We define knowledge management as a business activity with two
primary aspects:
* Treating the knowledge component of business activities as an
explicit concern of business reflected in strategy, policy, and
practice at all levels of the organization.
* Making a direct connection between an organization's
intellectual assets - both explicit (recorded) and tacit
(personal know- how) - and positive business results.
In practice, knowledge management often encompasses identifying
and mapping intellectual assets within the organization,
generating new knowledge for competitive advantage within the
organization, making vast amounts of corporate information
accessible, sharing of best practices, and technology that
enables all of the above - including groupware and intranets.
That covers a lot of ground. And it should, because applying
knowledge to work is integral to most business activities.
Knowledge management is hard to define precisely and simply. (The
definition also leapfrogs the task of defining "knowledge"
itself. We'll get to that later.) That's not surprising. How
would a nurse or doctor define "health care" succinctly? How
would a CEO describe "management"? How would a CFO describe
"compensation"? Each of those domains is complex, with many sub-
areas of specialization. Nevertheless, we know "health care" and
"management" when we see them, and we understand the major goals
and activities of those domains.
Business strategies related to knowledge management
As you explore other explanations of knowledge management - Bo
Newman's Knowledge Management Forum is a good starting point -
you'll detect connections with several well-known management
strategies, practices, and business issues, including
* Change management
* Best practices
* Risk management
* Benchmarking
A significant element of the business community also views
knowledge management as a natural extension of "business process
reengineering," a fact underscored by the recent announcement
that John Wiley's Business Change and Reengineering will become
Knowledge and Process Management in March, 1997. See
(http://www.mgmt.utoronto.ca/wensle/journal1.htm)
There is a common thread among these and many other recent
business strategies: A recognition that information and knowledge
are corporate assets, and that businesses need strategies,
policies, and tools to manage those assets.
The need to manage knowledge seems obvious, and discussions of
intellectual capital have proliferated, but few businesses have
acted on that understanding. Where companies have take action -
and a growing number are doing so - implementations of "knowledge
management" may range from technology-driven methods of
accessing, controlling, and delivering information to massive
efforts to change corporate culture.
Opinions about the paths, methods, and even the objectives of
knowledge management abound. Some efforts focus on enhancing
creativity - creating new knowledge value - while other programs
emphasize leveraging existing knowledge. (See below,
"Categorization of knowledge management approaches.")
What is "knowledge"?
Aren't we managing knowledge already? Well, no. In fact, most of
the time we're making a really ugly mess of managing information.
In practice, the terms information and knowledge are often used
interchangeably by business writers.
Let's choose a simple working definition and get on with it:
Knowledge has two basic definitions of interest. The first
pertains to a defined body of information. Depending on the
definition, the body of information might consist of facts,
opinions, ideas, theories, principles, and models (or other
frameworks). Clearly, other categories are possible, too. Subject
matter (e.g., chemistry, mathematics, etc.) is just one
possibility.
Knowledge also refers to a person's state of being with respect
to some body of information. These states include ignorance,
awareness, familiarity, understanding, facility, and so on.
Email from Fred Nickols, Executive Director - Strategic Planning
& Management, Educational Testing Service.
There are many thoughtful and thought-provoking definitions of
"knowledge" - including the important distinctions Gene Bellinger
et al. make in "Data, Information, Knowledge, and Wisdom".
Nevertheless, Nickols provides a good, sensible, functional
definition, and it is sufficient for our purposes.
Nickols' two kinds of knowledge parallel Michael Polanyi's often-
quoted distinction between explicit knowledge (sometimes referred
to as formal knowledge), which can be articulated in language and
transmitted among individuals, and tacit knowledge (also,
informal knowledge), personal knowledge rooted in individual
experience and involving personal belief, perspective, and
values. (Polanyi, Michael. The Tacit Dimension. London: Routledge
& Kegan Paul. See also Karl E. Sveiby's online description.
"Tacit Knowledge."
In traditional perceptions of the role of knowledge in business
organizations, tacit knowledge is often viewed as the real key to
getting things done and creating new value. Not explicit
knowledge. Thus we often encounter an emphasis on the "learning
organization" and other approaches that stress internalization of
information (through experience and action) and generation of new
knowledge through managed interaction.
In the opinion of the editors of Knowledge Praxis, quibbles about
fine distinctions in the meaning of knowledge are just not very
important. (See Rant #1: Thinking objectively about subjective
knowing) It doesn't matter whether a written procedure or a
subject matter expert provides a solution to a particular
problem, as long as a positive result is achieved. However,
observing how knowledge is acquired and how we can apply
knowledge - whether tacit or explicit - in order to achieve a
positive result that meets business requirements ... that's a
different and very important issue.
Why we need knowledge management now
Why do we need to manage knowledge? Ann Macintosh of the
Artificial Intelligence Applications Institute (University of
Edinburgh) has written a "Position Paper on Knowledge Asset
Management" that identifies some of the specific business
factors, including:
* Marketplaces are increasingly competitive and the rate of
innovation is rising.
* Reductions in staffing create a need to replace informal
knowledge with formal methods.
* Competitive pressures reduce the size of the work force that
holds valuable business knowledge.
* The amount of time available to experience and acquire
knowledge has diminished.
* Early retirements and increasing mobility of the work force
lead to loss of knowledge.
* There is a need to manage increasing complexity as small
operating companies are trans-national sourcing operations.
* Changes in strategic direction may result in the loss of
knowledge in a specific area.
To these paraphrases of Ms. Macintosh's observations we would
add:
* Most of our work is information based.
* Organizations compete on the basis of knowledge.
* Products and services are increasingly complex, endowing them
with a significant information component.
* The need for life-long learning is an inescapable reality.
In brief, knowledge and information have become the medium in
which business problems occur. As a result, managing knowledge
represents the primary opportunity for achieving substantial
savings, significant improvements in human performance, and
competitive advantage.
It's not just a Fortune 500 business problem. Small companies
need formal approaches to knowledge management even more, because
they don't have the market leverage, inertia, and resources that
big companies do. They have to be much more flexible, more
responsive, and more "right" (make better decisions) - because
even small mistakes can be fatal to them.
Roadblocks to adoption of knowledge management solutions
There have been many roadblocks to adoption of formal knowledge
management activities. In general, managing knowledge has been
perceived as an unmanageable kind of problem - an implicitly
human, individual activity - that was intractable with
traditional management methods and technology.
We tend to treat the activities of knowledge work as necessary,
but ill-defined, costs of human resources, and we treat the
explicit manifestations of knowledge work as forms of publishing
- as byproducts of "real" work.
As a result, the metrics associated with knowledge resources -
and our ability to manage those resources in meaningful ways -
have not become part of business infrastructure.
But it isn't necessary to throw up one's hands in despair. We do
know a lot about how people learn. We know more and more about
how organizations develop and use knowledge. The body of
literature about managing intellectual capital is growing. We
have new insights and solutions from a variety of domains and
disciplines that can be applied to making knowledge work
manageable and measurable. And computer technology - itself a
cause of the problem - can provide new tools to make it all work.
We don't need another "paradigm shift" (Please!), but we do have
to accept that the nature of business itself has changed, in at
least two important ways:
1. Knowledge work is fundamentally different in character from
physical labor.
2. The knowledge worker is almost completely immersed in a
computing environment. This new reality dramatically alters the
methods by which we must manage, learn, represent knowledge,
interact, solve problems, and act.
You can't solve the problems of Information Age business or gain
a competitive advantage simply by throwing more information and
people at the problems. And you can't solve knowledge-based
problems with approaches borrowed from the product-oriented,
print-based economy. Those solutions are reactive and
inappropriate.
Applying technology blindly to knowledge-related business
problems is a mistake, too, but What is knowledge...
the computerized business environment provides opportunities and
new methods for representing "knowledge" and leveraging its
value. It's not an issue of finding the right computer interface
- although that would help, too. We simply have not defined in a
rigorous, clear, widely accepted way the fundamental
characteristics of "knowledge" in the computing environment. (See
"Cooperative development of a classification of knowledge
management functions.")
A brief history of knowledge management
An overarching theory of knowledge management has yet to emerge,
perhaps because the practices associated with managing knowledge
have their roots in a variety of disciplines and domains. Special
thanks to Karl Wiig for supplying us with a pre-publication copy
of "Knowledge Management:Where Did It Come From and Where Will It
Go?" which will appear in The Journal of Expert Systems with
Applications. This section draws heavily on that work but
supplies only a small part of that value.
A number of management theorists have contributed to the
evolution of knowledge management, among them such notables as
Peter Drucker, Paul Strassmann, and Peter Senge in the United
States. Drucker and Strassmann have stressed the growing
importance of information and explicit knowledge as
organizational resources, and Senge has focused on the "learning
organization," a cultural dimension of managing knowledge. Chris
Argyris, Christoper Bartlett, and Dorothy Leonard-Barton of
Harvard Business School have examined various facets of managing
knowledge. In fact, Leonard-Barton's well-known case study of
Chaparral Steel, a company which has had an effective knowledge
management strategy in place since the mid-1970s, inspired the
research documented in her Wellsprings of Knowledge - Building
and Sustaining Sources of Innovation (Harvard Business School
Press, 1995).
Everett Rogers' work at Stanford in the diffusion of innovation
and Thomas Allen's research at MIT in information and technology
transfer, both of which date from the late 1970s, have also
contributed to our understanding of how knowledge is produced,
used, and diffused within organizations. By the mid-1980s, the
importance of knowledge (and its expression in professional
competence) as a competitive asset was apparent, even though
classical economic theory ignores (the value of) knowledge as an
asset and most organizations still lack strategies and methods
for managing it.
Recognition of the growing importance of organizational knowledge
was accompanied by concern over how to deal with exponential
increases in the amount of available knowledge and increasingly
complex products and processes. The computer technology that
contributed so heavily to superabundance of information started
to become part of the solution, in a variety of domains. Doug
Engelbart's Augment (for "augmenting human intelligence"), which
was introduced in 1978, was an early hypertext/groupware
application capable of interfacing with other applications and
systems. Rob Acksyn's and Don McCracken's Knowledge Management
System (KMS), an open distributed hypermedia tool, is another
notable example and one that predates the World Wide Web by a
decade.
The 1980s also saw the development of systems for managing
knowledge that relied on work done in artificial intelligence and
expert systems, giving us such concepts as "knowledge
acquisition," "knowledge engineering," "knowledge-base systems,
and computer-based ontologies.
The phrase "knowledge management" entered the lexicon in earnest.
To provide a technological base for managing knowledge, a
consortium of U.S. companies started the Initiative for Managing
Knowledge Assets in 1989. Knowledge management-related articles
began appearing in journals like Sloan Management Review,
Organizational Science, Harvard Business Review, and others, and
the first books on organizational learning and knowledge
management were published (for example, Senge's The Fifth
Discipline and Sakaiya's The Knowledge Value Revolution).
By 1990, a number of management consulting firms had begun in-
house knowledge management programs, and several well known U.S.,
European, and Japanese firms had instituted focused knowledge
management programs. Knowledge management was introduced in the
popular press in 1991, when Tom Stewart published "Brainpower" in
Fortune magazine. Perhaps the most widely read work to date is
Ikujiro Nonaka's and Hirotaka Takeuchi's The Knowledge-Creating
Company: How Japanese Companies Create the Dynamics of Innovation
(1995).
By the mid-1990s, knowledge management initiatives were
flourishing, thanks in part to the Internet. The International
Knowledge Management Network (IKMN), begun in Europe in 1989,
went online in 1994 and was soon joined by the U.S.-based
Knowledge Management Forum and other KM- related groups and
publications. The number of knowledge management conferences and
seminars is growing as organizations focus on managing and
leveraging explicit and tacit knowledge resources to achieve
competitive advantage. In 1994 the IKMN published the results of
a knowledge management survey conducted among European firms, and
the European Community began offering funding for KM-related
projects through the ESPRIT program in 1995.
Knowledge management, which appears to offer a highly desirable
alternative to failed TQM and business process re-engineering
initiatives, has become big business for such major international
consulting firms as Ernst & Young, Arthur Andersen, and Booz-
Allen & Hamilton. In addition, a number of professional
organizations interested in such related areas as benchmarking,
best practices, risk management, and change management are
exploring the relationship of knowledge management to their areas
of special expertise (for example, the APQC American Productivity
and Quality Council0 and ASIS American Society for Information
Science0).
Knowledge management: a cross-disciplinary domain
Knowledge management draws from a wide range of disciplines and
technologies.
* Cognitive science. Insights from how we learn and know will
certainly improve tools and techniques for gathering and
transferring knowledge.
* Expert systems, artificial intelligence and knowledge base
management systems (KBMS). AI and related technologies have
acquired an undeserved reputation of having failed to meet their
own - and the marketplace's - high expectations. In fact, these
technologies continue to be applied widely, and the lessons
practitioners have learned are directly applicable to knowledge
management.
* Computer-supported collaborative work (groupware). In Europe,
knowledge management is almost synonymous with groupware ... and
therefore with Lotus Notes. Sharing and collaboration are clearly
vital to organizational knowledge management - with or without
supporting technology.
* Library and information science. We take it for granted that
card catalogs in libraries will help us find the right book when
we need it. The body of research and practice in classification
and knowledge organization that makes libraries work will be even
more vital as we are inundated by information in business. Tools
for thesaurus construction and controlled vocabularies are
already helping us manage knowledge.
* Technical writing. Also under-appreciated - even sneered at -
as a professional activity, technical writing (often referred to
by its practitioners as technical communication) forms a body of
theory and practice that is directly relevant to effective
representation and transfer of knowledge.
* Document management. Originally concerned primarily with
managing the accessibility of images, document management has
moved on to making content accessible and re-usable at the
component level. Early recognition of the need to associate
"metainformation" with each document object prefigures document
management technology's growing role in knowledge management
activities.
* Decision support systems. According to Daniel J. Power,
"Researchers working on Decision Support Systems have brought
together insights from the fields of cognitive sciences,
management sciences, computer sciences, operations research, and
systems engineering in order to produce both computerised
artifacts for helping knowledge workers in their performance of
cognitive tasks, and to integrate such artifacts within the
decision-making processes of modern organisations." See Powers'
DSS Research Resources Home page.0 That already sounds a lot like
knowledge management, but in practice the emphasis has been on
quantitative analysis rather than qualitative analysis, and on
tools for managers rather than everyone in the organization.
* Semantic networks. Semantic networks are formed from ideas and
typed relationships among them - sort of "hypertext without the
content," but with far more systematic structure according to
meaning. Often applied in such arcane tasks as textual analysis,
semantic nets are now in use in mainstream professional
applications, including medicine, to represent domain knowledge
in an explicit way that can be shared.
* Relational and object databases. Although relational databases
are currently used primarily as tools for managing "structured"
data - and object-oriented databases are considered more
appropriate for "unstructured" content - we have only begun to
apply the models on which they are founded to representing and
managing knowledge resources.
* Simulation. Knowledge Management expert Karl-Erik Sveiby
suggests "simulation" as a component technology of knowledge
management, referring to "computer simulations, manual
simulations as well as role plays and micro arenas for testing
out skills." (Source: Email from Karl- Erik Sveiby, July 29,
1996)
* Organizational science. The science of managing organizations
increasingly deals with the need to manage knowledge - often
explicitly. It's not a surprise that the American Management
Association's APQC has sponsored major knowledge management
events.
That's only a partial list. Other technologies include: object-
oriented information modeling; electronic publishing technology,
hypertext, and the World Wide Web; help-desk technology; full-
text search and retrieval; and performance support systems.
Categorization of knowledge management approaches
The term "knowledge management" is now in widespread use, having
appeared in the titles of many new books about knowledge
management as a business strategy, as well as in articles in many
business publications, including The Wall Street Journal. There
are, of course, many ways to slice up the multi-faceted world of
knowledge management. However, it's often useful to categorize
them.
In a posting to the Knowledge Management Forum, Karl-Erik Sveiby
identified two "tracks" of knowledge management:
* Management of Information. To researchers in this track,
according to Sveiby, "... knowledge = Objects that can be
identified and handled in information systems."
* Management of People. For researchers and practitioners in this
field, knowledge consists of "... processes, a complex set of
dynamic skills, know-how, etc., that is constantly changing."
(From Sveiby, Karl-Erik, "What is knowledge management")
Sveiby's characterization is on target, but it may not capture
the full flavor of the important distinctions in approaches to
organizational knowledge management. At Knowledge Praxis, we have
adopted a three-part categorization: (1) mechanistic approaches,
(2) cultural/behavioristic approaches, and (3) systematic
approaches to knowledge management.
Mechanistic approaches to knowledge management
Mechanistic approaches to knowledge management are characterized
by the application of technology and resources to do more of the
same better. The main assumptions of the mechanistic approach
include:
* Better accessibility to information is a key, including
enhanced methods of access and reuse of documents (hypertext
linking, databases, full-text search, etc.)
* Networking technology in general (especially intranets), and
groupware in particular, will be key solutions.
* In general, technology and sheer volume of information will
make it work.
Assessment: Such approaches are relatively easy to implement for
corporate "political" reasons, because the technologies and
techniques - although sometimes advanced in particular areas -
are familiar and easily understood. There is a modicum of good
sense here, because enhanced access to corporate intellectual
assets is vital. But it's simply not clear whether access itself
will have a substantial impact on business performance,
especially as mountains of new information are placed on line.
Unless the knowledge management approach incorporates methods of
leveraging cumulative experience, the net result may not be
positive, and the impact of implementation may be no more
measurable than in traditional paper models.
Cultural/behavioristic approaches to knowledge management
Cultural/behavioristic approaches, with substantial roots in
process re-engineering and change management, tend to view the
"knowledge problem" as a management issue. Technology - though
ultimately essential for managing explicit knowledge resources -
is not the solution. These approaches tend to focus more on
innovation and creativity (the "learning organization") than on
leveraging existing explicit resources or making working
knowledge explicit.
Assumptions of cultural/behavioristic approaches often include:
* Organizational behaviors and culture need to be changed ...
dramatically. In our information-intensive environments,
organizations become dysfunctional relative to business
objectives.
* Organizational behaviors and culture can be changed, but
traditional technology and What is knowledge...
methods of attempting to solve the "knowledge problem" have
reached their limits of effectiveness. A "holistic" view is
required. Theories of behavior of large-scale systems are often
invoked.
* It's the processes that matter, not the technology.
* Nothing happens or changes unless a manager makes it happen.
Assessment: The cultural factors affecting organizational change
have almost certainly been undervalued, and
cultural/behavioristic implementations have shown some benefits.
But the cause-effect relationship between cultural strategy and
business benefits is not clear, because the "Hawthorne Effect"
may come into play, and because we still can't make dependable
predictions about systems as complex as knowledge-based business
organizations. Positive results achieved by
cultural/behavioristic strategies may not be sustainable,
measurable, cumulative, or replicable ... and employees
thoroughly "Dilbertized" by yet another management strategy may
roll their eyes. Time will tell.
Systematic approaches to knowledge management
Systematic approaches to knowledge management retain the
traditional faith in rational analysis of the knowledge problem:
the problem can be solved, but new thinking of many kinds is
required. Some basic assumptions:
* It's sustainable results that matter, not the processes or
technology ... or your definition of "knowledge."
* A resource cannot be managed unless it is modeled, and many
aspects of the organization's knowledge can be modeled as an
explicit resource.
* Solutions can be found in a variety of disciplines and
technologies, and traditional methods of analysis can be used to
re-examine the nature of knowledge work and to solve the
knowledge problem.
* Cultural issues are important, but they too must be evaluated
systematically. Employees may or may not have to be "changed,"
but policies and work practices must certainly be changed, and
technology can be applied successfully to business knowledge
problems themselves.
* Knowledge management has an important management component, but
it is not an activity or discipline that belongs exclusively to
managers.
Assessment: Unrepentant rationalists in the business world are
taking a systematic approach to solving the "knowledge problem."
You'll also find evidence of such approaches - as well as a less
formal use of the term systematic knowledge management -Karl
Wiig's Knowledge Research Institute Web site and Gene Bellinger's
Systems Thinking Web pages. Systematic approaches show the most
promise for positive cumulative impact, measurability, and
sustainability.
Conclusion
Where do we stand at the moment, and where do we go from here? We
conclude with a thought from Bo Newman, via email:
As attested to in numerous articles in the popular press,
knowledge management has already been embraced as a source of
solutions to the problems of today's business. Still it has not
been easy for this "science" to construct for itself that royal
road of self validation.
On the contrary, I believe that it is still, at least for the
majority of the practitioners and their customers, in the stage
of blind groping after its true aims and destination.
Enough said ... for the moment. Let's change the end of this
story.
