Making Knowledge Visible through Intranet Knowledge Maps: Concepts, Elements, Cases

Abstract

Establishes the conceptual and empirical basis for an innovative instrument of corporate knowledge management: the knowledge map. It begins by briefly outlining the rationale for knowledge mapping, i.e. providing a common context to access expertise and experience in large companies. It then conceptualizes five types of knowledge maps that can be used in managing organizational knowledge. They are: knowledge sources, assets, structures, applications and development maps. In order to illustrate these five types of maps, a series of examples is presented (from a multimedia agency, a consulting group, a market research firm and a medium-sized services company), and the advantages and disadvantages of the knowledge mapping technique for knowledge management are discussed. The paper concludes with a series of quality criteria for knowledge maps and proposes a five-step procedure to implement knowledge maps in a corporate intranet.

Full text

Making Knowledge Visible Through Intranet Knowledge Maps: Concepts,
Elements, Cases
Martin J. Eppler
mcm institute, University of St. Gallen, Switzerland
Martin.Eppler@unisg.ch
Abstract
This paper seeks to establish the conceptual and
empirical basis for an innovative instrument of corporate
knowledge management: the knowledge map. It begins by
briefly outlining the rationale for knowledge mapping,
i.e., providing a common context to access expertise and
experience in large companies. It then conceptualizes five
types of knowledge maps that can be used in managing
organizational knowledge. They are knowledge-sources, -
assets, -structures, -applications, and -development maps.
In order to illustrate these five types of maps, a series of
examples will be presented (from a multimedia agency, a
consulting group, a market research firm, and a medium-
sized services company) and the advantages and
disadvantages of the knowledge mapping technique for
knowledge management will be discussed. The paper
concludes with a series of quality criteria for knowledge
maps and proposes a five step procedure to implement
knowledge maps in a corporate intranet.
1. Introduction: the rationale for knowledge
mapping
A major weakness of the domain knowledge
management as it is discussed today in the business and
research community is its apparent lack of genuinely new
and effective instruments and methods to improve the way
individuals, teams, and organizations create, share and
apply knowledge (in the sense of know-how, know-what,
know-who, and know-why that individuals use to solve
problems; for this distinction see also [1]). In 1996,
Davenport, Jarvenpaa, and Beers argued that knowledge
work processes often lack adequate support by
information technology tools. This analysis still holds true
to a large degree today (see [2]). In contrast, this paper
views knowledge management not only as a new
perspective on information management problems, but as
a field that can provide new ways of improving
knowledge-intensive processes (such as market research,
consulting, or product development) by going beyond the
mere administration of electronic information and help
individuals make information actionable in new contexts,
connect it with previous experiences, identify relevant
experts, and enable organization-wide learning processes.
Knowledge maps, as will be shown below, serve exactly
this purpose.
Having said that this paper discusses a new, genuine
type of knowledge management tool, one must admit that
the terms knowledge map, knowledge cartography or
knowledge landscape are relatively new labels for an idea
that is rather old. This idea consists of representing our
vital environment in a graphic way to improve our actions
within this territory. The environment or territory in the
context of knowledge management is not geographic,
however, but intellectual. By constructing a visual
knowledge architecture, it should become possible to
examine the knowledge we depend upon on a global scale
and from different perspectives. Thus, a knowledge map
should assist an individual employee, a team or an
organizational unit in understanding and using the
knowledge available in an organizational setting. The
intellectual environment that is mapped through this tool
is mostly made up of referenced expertise, documented
experiences, and extracted and formalized processes or
procedures. It contains heuristic knowledge (know-how)
in the form of people (experts), processes (e.g., complex
workflows), and applications, rationales or experiences
(know-why) in the form of lessons learned or project
debriefings, and factual knowledge (know-what) in the
form of documents or database entries which in turn can
be linked to authors who can be asked for advice,
assistance or a clarification of their documented findings.
While the basic idea behind a knowledge map – to
construct a global architecture of a knowledge domain –
might be quite old, the application context, i.e., the
corporation, and the format, as an intranet hypertext
clickable map, are quite new. The reasons why knowledge
maps are now viewed as a necessary tool in a corporate
context are mainly the scope of (global) expertise that
resides within larger companies and the difficulty of
accessing this expertise through informal communication.
Today, these problems can be effectively resolved with
the help of knowledge maps. They not only make
expertise accessible through visual interfaces, but also
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 1
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

provide a common framework or context to which the
employees of a company can relate to in their search for
(or contribution of) relevant knowledge. As Fahey and
Prusak [3] stress in their analysis of common mistakes in
knowledge management projects, a prime goal of any
knowledge management initiative should be to create a
common context for the employees. Knowledge maps
provide this common context in an explicit common visual
model.
The technology that enables this kind of tool are
intranet-based software solutions such as Lotus
Development’s Raven, Autonomy’s Knowledge
Visualizer (this tool can be used to generate ad-hoc
knowledge source maps, see www.autonomy.com.),
Microsoft’s Visio or IBM’s KnowledgeX. All of these
software tools combine powerful visualization techniques
with database functionalities. Yet, while the technological
implementation of a knowledge map with the help of one
of these tools (or with simpler means such as
DHTML/JavaScript or XML) leads, as will be shown
below, to useful knowledge artefacts, the process of
creating a knowledge map is almost as important as the
final product itself. We will see that the technological
implementation is only half of the challenge of developing
and using knowledge maps in organizational knowledge
management. The other even more challenging task
consists of gathering the right reference information and
combining it in a framework that everybody can relate to.
Thus, the mapping process itself can already provide a
number of insights into the knowledge assets of a
company and its problems in allocating knowledge
effectively (a knowledge asset in this context is any
explicitly qualified source of knowledge that provides
potential benefits for the solution of problems relevant to
a company’s success). In our work with an intranet-based
knowledge map for a market research company, for
example, we conducted 35 interviews to gather the
relevant knowledge we needed to create a knowledge map
of the company’s methodological skills. These interviews
not only provided the necessary background information
for the knowledge map, but also revealed a lot about the
structural improvement areas of the company. Thus,
Galloway is right in concluding the following about
mapping (see [4], p. vii; for a similar conclusion see the
mapping examples provided in [5] or the argumentation in
[6]):
„Mapping is merely an enabler – a means to a more
important end. It is a vehicle for expressing and releasing
the knowledge, creativity, and energy that lies within
every group, regardless of its position or level within an
organization.“
In this paper, we would like to give examples of various
types of such knowledge enablers and describe the process
we followed to produce them. We do so by first outlining
the five types of knowledge maps we have found useful in
a corporate context. We then provide examples of such
maps and assess their advantages and disadvantages (if
used on a corporate intranet). Based on these insights we
outline a five step procedure to generate a high-quality
knowledge map.
2. The concept of a knowledge map
A knowledge map, as it is understood in this paper,
generally consists of two parts: a ground layer which
represents the context for the mapping, and the individual
elements that are mapped within this context. The ground
layer typically consists of the mutual context that all
employees can understand and relate to. Such a context
might be the visualized business model of a company
(e.g., the lending business model of a bank), the actual
product (e.g., a vehicle model in the case of a truck
company), the competency areas of a company (as in the
example of the multimedia company in section three), the
value chain of a firm (as in the example of the market
research group below), or a simple geographic map. The
elements which are mapped onto such a shared context
range from experts, project teams, or communities of
practice to more explicit and codified forms of knowledge
such as white papers or articles, patents, lessons learned
(e.g., after action reviews or project debriefings), events
(i.e., meeting protocols), databases or similar applications,
such as expert systems or simulations. Knowledge maps
group these elements to show their relationships,
locations, and qualities. In this paper, we refer to
knowledge maps as graphic directories of knowledge--
sources (i.e., experts), -assets (i.e., core competencies), -
structures (i.e., skill domains), -applications (i.e. specific
contexts in which knowledge has to be applied, such as a
process), or -development stages (phases of knowledge
development or learning paths). We focus on these five
types of maps as they answer the questions that came up
most frequently in our action research with six companies
over the course of two years. These questions were: how
do I find relevant knowledge, how can I judge its quality,
how can I make sense of its structure, and how do I go
about applying or developing it myself?
A different, more abstract set of map categories is used
by Huff in her anthology on the topic of mapping strategic
thought. In her mapping typology she focuses on cognitive
maps and distinguishes the following map types: text and
language analysis maps, classification maps, network
maps, conclusive maps, and schematic maps of cognitive
structures [5].
Many definitions of knowledge maps that we have
found in the descriptions of company projects or in
academic papers are similar, but less specific. Vail, for
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 2
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

example, defines a knowledge map as follows (see [7], p.
10):
„A knowledge map is a visual display of captured
information and relationships, which enables the efficient
communication and learning of knowledge by observers
with differing backgrounds at multiple levels of detail.
The individual items of knowledge included in such a map
can be text, stories, graphics, models, or numbers. [...]
„Knowledge mapping is defined as the process of
associating items of information or knowledge (preferably
visually) in such a way that the mapping itself also creates
additional knowledge.“
While this definition adequately describes the purpose
of knowledge mapping, it does not distinguish the various
types of knowledge maps that can be used in a corporate
context. Below we provide such a distinction with the
aforementioned five types of knowledge maps:
1. Knowledge source maps: They structure a population
of company experts along relevant search criteria, such as
their domains of expertise, proximity (for an example of
such a knowledge map, without hypertext links behind the
map however, see [8]), seniority, or regional distribution.
Knowledge source maps answer questions such as “where
can I find somebody who knows how to calculate a
company valuation” or “do we have people who have run
large e-commerce projects?”
2. Knowledge asset maps: This type of map visually
qualifies the existing stock of knowledge of an individual,
a team, a unit, or a whole organization. It provides a
simplified, graphic ‘balance sheet’ of a company’s
intellectual capital. Knowledge asset maps answer
questions such as “how many SAP-consultants do we
have, and how many SAP-projects have we completed?”
or “how many of our software engineers have been with
the company for more than five years?”
3. Knowledge structure maps: These maps outline the
global architecture of a knowledge domain and how its
parts relate to one another (for examples of this type of
map, see [9] or [10]). This type of knowledge map assists
the manager in comprehending and interpreting an expert
domain. Typical questions that can be answered by such a
map are “which are the skills needed to run a project, how
do they relate to one another, and what are the available
courses for every such skill?”
4. Knowledge application maps: These maps show
which type of knowledge has to be applied at a certain
process stage or in a specific business situation. Usually,
these maps also provide pointers to locate that specific
knowledge (documents, specialists, databases).
Knowledge application maps answer questions of people
who are involved in a knowledge-intensive process, such
as auditing, consulting, research, or product development.
They provide answers to questions such as “who do I talk
to if the market tests are inconclusive?” or “what are our
experiences in moving from a prototype to mass-
production?”
5. Knowledge development maps: These diagrams can
be used to depict the necessary stages to develop a certain
competence, either individually, as a team, or as an
organizational entity. These maps can serve as visualized
learning or development roadmaps which provide a
common corporate vision for organizational learning.
They answer questions such as “how do we achieve
business excellence for our unit?” or “how can we prepare
our unit (intellectually) for the entry into a new market?”
We have also used this type of map to visualize the
necessary steps to develop e-commerce competence (from
mere web designing skills, to community development
skills, to secure electronic contracting skills, to inter-
business networking skills).
Besides these five types of maps, one can also imagine
maps which combine some of the above types in one
single map. Typically, a knowledge application map and a
(partial) knowledge source map are combined into one
single image. This way, one can not only show what
knowledge is relevant at what project stage, but also how
to locate that knowledge.
There are numerous visualization techniques that can be
used to design such knowledge maps. They include, but
are not limited to, the following methods (see Table 1).
Table 1: Mapping techniques for the knowledge
map context
Simple Mapping Techniques Complex Mapping
Techniques
Mind mapping (T.Buzan) Concept maps (Novak)
Clustering Cause maps, see [5]
Matrices or portfolio diagrams Concentric circles or Venn
diagrams
Fishbone graphs Metaphoric maps (e.g., a
house, a balance, a compass,
or a park)
Cartesian and polar co-ordinate
systems
Process charts or flow charts
Pyramids Spider web graphs
Hierarchic trees Decision trees
Geographic maps 3D-environments (e.g.,
globes, landscapes)
The use of one of these techniques depends upon two
factors: first, the requirements of the conceptual
framework that provides the base layer for the knowledge
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 3
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

map (flow charts are feasible techniques for knowledge
application maps, while concept maps are more apt for
knowledge structure maps) and, second, the technological
infrastructure and software that is available to implement
a clickable map (there are numerous software packages
that support hypertext mind-maps, concept maps or
interactive flow charts, yet very few products that support
more complex forms of knowledge maps such as
metaphoric maps or 3D-environments).
In the following section we provide a number of
examples of how these visualization techniques can be
used to design knowledge maps of the five mentioned
types.
3. Intranet knowledge map examples
The following series of knowledge maps is a result of
our action research with six partner companies over the
course of two years (1998-2000).
3.1. Knowledge source map
In principle, a knowledge map can be implemented on
any technological platform that allows to combine
visualization engines with hypertext capability (i.e.,
linking sensitive, visual zones to other data points) and
database technology. Nevertheless, we have found the
intranet the most accessible medium to host knowledge
map applications. This is especially true since many
intranet users are familiar with clickable maps or touch-
sensitive visual zones from the Internet. Hence, the user-
acceptance of a knowledge map in a browser interface is
likely to be higher than in a proprietary application. The
first example of an intranet-based knowledge map is taken
from a multimedia company which mainly works in the
development of web-sites, CD-ROMs, and stand-alone
multimedia terminals. One of the company’s problems is
the adequate staffing of such projects with the right
experts.
Thus, a knowledge map was designed to help project
leaders and human resources professionals to assess the
current state of experts available within the company at its
three main locations (New York, Berlin, Basle).
The map divides the company’s competence into five
areas, namely graphic design, animation (i.e., shockwave
programming), database design, project management, and
technology know-how (i.e. server administration). The
people within the company apply these skills to the
aforementioned three areas of CD-ROM production,
stand-alone systems, and web development (the concentric
circles in the map below).
The partial view of the actual map below reveals that
the most expertise resides in the company’s headquarter in
Berlin (color-coded accordingly), especially for the
domain of web-site development. It also reveals that there
are very few project management or animation specialists
within the company. The map reflects the shift in activities
away from CD-ROM production to web development
where most experts in the map below apply their skills.
The technology behind this map is quite simple and
consists of three elements: the actual knowledge map
which is drawn in Microsoft Visio and linked to an Access
database that holds the expert profiles (expert area,
application area, location, phone number, etc.) and a CGI-
script that provides a browser interface for the map. The
experts update their profiles via a form which they can
access through their browser. This form is directly linked
to the Access database (which in turn is referenced in the
Visio chart).
While this particular map is still in the testing phase, an
identical application with the same architecture and map
design has already been used for several months in a
semiconductor company where it is also used to map
specialized communities of practice.
WebCD-
ROM
Stand-
Alone-
Systems
AlexMü ller
Ueli Sigg
Technicians
StefanWerd
Karl Toner
Graphic
Design
Ina Roehl
JuliaVenn
Holger Stier
MariaGal ata
EvaRohner
Database
ThomasSchmid
UteLe mp
MayaSenn
AnneWeick
Marion Pressl
MichaelGross
SteffiSieg er
PatrickAuer
MartinSik
MarkO tt
Project
Management
JosefGon er
Christian Teiler
Uli Rubner
Animation
MaxHi tz
Diane Strong
= Berlin
= Basel
= N.Y.
= mobile
Figure 1 : A knowledge source map for a multimedia company (excerpt)
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 4
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

3.2. Knowledge asset map
Knowledge asset maps provide a visual balance sheet
of a company’s capabilities. These capabilities can be
represented in the form of a core competency tree or as a
visual directory of individual or aggregated skills of the
work force. For a depiction of a core competence tree see
[11]. Based on such a map, a company can outline its
capability profile bottom-up instead of top-down (as in a
core competence tree). The following simple map
provides an overview of a consulting team in terms of the
competencies of its members. Large blocks represent
expert knowledge on the particular topic, small blocks
refers to basic knowledge in a particular domain. The
domains in which these consultants work are IT
(information technology), strategy, mergers and
acquisitions (M&A), accounting and marketing.
The knowledge asset map illustrates various properties
of the know-how constellation of the consulting staff. The
map reveals, for example, that Andi Ehrler is a central
asset for the company in terms of his skills (he is also the
practice leader for accounting as the shaded block in that
column indicates). He has substantial experience in all
five sectors of the consulting activities. The map also
reveals that there is a general lack of specialized
knowledge in the domain of mergers and acquisitions.
Thus, it might be advantageous to drop this consulting
service in the future. The map furthermore reveals training
and personnel development needs. It shows clearly that
Carl Brenner has not yet been able to gather an expert
status in any of the five domains. The future staffing
decisions need to take this fact into account.
Such a knowledge asset map can be valuable to plan
the allocation of staff members or their training needs or
assess the overall situation of a company’s intellectual
assets. By clicking on a name, further information on the
consultant is provided (such as e-mail, location, special
interests, etc.). By clicking on a block, the map reveals the
projects and courses that the consultant has completed in
the particular field.
A similar version of this map has been implemented
two years ago in a telecom consulting company.
According to the head of operations, it has proved
especially useful for the planning of training activities and
for emergency cases where experts have to be identified
quickly.
Consultants IT Strategy M&A Accounting Marketing
Tinner, Jeff
Borer, André
Brenner, Carl
Deller, Max
Ehrler, Andi
Gross, Peter
. . .
Figure 2: A knowledge asset map of a consulting company (excerpt)
3.3 Knowledge structure map
As mentioned earlier, knowledge structure maps divide
a skill domain into logical blocks. The map below shows
that a web publisher requires three levels of know-how.
First, basic editing knowledge is needed, which consists
of layout and sequencing skills, as well as the content
selection for an Internet site, plus an adequate vocabulary
and writing style for its communication. Secondly, the
publisher or editor requires knowledge on how to design
the interactive environment for the pages he or she
generated. Lastly, the web designer needs to know the
relevant HTML or Java codes (Web programming
languages) to make the site look attractive and implement
adequate security measures.
HTML/Java Programming (Code Level)
Navigation Design (Link Level)
Page Components (Content Level)
Layout
Skills
Sequencing
Skills
Selection
Skills
Language
Skills
Web Content
Knowledge
Figure 3: A knowledge structure map
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 5
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

Other examples of knowledge structure maps can be
found in product development (where the various
engineering, marketing, and management capabilities are
mapped around a product or process), project
management (where the various skills that are required to
manage a project are mapped in a framework that shows
which skills serve what purpose in the course of a
project), or financial engineering (where the inter-
relationship of insurance skills and banking skills are
mapped out).
3.4 Knowledge application map
The knowledge application map is probably the most
frequently used knowledge mapping format today. It
outlines which knowledge is required at a certain process
step, e.g., in the value chain of a company. In the example
below, the value chain of market research company is
divided into four processes: acquiring or generating data,
transforming it into information by analyzing it,
administrating and archiving this information, and
transforming it into knowledge by educating its clients
about its findings.
In every process step, various IT-based tools and
conceptual methods can be applied. If an employee is
interested in one of these (approx. seventy) instruments,
he can simply click at the name of a tool or method and a
short description of the tool, its application context,
functioning and contact person (as well as a rating of the
tool or method by the corresponding specialist) will
appear on the screen.
This is the actual map (with the exception of the terms
translated into English) that is currently installed in the
knowledge portal of the company’s intranet. Underlined
methods or tools are also sold as products to clients.
Every term in the map is linked to an explication- and
reference-page for the corresponding tool. This page
consists of three columns. The first column outlines the
tools title, the second column describes its application
context, functioning, experts and their rating of the tool,
the third column provides links to related tools or to
relevant Internet sites.
METHOD
Informationen
TOOL
EEP
Vorverkaufs-Kunden-
Workshop
Added Value
Trend-W orkshop
Interaktive Präsentationen
Präsentation ak tuell
Jour Fixe
Prosumentenclub Offertenvorlage
Support Programm
Team Manager
Kundissimo-Veranstaltungen
KuZu
Aktive Vollreferenz
HLT
Metaplan
Stakeholder Analyse
Massnahmenpl an
Stimmungsbarometer
WIDAS
AMOS SPSS
GfK Optimizer
SUCCESS
Pharma CD
MIC-Pharma Windows
Data View
IMSView
COSMO
Pharma Trend
Cognos PowerPlay
Rücklaufstatis tik nach
Themen gegliedert
Tachistos kop
Anal yzer
PowerView
Collage Technik
TC2000
Aktionsauswertung
Excel Express Wizard
Artikelliste
Apollo
Trendbarometer
Delphi-Methode
Tagebuch Methode
Online Panel
Telefonische
Befragung (CATI)
Stichprobenverfahren
TED/Moni tor
SAP
QforYou
Laufwerk IHADOCS
Intranet
Scheduler
Laufwerk IMSDOCS
Panelmanager
Geschäftspartner-
informationss ystem
SDI
Knowledge
Information
Data
generate/
acquire
analyze/
condense
present/ get
feedback
administrate/
organize
Radiocontrol
Kundenkreuz
Chancensonne
Verkäuferkreuz
Problemspinne
Kunden-Scores
Kundenübergabekonzept
Inmarkt
Statistische Methoden
und Verfahren
CAPI
Quantum
KonsumentenJury
Mystery Methoden
Visualisierungs-
techniken (Profil-
chart u.a.)
Figure 4: A knowledge application map of a market research company
3.5. Knowledge development map
The last type of knowledge map can be used to
visualize the necessary steps to develop a certain type of
competence. The relevant knowledge is mapped at various
(sequential) levels. By clicking on a certain level of
knowledge, the map reveals further details and describes
the necessary activities to acquire or develop that
competence. The example on the next page is taken from
an e-business context.
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 6
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

The map can be used to illustrate the necessary steps to
develop competence in e-commerce passing through
various stages from a mere web-presence that only
provides on-line information, to more complex sites that
offer community functionalities (e.g., forums, mailing
lists, polls, etc.) or full scale e-business sites that include
payment services. The map in its current form is not
hypertext based yet. Hence, the mapped steps cannot be
“zoomed in” on the corporate intranet.
3.6. Purposes of the maps
In summary, these examples illustrate that knowledge
maps can serve the following five purposes:
1. They increase the visibility of knowledge sources and
hence facilitate and accelerate the process of locating
relevant expertise or experience.
2. They improve the evaluation of intellectual assets (and
liabilities) in a corporation.
3. They assist employees in interpreting and evaluating
knowledge domains.
4. They connect processes with knowledge sources (and
thus go beyond the mere documentation of a process as
found in most quality manuals).
5. They sketch the necessary steps for knowledge
development in a certain area.
Having outlined the various functions of knowledge
maps, we will now examine their possible disadvantages
and advantages in more detail before outlining a five step
procedure to implement a knowledge map.
Figure 5: A knowledge development map for e-business competence
4. Evaluation of knowledge mapping:
advantages and disadvantages
Knowledge maps are one possible way to improve the
knowledge transfer and utilization in a company (other
ways include yellow pages type expert directories,
communities of practice, or knowledge fairs). Associated
with this way are various advantages and disadvantages
that one has to take into account in a knowledge mapping
project.
4.1. Advantages of knowledge maps
The general advantages of knowledge maps should have
become apparent by now. First, they render corporate
knowledge assets visible for all employees that have
access to the Intranet. Thus, they provide a systematic
context for the retrieval of reference information. Second,
they can connect experts with each other or help novices
or rookies identify experts quickly. As a consequence,
knowledge maps can speed up the information seeking
process and facilitate systematic knowledge development
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 7
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

since they connect insights with tasks and problems.
Another central advantage of knowledge maps is their
potential to make implicit knowledge explicit through the
use of visual metaphors and symbols. For illustrations of
this advantage see [12]. From a marketing point of view,
knowledge maps can be used to emphasize and sell the
competencies of a company to external groups (e.g.,
potential clients, potential employees, shareholders,
analysts etc.).
In summary one can say that knowledge maps can help
employees and clients remember, comprehend, and relate
knowledge domains through the insightful visualization
and aggregation of information about the company’s
experiences, skills, or intellectual resources in general. To
clarify this advantage, which is a direct result of the
visualization techniques used in knowledge mapping, we
can rely on the following explanation by hypertext
cartography expert Michael P. Peterson:
“Firstly, a diagram offers a synoptic, global
representation of structure or a process and this
contributes to the globality and the immediacy of
understanding. Secondly, a diagram is an ideal tool for
bridging between a conceptual interpretation and the
practical expression of a certain reality. A diagram is a
synthesis between these two apparently opposed types of
representations – the symbolic and the iconic. Diagrams
are not, generally, the direct image of a certain reality. It is
the figural expression of an already elaborated conceptual
structure, like any other symbolic system” (see [13], p.
34).
Although a knowledge map uses these two features of a
diagram – global understanding and conceptual
representation of reality – it is more than just a
diagrammatic representation. A knowledge map is more
than a simple diagram because it offers more dimensions
and richer semantics than a diagram. Specifically it offers
more elements that are relevant to solve a problem than a
diagram, such as time, location, quality levels, re-
lationships, and time aspects. Unlike a diagram it is not
inert and finished, but interactive and expandable.
In addition knowledge maps make extensive use of
metaphors (such as mountains, layers, routes, or
buildings). Finally, a map also shows pathways and
options and can relate to other maps (e.g., via embedded
hypertext links).
A knowledge map is therefore more similar to a
geographic map than a diagram. It answers the same basic
four questions that a geographic map seeks to answer,
namely:
1. Where am I, what is my context (environment)?
2. Where can I go, what are my options?
3. How do I get there quickly, and in the most direct way
possible?
4. What does it take to get there; what are the required
resources?
In consequence, a knowledge map provides systematic
orientation in the intellectual territory of a company and
helps to find directions, assess situations, or plan
resources.
4.2. Disadvantages of knowledge maps
As geographic maps, which still many people find
difficult to comprehend or use, a knowledge map also has
certain disadvantages. The disadvantages associated with
this method are summarized in table 2:
Table 2: Disadvantages of knowledge maps
Disadvantages for map
users
Disadvantages for map
designers
• the potential harmful
effects if the map is
seen by illegitimate
users (such as head
hunters or competitors)
• the danger of
misinterpretation
• the fixation or
‘reification’ of one
frame of reference (i.e.,
the layout of the
knowledge map)
• the danger of
information overload if
the map represents too
many elements or
dimensions of a
knowledge area
• the danger of using an
outdated map
• the commitment to one
scheme of order and
neglect of other
perspectives
• the difficult depiction
of dynamic processes
• the relatively high costs
for production and up-
dating
• the missing quantifi-
cation of inter-
dependencies
• the reduction of
complex structures to
graphic symbols
• the difficult and time-
consuming task of
ergonomic visuali-
zation
These numerous potential disadvantages have to be
weighed against the benefits that result from establishing a
knowledge map. In the following section, we propose an
implementation process and accompanying quality criteria
to avoid this extensive list of disadvantages.
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 8
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

5. Implementation and Quality Control
To construct and design a knowledge map that avoids
the aforementioned disadvantages, the following five steps
need to be sequentially completed:
1. Identify knowledge intensive processes, problems,
or issues within the organization. The resulting map
should focus on improving such knowledge intensive
areas. This step typically involves a screening of a
company’s value chain or main processes and various
interviews with key employees (involved in knowledge-
intensive business activities).
2. Deduce the relevant knowledge sources, assets, or
elements from the above process or problem. The question
which needs be answered at this point is: „In order to
manage the process or area well, what expertise and what
experience is needed or helpful, and where and how can
one access that knowledge?“
3. Codify these elements in a way that makes them
more accessible to the organization as a whole. Build
categories of expertise that are relevant to the process or
area identified in step one. If the process is, for example,
project management, possible categories might be experts
on project planning, controlling, project documentation,
or experts in IT-support for project management.
4. Integrate this codified reference information (i.e., the
different types of project management specialists or
resources) on expertise or documents into a visual
interface that allows the user to navigate or search
visually. Connect this navigation system to the process or
working environment itself (integrate it into the workflow
of the process or the homepage of an organizational unit).
This step involves the actual design and implementation
of the knowledge map. Here, a specific visualization
technique has to be chosen that best fits the objective of
the map. Knowledge application maps, for example, are
usually best visualized with process flow maps or decision
trees.
5. Provide means of updating the knowledge map. A
knowledge map is only as good as the links it provides. If
these links are outdated or obsolete, the map is useless.
Therefore, a map needs to be continuously updated by the
‘map maker’ or the people who are represented in it. This
step may involve designing an automatic workflow that
regularly asks experts to update their position in a
knowledge map.
These five steps provide a generic sequence of
activities to establish a knowledge map. In all of these
steps, one has to be aware of the quality of the final map.
In our experience in implementing three such maps, the
last step is often the most difficult and crucial one in order
to assure the quality of a knowledge map. Hence, it makes
sense to define a set of quality criteria for the resulting
map at the very beginning of the mapping process in order
to prevent quality issues in step five. Vail distinguishes
the following quality criteria for knowledge maps:
 participative: the mapping team creates the map
interactively and involves as many employees as possible,
 shared: the map represents a truly shared model that all
knowledge workers can relate to,
 synergistic: the experts all contribute their different
expertise to the map, in order to generate one logical and
comprehensive picture,
 systemic: the map’s elements can be combined logically
to an integrated whole,
 simple: the map can be overlooked at one glance,
 visual: the map uses a visual framework that is made up
of iconic elements,
 information rich: the map is informative in the sense
that it aggregates a great amount of noteworthy references
that help in the problem solving process. See [7], p. 14.
These criteria already provide some guidance in
gathering the elements and designing the framework for a
knowledge map. When completed, we suggest that the
map is again reviewed with the following quality criteria
(in the sense of a concise checklist that can be further
elaborated according to the company’s standards and
policies):
Table 3: Knowledge map quality criteria
Knowledge
Map
Quality
Dimensions
Review Questions
1. Functio-
nal map
quality
• Does the map serve an explicit purpose for a
specific target user group?
• Is there an implemented process to update
and review the knowledge map periodically?
• Is there a feedback mechanism through which
users can suggest improvements to the map?
2. Cognitive
map quality
• Can the map be grasped at one glance (not
overloaded)?
• Does it offer various levels of detail?
• Does it allow to compare elements visually?
• Are all elements clearly discernible?
3. Techni-
cal map
quality
• Is the access time sufficient (no time lags)?
• Can the map be used with a browser-
interface?
• Does the map appear legibly on various
screen resolutions?
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 9
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001

• Is the map securely protected against
unauthorized access?
4. Aesthetic
map quality
• Is the map pleasing to the eye (adequate color
and geometric form combinations)?
• Can the map’s visual identity be kept when
new elements are added (map scalability)?
If these criteria are met, an intranet knowledge map
may well become one of the killer-applications of a
corporate intranet since it provides a quick and
comprehensive overview of a company’s intellectual
assets.
6. Conclusion
Knowledge maps provide a visual orientation for
managers or specialists who wish to locate, evaluate or
develop knowledge in an organizational context. They
condense information about knowledge sources, assets,
structures, applications, or development needs in an
accessible way. As Wurman points out, the creative
organization of such information can create new
information and insights (see [14]). Each way that one
organizes information can create new knowledge and
understanding. This new understanding results from the
organizational context that knowledge maps can provide.
Nevertheless, knowledge maps have not yet lived up to
their potential in the corporate world. Management
literature on the topic has still only a few success stories
to illustrate the potential of such maps, such as the Swiss
pharmaceutical company Hoffmann-La Roche where a
comprehensive knowledge map was used to improve the
(knowledge-intensive) new drug approval process and
hence improve the time-to-market of new products (for a
depiction of this map see [15], p.172). However, with the
rapid development of intranet technology and its potential
to combine appealing visual interfaces with database
applications, knowledge maps may soon prove to be a
standard element in any company's knowledge
management repertoire.
7. References
[1] Quinn, J.B., Anderson, P., Finkelstein, S. (1996) Managing
Professional Intellect: Making the Most of the Best, in: Harvard
Business Review, March-April, pp. 71-80.
[2] Davenport, T.H, Jarvenpaa, S. L, Beers, M.C (1996),
Improving Knowledge Work Processes, in: Sloan Management
Review, Volume 37, No. 4, pp. 53-65.
[3] Fahey, L.; Prusak, L (1998).: The Eleven Deadliest Sins of
Knowledge Management, in: California Management Review,
Vol. 40, Nr. 3, pp. 265-276.
[4] Galloway, D. (1994), Mapping Work Processes, Milwaukee:
ASQC Quality Press.
[5] Huff, A.(Ed.) (1990), Mapping Strategic Thought, New
York: Wiley.
[6] Rhodes, J. (1994), Conceptual Toolmaking, Expert Systems
of the Mind, Oxford: Blackwell Publishers.
[7] Vail, E.F (1999), Mapping Organizational Knowledge, in:
Knowledge Management Review, Issue 8, May/June, pp. 10-15.
[8] Morris, Meed, Svensen (1996): Morris, S. and Meed, J. and
Svensen, N. (1996), The Intelligent Manager, Adding Value in
The Information Age, London: Pitman Publishing, 1996.
[9] Mok, C. (1996), Designing Business, Multiple Media,
Multiple Disciplines, San Jose: Adobe Press.
[10] Horn, R., (1989), Mapping Hypertext,Analysis, Linkage,
and Display of Knowledge for the Next Generation of On-Line
Text and Graphics, Waltham: The Lexington Institute.
[11] Prahalad, C.K., Hamel, G. (1990): The Core Competence
of the Corporation, in: Harvard Business Review, 68: 3, pp. 79-
91.
[12] Sparrow, H. (1998), Knowledge in Organizations,
Thousand Oakes: Sage.
[13] Peterson, M.P., (1995), Interactive and Animated
Cartography, New Jersey: Prentice Hall.
[14] Wurman, R.S. (1990), Information Anxiety, What to do
when information doesn’t tell you what you need to know, New
York: Bantam Books.
[15] Wurman, R.S. (1996), Information Architects, Zurich:
Graphis.
0-7695-0981-9/01 $10.00 (c) 2001 IEEE 10
Proceedings of the 34th Hawaii International Conference on System Sciences - 2001