump into each other. I think I have learned something from everyone.”
“The Square makes what everyone does more transparent, otherwise it’s more secret.”
“You can understand one thing in light of the other through discussions in the Square.”
“I can see now that it is easier to go and talk to those people that I get to know during lunch and coffee breaks.”
“I can read on the whiteboard what other things are being done.”

Four Concluding Observations

• Structured Socialization
We see from this example that people, even very smart people, are unable to anticipate the benefits of in-depth interaction with colleagues until they have experienced it for themselves. Before Researcher’s Square, the researchers at NIWL daily saw their colleagues coming in and out of the building, passed them on the staircase, or nodded to them in the hall, but they learned very little from one another. Moreover, when I asked them to anticipate how they might use Researcher’s Square, they could not envision a benefit beyond coffee and food. But when NIWL created a place for structured socialization* they not only learned from each other, the whole organization became more aligned and collaborative - in a word, more effective. Structured socialization named by my colleague, Robert Dalton, is the intentional design of processes and space that bring people together in conversational formats to create and share knowledge.
• Connection before Content
Before people can learn from each other or collaborate on issues, they need to build connections – that is, gain some understanding of who the other person is, including their skills, depth of knowledge, experience, and attitude toward others. People are unlikely to ask each other questions or ask for assistance, until they have built a connection that allows them to learn that the other person is knowledgeable enough and respectful enough to engage. In Researcher’s Square, the coffee, food, and small tables for chats, all provided an atmosphere in which the researchers were able to build the connections that then allowed content to flow.
• Cognitive Diversity
The researchers proved to be more interested in others’ projects than they thought they would be. They assumed that what others were doing would be of little interest to them and likewise that others would have little interest in what they are doing, after all the projects they were engaged in were greatly varied and appeared to have little in common with each other. However, in Researcher’s Square these differences also brought with them an attribute that boosts creativity and innovation within a group, cognitive diversity. When people are cognitively diverse they bring to any problem, a larger set of tools derived from multiple perspectives, problem solving tactics, heuristics and interpretations.
• Conversation Rather Than Presentation
The learning that occurred in Researcher’s Square did not come from presentations, rather the knowledge gained was through conversation. When we think about learning from others our first thought is to have someone make a presentation. But as ubiquitous as presentations are, they are a poor way to learn from peers. Typically, a presenter offers what happened in his or her own situation, but that is not what learners need to hear. Learners are interested in knowing how to adapt the lessons to their situation and for that they need to have a conversation so that the other person can understand their context, and they also can understand the context of the other.

NIWL created a Hallway of Learning that changed the organizational culture.

A Knowledge Management Strategy for Non-Profits Working in Developing Countries

I recently conducted a study of eleven non-profits to find out how knowledge management (KM) was being implemented in developing countries. In this article I have 1) outlined the findings from that study, and 2) developed a knowledge management strategy based on the findings.

Many of the Aid Organizations interviewed provided some form of healthcare aid (reducing mother and child mortality, AIDS/HIV) however several provided aid in a different area such as micro financing, agriculture, or environment. For convenience, in this article I will refer to the local workers who deliver the service to recipients as “care givers” regardless of the type of aid provided, and refer to those who support the care givers through training, technical assistance, coaching, and mentoring as “aid workers.”

Findings

1. Non-Profits working in Developing Countries Have Come Late to Knowledge Management
In many industry sectors knowledge management has been in place for 15 to 20 years. However, most of the non-profits interviewed are at the beginning of thinking about using knowledge management to leverage what they are learning. This is illustrated by their level of staffing, the lack of identified KM roles, and 3) the lack of a KM strategy.

One explanation for the late uptake of knowledge management is that most non-profits are science based so their members naturally hold a strong belief in the value of evidenced-based knowledge. Aid workers typically offer technical assistance to their clientele, which is largely a one-way dissemination process of evidence-based knowledge. By contrast, knowledge management places a high value on experiential knowledge, viewing it as different but of equal value to evidence-based knowledge. Experiential knowledge is that which is learned as care givers attempt to implement evidence-based knowledge in a local setting, producing “how to” knowledge. Knowledge management also maintains that the exchange of “how to” knowledge brings great value both to the giver and the receiver of the knowledge and that sharing knowledge is best accomplished in a reciprocal manner among peers. This striking contrast in what constitutes valuable knowledge is a possible explanation for why non-profits have come late to KM.

Notwithstanding the lag, non-profits working in developing countries have begun to actively seek ways to implement knowledge management. They face a number of obstacles in that implementation including, limited staffing, low band width in many developing countries as well as the absence of computers in rural areas, and perhaps most worrying, the lengthy chain of organizations through which most Aid Organizations accomplish their goals.

Even given these obstacles, non-profits working in developing countries have a positive attribute that favors successfully implementing KM. That attribute is the willingness on the part of aid workers, at all levels, to share what has been learned with anyone who asks or anyone whom they encounter in their daily work. The motivation for this openness is the primacy of the mission. A strong commitment to the mission leads to a willingness to do what ever is needed to accomplish that mission. When I asked interviewees what was preventing more sharing of knowledge, none of their responses referenced the obstacles I would typically hear in corporations. In corporations I hear that individuals tend to hoard knowledge or that employees are unwilling to share knowledge because it would reduce their own chances of promotion. None of these concerns were expressed by interviewees in the study.

I have noticed a similar willingness to share knowledge among other mission driven organizations that I have worked with, the US Army, NASA, and the World Bank. In such organizations, where the mission is clear and highly valued, knowledge management has been very effective.

Staffing KM Positions
Among the eleven organizations interviewed four had no KM staff, five had assigned KM responsibility as an additional duty to an existing role, one had a full time KM staff member, and one had eight KM staff members.

No KM Role
Four of the eleven organizations did not have anyone assigned to a KM role, although two of the four did have someone in a KM role up until the time of the economic downturn that resulted in many non-profits downsizing. In all four organizations, even in the two that had eliminated the KM role, there was one or more employees who had taken it upon themselves to help the organization share knowledge, at least in some limited way. For example,
• In one organization, when staff members return from visits to developing countries, a volunteer sets up teleconferences so that staff in other parts of the world can learn from the presentation of the returning staff member.
• In another organization the headquarters person for one of the subject matter topics, brings together the field staff working on the same topic across 12 different countries, to make a site visit to one of the countries. This allows field staff not only to learn from the site, but also from each other as they travel together during the visit. Travel also facilitates long term relationships that can be drawn upon for assistance after the site visit is over.

KM as Additional Duty
Five of the eleven organizations had assigned a staff member(s), currently functioning in a related role (e.g. evaluation, publication) to take on KM as an additional role. To deal with the time limitation of a part time role the KM staff members tended to limit the scope of their KM actions to one or two activities. They were, however, fully aware that there were many useful KM processes in addition to the few they were able to implement. Activities that various organizations chose to implement include:
• Conducting reflection meetings
• Providing the technology and basic instruction for setting up on-line forums, but without time to coach forum leaders in how to make the forums effective
• Drawing “lessons learned” from evaluations (the primary responsibility of the interviewee) the staff member was able to provide relevant lessons to colleagues when they were initiating a new project – playing a kind of broker role
• Establishing a repository for lessons learned but not being able to facilitate the capture or transfer of lessons
• Establish a joint learning network across countries that are working on the same issue. The joint learning network holds in person meetings annually and uses Facebook-type software on-line as well as monthly calls. However the interviewee noted, “Everyone in headquarters can access it but not very many people are using it.”

One Staff Member in the KM Role
One of the organizations interviewed had a full time person in a KM role. Through this staff member the non-profit was able to put into place a number of KM activities, many of which did not require the staff member’s on-going direct involvement.
• “Learning after” provides this organization a way to continually improve. The KM staff member set up a process for “learning after” meetings which were then facilitated in a number of ways; 1) the KM staff member facilitated “learning after” for strategic events, 2) there were key people who took responsibility for facilitating “learning after” sessions for some groups, and 3) some groups conducted “learning after” reflections on their own.
• Cluster groups had been established around specific topics. These groups, that were lead by subject matter experts, periodically came together for face-to-face knowledge sharing, but also sometimes use Skype, conference calls, and email groups.
• Disaster Relief – one part of this organization deals with disaster relief. When relief workers returned from assignments they were debriefed by other program staff and still other staff took it upon themselves to analyze the learning from the debriefs and use that knowledge to update the disaster relief manual. Knowledge sharing is made easier by the nature of disaster work, which is periodic. Workers have periods of “down time” to learn and share what they have learned and periods of “up time” to put that knowledge into action.

Eight Staff Members in the KM Role
This organization has a centralized group that primarily provides technology tools and trains staff to effectively use the tools. They are not responsible for content, which is generated in projects, but rather they are responsible for receiving and distributing that content through the technology tools. The technology tools include:
• Intranet - an online collection of resources for all employees that includes information about projects and programs, countries, technical expertise, employees, policies and procedures, and news. The KM staff tracks usage and coordinates with content managers.
• Newsletter - the KM staff publish a weekly electronic news update sent by email to all staff. It is designed to forward links to useful and timely information on the intranet to staff worldwide.
• Social Networks – the KM staff provides and continuously improves an online platform for staff to create their own social networks.
• Subject Matter Expert Networks – the KM staff provides technical assistance, capacity building, and assistance to the moderators of forums that are designed for subject matter experts. Forums include live virtual meetings and ongoing virtual discussions.
• The Institutional Memory database – The KM staff maintains an internal archive and document management system for project activities and technical information.

This sample of non-profits indicates that the KM staff have been most effective when they have established processes, both face-to-face and virtual, through which the organization can share knowledge, but have then left it to project personnel to facilitate the actual transfer.

Technology
There was a surprising lack of on-line Communities of Practice (CoPs) among the eleven organizations interviewed. Within most industries, communities have become the gold standard of knowledge management. It would be hard to find a Fortune 500 company that does not have effective communities to leverage the organization’s knowledge. Among the eleven organizations studied only one had a CoP where aid workers shared “how to” knowledge through Q&A. Two had tried communities of practice and finding that there was minimal activity, abandoned them. During the interviews both of these talked about the lack of computers in rural areas and as well as limited bandwidth as factors in the lack of activity. Two others had developed list serves that were largely one-way communication vehicles for headquarters to communicate with the field. However, both held periodic forums that lasted from 3 days to a week. Each forum was focused on a specific topic and was typically led by an expert. Participants would send in questions sometimes directed to the expert and sometimes to each other. In addition, to answering questions, the expert would ask stimulating questions of the participants to encourage conversation. A daily summary was produced and sent out to those who participated

Most of the organizations in the study had a presence on Facebook as well as an external website, both primarily used for public awareness and fund raising.

One organization had been very effective in producing toolkits, which are constructed as an electronic library on a topic. The organization formed working groups of subject matter experts to bring together evidence based practices and protocols on the topic. The experts met to review what others had developed so they were able to say, “Of the 100 research articles related to family planning, here are the 10 you want to read”. This organization now has 50 toolkits. The toolkits bypass the bandwidth problem since they can be sent as CDs.

With the exception of the one organization having 8 staff members, the use of technology appeared to be limited both by the lack of technology expertise at headquarters and by the lack of computers and bandwidth in the field. The bandwidth issue is gradually improving in many developing countries, which holds promise for developing effective communities of practice over the next few years.

The Lengthy Sequence of Organizations Through Which Aid Work is Accomplished
Non-profits working in developing countries are tasked with giving aid to hundreds or even thousands of people, often across several countries. Their aid work takes place through a chain of sub-organizations. A typical aid chain is described below although there are variations, some having more and some with fewer sub organizations involved.

1. An Aid Organization in a developed country, for example, the United States, Great Britain, or the Netherlands, receives funding from their government or other funding source to provide a specific type of aid (e.g. reducing mother mortality, teaching farming techniques) in specified countries
2. To accomplish this the Aid Organization makes an agreement with one or more local non government agencies (NGOs) that regularly work in the specified countries. Any one NGO may have contracts with several Aid Organizations
3. The Aid Organization employs a headquarters staff as well as staff in the field whose job it is to oversee the work of the NGOs
4. The NGOs accomplish the work by interacting with local government agencies, for example, the Ministry of Health, or Ministry of Education, with whom they have established long terms relationships. The NGOs provide training and technical assistance for the specific aid process to the local agencies
5. The local agency workers, in turn, provide training and technical assistance to care givers in the local setting, for example a clinic, or a school.
6. The care givers then provide aid to the recipients (patients, students, small business owners) using the processes designed by the originating Aid group.
7. To complicate matters still further, many different Aid Organizations may be working on the same issues within a country.

Typically, it is the Aid Organization headquarters that puts KM into place; however the lengthy chain of sub-organizations makes it difficult for Aid Organizations to implement knowledge management.

“We don’t have a mechanism for getting information about what is working well in the field to headquarters. The extent of the KM practice is a report-out from headquarters personnel who have been in the field.”

“We still have to learn how to share across countries, we don’t do that well now.”

“It is difficult to get learning going South to South.”

Each of the sub organizations receive money and must make reports to the organization providing their funding. It is in the interest of each organization writing reports to assure their funder that their money is being well spent so the reports are primarily activity reports (number of people trained, number of ARV drugs distributed to HIV patients). Each organization along the chain specifies its own report requirements for the subsequent organization it supports, and then in turn writes its report to the organization supporting it, often based on a different set of requirements. Reports from farther down the chain may not be made available to the originating Aid Organization. The originating Aid Organization, which is promoting knowledge management, is several organizational steps away from the care givers that are learning from their experience, leaving the care givers with little or no relationship with or felt responsibility to the originating Aid Organization.

The objectives Aid Organizations would like to achieve through implementing knowledge management include the following:

1. More effectively transfer knowledge from the field to headquarters including:

• Results for Recipients – accurate outcomes that have occurred for recipients as a result of the practice (e.g. TB patients screened and receiving medication, increase in bushels of produce per acre)
• Effectiveness of Dissemination - effectiveness of the teaching, communicating, and/or motivation provided to care givers by the NGOs and local agencies (e.g. number of care givers that have been trained; the number that are using the practice; how correctly the practice is being implemented)
• Implementation Lessons - what care givers are learning about how to implement practices in their local setting (e.g. put up reminder signs about hand washing to prevent MRSA, provide a private space to obtain sputum from TB patients)

2. More effectively transfer knowledge from headquarters to the field including:
• Evidence Based Practices - transferring practices that have proven to be effective (e.g. providing bottled milk for new born babies whose mothers who have HIV, so the mother will not transmit the disease through breast milk) to local care givers
• Synthesized Guidance - guidance about how to implement the practices in the local setting gleaned from the reports and stories from multiple care givers

3. More effectively transfer knowledge within and between countries including
• Effective Knowledge Sharing Processes - knowledge sharing activities that successfully transfer knowledge between care givers within and across countries about how to implement evidence based practices more effectively

Primacy of Mission
Writing and Learning from Written Lessons Learned
As discussed earlier, aid workers and care givers are willing to share what has being learned because they see the mission as primary. But the primacy of mission also tends to make aid workers reluctant to take the time to write up what they are learning for others. Interviewees explain that they are unclear who or how their knowledge would be made use of if they did take the time to write lessons up.

This concern reflects their own experience in being unsuccessful when they search for knowledge from repositories.
“Many lessons describe what someone did, but it doesn’t tell you enough to be able to do it yourself.” (from a potential user of lessons learned)

“We produced a format for submitting lessons learned from field staff but it has been difficult to obtain documentation from the field.” (from an Aid Organization)

“There is a visit report that is on the Sharedrive so it is available to everyone to access. The difficulty is getting people to use it or even to know where it is.” (from a sub organization)

Holding Reflection Meetings
Rather than asking care givers or aid workers to write up what they are learning, the Aid Organizations studied have been much more successful in transferring knowledge when they have brought people together to exchange their knowledge. For example, one of the organizations in the study holds quarterly learning sessions where care givers from facilities across a large region are convened. The meetings are an opportunity for local facilities to share their best practice with each other and to get help from peers as well. Unlike writing reports, such meetings are highly valued by care givers. Participants acknowledge that they learn a great deal from the exchange with others and look forward to attending. The difference between writing up lessons for others and face-to-face learning sessions is the essential element of reciprocity. Setting in an office writing up a lesson feels one way - just giving – whereas regional meetings are both give and take. Moreover when a care giver offers an idea to others in a small group, he/she can see the appreciation in the eyes of those who are listening and believes they will make use of what is offered. When Aid Organizations are able to bring care givers together to learn from each other, knowledge flows and relationships that can facilitate continuing exchange are built.

However, when a team is asked to reflect for the purpose of providing lessons learned for others, reflection meetings are considered a poor use of care givers time. This concern reflects the past experience of aid workers and care givers that have taken the time to participate in team or project reflection meetings but then had no way of knowing whether what they provided was of use to those others. Given this experience aid workers and care givers are unlikely to put the needs of anonymous others over the needs of clients who are in plain sight and who need their help.

Much valuable knowledge remains in the heads of workers and is available only to those who are in the same location. Many interviewees acknowledged, “A lot is trapped in people’s minds.”

A Knowledge Management Strategy for Non-Profits in Developing Countries

Based on the study I have proposed a design for implementing knowledge management in developing countries. I base the design both on the findings from the study and on the US Army’s implementation of knowledge management with deployed troops. Many of the issues Aid Organizations face in developing countries are similar to those faced by the US Army, which is also a mission driven organization.
• Soldiers need to quickly find out what other units have learned - what they can borrow without having to discover everything for themselves
• Soldiers do not have time to write up lessons learned or to search for them in a database.
• Soldiers meet together to reflect on actions they just took in order to improve their own actions for the next time.
• In battlefield situations, soldiers often do not have access to computers or the bandwidth to reach headquarters

The Army has addressed the flow of knowledge from and between deployed units by developing a brokering system called the Lessons Learned Integrators (L2i) network. The L2i network members are analysts; each based in a deployed unit and also stationed in the Army’s schoolhouses, training centers, and doctrine centers. With this distribution, what is learned in the field can be quickly transferred to what is being taught to soldiers in the schoolhouses and updated in manuals (guidance documents). To accomplish this exchange the group of 29 lessons learned integrators:
o hold a teleconference weekly where lessons that have been learned in one unit are shared with other units
o participate continuously in an on-line discussion forum where they ask each other questions on behalf of their unit and receive immediate replies, not from one person, but from many other LL integrators
o meet face-to-face three times a year, to strengthen their relationships within the network so that knowledge continues to flow among them.

To gather and distribute what they learn in these network interactions, the L2i network members interact with soldiers in their unit while in the mess hall, sit in on reflection meetings (AARs) and participate in the unit’s high-level meetings. Through these means they know the problems the soldiers in their unit are facing which helps them know what to ask other L2i’s about on the weekly calls, for example, “Has anybody found a solution to X?” Lessons learned integrators are brokers of knowledge in both directions - they bring the lessons that were learned elsewhere to their unit, and they share the lessons their unit has learned with other units. It is a reciprocal system.

There are differences as well as similarities between KM in developing countries and in the US Army. But in both situations the following KM principles apply:

• Processes for sharing knowledge should to be established. Although it is often and accurately said, “knowledge sharing is everyone’s job,” the reality is that it is necessary to put practices into place through which “everyone” can share their knowledge.
• To be useful lessons from the field need to be synthesized into guidance documents rather than remain as a collection of individual lessons.
• When the intended recipients of the knowledge lack access to computers and/or bandwidth, brokers are needed to facilitate the exchange of knowledge.
• Those who are learning lessons in the field are more willing to share what has been learned if they receive feedback that what they gave was received and used.
• It greatly assists a team or project to learn from its own experience, to periodically meet together to reflect on their actions and outcomes.
• The receivers and givers of lessons across a region gain the most when they are in face-to-face conversation with each other where they can ask the questions that allow them to modify each other’s lessons to fit their own context.
• To be sustained, knowledge exchanges need to be reciprocal. Over a period of time, people are uncomfortable if they always give without ever receiving, nor do people want to be on the receiving end without also being able to help others. Reciprocity needs to be designed into any knowledge sharing meeting or practice.

Suggested KM Processes and Roles for Non-profits Working in Developing Countries

Knowledge Manager Role
For each project topic, (e.g. infant mortality) that is established in each country or region the Project Manager appoints an in-country Knowledge Manager (a broker role) who the Project Manager holds accountable for seven knowledge management tasks. The Knowledge Manager role can be full or part time, depending on the size of the project. In either case these seven tasks should be included in the individual’s performance objectives. The Project Manager insures that the Knowledge Manager receives training to carry out the assigned tasks.

The practices and the Knowledge Manager’s role in establishing the practices are listed here:
1. Develop a Knowledge Plan - Meet with the project team (NGO level) at the beginning of a project to develop a knowledge plan for the project. Monitor the knowledge plan over the life of the project to update it and assure its completion. The knowledge plan is developed at the same time the project planning meeting is held. The knowledge plan includes:
a. The identification of 1) what knowledge the team needs to gain in order to do their work more effectively, 2) which team members will be responsible for seeking out each specific knowledge needed from other projects or experts, and 3) by when the knowledge will be obtained. The Knowledge Manager assists with the plan by identifying where needed knowledge might be located through his/her contacts in the Knowledge Managers’ forum (see below)
b. What knowledge the team is uniquely positioned to learn that will add to the knowledge base about how to do this work. What activities the team will engage in to make that learning a reality and by when.
2. Conduct Reflection Meetings
Meet with project team members as a group, according to the schedule identified in the knowledge plan, to facilitate reflection meetings about what the team is learning that could improve their own performance. Make audio and video clips of useful stories told in these meetings. Note any lessons that would be useful to others.
3. Conduct a Retrospect
Meet with the project team members as a group, according to the schedule identified in the knowledge plan, to assess what has been learned that would be of value to projects in other regions or countries. Make audio and video clips of useful stories told in these meetings
4. Participate in a Knowledge Managers Community Forum
Continuously engage with Knowledge Managers in other regions or countries through an on-line community forum and through monthly teleconferences
5. Provide New Knowledge to the Project team
Bring ideas gained in the Knowledge Manger community teleconference and forum to the Project Manager for possible implementation.
6. Facilitate Care Givers Knowledge Exchanges
Facilitate in-country/region knowledge exchanges among care givers to share what has been learned with each other. Make audio and video clips of useful stories told in these meetings
7. Provide Lessons to Subject Matter Owners
Provide the Subject Matter Owners (see below) detailed knowledge about practices when asked, including audio and video clips

Initially the Knowledge Manager will need to facilitate the team reflection meetings. However, once established the project team members can implement the reflection activity on their own.

Subject Matter Owner Role
The Aid Organization appoints a Subject Matter Owner for each improvement topic (infant mortality, crop yield). This position is part time and is held by someone with expertise in the subject area, typically stationed at headquarters. The Aid Organization holds Subject Matter Owners accountable for the following four responsibilities:
1. Facilitate the monthly Knowledge Managers teleconferences for their topic area and participate in the on-line discussion forum
2. Interact with individual Knowledge Managers to gather more detailed understanding of knowledge that has been gained through the monthly Knowledge Managers teleconferences and on-line discussion forum
3. Develop on-line, synthesized guidance for the topic based on what is learned across projects. Incorporate audio and video clips in this guidance document.
4. Keep the guidance document up to date on-line.

Synthesized guidance is written, not from the perspective of what a team has learned, but from the perspective of what a “user” would need to know. The synthesis is the summary, collation and integration of multiple sources of written knowledge into a single set of guidance documents (rather than multiple files of individual lessons). The guidance document is often enlivened by stories from experience, linked to people and documents for further detail, and structured in such a way as to be of maximum use to the reader. The synthesis of lessons into guidance documents serves several purposes:
1. to keep headquarters aware of what is being learned in the field
2. to on-board new employees,
3. to provide a source of up dated and accurate implementation knowledge to the field

Technology Role
o Provide suitable technology to make on-line guidance documents easily available
o Provide suitable technology to make on-line teleconferences (webinars, go to meeting) effective
o Provide suitable technology for Knowledge Managers to meet in an on-line community forum

Headquarters KM
o Provide training for Knowledge Managers
o Provide training for Subject Matter Owners
o Hold a yearly in-person meeting of Knowledge Managers that provides time for them to learn from each other and to build relationships that will facilitate their on-going learning during the year
o Hold a yearly in-person meeting of Subject Matter Owners that provides time for them to learn from each other and to build relationships that will facilitate their on-going learning during the year (The meetings of Knowledge Managers and Subject Matter Owners can be combined depending upon the number in each category)

Conclusion
It is essential that non-profit organizations working in developing countries have a viable strategy to manage their knowledge. Without such a strategy 1) much knowledge must be re-learned in each region or country, wasting costly time and resources, and 2) Aid Organizations do not learn what is working and what needs to be changed in order to make their aid efforts more effective. This model proposes a strategy with roles and processes that make knowledge management possible for non-profits working in developing countries.

How to Hold a Retreat or Meeting Where New Thinking Emerges

I attended a meeting a number of years ago that has remained in my mind as the ideal meeting where new thinking can emerge. I want to describe that meeting and then tease out the design characteristics that led to such memorable results.

The Monieson Centre at Queen’s School of Business in Ontario Canada hosted an invited meeting held over two days. The invitees were twenty-five recognized thought leaders in the field of knowledge management, both academics and practitioners. The stated goal of the retreat was to “push the boundaries of what we know about creating and managing knowledge-based organizations.” That sounds pretty ordinary so far, but here is where it started to get interesting. The invitation said:

"Attendees will be requested to submit an original “thought-piece” in advance of the retreat to be shared among the group. We are not looking for neatly packaged research articles. Rather, we hope you will share the leading edge of your best thinking on some important aspect of managing knowledge-based organizations. Your thought-piece should be in the form of a graphic poster accompanied by explanatory text, altogether no more than one page. We will use the posters to create an “idea gallery” as the means for exploring and linking ideas during the retreat."

I wrote my thought piece about a topic I was exploring at the time - the relationship between knowledge sharing and democratic organizations. I posed it as a chicken or egg question, "Does knowledge sharing thrive in egalitarian environments OR does knowledge sharing promote egalitarian environments?" The thought pieces were posted on-line and attendees were encouraged to read others' pieces before they came. The authors' names, however, were withheld, which provided the opportunity to consider the ideas without being biased by who wrote them, but which also created a bit of intrigue, prompting almost everyone to read the posted pages before they came to the meeting.

When we arrived on Friday evening each of our thought pieces had been made into large posters that displayed the main ideas of our one-pagers. The posters acted as a valuable reminder of the ideas we had already read on-line. We started with a dinner that first evening and were provided a brief explanation by the organizers James McKeen and Michael Zack, of what we were going to be doing.

The next day was simply a series of conversations. The organizers had grouped the ideas on the posters into themes and we were assigned to small groups of 6-9 that met to talk about that theme. After an hour a second set of themes was announced and we re-grouped and were again back in conversation. And so it went, with time out for snacks and meals.

Each of us carried our poster from room to room. The meeting rooms were cozy, with overstuffed chairs and small tables for a cup of coffee. Each time we entered a room we placed our posters in a circle around the conversation area so the posters were always in view. As interesting ideas sprang up from the discussion, attendees would write the idea on a post-it and stick it on the appropriate poster. By the end of the two days, our posters looked like they had the measles. A couple of the post-its on my poster are still relevant to me:
● Knowledge sharing and egalitarianism is a loop
● We have to be bound by some notion of the common good

We held a full group meeting the last morning in order to discover the themes that had emerged across all the conversations.

The initial papers were interesting but it was, of course, the conversations where the knowledge got exchanged and learning happened. The idea I had come with grew through the discussions as others suggested lines of thought that had not occurred to me. I caught myself frequently thinking, if not saying, “Ah ha!” as insights came together for me. Others offered the names of books and articles that, long after the retreat, have continued to expand my thinking. Recently I have found myself re-visiting the implications of those ideas as a result of work I have been doing for several months in a communist country, China.

I found the topics others brought to the retreat equally stimulating and was able to contribute ideas to their work, which felt very affirming. It was, hands down, the most intellectually stimulating conference I have ever attended.

I have tried to think why this meeting has stayed with me as the ideal of what meetings ought to be. Some of my conclusions are:

• There were no presentations to sit through at this meeting. Often when I attend a conference, less than ten percent of what the presenters offer is either new for me or fits my current needs and interests. I know I am not alone in this regard, because I often overhear others at conferences proclaim that, “If I get one new idea from a conference I’m happy.” But at this meeting I would put my applicability meter at least at 80%. After all, in a small group, the group members influence the direction of the conversation. At Monieson I could delve deeper into areas that peaked my interest by requesting that the speaker elaborate a point or I could get multiple views on what a speaker offered by asking what others in the small group what they thought about that idea. As a group we tailored the conversation to our immediate interests.

• That 80% percentage may also reflect that we were all well versed in the topic. We had the absorptive capacity to learn from each other as we explored innovative ideas that required a thorough understanding of knowledge management as a foundation.

• No one was trying to impress anyone else or sell others on their services as happens all too frequently in conferences. That was refreshing in and of itself, but more importantly it meant attendees were willing to reveal doubts about their own ideas and to test their thinking rather than just making declarative statements about it.

• The conversation groups were the right size for participants to inquire, challenge, elaborate, and clarify - in short, all the indispensable types of interaction that serve to deepen conversations.

• The retreat was built on the idea of reciprocity – Jim and Mike assumed everyone was there both as learner and teacher, so they carefully designed the meeting in a way that both could occur.

• We were asked to write about our new ideas, those that were half formed and not yet ready for prime time. That meant we were still open to others' thinking about what we had written; no one felt a need to defend their current thinking.

• The group was small, twenty-five. That was large enough for there to be cognitive diversity but small enough that we could build connections between our ideas.

• The dinner on the first night provided the opportunity to renew acquaintances before we started the hard work of content - connection before content as Peter Block says.

• The setting was informal with easy chairs, lamps on the small tables where you could also rest a cup of tea. There was no head of the table or podium that drew our eyes away from each other during the conversation.

• The themes that formed the small groups were pertinent to the attendees’ interests, given that they were constructed from the thought pieces attendees had written, rather than being predetermined by a conference agenda.

The Monieson Centre meeting was designed to engender new thinking. Of course, all meetings are designed; it is just that for most meetings we give little thought to our choice of design elements, rather we simply relying on the familiar presentation format. We would be well served if we took the time the Monieson Centre staff did to match the meeting design to the desired outcome. That care would create a great deal of change in how we use space, furniture, the size of groups, who is learning from whom, how we interact, and of course what we are able to learn.

Why Knowledge Management Didn’t Save General Motors: Addressing Complex Issues By Convening Conversation

Starting about 2000 GM put together a vibrant KM program. At one point GM had 138 best practice teams and 33 centers of expertise working with identified subject matter experts. The KM program produced over 5000 best practices that impacted both quality and schedule and led to millions in cost avoidance. In 2008 KM was alive and well at GM. In the summer of 2009 GM filed for bankruptcy. Why didn’t KM save General Motors? GM was brought down by a flawed strategy, but an organization’s strategy is clearly a product of the knowledge that exists within its walls. The knowledge existed within GM to develop a more competitive strategy. But between 2000 and 2008 knowledge management did not help GM bring that organizational knowledge together in a way that could have saved it from bankruptcy.

Knowledge Management has the capability to address an organization’s very difficult and complex issues, for example how to merge two cultures or how to refocus an organization from selling products to selling service. For these, and many other complex issues, no standard operating procedure exists and no history within the organization is available to draw on for guidance. The knowledge to solve complex problems does not reside in individuals nor even in the executive team – but the knowledge does exists in the collective. The knowledge management task is to bring together the collective knowledge of the organization to bear on complex issues.

To date, Knowledge Management has primarily been in the service of cost cutting or its twin, time saving. The GM example illustrates, that as KM professionals we have not garnered the collective knowledge to address the complex problems that our organizations are facing. That has certainly been true of my own work over the past 15 years. My work has primarily been about creating more effective knowledge sharing processes and communities. But in the last couple of years, more and more I find myself being asked to help convene organizations to address the complex issues they are facing. The realization about how my own work is changing has led me to think that the time is ripe to move KM to a new level and has encouraged me to write this post.

James March wrote a much acclaimed article in 1991 called “Exploitation and Exploration in Organizational Learning.” In that article he says that the exploitation of existing knowledge (which is what knowledge management has been doing) must be balanced with the exploration for new knowledge. Both exploration and exploitation are essential for organizations, but they compete for scarce resources. Exploration is more costly because new ideas often don’t work out. Exploitation of existing knowledge is cheaper because existing knowledge has already been tested, so there are fewer failures.

But the downside of exploitation is that it is a closed loop that can sub optimize organizational knowledge. An organization, exclusively relying on exploitation, will not move beyond the best practice that already exists within its walls. For this reason exploitation can produce a valued step change, but cannot produce the kind of discontinuous change that exploration can yield. KM has been focused on exploitation; we need to now balance the KM effort to help organizations explore new knowledge.

Johnsonville Foods an Example of Using Collective Knowledge to Address Complex Issues
Ralph Stayer, the CEO of Johnsonville Foods was faced with a difficult issue that was also an opportunity. Johnsonville Foods was a small, family owned sausage company of 1000 people, well known in Wisconsin for its quality. Stayer had received an offer from a food-processing company to buy large quantities of product on a regular basis.
Stayer saw this as a complex issue because Johnsonville Foods did not have the capacity to handle such a large job, so accepting it would mean everything would have to change, e.g. shipping, purchasing, equipment, production, packaging, new people, and training. To deal with this very complex issue Stayer called a meeting of the whole organization, giving them all of the information he had about the offer.

He asked employees to work in teams to answer three questions: What will it take to make it work? Is it possible to reduce the downside? Do we want to do it? Teams all across the organization struggled with the questions for almost two weeks, holding meeting after meeting to talk through the risks - which were considerable - and figuring out how they would have to operate differently if they were to increase production that much.
When Stayer brought the teams back together, each team laid out what would be needed and what the problems would likely be for the others. In the end there was almost unanimously agreement that they should take on the new business. Reflecting on the process later, Stayer said, "If you issue orders you’re telling people, "Don’t think; just do." But if you’ve got 1000 people, you’ve got 1000 minds. And if you issue orders from the top, you’re using only 3 of them, or 2, or one. That’s stupid.”

Elements Needed to Address Complex Issues
There are three elements that are necessary if organizations are to use all their knowledge to address complex problems. All three elements are present in the story about Johnsonville Food.

1. Convening
• How to convene the organizational conversation in a way that allows new thinking to emerge
• How to frame the question for the conversation
• The types of issues that require leveraging the collective knowledge
2. Transparency
• Making all the knowledge available so that those who have been convened are as well informed as possible
• The willingness of the leader to say, “I don’t know.”
3. Cognitive diversity
• Who needs to be invited to those conversations
• How to obtain cognitive diversity

1- Convening

How to convene
Convening is about designing a meeting in a way that allows new thinking to emerge. That requires reconsidering all elements of a meeting, e.g. the activities participants engage in, how the seating is configured, the nature of the invitation that participants receive, the length of the meeting, and so on. I have written about the task of the leader as “conversation architect” in other posts. Here I want to focus on the importance of small groups as the source of new ideas.

In a small group each participant can not only offer his or her own thinking, each can ask others to clarify their meaning, elaborate on what someone else has said, challenge an idea, and offer data to support his/her own position – in other words, a small group can have a conversation. In a large group participants are limited to declaring their position on an issue. Participants are all aware that "turn taking" is considered courteous and that to take up additional airtime by engaging others in the back and forth of conversation is sure to be viewed by others as discourteous.

Johnson and Johnson, researchers at the University of Minnesota, who study collaboration, say that the synthesis of diverse perspectives comes from being able to hold both one’s own and another’s’ perspective in mind at the same time. Holding both perspectives implies more than having just heard another’s perspective.
It means fully understanding the reasoning behind a perspective as well as the implications that would result from it. That necessarily means there has been enough back and forth between participants for each to gain that level of understanding.

The format for convening is a mix of three elements:

Framing the Conversation
Convening occurs around a question that frames the problem or issue in a way that has relevance to the convened group and that helps to move the group beyond obvious or known positions. One of the difficulties of framing a question is that questions often have an assumption embedded within them. For example, an organization may have a turnover rate of 15% among employees in their first two years of employment. The question, “How do we reduce the turnover rate of employees who have been with the organization less than 2 years?” assumes that a 15% rate of turnover is a problem. The framing of the question prevents participants from thinking about other possibilities such as viewing 15% as a useful weeding out of unsuitable employees, or as a way to bring new ideas into the company. A less biased question might be “What is the optimum rate of new employee turnover for our company and how do we achieve it?” Or even less biased might be, “What experience do we want new employees to have in their first two years, that prepares them and us to make a sound decision about their future with the company?” The task is to frame the question in a way that opens up the space to invite new perspectives on the topic.

Issues that Benefit from Convening
Here I am drawing on the work of Ron Heifitz, who has written a number of books on this issue, particularly “Leadership Without Easy Answers.” Heifetz talks about two kinds of challenges organizations face, adaptive challenges and technical problems.

Adaptive challenges are complex and unpredictable; they have no known answers. People in the organization may even disagree on what the problem really is. To address adaptive challenges organizations must invent their way to a solution. Examples of adaptive challenges are, hospital systems faced with an interminable nursing shortage; the anticipated retirement of thousands of workers in the government sector; the relocation of 45% of an organization’s employees to a new location.” My mentor, Reg Revans use to say, in his rather outdated way, “These are problems about which reasonable men can disagree.”

By contrast, technical problems are predictable and usually solvable by methods and tools that the organization has already developed to address them. If, for example, a well is to be dug in a new oil field, there are standard operating procedures; if a special issue of a journal is to be published, there are steps outlined that will put the issue out on schedule and within budget; if a new cell phone is to be developed, there are well defined phases of the development process. Technical problems do not require a fundamental change in methods or tools. More importantly, the people who solve technical problems can function within a known set of assumptions about how work gets done in the organization, as well as what kind of behaviors are needed to make that work happen.

Where convening brings great value is in addressing adaptive challenges. Heifitz says, “The power of a leader to address adaptive challenges does not lie in inventing solutions; rather it lies in using leadership authority to convene the conversations.”

Johnsonville Foods faced an adaptive challenge and so did Ecopetrol a few years ago.

Ecopetrol a Second Example of Using Collective Knowledge to Address Complex Issues
Five years ago I got a call from Ecopetrol, the National Oil Company of Columbia. The conference organizers said, “We are having a meeting to develop a knowledge strategy. We have not had a knowledge strategy and we think we need one now as we get ready to be publicly traded on the stock market. Could you come to our three day meeting and make a presentation about what makes an effective knowledge management strategy?” “But,” they said, “we only want you to talk for 15 minutes.” Columbia is over 2000 miles from my home in Austin Texas, so my initial reaction was that 15 minutes seemed a very short time for such a long journey. But when I heard what they were trying to do I got very excited. They were convening the top 200 people from across all parts of the company to figure out what they needed to do about knowledge - a complex issue for which there is no “right” answer.

To obtain the diversity of thought they needed at the meeting, they invited representatives from fifteen MAKE winners to speak on the first day of the meeting. And, like my presentation, each of them spoke for only 15 minutes. After each presentation time was set aside for small group discussions – not Q&A - but pairs or trios talking together to make sense of what they had just heard.

The second day was devoted to a Knowledge Café (some times called the world cafe) where the Ecopetrol managers worked in small groups to incorporate what they had heard the first day. The Knowledge Café questions were, “What are the critical areas of knowledge that Ecopetrol needs to manage?” and “What KM processes should we focus on to manage that knowledge?”
Those of us who had made presentations the day before also participated in the knowledge café, moving from table to table along with all the managers. After six rounds of the knowledge café, with managers listening to and building on the ideas of other managers from different parts of Ecopetrol, the group began to coalesce around twelve knowledge areas and had also developed ideas about what KM processes would be most useful.

The third day was Open Space where participants selected which of the twelve topic groups they wanted to join, based on their own expertise and interest. The goal of each group was to set up action plans for that critical knowledge area. By the end of the third day the Open Space groups had a plan to implement each of the initiatives they had agreed upon during the Knowledge Café discussions the day before.

Ecopetrol had a knowledge management strategy after three days of convening and conversation – one that the 200 top managers were committed to.

2 - Transparency
I have noted the way the Ecopetrol meeting involved the elements of convening and cognitive diversity. The meeting also incorporated the third element needed to solve complex issues, transparency. There are two issues connected to transparency, 1) the willingness to say, “I don’t know,” and 2) providing all the knowledge available about the issue under discussion

The willingness to say, "I don't know."
The President of Ecopetrol, Javier Gutiérrez, attended all three days of the meeting. During that time he never made a speech about the outcome he wanted. Rather he listened to the presentations on the first day and on the second day he moved from table to table in the knowledge café with the other 199 managers to think together about how to build a knowledge strategy. What he said through his actions was, ‘I don’t know how to build a knowledge strategy, but I'm confident we can figure it out if we use our collective knowledge.”

Over the last ten years, KM has been able to develop horizontal transparency in organizations through the implementation of communities. In many organizations it is now acceptable for a peer to say to others, “I don’t know how to do this, can anyone help?” However, KM has not addressed the issue of vertical transparency in either direction, neither from frontline to senior leadership, nor from senior leadership to the frontline. Yet, as KM Professionals, we are fully aware that transparency is necessary for leveraging the knowledge of the organization.

Providing all the knowledge available on the issue
The second part of transparency is providing all the knowledge. And this President Gutierrez did as well. Ecopetrol was considering becoming a publicly traded company and everyone in the room understood the implications of that action for Knowledge Management. There can be no integration of collective knowledge without transparency. Unless everyone who is jointly attempting to make sense of an adaptive challenge has access to all the available data, there is no way to successfully address the problem.

Transparency is two sided, leadership must be willing to reveal what they know that may impact the issue, and likewise, frontline employees must be willing to reveal the difficulties they anticipate facing. One of the benefits of using proven formats for convening, such as Knowledge Cafés, Open Space, Storytelling Circles, and Appreciative Inquiry, to name a few, is that these formats are designed to establish an environment conductive to trust and disclosure. More typical meetings like retreats and townhalls tend to discourage disclosure, because if participants have ideas to offer they must say those ideas in front of a large group. But participants are often unsure about how their ideas will be accepted by a large group. Small groups provide the advantage of, 1) being able to test an idea with peers before offering it more broadly and, 2) having a sense of where others stand on an issue before offering a new perspective.

Ecopetrol successfully implemented the strategy they put together five years ago. The meeting I described was not a one time event. It is the way Ecopetrol continues to manage its knowledge using Communities, Technology Forums, Expert meetings, Knowledge Cafés, Conversation Circles, and Open Space to create new knowledge. In both 2010 and 2011 Ecopetrol was nominated for the MAKE award. I predict in 2012 they will be a MAKE award winner.

3 - Cognitive Diversity
I have spoken about convening and transparency, the third element is cognitive diversity.

How to obtain cognitive diversity
There was a great book published in 2007 by Scott Page called, “The Difference.” Page reminds us that people can be diverse in many ways, race, gender, religion, etc. but Page's research shows that the kind of diversity that makes a difference in solving complex problems is a difference in the way people think, which he calls cognitive diversity. That includes the heuristics they use, their interpretations, problem solving strategies, and their prediction models. A physician thinks in a different way than an engineer and uses different strategies to tackle a problem.
A historian thinks in a different way than a mathematician – their training and experience provide them different approaches. When we convene the members of an organization to address a complex problem, we are not so much drawing on the knowledge they use in their job, as we are seeking the diversity in how they think about issues. Research that Page quotes in his book, shows that bringing diversity into the conversation increases the possibility that a group will generate new and more creative ideas.

A number of organizations have taken this idea of cognitive diversity very seriously. The Intelligence Cells in Iraq and Afghanistan, for example, are US joint knowledge production efforts to address a specific intelligence challenge. The makeup of a cell is multi-INT (i.e. SIGINT, HUMINT, IMINT). A cross section of disciplines (e.g. analysis, collection, legal, enterprise architecture, technology development) are also represented in each cell, as well as, government, industry, and academia. To address the very difficult intelligence issues about terrorists, requires more than spies, intercepted signals and picture taking drones, it requires a conversation among people who think in diverse ways.

Who needs to be invited
When I am asked by an organization to convene a meeting to address a complex issue, I bring the whole system into the meeting. That doesn’t mean every employee, but it does mean representatives from every part of the organization as well as from every level. And often it also means involving outsiders, customers or suppliers who bring a different perspective on the issue we are addressing.

Teams, departments and divisions can face complex issues, as well as issues that impact the whole organization. For example, a Director, in a government organization I was working with, came to me with a problem. He said, “My office chiefs don’t collaborate. They don’t think about what is good for the whole organization, they are just each out for themselves.” I had worked with many of his direct reports and I knew them to be well meaning people that did have the good of whole organization in mind. So I knew the answer lay somewhere other than the poor motivations of his direct reports.

I asked if I might observe several of his staff meetings, which he regularly held on Tuesday morning. What I saw happening was a very typical staff meeting - the Director started off with announcements, sometimes followed by a presentation from some part of the organization that wanted to get their ideas in front of the Office Chiefs. The largest part of the meeting involved the Director asking each Office Chief in turn for an update of what was happening in his or her area. There was very little conversation between Office Chiefs. Almost all of the talk was aimed at the Director – he talked to each in turn and each in turn responded to him.

The next time I met with the Director I asked him if there were any complex issues that he was dealing with but had not yet been able to find a solution for. He assured me there were a great many and named several. I suggested that he 1) devote one staff meeting a month to asking the group of Office Chiefs to help him address one of those issues, 2) that he think carefully about how to frame the question for them so as not to bias the answers they might come up with, 3) that he provide them with all the information he had on the issue, preferably ahead of time so they would be prepared, 4) that he have them work on the issues in small groups for an hour, and 5) that for the last hour he bring them together to have a large group discussion about their ideas.

Some times we held those small group conversations in a knowledge café format, sometimes in an interview format, sometimes we invited in someone who had a particular expertise. After about six months the group had begun to learn how each other thought. They had started talking with each other outside the meetings on other issues, and they had produced thinking on several issues that the Director could not have come up with by himself. The Director was pleased with what he saw as the Office Chiefs collaborating. For me the most significant move toward collaboration was on the part of the Director himself, he was willing to say, ”I don’t know,” and to draw on the collective knowledge of his very diverse and very capable team to address some very complex issues.

Conclusion
It is possible to think about convening conversations as an event – something to do once in a while to change things up. But from my perspective convening conversations to address complex issues is a new way of thinking about knowledge. KM professionals experienced a shift in how they thought about knowledge around 2005 when they expanded their definition of knowledge from explicit knowledge in a repository, to the realization that much of the important knowledge was in people’s heads and could not be captured. To make that shift a reality KM Professionals had to put new processes into play, like communities and peer assists that supported knowledge flow. As KM professionals we are again on the cusp of broadening how we think about knowledge. The shift is from limiting ourselves to drawing on the knowledge that organizational members possess, to also drawing on their sensemaking capability – their ability to bring heuristics, interpretation and perspectives to critical organizational challenges.

Are On-line Discussion Forums Conversations?

Conversation is a topic I often present about. On those occasions I am frequently asked, “Do you consider on-line discussion forums conversation?” I usually fumble and hedge my answer. But now, thanks to a study done by Yongsuk Kim, when he was a doctoral student at the University of Texas, I have at least a partial answer. Before I get to Kim’s study let me explain both my experience and my concerns about on-line forums that cause me to hedge my answer to that frequently asked question.

First, I fully support on-line forums. I have had the opportunity to set up on-line forums in many organizations; I co-authored a book about CompanyCommand which is one of the best on-line forums around; and I encourage organizations to make use of on-line forums every chance I get. In my opinion on-line forums are the gold standard of knowledge sharing.

My concerns are related to what I observe happening in many on-line discussion forums. I observe that a member asks a question and then various other members provide an answer. But there is little back and forth among the members. Rather, each responder simply makes a declarative statement that represents his or her own position. Responders may not have even read others’ answers before stating their own position.

Of course, back and forth exchanges are not necessary if the question is just asking for factual information, for example, “How much pressure should valve 25 hold?” A clear response from a knowledgeable peer is all that is needed. But too often I observe the ask/answer format occurring even when the asker is seeking the lessons of experience from other members. And I am hard pressed to call those exchanges a conversation.

I recognize there are many ways to define the term “conversation.” The way I am using it here is: “One or more persons actively working to understand the meaning another is trying to convey.” There is tremendous value gained, both in terms of learning and development, from an exchange that meets that definition. Through conversation with knowledgeable peers, a member is able to gain a different perspective on an issue or a broader way to think about a problem. It is possible to develop one’s judgment through such conversation.

Johnson and Johnson, researchers at the University of Minnesota, who study collaboration, say that the synthesis of diverse perspectives comes from being able to hold both one’s own and another’s perspective in mind at the same time. Holding two perspectives in mind necessarily means there has been enough conversation for each person to thoroughly understand the perspective of the other - not only the conclusions, but to fully comprehend the thinking behind those conclusions. And, having both perspectives in mind, they are able to then hold an intelligent dialogue that explores different interpretations of the problem. And that exploration even has the potential to create new knowledge.

For an on-line exchange to meet my definition of a conversation, much more interaction between parties would have to occur than the ask/answer format that is typical of many on-line exchanges. For example some of the following actions would need to take place:
• The question asker would reply to an answer that he or she receives by,
o Asking for an example,
o Asking for the data that supports the responder’s conclusion,
o Inquiring why that member’s response is better than an earlier response,
o Challenging the answer,
o Or explaining why the answer made sense from the asker’s perspective.
• Responders would have similar interactions with each other by,
o Referencing what another member said,
o Elaborating on another member’s answer,
o Posing a question to another member,
o Suggesting conditions under which another member’s answer is accurate,
o Asking for clarification of another member’s response

In his study Yongsuk Kim referenced this type of interaction as “collective elaboration.”

Kim analyzed 190 discussion threads that involved 1,200 participants in ConocoPhillips’ Networks of Excellence. He surveyed the members who had posted the questions that started each discussion thread, asking them to rate their learning from high to low and to specify the performance that resulted from the answers they received.

Kim found two factors in the discussion threads that were correlated with both learning and performance for the askers:
1) how much collective elaboration occurred in the discussion thread
2) how diverse the responders were.

In terms of the second factor, research has shown that diverse groups are often more creative, innovative, and productive than homogenous groups. Based on this understanding, ConocoPhillips Network leaders frequently post questions to a number of networks rather keeping the question within a single network. And ConocoPhillips’ network structure has made it relatively simple for a network facilitator to post a question in related networks.

This following graph displays the relationships Kim found in his study.(click to enlarge)

As the chart on the left shows, Kim found the most learning occurs when responses have high levels of collective elaboration and when members of other networks join in the discussion. Interestingly, there is less learning when members of other networks provide answers but there is little or no collective elaboration. The same pattern occurs in terms of performance shown in the right hand chart, collective elaboration increases the value. Likewise diversity without discussion is less helpful.

I would add to Kim's findings a lesson I learned working with CompanyCommand; when the facilitator models collective elaboration by his or her own questions, others will follow suit. I would also note that since most organizations track metrics for their forums, e.g. number of questions, or the number of responses to a question, it would be useful to add a metric related the amount of collective elaboration that occurs in the forums.

I started with the question, “Are on-line discussion forums conversations?” My conclusions are that:
1. It is possible to have conversations in on-line forums.
2. But many on-line exchanges are not conversations, they are just declarations of each responder’s position
3. When conversations do occur in on-line forums the learning and performance results go up.

I have a great deal of interest in the topic of how to encourage greater collective elaboration in on-line discussion forums. I would be very interested in hearing from organizations that have found ways to increase conversation with their on-line forums.

Tips for Making Virtual Meetings Effective

Before the Meeting

• Keep virtual meetings small; no more than twelve, even if that means you have to hold five different virtual meetings. You can hold a discussion with a small group but that is not possible if there are thirty or more people on line.

• If you hold a series of virtual meetings keep the same small groups together each time. After several virtual meetings twelve people can begin to feel like a community, recognizing each other’s voices, remembering what each person sounds like, what ideas are important to him or her. In other words members get to know each other, which in turn makes them more willing to offer their ideas and thinking on the call.

• Contact people personally before the meeting to build rapport. Reach out to participants whose participation is particularly important.

• Create a picture collage of people who will be on the call. Pictures help make the others on the call real to us. We have a much better memory for faces than for names. We have a sense that we “know” another person if we can recall how they look. If a few people don’t send a picture, you can still include their names in the circle.

If pictures are not possible use a clock face to help yourself and others know who is on the call.

At the Beginning of a Virtual Meeting
• Have the pictures of the people on the call up on the webinar screen. If it is a teleconference, send an email of the collage ahead of time and ask people on the call to have it up on their screen.

• Ask people to always say their name before they speak. The format is: “This is Joe Fitz, I agree that ………” If someone forgets, ask them to say their name after they have finished. “That was very helpful, who was just speaking?” The facilitator should also say his/her name each time he/she speaks to help to set the norm.

• Provide time at the beginning for a check in. Get each person’s voice in the room at the beginning. Checking in creates a friendly atmosphere. People need an example to know how long they should talk. As the facilitator, provide your answer first so others will have a model for how detailed they should be. But before you provide your example, give everyone a moment to think about what they want to say. Here are some possible check in questions:
o What is something about (x) you’ve learned in the last week?
o What do we need to be working on together?
o What was the highlight of your month?
o What have you learned from others that you have tried?
o What is the most interesting thing that happened in your shop this week?
o What are you most grateful for this morning?
o What has inspired you this week?

• Call on people individually early in a meeting which sets the expectation that you might call on them at anytime.

During a Virtual Meeting
• Many virtual meetings are scheduled as presentations by either faculty or members. However virtual meetings can have a broader range of goals than just informing. Virtual meetings can also:
o Problem solve
o Make decisions about how or what work will be done
o Generate ideas

• In actuality, virtual meetings are a poor medium for presentations. As listeners, we simply don’t experience enough cognitive stimulation to keep our attention on an audio presentation. However, there are alternatives to having a speaker present for an hour.

o Send an article or report out ahead of time so the call can be a discussion of the content. When you send out content put it in a separate email and put in the subject line, “Action required.” Send the content out about 3 working days before the call. Let members know, “You will need about 20 minutes to read this.”

o Send an article or report out ahead of time and ask participants to prepare questions for the presenter. Then during the virtual meeting call on each member to ask his or her question. Because questions are often based in a specific context, give the question asker time to provide enough context for the speaker to develop a useful and thoughtful response. Some questions may even become more of a dialogue as the speaker asks other members for their response to that question.

o If you must have on-line presentations keep them to no more than 15 minutes before giving the group an opportunity to interact for at least 10 minutes. Group interaction does no mean asking the presenter questions. Rather the facilitator asks the group a question (see attached questions) that gives them space to offer their thinking about what the speaker has said. Then the presenter might continue again for 15 minutes, then again group discussion. "Ted Talks" must surely have convinced us all that 15 minutes is adequate to present even the most complex of ideas.

o Keep a list of participants in front of you and check names off each time someone speaks. That way you will know who you might need to call on to get their thinking.

Near the Meeting End
• During the meeting create a list of “to dos” and list them verbally at the end of the call.

• One of the most effective ways to end a meeting is to ‘check-out’ with each member to gain closure.
- Name one take away from this meeting
- Give a number from 1 to 5 that represents your confidence with this solution
- Say one thing you are going to do about this issue before our next meeting

After the Meeting
• Summarize the call and send the notes to everyone. If the chat function is used clean it up (clean up names, correct spelling, remove small talk) and send the script.

• Send the to-do list to everyone on the call.

Questions to ask during a Webinar or teleconference to get a discussion started after a presentation or after another member’s comment.

• “John has just raised an interesting issue. I’d like to hear other’s thinking about that issue.”

• “Mary has told a story about a frustrating experience. I’m sure many of you have had similar experiences. I’d like to hear a couple more examples.”
• Jack has explained how his team does “X.” What are other ways you have accomplished “X?”
• What are the 3 most important things to consider when doing “X?”
• “Fred has given us a very interesting presentation. What did you hear that really stuck with you?”
• “How would you complete this sentence? ‘The greatest difficulty we face with XXX is _________________.”
• “Sue has told us about an improvement in a rural hospital.
o What did you hear that would work in your setting?
o What did you hear that would need to change because your setting is so different?”
• “What did you hear in Joanna’s presentation that you could put into practice next week?”
• “We’ve heard three examples, what did you hear that was similar in all three examples?”
• “What possibilities come to mind based on what you've heard?”

These questions maybe equally useful in a face-to-face format.

Methods to Use to Ask the Questions
• Ask one of the questions above, and then say, “Let me give you a minute to write a few notes to yourself before people start to answer. (then count to 60 before opening the microphone or calling on people)
• Poll the group – If the group is under 10, call on each person in turn. Then summarize what you heard in the poll. “It sounds like we are divided between (X) and (Y).” or “I’m hearing general agreement about XX.”
• Poll in a random fashion if the group is large – “I’m going to call on a few people to get a sense of the reaction to (Bill’s) idea.” Then call on 5 people. Then ask, “Who else has an idea to offer?” And wait at least 20 second before moving on.
• If all the responses to a poll are similar ask, “I’d like to hear from anyone with a different view. Who sees it differently?” Then give it 20 seconds.
• Present a simple scenario and ask each participant to share how they would respond in that situation
• Use the chat function to get responses to a question on a 1-5 scale. “Just type a number from 1 to 5 into the chat area.” Then ask, “Most of you put in a 4 or 5 but there are a few people who were at a 1 or 2. I’d like to hear your thinking”
• Use the voting function on a webinar with pre-designed questions based on the presentation.
• Pause to have a paired conversation. Say, “If there is someone else at your site, talk with them for a few minutes before we call on people for their response. If you are by yourself then make some notes.”
• In the chat function, ask participants to write in brief notes during the presentation about what they agree and disagree with, for example, “X’ was a good point.” Or “I agree with “XX.” These notes are not for the presenter to read, but for others on line to see the reaction of their colleagues. It also is useful for starting the discussion after the presentation. “Many of you agreed with “XX,” Bella, what was your thinking that caused you to agree?”
• Don’t feel you have to answer every question asked of you. It is often useful to ask the group members what their thinking is on a question addressed to you.

Conversational Patterns That Support Telling Truth to Power

The Defense Intelligence Agency’s (DIA) Knowledge Lab faced the challenge of how to move accurate intelligence up the chain of command. Too frequently the intelligence analysis, painstakingly generated by front line analysts, was delayed and often severely modified by a chain of superiors before it reached policy makers who could act upon it.

Zeke Wolfberg, the Director of the Knowledge Lab and I, as the knowledge management consultant to the Knowledge Lab, recently published an article about how we addressed that problem at DIA. The article appeared in Reflections, the SOL Journal on Knowledge, Learning and Change (Vol. 10, 4). I provide a summary of the article in this post. The full article can be downloaded under My Publications on the left.

DIA provides military intelligence to prevent strategic surprise and deliver a decision advantage to warfighters, defense planners, and policymakers. The analysts who develop intelligence assessments are divided into teams of 30 plus with each team focused on a different region (e.g. South East Asia) or a different functional area (e.g. missiles), as well as task forces that are periodically assembled to address crises (e.g. the Mumbai terrorist attack).

The intervention we undertook to enable telling truth to power, was called “Critical Discourse.” The format was to bring together a team of analysts along with their supervisor, to jointly analyze the actual conversations that occurred between members of the team and between team members and their supervisor. The conversation analysis was based on the work Chris Argyris, using his left and right hand case format to create scripts of difficult past conversations. Each analyst and supervisor selected and wrote out the dialogue that occurred in three difficult conversations they had engaged in. Each team met over a period of three months to analyze each other’s cases and to practice Argyris’ Model II skill set with the goal of reducing the misunderstandings occurring in their conversations. Each team member also received individual coaching between group meetings.

Model II is a way of interacting that results in knowledge and learning for everyone who is engaged in the conversation. The goal of Model II interaction is not to win an argument, but to create a space for joint learning. A person using the Model II skill set holds in their mind the underlying assumption that, “I have part of the information and others have other parts that I may not be aware of.” The skills are used to get all the information that is available and pertinent into the conversation. The Model II skills that produce full and accurate knowledge are:
 Advocate your own position and share all relevant information related to the issue, not just the data that supports your view
 Encourage others to question your position so if your reasoning is faulty or if there is information you have missed you will discover it
 Ask questions about the position of others, not to find where you might “catch them” in a mistake, but to fully understand their reasoning and data.

The Argyris Model II skills are simple to state, but take a great deal of practice to put into place in a real conversation. The difficulty occurs because the default way of interacting in difficult situations in nearly every organization is Model I, which defeats gaining learning through conversation. Model I skills are:
 Advocate my position in a way that discourages inquiry into it
 Keep my reasoning private so others cannot challenge it
 Don’t ask others about their reasoning
 Assume I understand the situation and anyone who sees it differently does not

A team using Model I skills in a contentious or potentially embarrassing situation would exhibit interactions that are familiar to all of us. For example, 1) a team leader states his opinion on an issue in such a way that it is clear he doesn’t want to be challenged; or 2) a team member offers an idea that no one understands, but other team members don’t ask him the questions that would make his thinking available to them - his idea just dies on the table, or 3) a team member offers a solution, providing convincing evidence about why it would work, but doesn’t reveal the very real risks he knows are involved.

Using Model I interaction teams make poor decisions because all of the available data about the issue does not get into the discussion. Team members are more interested in winning their point than in learning. Yet Model I interaction is so ubiquitous that team members are largely unaware they are employing it in a conversation. To use Model II interaction skills, team members first have to become aware of their Model I interactions. Fortunately, others can see what we cannot see in ourselves. So a team’s joint analysis of their cases allows members to help each other become aware of and then unlearn Model I skills. And provides them a safe place to practice using Model II skills.

Over a two year period, I worked with 15 DIA teams using the Critical Discourse intervention, including teams at the highest executive level. During that time team members wrote hundreds of conversational cases. The analysis of those cases resulted in two outcomes that shifted the culture of DIA; 1) both seniors and analysts developed the skills to challenge each other effectively – to tell truth to power, 2) dysfunctional patterns that were preventing accurate information from being exchanged across the whole of DIA were identified. These patterns, that Argyris calls defensive routines, had become embedded in the fabric of the culture in such a way that made it unlikely any that any individual, on his or her own initiative, would be able to act in a manner other than Model I.

To illustrate the process through which the culture changed I will focus on one of the teams that was set up to look at an emerging intelligence issue. The code name for this team was “Fresh Look.” Fresh Look is discussed in the article but in less detail than I write about it here.

The Situation
Fresh Look was given the task of finding a solution to a long-standing problem that many other teams had been unable to resolve. The team was to report their findings to a high level executive team – tell truth to power.

Previous DIA teams that had struggled to find a solution to this issue had worked virtually, dividing the work up so that people could function as individual contributors with minimal interaction. Fresh Look, however, was co-located to give them the advantage of bringing together their collective knowledge to solve this difficult problem. But co-location meant that to be successful in using all the team’s knowledge, the team would have to interact in a Model II manner to jointly address content issues. To assist the team, early in their time together, the Knowledge Lab provided the Critical Discourse intervention described above.

Using the Critical Discourse format, each team member wrote three cases about their interactions with other Fresh Look team members. By reading and discussing each other’s cases, team members were able to see, laid out in black and white, the negative impact of their actions. With practice they began to recognize the conversational tactics that were in the cases being exhibited in their real time conversations and begin to call each other on it when they recognized Model I tactics. With further practice of the Model II skills, they began to be able to share their knowledge more fully and to speak openly about their concerns related to the problem they were addressing.

A turning point in terms of new thinking for the Fresh Look team came about six weeks into the project. For some time the team had been struggling with how to organize themselves to deal with the central issue of their analysis. This issue was not only at the heart of the Fresh Look topic, but was an issue that touched on the very nature of how DIA does analysis – so how the team structured themselves and their work was a very relevant topic for the team.

At a Friday meeting, with two of the team members absent, the team finally made a decision to organize into several smaller teams, each devoted to a piece of the larger question. The norm within the intelligence community is that when a meeting is convened and decisions are made, those who were absent are expected to accept the decision and support it. The rationale being that there is not enough time to rehash decisions. There is also an implied perception that if the team member had really wanted to be present at the decision meeting, they would have attended.

On the following Monday the absent team members were back in place for the morning meeting. The absent team members respectfully challenged the decision-making norm by asking the rest of the team to go over their logic for reaching the decision of dividing into several smaller teams. Instead of rebuking the two, the team not only discussed their logic of the previous Friday, but the entire topic of methodological pursuit was reopened for debate: should the team employ the status quo method of structuring itself or should they behave differently? During that discussion, the team members who were present on Friday acknowledged that their decision was likely driven by being tired at the end of the week and wanting to end the day to start the weekend. The willingness to challenge and be challenged that had begun to develop within the team during the previous weeks of practice using Model II was for the first time being fully exercised.

Encouraged by the group’s response to this initial challenge, a few days later another team member was willing to raise a much more critical issue that had been troubling her - how the team had framed the analysis question. Her position was that the team was restricting the analysis by unthinkingly accepting as their starting point the assumption currently held among subject matter experts. The team took her concern seriously and was able to hold an honest and open interaction that resulted in significantly broadening the scope of their analysis. The reframing of the question eventually led the team to discover a remarkable finding - that the intelligence community had ignored a plausible counter-assumption without ever collecting data to prove or disprove it! Essentially leaving the intelligence community blind on a critical issue.

The on-going practice the Fresh Look team had in using their inquiry skills to elicit knowledge from each other and to respectfully challenging each other prepared them to speak truth to power when their scheduled briefing with high level DIA seniors occurred at the end of the project. By the time of the executive briefing, team members had grown confident in using the Model II skills in increasingly challenging situations.

The following are two descriptions of that meeting, first from the perspective of a senior executive who was in attendance. The second from the perspective of one of the Fresh Look team members who engaged the seniors in conversation.

Senior Executive Perspective

"I was one of five senior executives attending the debrief of Fresh Look. When I walked into the meeting room I could see that the physical arrangement of the seating area was very different than what I was used to. Normally, seats were arranged in rows facing a podium. On this day the seats were arranged around a set of tables so that everyone sitting at the table could see each other. During the team’s introductory remarks they mentioned how they intentionally designed the physical space to promote mutual learning between the seniors and themselves rather than the typical transmit then respond mode of communication.

The Fresh Look team explained that their briefing content would be different from the usual format as well. The team said that the goal they had in mind was that everyone around the table would learn with and from each other. I felt okay with this approach but wondered if they could pull it off. They started with brief presentations of their findings during which several members of team offered their own perspective, providing a variety of views, again not a typical debrief. Instead of just saying, “Here are our findings.” the team spoke about the challenges they had faced and how they had navigated those challenges, providing context for how their conclusions had been reached. Having laid out their findings, the team indicated they were ready to open the floor for dialogue.

I noticed that during the give and take of the conversation the team members were able to strongly advocate the findings they had developed as well as demonstrate their openness to our (the executive team) views. However, one of the team members, Carrie (not her real name), said something that really bothered me. She made a claim that a certain piece of information was non-existent. Now, I was aware that she was wrong, that piece of information was available. I said just that, perhaps a bit more forcefully than I intended. And I said that if the team had been thorough enough in their search they would have found it. However, rather than becoming defensive, Carrie said she found my comment interesting, because even the senior member of their team who had full clearance to look at all the available data did not come across it. She began a discussion among us all about the availability of information across silos. Out of that discussion I gained a new understanding about the issue of information availability from the perspective of analysts doing a search. It was a very stimulating conversation."

The Fresh Look Team
"The meeting with the executives was even more successful than we had thought it would be. The executives were impressed with the quality of our findings. And we were pleased with the skills we were able to exhibit in being able to respectfully detect and overcome the “old guard” approach.

As a team we were particularly proud of our interaction with Sam, the senior executive who reacted critically when he realized that the team did not know about the existence of a critical piece of information, saying, “Everyone knows it exists.” We were able to recognize the role-based behavioral pattern in Sam’s remark and were able to avoid the emotional trap that could have deflated and sidetracked us away from the issues. We were able to detail the process we went through to find the data, what was made available to us, and what had been hidden from even the most senior analyst member of our team who specialized in the issue. This detailed explanation led to a more general discussion of how information is shared and accessed which brought new insights both to us and to the high level executives present. Through using the critical discourse skills, our goal of using review meetings to jointly develop knowledge was realized.

The Director of DIA attended the debrief and told us later that what he witnessed was the embodiment of team behavior - the kind of team behavior DIA needs for detecting new threats. He said the discussion with the team was cathartic for him."

Fresh Look is one of many examples of changes in the ability to speak truth to power at DIA. To do so analysts and seniors first needed to recognize their own contribution to the dysfunctional patterns that existed inside DIA. Secondly, they needed to learn a new way of interacting and practice those new skills in a relatively safe environment. Finally, they needed to use the skills to interact in real world situations to gain confidence in their own ability to interact more openly and honestly. Through this slow developmental progression they were able to bring knowledge into a conversation that otherwise would have been hidden and they were able to speak truth to power.

Guidelines for Leveraging Collective Knowledge and Insight

Knowledge workers in any organization have a wealth of insights that are available to their organization to address the difficult issues the organization is facing. Drawing out those insights requires bringing knowledge workers together in meetings that are expressly designed to take advantage of collective knowledge. Over the years, as I have designed such meetings, I have come to rely on seven principles that work together to make the most of collective knowledge in conference settings as well as in-house meetings. The principles have been assembled from the work of many researchers and thought leaders. Where possible I have identified the sources.

1. Connection before content
Knowledge workers attending a meeting or conference need to get connected to each other before they try to construct new ideas together (Peter Block). In order to work effectively with others, they need to know:
a. who is in the room
b. what knowledge others have
c. how others think about the issue of the meeting, and
d. the group’s strengths and weaknesses

In a business context, connecting is best accomplished by engaging knowledge workers in small group conversations about strategic organizational issues. To build connections, those conversations have to be structured in a way that allows knowledge workers to frame themselves in a positive light, relative to their experience and successes. Icebreaker type exercises are less useful when the intent is to leverage collective knowledge because they do not provide adequate understanding of others’ knowledge and experience.

When a group has come together many times, the period of connecting can be brief, but not neglected altogether. Just as two friends typically engage in “small talk” for a few minutes each time they meet, any group that comes together regularly also needs a brief period of re-connecting before turning to content. In both situations the “small talk” affirms the relationship and the readiness to engage the topic.

2. Circles connect
Circles represent unity. They help individuals in the group view themselves as part of the whole. For example, the UN meeting hall is designed in concentric circles to provide a visual representation of what the UN stands for – unity among nations.

I was reminded of the benefits of seeing oneself, and being seen, as a part of a whole, when I conducted an exit interview with Lieutenant General Maples, then Director of the Defense Intelligence Agency. He told me that one of his first actions as Director had been to remove the large rectangular conference table in his office and have it replaced with a round table. He very clearly understood the difference a circle would make to the many, very difficult conversations he would have in the coming years.

It is a useful symbolism to begin and end meetings with chairs in a circle. It can be a big circle of 30+ or many small circles of 5. Ideally it is a circle of just chairs, without a table. Knowledge workers have a profoundly different experience when they converse in a group, absent a table.

3. Learn in small groups – integrate knowledge in large groups
We learn and create new ideas through our conversation with others in small groups. A small group is 3-5 members. This is the size that produces the richest and most in-depth thinking. It is large enough to contain diverse views yet small enough for members to engage each other.

Engaging each other means asking questions to clarify the meaning another has expressed and it means challenging as well as building on others’ ideas. The give and take of the small group serves to exchange existing knowledge as well as generate new knowledge.

In meetings designed to draw on the organization’s collective knowledge, after small groups have been in conversation, their ideas are brought together in a large group setting to integrate their insights into the thinking of the whole. In a lengthy meeting, small and large group discussions regularly alternate.

4. Diverge then converge
Any meeting that is focused on collective knowledge must first diverge in order to draw new ideas out and to stimulate thinking. “Knowledge diversity facilitates the innovative process by enabling the individual to make novel associations and linkages” (Cohen and Levinthal). Meeting leaders who want to leverage collective knowledge, intentionally invite people from other disciplines, stakeholder groups, and outside experts to introduce that diversity.

Without time for divergence the many differences within a group are not expressed and without their expression they cannot be made use of by the group. It is in the critical space that lies between divergent ideas that innovation often emerges (Scott Page).

Knowledge workers are reluctant to turn their thinking to what they have in common until they have had an opportunity to give voice to their differences (Marvin Weisbord). The way the human brain works is to first recognize differences and only after those are clarified to focus on similarities.

The first part of a meeting, whether it is half a day or 3 days, is dedicated to divergence. The second part of a meeting converges to put those divergent ideas to work in terms of a complex issue that needs to be addressed.

5. Outside experts inform the thinking of others, not provide them answers
Bringing outside experts into a meeting can provide much needed cognitive diversity, but experts cannot provide solutions. It is the knowledge workers of an organization, who function within its context and who together understand the complexity of its issues, who are uniquely capable of developing workable solutions.

Experts can stimulate the thinking of knowledge workers by talking about what others have done in similar situations. But any practice from another organization or team always has to be adapted, not adopted. Even the best of examples have to be modified to fit a new context.

Fifteen minutes is adequate time for an expert to stimulate a group’s thinking. A presentation by an expert needs to be immediately followed by a period of time dedicated to participants connecting what the expert has said to their own knowledge and to thinking with others about how that knowledge can be used. Without committed time for processing, an expert’s ideas are forgotten within a couple of hours. Unfortunately, the ubiquitous Q&A does not provide the needed processing time; rather what is needed is time for colleagues to talk with each other in small groups.

6. Connect new ideas to what knowledge workers already know
Processing is about moving new ideas from short term to long-term memory by connecting the new ideas to knowledge worker’s existing knowledge. The term, absorptive capacity (Cohen and Levinthal), references the ability of knowledge workers to recognize the value of new, external information, assimilate it, and apply it to their organization’s issues. An individual’s or an organization’s absorptive capacity, is a function of their prior knowledge. That means that without related knowledge to connect new ideas to, new knowledge will not be absorbed.

In meetings designed to exploit collective knowledge, a more robust environment for knowledge linkages to be made is provided when knowledge workers from varying disciplines are present. A knowledge worker from engineering, for example, will have different prior knowledge than one from R&D, and different yet from a knowledge worker from legal. The greater the cognitive diversity of prior knowledge in the room the more likely that new knowledge, from outside or internally, will be connected to prior knowledge which will spur insight. To process new knowledge, group discussion, made up diverse participants, is most effective. If the group is small (3-5 participants) each knowledge worker has enough airtime to put their ideas into words – which leads to the last principle.

7. We learn when we talk
This principle is central to leveraging the collective knowledge of a group. Listening provides us new ideas but as long as those ideas are just swimming around in our heads, they are neither fully formed nor implementable. It is only when a knowledge worker puts an idea together in a way that allows him or her to explain the idea to others, that the idea takes shape for the knowledge worker, as well as for the person the knowledge worker is talking with.

Johnson and Johnson, researchers at the University of Minnesota, have shown that we organize information in a different way when we are preparing to explain our thinking to others. The information not only becomes more logically organized, but new connections are made, often in the act of speaking. It is fair to say, “We don’t learn when we listen, we learn when we speak,” or write, or even create a visual representation of our understanding. Giving knowledge workers the time needed to put their thinking into words, not only shares the organization’s knowledge, but creates it.

Implementing these seven principles requires considerable change in how we design meetings and conferences - but then the idea of leveraging collective knowledge itself requires considerable change. More expansive thinking about who within the organization has valuable knowledge with which to address organizational issues, necessitates new approaches to how we meet and work together. These seven principles, tested through research and practice, provide a way to design those new approaches.

Conversations That Share Tacit Knowledge

I was doing some work on knowledge transfer at Erickson in Sweden. As I interviewed technicians I kept hearing Hans’ name come up as the person to go to if a technician had a question about the computer language C++. Finally I checked with someone, “So Hans must be the resident expert on C++?” and was quickly told, “Oh no, the resident expert is Joachim, but he’s not very good at explaining. Now Hans, he can really help you understand how C++ works with our switches.”

Later when I interviewed Hans I told him that I kept hearing his name as the person a technician should seek out if they want to understand C++. I asked him how he got that reputation. Modestly he said, “Well, I’m not sure how I got the reputation, but I’ll tell you how I try to help people who come to me with a question. Is that what you want to know?” I agreed it was what I was after. So he explained, “When a technician comes to my office he usually starts by explaining what he wants to know about. But before I try to answer his question, I ask a few questions to see how much he already knows about C++. Then I ask a few more questions to find out what he’s planning to do with the information. By that time I’ve formulated an idea about where to start and what I need to say. And we just go from there talking back and forth.”

Now when I’m asked, “What's the most effective way for people to share their tacit knowledge?” I always think of Hans and the answer I give is: “Tacit knowledge needs to be shared through conversation.” My reasoning is as follows.

Our tacit knowledge is drawn from our experience as well as our years of study and is stored in bits and pieces in our brain, that is, it is not stored as answers or explanations but as fragments. What we call “tacit knowledge” is the human ability to draw on those fragments to construct a response to a new problem or question.

Tacit knowledge is particularly useful when we are faced with a complex problem. By complex I mean a problem that does not have a factual, right or wrong answer, for example, "What architectural design would best fit this physical space and meet the needs of the client?" or “How would you stop an oil leak 5000 feet under water?” When an expert like Joachim faces a complex problem he brings together those bits and pieces of his experience and study that are relevant to that specific problem situation and puts those together to form a solution. Because he is embedded in the situation he knows the context and the end goal. In bringing together those bits and pieces that are in his head, he conducts, what Don Schon would call, a “reflective conversation with the situation.”

But when someone like Hans is asked a complex question he faces difficulties that Joachim does not have to deal with. Hans knows very little about the asker’s situation. Without that knowledge he can only provide a general answer - rules of thumb, for example. But he can’t give an answer that takes into account the asker’s situation. If Hans wants to respond to a specific situation he will have to learn much more about it. To share his tacit knowledge he will first have to have a conversation with the asker and then create the knowledge that applies to that situation in the moment of answering the question. The knowledge he supplies never existed in that form before he spoke it aloud. If he were to be asked the same question tomorrow his response would likely be different because he would have new fragments of knowledge stored in his brain. One of those fragments might be more recent and so more readily come to the fore. Tacit knowledge, then, is constructed in response to a question or to a problem at a specific moment in time. It is a magnificent human capability we have to be able to continually reconstruct what we know into new forms to face new situations.

The conversation between an asker and an expert is necessary because it affords the opportunity for a number of important exchanges:
o the asker can offer information about the situation in which the knowledge will be applied,
o the expert can probe deeper about that situation,
o the expert can gain a sense of what the asker already knows so the expert can determine at what level to construct his/her answer,
o the asker can ask the expert about the meaning of a term or concept the asker is not familiar with,
o the asker can seek the reasoning behind a conclusion that the expert has offered, when it is not evident to the asker,
o the expert can correct any false assumptions the asker is making about what the asker needs to do or about concepts the asker holds,
o the asker can correct any assumptions the expert is making that do not fit the asker’s situation,
o and on and on

It is in the back and forth of conversation, that is, both parties actively trying to understand the meaning the other is attempting to convey, that tacit knowledge is exchanged.

The reasoning I’ve just offered begs the question of whether the conversation can take place virtually, in other words, “Does it have to occur face-to-face?”

From my perspective, there is a great deal of useful information that people can share virtually, for example, in an on-line discussion or a list-serve. I’m a great advocate of such discussions and have written in CompanyCommand about their value. Members of the community can share their experience, support others, and provide solutions or answers to some types of questions. For example, a member can write, “I need some new ideas about how to do a safety briefing, because having given them too many times, my briefings have become old and stale.” then others can respond by describing how they do safety briefings. Or a member can ask, “Has anyone had trouble making an HP printer scan from a Mac?” and others can offer their own suggestions.

However, my experience with these kinds of on-line discussions is that members who respond to a question rarely ask for any context. Rather they respond in declarative statements about their own experience or they offer their own rules of thumb. Seldom is there an attempt for asker and responder to probe the meaning that the other is attempting to convey. For this reason I think on-line discussions are not an effective way to share tacit knowledge.

What about phone or email exchanges?
A phone call has greater possibilities for sharing tacit knowledge than does email, but still has limits that can reduce the extent of the back and forth of conversation. Any conversation has two levels of meaning that are constantly being conveyed. One is about content and the other is about who the asker and responder are in relation to each other. The second conveys such information as, “Can I trust you to not think less of me if I don’t understand immediately?” or “I don’t really have time to answer your question in detail.” That conversation is expressed through intonation, gestures, and facial expression, little of which can be conveyed over the phone and none of which can be conveyed by email – thus the profusion of misunderstandings that email generates! Either medium is more likely to serve the transfer of tacit knowledge if the two people are well acquainted, which provides past experience with which to fill in or interpret the second conversation. But by-and-large, email and phone are less effective mediums for transferring tacit knowledge than is a face-to-face conversation like Hans had with the technician.

As I studied the way Hans transferred tacit knowledge, I identified two simple elements that helped Hans earn his reputation for being good at explaining C++, 1) he conducted his conversations face-to-face, and 2) it was an exchange in which Hans took the time to understand the current knowledge level and the context in which the question was situated.

A Model Lessons Learned System – The US Army

The US Army Lessons Learned system has evolved over 40 years to become a model lesson learned system. What began as an AAR process in the 1970s has become a robust system of identifying, collecting, analyzing, transferring, and moving lessons learned at all levels of command. I have detailed the progression of this system using the model I constructed for The Three Eras of Knowledge Management.

The Center for Army Lessons Learned (CALL) in Ft Leavenworth is the center of the Army’s LL program. Since the beginning of the Iraq war CALL has become a subset of the larger Army Combined Arms Center’s Battle Command Knowledge System located in Ft Leavenworth. The goal of the US Army Knowledge Management System is to capture, integrate and use organizational knowledge to gain an advantage over the enemy.

Elements of a Robust Lessons Learned System

There are five strategic elements in a robust and effective lessons learned system; Collection, Repository, Transfer Process, Implementation, and Analysis and Data Mining. Each is illustrated here in the description of the Army’s lessons learned system. For many of the elements the US Army has multiple processes through which that element is accomplished, adding to the robustness of the system.

1. Collection: A robust LL system has multiple ways to identify and collect lessons learned

• After Action Reviews (AAR’s) are systematically conducted at all levels on all events. In 2008 alone 20,000 observations, insights, and lessons were collected.
• Collection and Analysis Teams (CAATs ) – Teams (of 10 or fewer) are sent to the field for 1-2 weeks to study issues identified by the command element. Ten to fourteen teams are deployed annually. Members of these teams are from proponents related to the issue being studied. They are directed to work together, assuring that multiple perspectives are considered. Members go to Ft Leavenworth for training in the collection process. While there, they develop a collection plan with sub issues, a question list and a list of people and units to interview. Before returning to Ft Leavenworth to write the final report, the team provides an initial report to the unit Commander for approval. After the final report is written at Ft Leavenworth collection and analysis teams brief the Commanding General of CAC. All of the issues collected are quickly moved out through the L2I network (see below).
• On-line community discussions - peer-to-peer knowledge exchanges that allow soldiers to quickly adapt to rapidly changing situations. There are 60 BCKS functional communities (CompanyCommand, Platoon Sergeant) and as many field communities (CAVNET). A fulltime facilitator supports each community.
• Lessons Learned Integrators (L2INET) – Analysts are deployed with units, as well as being stationed at Schools, the National Training Center, TRADOC and Headquarters. L2INET members are responsible both to actively collect what is being learned in the commands and to disseminate the lessons learned from other commands. The network of 29 members meet regularly on-line and in monthly teleconferences to exchange what they are learning in the field and to understand what other members need.
• Theater Observation Detachment Program (TOD) – Primarily reserve volunteers deployed with the commands for a period of 6-12 months. They are the eyes and ears of CALL. Observers write topical products based on collection in theater and have reach back capability for units. They share information electronically with other units in-country and between theaters. The network of 48 DC0’s hold weekly sessions for the exchange of knowledge.
• Umbrella weeks – Returning deployed units remain in garrison for one week to be debriefed by any proponent that needs their up-to-date information, e.g. replacement units, seniors, policy makers. The event is run by the unit, which solicits interested parties to collect data and makes key people available in response.

2. Repository: A central repository that stores documented lessons learned and makes them available to the whole system.
CALL serves as the central repository for the US Army. Two hundred employees work in the CALL center checking documents for classification, adding metadata and archiving data. CALL center employees also respond to RFis, monitor web sites in operational theaters, and download documents.

3. Knowledge Transfer Processes: Active as well as passive processes for moving the knowledge to targeted areas
o RFI – CALL has 11,000 requests a year that are answered drawing on knowledge stored within the repository. RFIs from deployed units are answered within 24 hours. Sixty percent of RFIs are at the battalion unit level or below.
o L2I network - The L2I analysts deployed with units disseminate lessons learned through other L2I members. L2INET has made a significant difference in the dissemination (flash to bang) of observations, insights and lessons (OIL) and best practices
Example
The 2nd BCT, 10th Mountain Division reported they were using a device called the Rat Claw, a specially designed steel hook, to rapidly pull open HUMVEE doors in order to rescue soldiers trapped in an overturned vehicle. The L2I analysts at 10th Mountain posted a video with specifications and within 24 hours other operational units were able to construct similar devices
o Community Discussions – both functional and in-theatre
Examples
An Army Reserve major in Wisconsin wanted to develop a training capability for logistics units deploying to Iraq. The officer contacted a LogNet facilitator and discovered that the Army had no written doctrine or useful instructional materials, so he summoned the logistics community via LogNet. The responses pointed him to a major at Fort Hood, Texas, who had set up a logistics support area in Iraq. The two connected, and the Wisconsin officer found the insight he needed.

Iraqi insurgents placed an IED behind a poster with anti-American slogans. A soldier noticed that the poster looked different from others he had observed, so he entered information about the suspicious sighting into CAVNET. A threaded discussion developed on-line while specialists evaluated the potential threat. When they confirmed the soldier’s suspicions, the Army sent a message via the system to alert other units about the insurgents’ new method of concealing IEDs

4. Implementation methods: Processes to put lessons learned into practice and resolve issues raised in the lessons
• L2I network members have a direct line of communication to unit Commanders who have authority to implement changes within their units
• Community members implement lessons from their peers in the community.
• Issue Resolution Cycle: a quarterly cycle of councils sort, make recommendations, and take action on identified issues

If through collections and observation there is enough evidence that there is a problem, the first level action officers working group vote to take it on, for example instruction in new TTPs. They decide who should be the lead and forward the issue to the Council of Colonels. Issues that cannot be addressed at a lower level go the General Officer Steering Committee. Since funding is above the army level the COS submit some issues to the joint committee

5. Analysis and data mining: Processes, Analytic tools and resources for reviewing and analyzing large numbers of lessons to gain insights that would not be obvious from examining a specific lessons learned. Analysis and mining data:
• Allows for the identification of weak signals that can provide early warning about an issue
• Facilitates finding trends across units and across periods of time
• Identifies gaps in knowledge
Example:
One of the troubling realities of combat is that soldiers are in the greatest personal danger at the beginning of their combat experience. Lessons about how soldiers should protect themselves are embedded in many of the AARs CALL receives. But it would be unreasonable to expect soldiers to search through thousands of AARs to find the knowledge needed for their own safety. CALL gleaned lessons from unit AARs to produce a handbook, The First Hundred Days. The handbook is provided to all new combatants before they deploy.

The Army lessons learned system has evolved over time as the model illustrates. As new technology becomes available the lessons learned system has added capability. As new challenges arise in theater, the system adapts. New processes are introduced to meet new situations but the five strategic elements Collection, Repository, Transfer Process, Implementation, and Analysis and Data Mining remain as a constant.

How to Make Use of Your Organization’s Collective Knowledge – Accessing the Knowledge of the Whole Organization - Part I

Collective knowledge is the next step beyond knowledge sharing for organizations. More than just making use of the existing knowledge employees have, collective knowledge makes use of the sensemaking capabilities of employees.

Sensemaking is the very human act of creating meaning out of the incredible amount of data and input that continuously surrounds everyone within an organization. Sensemaking is a creative act, not an act of discovering or uncovering what is already there. The meaning created through sensemaking did not exist before the conversation that created it. In conversation the meaning that is created alters as new data and new patterns in the data emerge. The understanding that results from sensemaking is not a definitive answer, rather it is an understanding that is adequate for the organization to plan and take its next action.

The need for leveraging collective knowledge arises when organizations face difficult, complex issues – the kind of situations that are marked by disagreement on what the problem even is and certainly disagreement on what would constitute an acceptable solution. I have written a number of blog posts about organizations that have faced such issues, e.g. the US Army facing the issue of changing how doctrine is produced, the World Health Organization’s eradication of smallpox, the cancelation of NASA’s Constellation program.

Understandably most leaders see it as their responsibility to come up with answers for the complex issues their organizations face, ascribing to the manager-as-problem-solver model of leadership. That has certainly been the predominant model for thirty years - taught by most (but not all) business schools, management development programs and propounded by best-selling management books. However, over the last few years a new way of thinking about leadership has been gradually emerging. The leaders of the organizations I have written about in this blog are people who have come to understand that no one person has the sensemaking capacity needed to deal with truly complex issues, no matter how highly placed the leader is. These leaders have redefined their role from problem solver to conversational architect, designing ways to leverage the knowledge of the whole organization when faced with complex issues.

Collective Knowledge can be accessed at different levels within an organization, that is, the whole organization, divisions, teams/groups, or even between two individuals. Here I am focusing on accessing the knowledge of the whole organization.

I told the first story about a leader recognizing that he needed the sensemaking capability of the whole organization in a book I published in 1994, The Organizational Learning Cycle. It was the story of Ralph Stayer, the CEO of Johnsonville Foods. Johnsonville Foods was faced with a complex issue in the form of an opportunity. It was a small, family owned sausage company known in Minnesota for its quality. The issue the company faced was whether to “accept an offer from a food-processing company to buy large quantities of product on a regular basis. It was a complex issue because Johnsonville did not have the capacity to handle the job so accepting it would mean everything would have to change.

Stayer called a meeting of the whole organization, giving them all of the information he had and asking them to work in teams to answer three questions: What will it take to make it work? Is it possible to reduce the downside? Do we want to do it? The teams struggled with the questions for almost two weeks, wrestling with the risks, which were considerable, and figuring out how they would have to operate to accomplish that much increase in production. In the end the teams almost unanimously recommended taking on the new business. Reflecting on the process, Stayer said, ‘If you issue orders you’re telling people, don’t think; just do. But if you’ve got 1000 people, you’ve got a 1000 minds. And if you issue orders from the top, you’re using only 3 of them, or 2, or one. That’s stupid.’”

The second story is one that Ron Heifetz wrote about in Leadership Without Easy Answers. I’ve summarized it here but the full story is woven throughout his book and he uses it to build the case for leaders moving from the role of problem solver to convener of the conversation.

William Ruckelshaus was the head of the US Environmental Protection Agency (EPA) in 1983 when he dealt with a case involving a copper plant near Tacoma Washington. The Asarco plant was the only one in the nation to use copper ore with a high content of arsenic. Ruckelshaus was expected to decide what to do about the plant; in particular, he was expected to determine what constituted an "ample margin of safety" in the plant's operation to protect public health.

The plant had spent 40 million dollars on equipment to reduce emissions even before the EPA got involved. They were ready to install new converters at a cost of four million dollars that would reduce deaths from cancer from four persons a year to one. But there was no technology available to reduce the emissions further.

If Asarco were to close the plant it would be a devastating economic blow to the region. The former mayor of near-by Ruston, told reporters, "I've worked in the plant all my life. So have my brothers, and so have my neighbors. We're not sick. This town was built around that plant. People came here looking for fire and smoke in the 1900's to find work. Now the government's complaining about that same smoke and trying to take our children's livelihood away."

Ruckelshaus refused publicly to decide on his own. He said, "For me to sit here in Washington and tell the people of Tacoma what is an acceptable risk would be at best arrogant and at worst inexcusable." He wanted to solicit the views of those that would be most affected by the EPA ruling. He decided the usual public hearings would be preceded by public workshops. The three public workshops held that August were controversial and packed with people, including a large number of smelter workers, union representatives, local citizen organizations, and environmental groups. After a formal presentation by EPA, the audience was divided into small groups to discuss the issues while the EPA staff circulated to answer questions.

The workshops and hearings surprised the EPA staff. As Ruckelshaus put it, local citizens had shown that they were "capable of understanding [the problem of the smelter] in its complexities and dealing with it and coming back to us with rather sensible suggestions." After the three EPA sponsored workshops, local groups began holding their own workshops.

It was not easy for Ruckelhaus to take such a radical position. Throughout the months of discussion Ruckelhaus was frequently vilified in the press. He was likened to Caesar who was asking the crowd to signal thumbs up or down on whether a defeated gladiator should die. The Sierra Club said it was the EPA’s job to protect the public health not to ask them how many people should die from cancer.

Through the EPA workshops and the community’s own workshops, local people began to see the situation in a new light. Rather than view it solely as a conflict between jobs and health, many people began to see a new possibility, the diversification of the local economy. That idea was obvious in retrospect, but at the time no one had seen it. By the time the plant closed in 1985, (due to falling copper prices) Ruckelhaus had still not made a decision, but Tacoma and Ruston had already begun the task of diversifying its economy. People had come to the early workshops displaying buttons labeled either "Jobs" or "Health." By the final workshops, people were sporting buttons that said "BOTH."

What a conversational architect does, as illustrated in these two stories, is to convene a conversation about the tough issues that the organization needs to face. The leader brings together those who are impacted by the issue and creates a setting in which he/she can draw on the collective knowledge to arrive at a workable solution.

In the post The Power of the Conversation Architect I discuss the critical tasks of the leader/convener in detail:
• Frame the conversation
• Identify who needs to be in the conversation
• Design highly interactive activities to fuel the conversation
• Use small groups as the unit of conversation
• Forge connections before discussing content
• Configure the physical space to serve the conversation

In Part II I will describe some of the conversational processes (e.g. Future Search, Appreciative Inquiry, Knowledge Café, etc.) that organizations have found useful in leveraging the collective knowledge of the whole system.

A Knowledge Management Conference that Actually Used KM Principles

Like many organizations, the US Army holds a Knowledge Management Conference each year. And like most conferences there are three kinds of activities, 1) keynote speakers, 2) practitioners telling about their KM successes, and 3) long breaks for people to network. The speaker format is also typical - three or four speakers in a row with a few minutes of Q & A at the end of each speaker’s time and then on to the next speaker.

I’ve had the honor of speaking at all of the Army KM conferences since they began in 2005, including the one just recently held. For at least the last three of those conferences, I have patiently explained to the conference organizers that having speaker after speaker is no way to run a KM conference because it violates everything we know about how people learn from one another.

This year, the 6th annual conference, was run by a very savvy group of folks at Strategic Knowledge Solutions (SKS). They heard what I had been saying year after year and asked me to help them design the conference so it would be more in keeping with the principles of KM.

The 6th annual US Army KM Conference was indeed different! On six occasions during the two and a half day conference, there was time for the audience to process what they had just heard – time to connect what the speaker said to their own knowledge and experience.

The way it worked was that after a major speech I or my colleague, Kent Greenes, would pose a provocative question based on the content. Then we would ask the members of the audience to talk to each other about that question. As I have written (see We learn when we talk) putting what we have just learned into our own words helps us build new connections in our minds and not surprisingly results in a huge jump in retention.

Each time that we took the stage we asked the audience to arrange themselves into a different configuration, (e.g. trios, pairs, etc.). And each time we gave them a different way to process what they had just learned from the speaker. Each time the room of 400 people was abuzz with noise. And each time we had great difficulty in getting them to end their group conversations – they had a lot to say to each other! When we did finally manage to get their attention back, we used the popcorn (see A Rant on Report Outs) approach to get a quick sense of what had been talked about in their small groups.
I preceded each discussion period with a brief four sentence lecture on what we were going to be doing and the principle behind the activity. This is a critical part of my practice, that is, explaining not only what I am asking audience members to do, but also the learning principle it is based upon. That practice enables audience members to use the activity in their own settings.

After the final period of time that had been set aside for processing, I asked the audience to reflect on how the experience had been for them, explaining that it had been an experiment and we wanted to know how to improve it. What I got back was an overwhelmingly positive response, surprising even the folks at SKS. We heard, “The small groups were the best part of the conference.” and “I liked how the discussion format changed each time.” The audience also had ideas for taking it even further, for example, “Some of the time put us in cross functional groups and at other times group us by similar job functions.” and simply, “Give us more time in the small groups.”

The audience’s positive response means that next year we can make the 7th Army KM conference based even more on KM principles. I can’t wait!

The Three Eras of Knowledge Management - Summary

I have posted lengthy descriptions of each of the three eras of knowledge management and here I have made a brief summary of all three. I have also made substantial changes to the third era in this post. If you would like to view each era in more detail just click on the heading of that era in this post.

Since the term “knowledge management” came into popular usage, there have been three significant changes in how organizations have thought about their knowledge. Each successive era has expanded the type of knowledge that organizations considered important without eliminating the need for and use of the previous type of knowledge.

Knowledge management began in the mid 1990’s. Before that time knowledge was typically considered the province of training and was thought of as an individual capability. However, in the mid-90s Peter Drucker began to write about “knowledge workers” and the “knowledge economy” and proposed the idea that knowledge was a critical organizational asset that was as important as capital or property.

Leveraging Explicit Knowledge
The initial idea of knowledge management was that an organization’s knowledge needed to be documented and then placed in a database where everyone could access it whenever they needed it - no longer would employees only be able to learn when attending a training class. Efforts were made to capture an organization’s best practices and lessons learned. Organizations spent large sums of money creating repositories and databases and employees were encouraged, sometimes even badgered, into contributing to them. The prevailing way of thinking about knowledge management was as a library or a warehouse with inputs and outputs, the more inputs the better.

By 2000 the limitations of content management were becoming evident:

1. Organizations found it difficult to get people to document their knowledge, and even more difficult to get others to make use of what had been documented and stored. Users found tools (checklists, steps in a process) useful as well as reusable documents (PP presentations, proposals) but lessons learned and best practices were largely ignored.

2. Organizations began to recognize that they had only been supporting explicit knowledge that could be written down - they had disregarded what might be the most critical knowledge to organizational success, tacit knowledge. Employees would not, and in fact could not, write all of their tacit knowledge down, first because the tacit knowledge that resides in an employee’s mind is immense and secondly because tacit knowledge is primarily useful in response to specific contexts, and much less useful when written as generalizations.

Leveraging Experiential Knowledge
Given the limitations of content management, by 2000 there began to be glimmers of a new perspective on knowledge within organizations. This new perspective held that:

1. much of an organization’s knowledge is in the heads of employees, with only a small percentage residing in documents, still recognizing that some explicit knowledge is needed and should be maintained

2. much of an organization’s knowledge is dynamic and rapidly changing so that what is “captured” is soon out-of-date

3. knowledge is essentially social and is developed and held by groups of people who engage together in a specific practice.

Etienne Wenger’s book Communities of Practice, that named and explained this phenomena, came out in 2000. Organizations began to support communities of peers, providing a way for them to ask for and receive knowledge on a just-in-time basis and thus keeping fast changing knowledge up-to-date. The Q&A that is ubiquitous in communities provided a way for employees to share their tacit knowledge in response to specific situations. By 2005 nearly every Fortune 500 Company had established Communities of Practice, acknowledging the growing understanding that knowledge is largely a property of groups of people.

My book Common Knowledge also came out in 2000 and talked about the knowledge management processes through which teams and projects could share their knowledge. Organizations put After Action Reviews into place to promote continuous learning in teams and projects so that what was being learned in the field could be continually updated. Expertise Locator systems helped employees draw on the knowledge of experts across an organization and Knowledge Harvesting helped to move knowledge from one team/project to another.

However by 2005 the limitations of even this expanded perspective on organizational knowledge began to be recognized:

1. It became obvious that knowledge was flowing primarily between peers and was largely limited to frontline employees. Senior and even middle management were supporters of knowledge management but not users of knowledge management processes.

2. Knowledge management was primarily dealing with existing knowledge, attempting to bring all units up to the best in class of the organization. But it did not help organizations create new knowledge or spur innovation.

3. The focus of knowledge management was tactical issues to the exclusion of strategic issues. Thus, although GM had an outstanding knowledge management program, that program did not address the difficult strategic issues that put GM into bankruptcy.

Leveraging Collective Knowledge
Although the first thinking about Leveraging Collective Knowledge began to appear around 2005, there are only a few leading edge organizations that have developed new practices for making use of their organization’s collective knowledge. Most organizations are still centered in the perspective of the second era and some, who have come late to knowledge management, are still struggling with getting good content management in place.

Those that are inventing processes for collective knowledge are finding ways to bring the whole organization to bear on strategic issues. Process like Knowledge Café’s, Appreciative Inquiry, and Search Conferences bring together all levels of the organization – the whole system in the room. The processes used to leverage collective knowledge are conversation based, alternating between small group and large group configurations. Even regularly held organizational meetings such as staff meetings, team, and project meetings in these organizations are turning to conversational forms to address their most difficult organizational issues. There is a growing understanding that in an age of increasingly complex organizational issues, leaders cannot be expected to have all the answers; rather the task of leaders becomes convening the conversations that can come up with new answers.

Leading edge organizations are taking advantage of Web 2.0 social media, building more user controlled platforms such as Wiki’s, blogs and social networking (e.g. Facebook) that bring with them greater organizational transparency and give rise to more diverse perspectives in the organizational conversation. The use of crowd sourcing, cognitive diversity, and predictive markets draw on a wider base of thinking, both internally and externally, that increases organizational innovation.

Over the three eras, each new set of knowledge management practices has been created in response to an ever-expanding understanding of 1) where knowledge lives within organizations and 2) what knowledge is important to organizational success. We can anticipate yet greater understanding as organizations move further into the third era.

Leveraging Collective Knowledge: NASA’s Constellation Program

Every organization has the problem of how to save the knowledge it has created, but after the cancelation of the Constellation program (CxP), NASA has that problem in spades. NASA has been working on Constellation, the human space flight program that was to replace the Shuttle, for 5 years now, at a cost of 9 Billion dollars – so saving that knowledge is
critical. On February 1, 2010 the Obama administration announced the cancelation of Constellation. The administration’s intention is to pay private companies to shuttle astronauts to and from the Space Station, while NASA is suppose to turn its attention to developing advanced technologies and demonstrations, including heavy-lift propulsion research research.

In the 5 years NASA has been working on Constellation it has accrued an enormous amount of new knowledge, for example, new habitats for astronauts living on the moon or Mars, more sophisticated space suits, and of course many new vehicles including, Ares I and V, the
Orion crew capsule, and Altair Lunar Lander. With the ending of the Constellation program, the engineers and scientists who created all that knowledge will now disperse to other NASA projects, or in many cases leave NASA altogether to work for other organizations. Without some direct intervention the “know how” accumulated over 5 years will be lost.

NASA learned its lesson about losing knowledge early in 1990. They experienced the sad recognition that much of the knowledge about how to build the Saturn V rocket that took the astronauts to the moon, had retired along with the engineers who had been encouraged to take early retirement. David Delong wrote about NASA' loss in Lost Knowledge, Oxford Press 2004.

Dave Lengyel, who heads NASA’s Risk and Knowledge Management Program is committed to ensuring that Constellation’s knowledge will not be lost. But the task of saving it is an enormous one. The program is spread across NASA’s ten centers from the Kennedy Space Center in Florida to the Jet Propulsion Lab on the West coast, each working on a different aspect of the program.

What Lengyel needed in order to meet his commitment, was a knowledge capture strategy that would provide direction over the next year as the program shuts down. The capture strategy needed to include:
* how to identify the most critical knowledge to be retained
* effective methodologies for capturing knowledge
* how the captured knowledge should be formatted so it would be most useful to other parts of NASA or to the commercial companies that might eventually use it
* effective knowledge transfer techniques for a wide range of explicit and tacit knowledge
* an estimate of the potential cost of capturing and storing five years of work
* a way to prepare engineers with the skills to effectively capture and then transfer what they have learned

Lengyel chose to address that need by leveraging NASA’s collective knowledge to create a knowledge capture strategy. He invited 35 people who had worked on the Constellation program to a two day meeting in Huntsville Alabama for the purpose of jointly developing the knowledge capture strategy. Before arriving each had been asked to construct a knowledge map that identified and prioritized the knowledge in their part of the project. The group was given only minimal instruction in how to construct such a map, which resulted in a great variety of formats and content.

The first afternoon of the meeting, following the usual introductions and welcomes, the group did a walk-around their maps, each of which had been blown up to poster size. During their walk-around, which was formatted much like a poster session, they examined each other’s maps and gained ideas about how they might revise or add to their own. They learned different ways to prioritize the knowledge, ways to illustrate levels of
expertise, the need for more than one POC, etc. The knowledge map session ended with high-spirited voting for the most useful map. Each attendee was given three dots to distribute among the maps based on what each found to be most useful. The many discussions about the merits of the maps led to a number of insights about the elements a knowledge map needed to display in order to make it a valuable tool for knowledge capture.

A second activity that same afternoon was a panel discussion of KM thought leaders, myself among them, who were asked to talk about best practices in knowledge capture and transfer from other organizations. The panel got the NASA engineers thinking about what did and did not work in other organizations and generated a lively conversation about what had worked at NASA in the past.

The next day was a day-long knowledge café, complete with all the trappings including red and white checkered tablecloths and menus. The menus had nothing to do with food, rather
each menu was a list of questions to be discussed at that table. We called the table facilitators, Sous Chefs, and the Maître D’ was Dave Lengyel himself, wearing a chef’s apron. The idea behind a knowledge café is for participants to have a conversation that is as open and as fervent as they would have at a sidewalk table on the streets of Paris. And that is exactly the kind of conversation that happened in Huntsville. Each table addressed a different issue related to knowledge capture and transfer, with participants moving from table to table until they had engaged in all of the topics. Each facilitator stayed at his own table to help jump-start each new conversation and to take notes on what was being said.

At the end of a very busy day a number of the engineers commented on the experience:

“I got 6-8 new things to put in my plan.”

“It was huge dose of reality and grounded me in the difficulty in doing this. It needs to be carefully planned out.”

“Very powerful. There were experts from different disciples and I sure took away more than I gave. The networking was good.”

“I was blown away by the diversity of ideas on frameworks.”

By the following morning the table facilitators were ready to formulate a draft plan based on their table discussion. A sophisticated framework for the knowledge capture emerged as
well as detailed steps in the process. Even in this final step everyone in the room was able to comment and improve upon what the facilitators offered using Think Tank. The Think Tank software enabled each person to use their own laptop to project their reactions and comments for everyone to see during the facilitators’ summaries.

This meeting was an excellent example of leveraging collective knowledge and illustrates the three elements that need to be in place to make use of the knowledge that resides in the minds of those doing the work, 1) joint sensemaking, 2) cognitive diversity, and 3) organizational transparency.

1. Joint Sensemaking.
Dave Lengyel could certainly have sat at his desk at NASA headquarters and drawn up a knowledge capture strategy, but that plan would not have been able to take into account the unique aspects of each of the Center’s needs. It would not have been as rich nor as comprehensive as the plan the group was able to develop together. Moreover, had Lengyel constructed it on his own, he would then have had the job of selling the plan to those who would implement it – never an easy task with a plan conceived at headquarters! The Knowledge Café gave everyone the opportunity to fully express their thinking and needs and to understand the needs and thinking of their colleagues.

For joint sensemaking to occur the leader, in this case Dave Lengyel, has to take responsibility of convening the conversation – and that task itself requires a number of skills.

Framing the conversation: Before the meeting was held Lengyel had thought through what the issues were that were facing this group. Early on he provided the rationale for the meeting, then developed the topics to be addressed, and just before the meeting posed questions for the facilitators to ask.

Identifying who needs to be in the conversation. In more traditional meetings people are invited on the basis of who needs to be informed. For joint sensemaking the criteria is different, it is who can inform the conversation. Lengyel convened people who had been working on the project and would therefore know the most about what needed to be captured. He also invited engineers from the space shuttle program who had already been engaged in knowledge capture for shuttle. Others that would be tangentially impacted by the plan were invited, e.g. a CxP information systems representative, the head of CxP records management and the head of CxP security. The meeting was by invitation only - no one was required to attend - so of course Lengyel had more people wanting to come than he had room for.

Designing high interaction activities: The meeting involved several different interactive processes; the walk-around and voting for the maps, the knowledge café that
generated content, and Think Tank to refine the plan. This was the first time Lengyel had used walk-arounds, but knowledge cafés were a process he had used several times before. Several of the participants had also been involved in other knowledge cafés.

Using small groups as the unit of conversation: Smaller groups create conversations with more depth, authenticity and rigor. The café groups in Huntsville ranged from 3-7 and were different each time the group moved from table to table, which greatly enhanced networking. One of the practices of joint sensemaking is to alternate between small group conversations, where the hard work of knowledge creation is accomplished, and large group meetings where the knowledge can be integrated. This alternation was practiced and effective at the Constellation meeting.

Forging connection before content: To work on difficult issue, participants need first to gain a sense of who others are, the skills they bring, the experience they represent, and the hopes they have. Round table introductions, as useful as they are, are never enough to build the kind of connections needed to do difficult work. The Constellation meeting started in the afternoon, with the panel and the knowledge maps, and then spent the first evening with a group dinner and drinks to that built the connections this group needed to do the hard work in the knowledge café the following day.

Configuring the physical space: One of the responsibilities of a convener is to make sure the space is conducive to the type of activity planned. As happens in many situations, the
meeting space in Huntsville turned out to be too small for the number of people who showed up. Lengyel quickly re-configured the physical space by rounding up enough patio tables so that several of the café sessions were held outside in true sidewalk café style.

2. Cognitive Diversity
Cognitive Diversity (as differentiated from identity diversity) increases the possibility that a group will generate new and more creative ideas. The inclusion of people from different disciplines provides a larger set of problem solving strategies and perspectives on an issue.

The Constellation group was naturally diverse coming, as they were, from many NASA Centers across the country. Even so their experience was limited to NASA’s culture and an engineering approach to issues. By bringing in KM thought leaders, myself and Larry Prusak, Lengyel provided the group with new ways of thinking about knowledge capture and transfer. Lengyel invited in several other sources of cognitive diversity:
• Ed Hoffman from NASA’s Project Management program, APPEL, who brought a social science perspective
• a representative from Lockheed (a likely recipient of Constellation knowledge)
• a Marshall Space Flight Center History Office representative
• a NASA Engineering and Safety Center representative
• John Adams from the Defense Acquisition University (DAU) who brought in-depth knowledge about how other government organizations have dealt with shut downs based the work DAU has been doing developing insight into smart shut downs. The cognitive diversity at the meeting made a difference. As one participant noted,

“It opened my mind. I had in my head what KM was and now I have a very different way of thinking about it.”

• Organizational Transparency
Organizational transparency is the willingness of an organization to be open both about its knowledge and its problems. It is, of course, not possible to leverage the collective knowledge of an organization unless the collective is fully apprised of the issues the organization is facing. Being a government agency funded by Congress, NASA had the advantage of all employees being fully apprised of the pending shut down of Constellation. So in this situation a certain level of transparency was built in.

In two additional ways the meeting in Huntsville demonstrated organizational transparency. First, as just described, the meeting made itself open to ideas from outside NASA. This meant that those of us from outside became privy to the problems and complaints voiced by the participants, but more importantly allowing us to offer a new way of thinking about many of the issues participants raised.

Secondly, organizational transparency was demonstrated by Lengyel himself who was willing to say to NASA seniors as well as to participants at the meeting, “I need to draw on the collective knowledge of the organization to address this issue.” The acknowledgement by leadership that “I don’t have all the answers” is the pre-requisite and the first step in leveraging collective knowledge.