Content uploaded by Thomas H. Davenport
Author content
All content in this area was uploaded by Thomas H. Davenport on Aug 01, 2014
Content may be subject to copyright.
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
1 of 15 2/11/05 7:00 PM
Invitation
Send a Comment
Submit an Article
Subscribe
Working Knowledge: How Organizations
Manage What They Know
By Thomas H. Davenport and Lawrence Prusak.
In the end, the location of the new economy is not in the
technology, be it the microchip or the global telecommunications
network. It is in the human mind.
-- Alan Webber
What Do We Talk about When We Talk about Knowledge?
KNOWLEDGE is neither data nor information, though it is related
to both, and the differences between these terms are often a matter
of degree. We start with those more familiar terms both because
they are more familiar and because we can understand knowledge
best with reference to them. Confusion about what data,
information, and knowledge are -- how they differ, what those
words mean -- has resulted in enormous expenditures on
technology initiatives that rarely deliver what the firms spending
the money needed or thought they were getting. Often firms don't
understand what they need until they invest heavily in a system
that fails to provide it.
However basic it may sound, then, it is still important to emphasize
that data, information, and knowledge are not interchangeable
concepts. Organizational success and failure can often depend on
knowing which of them you need, which you have, and what you
can and can't do with each. Understanding what those three things
are and how you get from one to another is essential to doing
knowledge work successfully. So we believe it's best to begin with
a brief comparison of the three terms and the factors involved in
transforming data into information and information into
knowledge.
A Working Definition of Knowledge
A word of qualification before we proceed with our definitions.
We're aware that some researchers identify more than the three
entities of data, information, and knowledge -- going on, for
example, to describe wisdom, insight, resolve, action, and so forth.
Since we've noticed that firms have enough difficulty
distinguishing among three related concepts, however, we're not
inclined to address more. For practical purposes, we'll lump
higher-order concepts such as wisdom and insight into knowledge.
And things like "resolve" and "action," while desirably pointing to
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
2 of 15 2/11/05 7:00 PM
the need to do something with knowledge, we'd put into a
different category of "things you do with knowledge" rather than a
variation on knowledge itself. With that caution, let's proceed to
some definitions.
Data
Data is a set of discrete, objective facts about events. In an
organizational context, data is most usefully described as
structured records of transactions. When a customer goes to a gas
station and fills the tank of his car, that transaction can be partly
described by data: when he made the purchase; how many gallons
he bought; how much he paid. The data tells nothing about why
he went to that service station and not another one, and can't
predict how likely he is to come back. In and of themselves, such
facts say nothing about whether the service station is well or badly
run, whether it is failing or thriving. Peter Drucker once said that
information is "data endowed with relevance and purpose," which
of course suggests that data by itself has little relevance or
purpose.
Modern organizations usually store data in some sort of
technology system. It is entered into the system by departments
such as finance, accounting, and marketing. Until recently it has
been managed by central information systems departments that
respond to requests for data from management and other parts of
the company. The current trend is for data to be somewhat less
centralized and available on demand from desktop PCs, but the
basic structure of what it is and how we store and use it remains
the same.
Quantitatively, companies evaluate data management in terms of
cost, speed, and capacity: How much does it cost to capture or
retrieve a piece of data? How quickly can we get it into the system
or call it up? How much will the system hold? Qualitative
measurements are timeliness, relevance, and clarity: Do we have
access to it when we need it? Is it what we need? Can we make
sense out of it?
All organizations need data and some industries are heavily
dependent on it. Banks, insurance companies, utilities, and
government agencies such as the IRS and the Social Security
Administration are obvious examples. Record keeping is at the
heart of these "data cultures" and effective data management is
essential to their success. Efficiently keeping track of millions of
transactions is their business. But for many companies -- even
some data cultures -- more data is not always better than less.
Firms sometimes pile up data because it is factual and therefore
creates an illusion of scientific accuracy. Gather enough data, the
argument goes, and objectively correct decisions will
automatically suggest themselves. This is false on two counts. First,
too much data can make it harder to identify and make sense of
the data that matters. Second, and most fundamentally, there is no
inherent meaning in data. Data describes only a part of what
happened; it provides no judgment or interpretation and no
sustainable basis of action. While the raw material of decision
making may include data, it cannot tell you what to do. Data says
nothing about its own importance or irrelevance. But data is
important to organizations -- largely, of course, because it is
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
3 of 15 2/11/05 7:00 PM
essential raw material for the creation of information.
Information
Like many researchers who have studied information, we will
describe it as a message, usually in the form of a document or an
audible or visible communication. As with any message, it has a
sender and a receiver. Information is meant to change the way the
receiver perceives something, to have an impact on his judgment
and behavior. It must inform; it's data that makes a difference. The
word "inform" originally meant "to give shape to" and information
is meant to shape the person who gets it, to make some difference
in his outlook or insight. Strictly speaking, then, it follows that the
receiver, not the sender, decides whether the message he gets is
really information -- that is, if it truly informs him. A memo full
of unconnected ramblings may be considered "information" by the
writer but judged to be noise by the recipient. The only message it
may communicate successfully is an unintended one about the
quality of the sender's intelligence or judgment.
Information moves around organizations through hard and soft
networks. A hard network has a visible and definite infrastructure:
wires, delivery vans, satellite dishes, post offices, addresses,
electronic mailboxes. The messages these networks deliver include
e-mail, traditional or "snail" mail, delivery-service packages, and
Internet transmissions. A soft network is less formal and visible. It
is ad hoc. Someone's handing you a note or a copy of an article
marked "FYI" is an example of information transmission via soft
network.
Quantitative measures of information management tend to include
connectivity and transactions: How many e-mail accounts or Lotus
Notes users do we have? How many messages do we send in a
given period? Qualitative measures measure informativeness and
usefulness: Did the message give me some new insight? Does it
help me make sense of a situation and contribute to a decision or
the solution to a problem?
Unlike data, information has meaning -- the "relevance and
purpose" of Drucker's definition above. Not only does it
potentially shape the receiver, it has a shape: it is organized to
some purpose. Data becomes information when its creator adds
meaning. We transform data into information by adding value in
various ways. Let's consider several important methods, all
beginning with the letter C:
Contextualized: we know for what purpose the data was gathered
Categorized: we know the units of analysis or key components of
the data
Calculated: the data may have been analyzed mathematically or
statistically
Corrected: errors have been removed from the data
Condensed: the data may have been summarized in a more
concise form
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
4 of 15 2/11/05 7:00 PM
Note that computers can help to add these values and transform
data into information, but they can rarely help with context, and
humans must usually help with categorization, calculation, and
condensing. A problem we will deal with throughout this book is
the confusion of information -- or knowledge -- with the
technology that delivers it. From Marshall McLuhan's The Medium
Is the Message, with its assertion that television would bind
humanity into a global village and end world conflict, to recent
statements about the transforming power of the Internet, we have
heard that information technology will change not only how we
work but who we are. One important point we will make in this
book is that the medium is not the message, though it may
strongly affect the message. The thing delivered is more important
than the delivery vehicle. Having a telephone does not guarantee
or even encourage brilliant conversations; owning a
state-of-the-art CD player is pointless if you use it only to listen to
polkas played by a kazoo ensemble. In the early days of
television, many commentators said that the new medium would
raise the level of cultural and political discourse in the nation, a
prediction that clearly did not come true. The corollary for today's
managers is that having more information technology will not
necessarily improve the state of information.
Knowledge
Most people have an intuitive sense that knowledge is broader,
deeper, and richer than data or information. People speak of a
"knowledgeable individual," and mean someone with a thorough,
informed, and reliable grasp of a subject, someone both educated
and intelligent. They are unlikely to talk about a "knowledgeable"
or even a "knowledge-full" memo, handbook, or database, even
though these might be produced by knowledgeable individuals or
groups.
Since epistemologists spend their lives trying to understand what it
means to know something, we will not pretend to provide a
definitive account ourselves. What we offer is a working definition
of knowledge, a pragmatic description that helps us communicate
what we mean when we talk about knowledge in organizations.
Our definition expresses the characteristics that make knowledge
valuable and the characteristics -- often the same ones -- that make
it difficult to manage well:
Knowledge is a fluid mix of framed experience, values, contextual
information, and expert insight that provides a framework for
evaluating and incorporating new experiences and information. It
originates and is applied in the minds of knowers. In
organizations, it often becomes embedded not only in documents
or repositories but also in organizational routines, processes,
practices, and norms.
What this definition immediately makes clear is that knowledge is
not neat or simple. It is a mixture of various elements; it is fluid as
well as formally structured; it is intuitive and therefore hard to
capture in words or understand completely in logical terms.
Knowledge exists within people, part and parcel of human
complexity and unpredictability. Although we traditionally think
of assets as definable and "concrete," knowledge assets are much
harder to pin down. Just as an atomic particle can appear to be
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
5 of 15 2/11/05 7:00 PM
either a wave or a particle, depending on how scientists track it,
knowledge can be seen as both process and stock.
Knowledge derives from information as information derives from
data. If information is to become knowledge, humans must do
virtually all the work. This transformation happens through such C
words as:
Comparison: how does information about this situation compare
to other situations we have known?
Consequences: what implications does the information have for
decisions and actions?
Connections: how does this bit of knowledge relate to others?
Conversation: what do other people think about this information?
Clearly, these knowledge-creating activities take place within and
between humans. While we find data in records or transactions,
and information in messages, we obtain knowledge from
individuals or groups of knowers, or sometimes in organizational
routines. It is delivered through structured media such as books
and documents, and person-to-person contacts ranging from
conversations to apprenticeships.
Knowledge in Action
One of the reasons that we find knowledge valuable is that it is
close -- and closer than data or information -- to action.
Knowledge can and should be evaluated by the decisions or
actions to which it leads. Better knowledge can lead, for example,
to measurable efficiencies in product development and
production. We can use it to make wiser decisions about strategy,
competitors, customers, distribution channels, and product and
service life cycles. We'll describe the characteristics of
knowledge-intensive organizations later in this chapter and
throughout the book. Of course, since knowledge and decisions
usually reside in people's heads, it can be difficult to trace the path
between knowledge and action.
We've observed and analyzed over a hundred attempts to manage
knowledge in organizations. To the managers of most of them
we've posed the question, "How do you make the distinction
between data, information, and knowledge?" Many make no hard
distinction in practice, and most of these initiatives involve a
mixture of knowledge and information, if not some data as well.
Many pointed out that they just tried to add value to what they had
-- to move it up the scale from data toward knowledge.
Chrysler, for example, stores knowledge for new car development
in a series of repositories called "Engineering Books of
Knowledge." The goal of these "books," which are actually
computer files, is to be an "electronic memory" for the knowledge
gained by automobile platform teams. The manager of one such
"book" was given a series of crash test results for inclusion in the
repository. However, he classified the results as data and
encouraged the submitter to add some value. What was the context
of the results -- why were the crash tests performed? How about
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
6 of 15 2/11/05 7:00 PM
comparisons to the results of other models, previous years, and
competitors' cars? What consequences did the results suggest for
bumper or chassis redesign? It may be difficult to note the exact
points at which data becomes information or knowledge, but it's
easy to see how to move it up the chain.
Knowledge can also move down the value chain, returning to
information and data. The most common reason for what we call
"de-knowledging" is too much volume. As one Andersen
Consulting knowledge manager told us, "We've got so much
knowledge (not to mention a lot of data and information too) in
our Knowledge Xchange repository that our consultants can no
longer make sense of it. For many of them it has become data."
Aeschylus made a similar point clearly twenty-five centuries ago:
"Who knows useful things, not many things, is wise."
Because knowledge is such a slippery concept, it's worth reflecting
a bit on some of its key components, such as experience, truth,
judgment, and rules of thumb.
Experience
Knowledge develops over time, through experience that includes
what we absorb from courses, books, and mentors as well as
informal learning. Experience refers to what we have done and
what has happened to us in the past. "Experience" and "expert" are
related words, both derived from a Latin verb meaning "to put to
the test." Experts -- people with deep knowledge of a subject --
have been tested and trained by experience.
One of the prime benefits of experience is that it provides a
historical perspective from which to view and understand new
situations and events. Knowledge born of experience recognizes
familiar patterns and can make connections between what is
happening now and what happened then. The application of
experience in business may be as simple as an old hand's
identifying a down-turn in sales as a seasonal phenomenon and
therefore no cause for alarm. It may be as complex as a manager's
noticing subtle signs of the corporate complacency that led to
problems in the past, or a scientist's having a sense of which new
avenues of research will likely lead to useful results. These
experience-based insights are what firms pay premiums for; they
show why experience counts.
Ground Truth
Experience changes ideas about what should happen into
knowledge of what does happen. Knowledge has "ground truth,"
to borrow the phrase the U.S. Army's Center for Army Lessons
Learned (CALL) uses to describe the rich truths of real situations
experienced close up: on the ground, rather than from the heights
of theory or generalization.
For obvious reasons, effective knowledge transfer is a critical issue
for the army. Knowing what to expect and what to do in military
situations can be literally a life-or-death matter. Ground truth
means knowing what really works and what doesn't. Experts from
CALL take part in real military operations as learning observers
and disseminate the knowledge they gather through photos, video
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
7 of 15 2/11/05 7:00 PM
tapes, briefings, and simulations. Lessons learned in Somalia and
Rwanda in the early '90s, for example, were passed on to the
troops involved in the 1994 Haitian mission. The experiences of
the first units in Haiti that went from house to house looking for
weapons were also videotaped to provide guidance to those who
followed.
A key aspect of the army's success at knowledge management was
its "After Action Review" (AAR) program. This exercise involves
an examination of what was supposed to happen in a mission or
action, what actually happened, why there was a difference
between the two, and what can be learned from the disparities.
Enlisted soldiers and officers meet together in a climate of
openness, collaboration, and trust. Results from the AAR are
quickly incorporated into army "doctrine," or its formally
documented procedures, and training programs. The AAR
program was developed not as a knowledge management vehicle
but rather as a means to return to values of integrity and
accountability. These values had suffered considerably during the
Vietnam War, and army leaders adopted the AAR and an
orientation to ground truth to restore them -- initially in training
missions, and later for all types of missions. Over the past few
years the army has realized that it had a knowledge and learning
tool in the AAR.
Another breakthrough in the army's extensive knowledge
experience grew out of the reflections of a senior officer who, late
in his career, read Tolstoy's War and Peace. He was struck by the
difference between Tolstoy's depictions of Napoleonic War battles
and the way those battles were taught in classes at military
academies. How rich, true, and grounded were Tolstoy's
descriptions (he had actually interviewed veterans of those
campaigns) compared with the bloodless, rational abstractions
taught in the classroom! The gap between ground truth and
rational analysis prompted such innovations as CALL.
We could make a similar distinction between how business strategy
actually happens and how it is taught in business schools.
However, we believe that managers recognize the importance of
real-life knowledge or ground truth. This is suggested by some of
the language they use. They exchange "war stories" and talk about
"life in the trenches." In other words, they share the detail and
meaning of real experiences because they understand that
knowledge of the everyday, complex, often messy reality of work
is generally more valuable than theories about it.
Complexity
The importance of experience and ground truth in knowledge is
one indication of knowledge's ability to deal with complexity.
Knowledge is not a rigid structure that excludes what doesn't fit; it
can deal with complexity in a complex way. This is one essential
source of its value. Although it is tempting to look for simple
answers to complex problems and deal with uncertainties by
pretending they don't exist, knowing more usually leads to better
decisions than knowing less, even if the "less" seems clearer and
more definite. Certainty and clarity often come at the price of
ignoring essential factors. Being both certain and wrong is a
common occurrence. In Sensemaking in Organizations, Karl
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
8 of 15 2/11/05 7:00 PM
Weick observes that "it takes a complex sensing system to register
and regulate a complex object," and elsewhere he remarks:
The illusions of accuracy can be created if people avoid
comparison . . . , but in a dynamic, competitive, changing
environment, illusions of accuracy are short-lived, and they fall
apart without warning. Reliance on a single, uncontradicted data
source can give people a feeling of omniscience, but because those
data are flawed in unrecognized ways, they lead to nonadaptive
action.
Knowledge is aware of what it doesn't know. Many wise men and
women have pointed out that the more knowledgeable one
becomes, the more humble one feels about what one knows. Since
what you don't know can hurt you, this awareness is extremely
important. Recently, a genetic-engineering firm created a new
tomato that farmers could pick and ship later than current varieties
and that therefore would be more flavorful than the tomatoes
available in supermarkets. The firm's scientists had all the expertise
needed to develop the new tomato but didn't know enough about
farming to know that there were essential things they didn't know.
For instance, any farmer with experience growing tomatoes could
have told them that any given single variety does not do equally
well in all climates. Their new tomato was derived from only one
variety. It grew successfully in some areas but not in others, and
their scientific triumph was a commercial failure.
Judgment
Unlike data and information, knowledge contains judgment. Not
only can it judge new situations and information in light of what is
already known, it judges and refines itself in response to new
situations and information. Knowledge can be likened to a living
system, growing and changing as it interacts with the environment.
Of course, everyone has met "experts" whose knowledge seems to
consist of stock responses and who offer the same old answer to
any new question: every problem looks like a nail to a person who
has only a single conceptual hammer in his toolbox. We would
argue that the expertise of these experts ceases to be real
knowledge when it refuses to examine itself and evolve. It
becomes opinion or dogma instead.
Rules of Thumb and Intuition
Knowledge works through rules of thumb: flexible guides to
action that developed through trial and error and over long
experience and observation. Rules of thumb (or, in the language
of the artificial-intelligence community, heuristics) are shortcuts to
solutions to new problems that resemble problems previously
solved by experienced workers. Those with knowledge see known
patterns in new situations and can respond appropriately. They
don't have to build an answer from scratch every time. So
knowledge offers speed; it allows its possessors to deal with
situations quickly, even some very complex ones that would baffle
a novice.
Roger Schank, a computer scientist at Northwestern University,
calls these internalized responses "scripts." Like play scripts (or
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
9 of 15 2/11/05 7:00 PM
computer program codes), they are efficient guides to complex
situations. Scripts are patterns of internalized experience, routes
through a maze of alternatives, saving us the trouble of
consciously analyzing and choosing at every step along the way.
Scripts can be played so quickly that we may not even be aware of
them: We arrive at an answer intuitively, without knowing how we
got there. That does not mean the steps do not exist -- intuition is
not mystical. It means we have so thoroughly learned the steps that
they happen automatically, without conscious thought, and
therefore at great speed. Karl Weick calls intuition "compressed
expertise," a phrase that vividly suggests how knowledge works
and what it can do.
The skill of an experienced driver provides an example of this
kind of intuition. She knows how to drive, rapidly accomplishing a
series of complex actions without having to think about them, as a
beginner would. The veteran driver also develops an intuitive
sense of what to expect on the road. Hundreds of hours of driving
have led her to "know" that another driver is going to pull out of a
side street or change lanes without looking. Experience has made
her aware of minute signs that the beginning driver would almost
certainly miss and that may be too subtle to verbalize. Like an
experienced businessperson, she sizes up a situation quickly
without going through a definable process or even being able to
explain her "reasoning."
Values and Beliefs
It may seem odd to include values and beliefs in a discussion of
knowledge in organizations. Many people assume that
organizations are objective and neutral; their purpose is to create a
product or provide a service, and that goal may seem unrelated to
values. In fact, people's values and beliefs have a powerful impact
on organizational knowledge. Organizations are, after all, made up
of people whose values and beliefs inescapably influence their
thoughts and actions. The organizations themselves have histories,
derived from people's actions and words, that also express
corporate values and beliefs.
Values and beliefs are integral to knowledge, determining in large
part what the knower sees, absorbs, and concludes from his
observations. People with different values "see" different things in
the same situation and organize their knowledge by their values.
Someone who values the bustle of urban life may find energy and
variety in a crowded city street. Someone who prefers rural quiet
may see only chaos and danger in the same scene. A publishing
executive who values risk and change may see a new opportunity
in the same online technology that a competitor views as a threat
to traditionally successful print products.
Nonaka and Takeuchi say that "knowledge, unlike information, is
about beliefs and commitment." The power of knowledge to
organize, select, learn, and judge comes from values and beliefs as
much as, and probably more than, from information and logic.
Knowledge as a Corporate Asset
People in organizations have always sought, used, and valued
knowledge, at least implicitly. Companies hire for experience
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
10 of 15 2/11/05 7:00 PM
more often than for intelligence or education because they
understand the value of knowledge that has been developed and
proven over time. Managers making difficult decisions are much
more likely to go to people they respect and avail themselves of
their knowledge than they are to look for information in
databases. Studies have shown that managers get two-thirds of
their information and knowledge from face-to-face meetings or
phone conversations. Only one-third comes from documents.
Most people in organizations consult a few knowledgeable people
when they need expert advice on a particular subject. As we have
said, knowledge is what makes organizations go. Knowledge is not
new.
Explicitly recognizing knowledge as a corporate asset is new,
however, as is understanding the need to manage and invest it with
the same care paid to getting value from other, more tangible
assets. The need to make the most of organizational knowledge, to
get as much value as possible from it, is greater now than in the
past.
The Changing Global Economy
Fifty years ago, the United States accounted for about 53 percent
of the world GDP. The demand for American goods at home and
abroad was so great that almost any product could find a market.
Today, the U.S. share of the world GDP is approximately 18
percent. Although the "pie" is much bigger than it was, American
companies no longer dominate the world market. There is fierce
international competition for every marginal dollar of profit. A
rapidly globalizing economy unified by improved communication
and transportation gives consumers an unprecedented choice of
goods and services and an endless cavalcade of new and better
offerings from global companies.
In short, companies can no longer expect that the products and
practices that made them successful in the past will keep them
viable in the future. Pricing pressures leave no room for inefficient
production. The cycle time for developing new products and
getting them on the market is becoming more and more
compressed. Companies now require quality, value, service,
innovation, and speed to market for business success, and these
factors will be even more critical in the future.
Increasingly, companies will differentiate themselves on the basis
of what they know. A relevant variation on Sidney Winter's
definition of a business firm as "an organization that knows how to
do things" would define a business firm that thrives over the next
decade as "an organization that knows how to do new things well
and quickly."
In their search for new efficiencies, global corporations have
outsourced much of the labor of manufacturing to countries where
the cost of labor is still relatively low. Clearly, the
knowledge-based activities of developing products and processes
are becoming the primary internal functions of firms and the ones
with the greatest potential for providing competitive advantage.
Product and Service Convergence
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
11 of 15 2/11/05 7:00 PM
Increasingly, knowledge and related intangibles not only make
businesses go but are part or all of the "products" firms offer. Old
distinctions between manufactured objects, services, and ideas are
breaking down. Not surprisingly, distinctions between
manufacturing and service firms are disappearing too. Alan
Webber described the change in a 1993 article:
Not so long ago, observers predicted with confidence the arrival
of a "postindustrial" service economy, where the central role
played by manufacturing in the economy would be steadily
replaced by new service industries and service jobs. Now we know
that the real impact of the information economy is to explode the
distinction between manufacturing and services altogether.
Fortune magazine recognized the same trend in 1993, when it
replaced its separate Fortune 500 industrial-firm and service-firm
issues with a combined issue. The decision to make that change
resulted from an internal debate about whether Microsoft was an
"industrial" or a "service" firm, and furthermore, whether it
mattered. The editors saw that it was no longer meaningful or even
possible to decide which firms fit which category.
Software companies sell products that are essentially ideas --
intellectual property -- embodied in lines of code. We can classify
software as a service: a set of functions delivered in digital form.
It's no wonder that Microsoft works so diligently to hire smart
workers. The software business is a new kind of knowledge-based
industry, but even traditional manufacturing firms are increasingly
both users and sellers of knowledge. Once-traditional
manufacturing firms differentiate themselves from competitors by
offering "smart" products ranging from automatic breadmakers to
cars that sense driver habits and adjust to them. Xerox calls itself
"the document company," not "the copier/printer company." It
sells solutions to business problems, not just office machinery.
Ford focuses on "quality." IBM markets "industry-solution units."
3M calls itself a knowledge company, and Steelcase, the office
equipment firm, has placed full-page ads touting itself as selling
"knowledge." These self-definitions are not just market hype but a
genuine recognition of the type of value these firms need to offer
their customers.
These changes and pressures make knowledge vital to
organizations. As James Brian Quinn points out, the intangibles
that add value to most products and services are knowledge-based:
technical know-how, product design, marketing presentation,
understanding the customer, personal creativity, and innovation.
The powers of knowledge that we have described -- speed,
complexity, a sense of history and context, judgment, and
flexibility -- are precisely those needed in a rapidly changing,
increasingly competitive global economy.
A small but telling case in point: The NEC factory in Honjo,
Japan, has been replacing assembly-line robots with human
workers, because human flexibility and intelligence make them
more efficient at dealing with change. Assembling a new model of
mobile phone, humans reached target efficiency after making
8,000 units (compared with the 64,000 units robots needed) and
were 45 percent more productive than the machines after both
reached peak efficiency. The cost of a model change fell from
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
12 of 15 2/11/05 7:00 PM
$9.5 million to between $1 million and $2 million, a significant
savings given that NEC is making model changes every six
months rather than every two years, as in the past. Tomiaki
Mizukami, president of NEC's Saitama plant, says, "Before, we
ended up using people as robots. But now we must use their
intelligence. Using robots was good, but now we're discovering
that using people is actually faster." Even assembly-line work,
often considered merely mechanical, benefits from the experience,
skill, and adaptability of human expertise.
Similarly, firms that have replaced some accounts-payable
personnel with computers are finding that overpayments have
increased because automated systems don't catch errors that would
be obvious to experienced employees. Although the financial cost
of additional overpayments is in many cases more than offset by
the savings in salaries and benefits, the errors can cause strained
relationships between firms and suppliers. Again, the human
dimensions -- the knowledge dimensions -- of a supposedly
mechanical task become apparent when machines try to
accomplish them. Richard Loder, president of Loder Drew &
Associates, a payables consulting firm, comments, "Payable clerks
are blessed with intuition, memory recognition and the ability to
make educated guesses. Computers are dumb and dumber in these
areas."
Konosuke Matsushita, founder of Matsushita Electric, Ltd., has
said, "Business, we know, is now so complex and difficult, the
survival of firms so hazardous in an environment increasingly
unpredictable, competitive and fraught with danger, that their
continued existence depends on the day-to-day mobilization of
every ounce of intelligence." Managers around the world have
come to realize that they need to understand what they know how
to do well and take advantage of that knowledge as effectively as
possible.
Sustainable Competitive Advantage
Centuries ago, manufacturers and nations maintained commercial
supremacy by keeping material and processes secret. Guilds
protected their special knowledge; governments prohibited the
export of economically important skills. France, for instance,
made exporting lace-making expertise a capital crime: Anyone
caught teaching the skill to foreigners could be put to death.
Today, real trade secrets are a rarity. There are a few well-known
examples (like the formula for Coca-Cola) and a few specialized
ones (the Zildjian cymbal company, owned by the same family
since its origin in alchemical experiments centuries ago, still
guards the formula for the exact composition of the alloy used in
its cymbals). For the most part, though, it is virtually impossible to
prevent competitors from copying and even improving on new
products and production methods fairly quickly in an era
characterized by mobility, the free flow of ideas, reverse
engineering, and widely available technology.
Alan Webber, the editor of Fast Company magazine, has referred
to this phenomenon as the "self-canceling technological
advantage." "As technology transforms the logic of competition,"
he explains, "technology disappears as a sustainable source of
competitive advantage." Because essentially the same technology is
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
13 of 15 2/11/05 7:00 PM
available to everyone, it cannot provide a long-term edge to
anyone. A global marketplace for ideas has developed and there
are very few concepts and formulae that are not generally
available. Competitors can quickly duplicate most products and
services. When only Citibank and Chemical had automated teller
machines, they briefly had a significant advantage over their
competitors, offering a service that customers wanted and they
alone could provide. But ATMs soon became available throughout
the industry, and what had been a competitive advantage was
simply a baseline requirement for consumer-oriented banks. There
is no way to make the ATM or any other piece of technology a
trade secret for long -- even if you build it yourself, as Citibank
did.
The advantages of new products and efficiencies are more and
more difficult to sustain. VF, the company that sells Lee Jeans and
other apparel, has experienced 20 percent annual growth for five
years, thanks in part to technical innovations. These include an
electronic market response system that informs both the company's
shipping and manufacturing departments of every sale made
within hours. But Jerry Johnson, VF's chief financial officer, says,
"The half-life of innovation is getting shorter and shorter. A
couple of years ago we thought we had established a definitive
lead in service to our customers. Now it's become the industry
standard." Robert Stasey, the director of quality improvement for
Analog Devices, another growing company, expresses a similar
idea when he says that Analog "is basically a new product engine.
Life cycles are short and we want to obsolete our own products
before the competition does."
Knowledge, by contrast, can provide a sustainable advantage.
Eventually, competitors can almost always match the quality and
price of a market leader's current product or service. By the time
that happens, though, the knowledge-rich, knowledge-managing
company will have moved on to a new level of quality, creativity,
or efficiency. The knowledge advantage is sustainable because it
generates increasing returns and continuing advantages. Unlike
material assets,which decrease as they are used, knowledge assets
increase with use: Ideas breed new ideas, and shared knowledge
stays with the giver while it enriches the receiver. The potential for
new ideas arising from the stock of knowledge in any firm is
practically limitless -- particularly if the people in the firm are
given opportunities to think, to learn, and to talk with one another.
Paul Romer, who has worked at the leading edge of knowledge
economics, argues that only knowledge resources -- ideas -- have
unlimited potential for growth:
In a world with physical limits, it is discoveries of big ideas (for
example, how to make high-temperature superconductors)
together with the discovery of millions of little ideas (better ways
to sew a shirt), that make persistent economic growth possible.
Ideas are the instructions that let us combine limited physical
resources in arrangements that are ever more valuable.
And, he goes on to say, the number of potential combinations of
the steps that make up processes or the components of a product is
virtually inexhaustible.
Corporate Size and Knowledge Management
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
14 of 15 2/11/05 7:00 PM
At a time when firms need to "know what they know" and must
use that knowledge effectively, the size and geographic dispersion
of many of them make it especially difficult to locate existing
knowledge and get it to where it is needed. In a small, localized
company a manager probably knows who has experience in a
particular aspect of the business and can walk across the hall and
talk to him. Our studies have shown that the maximum size of an
organization in which people know one another well enough to
have a reliable grasp of collective organizational knowledge is two
hundred to three hundred people. The stock of knowledge in a
global enterprise with scattered offices and plants and a complex
mix of products and functions is vast, but that potential boon is
part of the problem. How do you find what you need? The mere
existence of knowledge somewhere in the organization is of little
benefit; it becomes a valuable corporate asset only if it is
accessible, and its value increases with the level of accessibility.
Managers in large corporations know how common it is to
reinvent the wheel, solving the same problems from scratch again
and again, duplicating effort because knowledge of already
developed solutions has not been shared within the company. This
was one of Chrysler's motivations in formulating its "Engineering
Books of Knowledge"; the company had forgotten some things it
had previously learned about building cars. If there is no system in
place to locate the most appropriate knowledge resources,
employees make do with what is most easily available. That
knowledge may be reasonably good, but in today's competitive
environment reasonably good is not good enough. Hence the
attempts by many companies, including one described below by
the worldwide oil firm BP, to apply technology to the problem of
global knowledge transfer.
Computer Networks and Knowledge Exchange
The low cost of computers and networks has created a potential
infrastructure for knowledge exchange and opened up important
knowledge management opportunities. The computational power
of computers has little relevance to knowledge work, but the
communication and storage capabilities of networked computers
make them knowledge enablers. Through e-mail, groupware, the
Internet, and intranets, computers and networks can point to
people with knowledge and connect people who need to share
knowledge over a distance. Desktop videoconferencing and
multimedia computing that transmits sound and video as well as
text make it possible to communicate some of the richness and
subtlety of one person's knowledge to another.
What we must remember is that this new information technology is
only the pipeline and storage system for knowledge exchange. It
does not create knowledge and cannot guarantee or even promote
knowledge generation or knowledge sharing in a corporate culture
that doesn't favor those activities. The proverbial phrase "if we
build it, they will come" does not apply to information
technology. The availability of Lotus Notes does not change a
knowledge-hoarding culture into a knowledge-sharing one, alas.
The medium turns out not to be the message and does not even
guarantee that there will be a message.
ACM: Ubiquity - Working Knowledge: How Organizations Manag... http://www.acm.org/ubiquity/book/t_davenport_1.html
15 of 15 2/11/05 7:00 PM
********
Reprinted by permission of Harvard Business School Press.
Excerpt of Working Knowledge: How Organizations Manage
What They Know by Thomas H. Davenport and Lawrence Prusak.
Copyright 2000 by the President and Fellows of Harvard College;
All Rights Reserved.
[Home] [About Ubiquity] [The Editors]
Ubiquity welcomes the submissions of articles from everyone interested in the
future of information technology. Everything published in Ubiquity is
copyrighted ©2000 by the ACM and the individual authors.