See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/221581342

Towards knowledge needs-technology fit model for knowledge management

systems

Conference Paper · May 2009

DOI: 10.1145/1555619.1555646 · Source: DBLP

CITATIONS READS
13 362

2 authors:

Peter Baloh Kevin C. Desouza

33 PUBLICATIONS   890 CITATIONS   
Queensland University of Technology
278 PUBLICATIONS   7,480 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Charting the coevolution of cyberprotest and counteraction: The case of former Soviet Union states from 1997 to 2011 View project

All content following this page was uploaded by Kevin C. Desouza on 19 February 2014.

The user has requested enhancement of the downloaded file.

 TOWARDS KNOWLEDGE NEEDS-TECHNOLOGY FIT
MODEL FOR KNOWLEDGE MANAGEMENT SYSTEMS
Peter Baloh Kevin C Desouza
Faculty of Economics Information School
University of Ljubljana Mary Gates Hall, Suite 330D
Kardeljeva ploscad 17, Ljubljana UW Box 352840
Slovenia 98195-2840 Seattle, WA
+386 41 711 317 +1 206 616 0880
peter@baloh.net kev.desouza@gmail.com

ABSTRACT related to achieving knowledge of a problem domain and its

The goal of this paper is threefold. The first goal is to provide an solution through construction and application of an IS artifact. As
illustrative example of design science research. Readers will such, design-science is proactive towards technology and does not
benefit from seeing how design science research guidelines, as take it as “given” (Hevner et al., 2004).
proposed by Hevner et al. (2004), can be rigorously followed in a This paper seeks to provide an illustrative example of a design
practically relevant study. The second goal is to describe the science research. The goal of the research was to create a model
novel artifact constructed during the research project: a for designing Knowledge Management Systems (KMSs).
knowledge management system design model. The third goal is to Research, inspired by the design science paradigm, was
provide a methodological contribution to the design science conducted in two global organizations: Parsons Brinckerhoff
community. The paper calls for adding an exploratory step in the (USA) and Samsung Electronics (Korea). The paper provides a
build phase when designing a new artifact. methodological contribution to the design science community.
Based on the experiences of the research project, we call for an
Categories and Subject Descriptors additional exploratory step in the ‘build’ phase of design science
H.4.2 [Information Systems Applications]: Types of Systems--- research.
decision support; K.6.1 [Management of Computing and
Information Systems]: Project and People Management 2. ABOUT THE DESIGN SCIENCE
General Terms PARADIGM
Management, Performance, Design, Standardization, Theory. Design and behavioral science are two distinct but
complementary research paradigms [1, 2]. The behavioral-science
paradigm has its roots in natural sciences. It seeks to develop and
Keywords justify theories that explain and/or predict phenomena
Knowledge management systems, design science, design model. surrounding the analysis, design, implementation, management,
and use of information systems. Such theories ultimately inform
1. INTRODUCTION us of the interactions among people, technology, and
Each research paradigm has its own nuances. Selecting the organizations that must be managed if information systems are to
appropriate paradigm, research approach, and methodology to achieve their stated purpose, namely increasing the effectiveness
arrive at scientific knowledge depends on the nature and scope of and efficiency of an organization. These theories impact, and are
inquiry. In principle, information system (IS) research can be impacted by, design decisions made with respect to the system
classified under two paradigms: behavioral science and design development methodology used and the functional capabilities,
science. Behavioral science is concerned with building theories information content, and human interfaces implemented within IS
using positivist, interpretivist, or critical lenses, seeking “what is [2]. The design-science paradigm has its roots in engineering and
true”. The goal is explaining and predicting phenomena related to the sciences of the artificial [3]. It is fundamentally a problem-
acquisition, implementation, management and use of solving paradigm. It seeks to create innovations that define the
technologies. Design science seeks “what is effective.” It is ideas, practices, technical capabilities, and products through
which the optimal analysis, design, implementation, management,
and use of IS can be undertaken [4]. Most often, the behavioral
Permission to make digital or hard copies of all or part of this work for science paradigm is passive in respect to technology, often
personal or classroom use is granted without fee provided that copies are
ignoring or “undertheorizing” the artifact itself [5]. Its focus is on
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, or describing the implications of technology. Design science is
republish, to post on servers or to redistribute to lists, requires prior active with respect to technology, engaging in creation of
specific permission and/or a fee. technological artifacts that impact people and organizations [6-7].
DESRIST'09, May 7-8, 2009, Malvern, PA, USA.
Copyright 2009 ACM 978-1-60558-408-9/09/05…$5.00. March and Smith [1] identify two design processes and four
design artifacts. The two processes are build and evaluate.
Purposeful artifacts are first built to address unsolved problems. Hevner et al. [2], in their guidelines for design-science type of
Then they must be evaluated with respect to the utility provided research, emphasize the need for research rigor (Guidelines 3 and
in solving those problems. Artifacts are interdependent with 5). With respect to construction of the artifact, rigor is derived
people and organizations [1, 2]. The artifacts are constructs, from 1) effective use of the knowledge base (theoretical
models, methods, and instantiations, which define ideas, foundations), and 2) selecting, and executing, research
practices, technical capabilities, and products, through which methodologies to build the artifact. In our case, we accounted for
analysis, design, implementations, and use of information systems prior research in knowledge management and its allied
can be effectively and efficiently accomplished [4]. disciplines. The analysis of the literature aided in
conceptualization of constructs. However, in many cases there is
Constructs provide the language in which problems and solutions very little, if any, prior research that was directly relevant to our
are defined and communicated [6]. They are concepts that problem. The major contribution of this work lies in an innovative
characterize a phenomenon [7]. Models use constructs to recombination of the existing IS literature leading to the proposed
represent a real world situation, the design problem, and its KMS design model.
solution space [3]. Models frequently represent the connection
between problem and solution components enabling exploration The process by which the artifact is created, and often the artifact
of the effects of design decisions and changes in the real world. itself, incorporates or enables a search process whereby a problem
They describe tasks and situations [7]. Methods provide guidance space is constructed and a mechanism is devised to find an
on how to solve problems. These can range from formal, effective solution (Guideline 6). Design science is inherently
mathematical algorithms that explicitly define the search process iterative where the initial artifact needs to be evaluated and
to informal, textual descriptions of best practice approaches, or redesigned until a satisfactory solution is found. The search for
some combination. Instantiations are physical implementations solutions is guided by satisficing. Searching stops when an
[7]. They show how constructs, models, or methods can be artifact that “works well for the specified class of problems” is
implemented in a working system. They demonstrate feasibility, found [2]. Finally, the results of the design-science research must
enabling concrete assessment of an artifact’s suitability to its be communicated effectively (Guideline 7) both to a technical
intended purpose. Researchers can also learn about the real world, audience (researchers who will extend them and practitioners who
how the artifact affects it, and how users appropriate it [2]. Such will implement them) and to a managerial audience (researchers
artifacts are not exempt from natural laws or behavioral theories. who will study them in context and practitioners who will decide
To the contrary, their creation relies on existing kernel theories if artifacts should be implemented within their organizations) [2].
that are applied, tested, modified, and extended through the
Fundamental questions on utility (what does it do?) and its
experience, creativity, intuition, and problem solving capabilities
demonstration (does it really do that?) are salient: “Contribution
of the researcher [2].
arises from utility. If existing artifacts are adequate, then design-
“The resultant IT artifacts extend the boundaries of human science research that creates a new artifact is unnecessary (it is
problem solving and organizational capabilities by providing irrelevant). If the new artifact does not map adequately to the real
intellectual as well as computational tools. Theories regarding world (rigor), it cannot provide utility. If the artifact does not
their application and impact will follow their development and solve the problem (search, implementability), it has no utility. If
use…” [2]. utility is not demonstrated (evaluation), then there is no basis
upon which to accept the claims that it provides any contribution
To ensure the quality of the design-science research, Hevner et al. (contribution). Furthermore, if the problem, the artifact, and its
[2] provided seven guidelines, to guide the execution of design utility are not presented in a manner such that the implications
science research. The fundamental principle being knowledge of a for research and practice are clear, then publication in the IS
design problem and its solution are acquired through building, literature is not appropriate (communication)” [2].
application and evaluation of an artifact.
Design-science research requires the creation of an innovative, 3. THE STUDY OF KMS DESIGN MODEL
purposeful artifact (Guideline 1) for a specified and relevant
problem domain (Guideline 2) [2]. Because the artifact is
CREATION
purposeful, it must yield utility for the specified problem. Hence, This section discusses the design study performed, and how
thorough evaluation of the artifact using established methods is design science guidelines were followed in its execution. We start
crucial (Guideline 3). IS artifacts can be evaluated in terms of with the crafting of the research question and continue with
functionality, completeness, accuracy, fit with the organization, detailed explanation of research methodology.
and other relevant quality attributes. Novelty is crucial since the
artifact must be innovative, solving a heretofore unsolved 3.1 The context and the research question
problem or solving a known problem in a more effective or Organizations continue to struggle with questions, such as how
efficient manner (Guideline 4). In this way, design-science can they improve what employees know, how can they add
research is differentiated from the practice of design [2]. The creative insights to business decisions, how can they capitalize
artifact itself must be rigorously defined, formally represented, upon what others have learned before in doing same or similar
coherent, and internally consistent. “Research into KMS design tasks, and how can they stop employees from reinventing the
should be built on the foundations of the cumulative body of same or even suboptimal solutions to problems that were already
research in the IS field and not by reinventing the wheel in a solved by someone else?
research context” [8].
Senior executives, analysts, and policymakers from an Economist
2006 survey feel that improving the productivity of knowledge
workers through technology, training, and organizational change 3.2 The research methodology and
will be the major boardroom challenge of the next 15 years. In
another survey, CEOs ranked knowledge management (36%) development of the artifact
second to sales and marketing (56%) as the business function that 3.2.1 Literature review
will be the most important in realizing corporate strategy goals A systematic review of existing literature on knowledge
over the next three years [9]. In addition, KM has become the top management was conducted. Analysis and synthesis of the
priority for strategic IT investments. Two-thirds of Western literature from different research fields was conducted in order to
European senior executives in the 2005 Economist survey recognize the importance of the problem, to uncover current
answered that KM and business intelligence tools will be the most understanding of the areas of concern, and to theoretically ground
important technology underpinning their company’s goals over the research. The literature review was also necessary to help
the next three years [10]. Similarly, CEOs from the 2006 identify promising areas for future research.
Economist survey have expressed certainty that investments in
technology by the year 2009 will mostly result in improved 3.2.2 Building the design model
customer relationship management (39%), improved sales and The KMS design model was proposed up-front in a positivist
marketing (34%), and improved KM (33%) [9]. Accordingly, manner. This artifact was constructed in the form of “ideal fit
30% of CEOs have stated that knowledge management is the most profiles,” which advise appropriate technology solutions for
important investment for the year 2007, second to marketing and particular knowledge work contexts. Figure 1 shows the
sales improvement investment (36%) [11]. With the purpose of conceptual framework of the model.
improving the organization’s efficiency and effectiveness through
Three major theoretical lenses guided the model development.
better decisions, organizations consciously design and deploy KM
First, Task Technology Fit Theory [38, 39] was taken as the
solutions that instigate utilization of existing knowledge and new
conceptual lens for construction of the KMS design model and
knowledge creation [12, 13]. Measuring intellectual capital,
was amended to the fit KM context. Second, Evolutionary
establishing corporate libraries, building intranets, sharing best
Information Processing Theory [40] was used to describe how
practices, leading cultural change, building databases, leading
knowledge is created and utilized in the course of individual
training programs, installing groupware, fostering collaboration in
decision-making. This was important as we wanted to open up the
communities of practice, and creating virtual organizations are
‘black box’ of ‘knowledge creation and utilization’ process.
only a few examples of what companies implement with hope of
Knowledge creation and utilizations was framed as a six-stage
improving how knowledge is created and used in individual and
iterative process of 1) problem recognition, 2) goal setting, 3)
group decision making. However, benefits remain elusive for
generation of tentative knowledge variation, 4) knowledge
most of the companies; failure rate of KM initiatives is over 70%
selection, 5) knowledge retention, 6) resource management.
[14-16]. Some report even higher failure rate (84%) [i.e. 84% by
Third, we drew on the process value of IT Theory [41-43], which
17]. While there was a considerable body of literature that deals
suggest organizational and technological innovations should be
with KM projects, it more often than not merely lists variety of
introduced and evaluated at the business process level and not
factors as being critical for their success [8]. Many factors have
organization-wide.
been explored in detail in the literature, such as:
• clear KM vision and strategy [18, 19]
• alignment of KM strategy to business goals [20, 21]
• promoting a learning culture [22-24]
• incentives for knowledge creation and reuse [25]
• a community that provides a context for KM to flourish [26]
• continuous top management support [27]
• employee empowerment [28]
• a positive attitude to knowledge sharing [29, 30]
• a flexible organization structure [31]
• usable and up-to-date KM systems [32-35]
• knowledge governance structure for maintaining quality of
knowledge content [36]
However, there remains a lack of actionable know-how that
describes how to actually build a meaningful and business-value-
adding KM solution [16]. KM activities and technologies are
indiscriminately deployed, without regard to the actual context
into which they are being brought [37]. Based on thorough
scrutiny, synthesis and combination of existing literature we
posed the following research question: How do workers’
knowledge needs influence the design of KMS for supporting
and enabling knowledge creation and utilization? To this end, Figure 1: Concept of the KMS design model
we created a model for designing KMSs. KMS characteristics represented the “technology” part of the
construct. Here, the literature was analyzed [i.e. 44, 45-48] and
synthesized to arrive at five types of KMS: 1) explicit knowledge
exchange type, 2) tacit knowledge exchange type, 3) connectivity
between humans type, 4) collaborative work type, and 5) Thus to complete model building an exploratory case study was
knowledge discovery type. KMSs have to support individuals in performed. Parsons Brinckerhoff is a knowledge intensive
the course of their knowledge work. The basic tenant of our engineering company which was selected based on its reputation
research is types of technology that are appropriate for particular of being the top American company in road and highway design,
work contexts will differ within, and across, organizations. We mass transit and rail design, and second best in bridge design and
defined a contingent variable “knowledge needs”. The variable airport design [49]. At Parsons Brinckerhoff, issues of knowledge
was conceptualized by describing: 1) task domain, 2) type of management and learning are recognized and appreciated.
knowledge and 3) volatility of knowledge. Different combinations Organizational and technological mechanisms (i.e. organizational
of these three elements gave us eight organizational contexts in structure, goals, processes and KMS) to enable and support KM
which knowledge work happens. had already been introduced. KM executives were interviewed to

Table 1: Initial, theory-based model

WORK CONTEXT KNOWLEDGE WORK PHASES

G eneration of Selection of
Task Knowledge Type of Problem Go al Ret ention of Resource
tentative t entative
domain volatility knowledge recognition setting new knowledge management
know ledge knowledge

1 focused low inf ormational EKX EKX EKX, KD EKX EKX EKX, CON

2 focused low procedural EKX EKX EKX, KD EKX EKX, TKX EKX, CON

Types of KMS
3 focused high inf ormational EKX, COL EKX, COL EKX, COL, KD EKX EKX CO N , EKX

4 focused high procedural EKX, COL EKX, COL EKX, TKX, COL EKX, COL EKX, COL, TKX EKX, CON

5 broad low inf ormational COL, EKX EKX, COL COL COL EKX, CO L EKX, CON

6 broad low procedural COL, EKX, TKX EKX, COL COL, TKX COL EKX, CO L EKX, CON

7 broad high inf ormational COL, EKX EKX, COL CON, COL COL CON, COL EKX, CON

8 broad high procedural COL, EKX, TKX EKX, COL CO N, COL, TKX COL CON, TKX EKX, CON

KMS Types TKX: Tacit knowledge exchange

EKX: Explicit knowledge exchange CON: Connectivity between human knowledge sources
COL: Collaborative knowledge creation KD: Knowledge discovery from explicit sources

After conceptualization of each of the constructs, relations Table 2: Examples of eight knowledge needs profiles in
between them were substantiated drawing on the literatures of Parsons Brinckerhoff, that were discussed from the
management, KM, and IS. Existing fit prescriptions and empirical viewpoint of KMS support to the knowledge worker
research provided a starting point for assessing key constructs and
their potential relationships. Effort was invested in finding mini WORK CONTEXT EXAMPLES
cases for each of the relationships. However, in some cases, there
Task K
was very little, if any, prior research that is directly relevant to Type of K Short description
domain volatility
particular proposition. In such occurrences, theoretical concepts
were searched for and applied analytically to the given situation. Determine the average rainfall in a region for the
focused low informational
Fit profiles connecting knowledge needs, knowledge work and last 50 years.
KMS technology choices were then proposed. They represent the Determine the proper method to measure a river’s
focused low procedural
essence of the model. The initial, theory based model, is shown in bank erosion over time.
Table 1. Establishing the overall cost of steel for a bridge
focused high informational
project.
As per design science guidelines, the artifact was proposed in the
build phase strictly based on existing research. Before assessing Determining the best way to handle PR in a crisis
focused high procedural
usability of the proposed (theory-based) model, which is the goal situation.
of the “evaluate” phase in design science, we decided to conduct a Determine the overall volume of materials required
broad low informational
preliminary check of the model. The goal was to assess and for a building of specific architectural design.
validate the core of the model before asking about its actual Establish a change request process for a project
usability. Initial interviews with the key informant, the KM broad low procedural
that involves multiple stakeholders.
executive of the company in which we wanted to do the
Determine the cost of project depending on how the
evaluation study, were conducted. Interviews revealed that there broad high informational
weather affects the schedule for the project.
were stark differences in the knowledge management practices of
found in business organizations versus the academic literature. Determine the best way to handle public comments
broad high procedural and protests as they arise throughout the 5-year
lifespan of a project.
 Table 3. Proposed knowledge needs-KMS fit profiles after the exploratory study in Parsons Brinckerhoff

WORK CONTEXT KNOWLEDGE WORK STAGES

G eneration of
Select ion of
Task Knowledg e Type of Problem Goal tentative Retention of new Resource
tentative
domain volat ility knowled ge recognition setting know ledge knowledge management
knowledge
variation

1 focused low information al EKX EKX EKX,KD EKX EKX EKX

2 focused low procedural EKX EKX EKX EKX EKX,TKX EKX

3 focused high information al EKX,COL EKX,COL EKX,COL,KD EKX,COL EKX EKX,CON

Types of KMS
4 focused high procedural EKX,COL EKX,COL EKX,COL EKX,COL EKX,TKX,COL EKX,CON

5 broad low information al EKX,COL EKX EKX,COL COL EKX,CO L EKX,CON

6 broad low procedural EKX,COL EKX EKX,COL,TKX COL EKX,COL,TKX EKX,CON

7 broad high information al COL,EKX EKX,COL COL,CO N,EKX COL EKX,CON,COL EKX,CON

8 broad high procedural COL,EKX COL,EKX CO L,EKX,TKX COL EKX,COL,TKX,CO N EKX,CON

KMS Types TKX: Tacit knowledge exchange

EKX: Explicit knowledge exchange CON: Connectivity between human knowledge sources
COL: Collaborative knowledge creation KD: Knowledge discovery from explicit sources

gain insights about relevance of the core of the model and its people, but in a way that he will be able to map that
applicability. As the interviewees’ everyday work was leading everyday discussion onto the model’s concepts.”
KM teams, the KM executives had highly developed theories-in-
use, which they were forced to articulate and make explicit. They As noted by the interviewee, utility of the model would indeed be
represented ‘well-informed informants’ who were able to reflect lower if it was difficult to use. We decided to craft a short list of
upon and discuss KMS and KM issues [50]. questions that complement the fit profiles table, and that a KMS
designer could pose to improve the quality of how knowledge
Eight different organizational settings, which correspond to the
analysis questions are interpreted by the interviewee (knowledge
design model profiles, were analyzed to deduce the most
worker). The questions needed to reflect the overall concept of the
appropriate technology for each of the knowledge work stages. As
design model and touch its most important relationships and
such, these eight different contexts represent mini cases within a
constructs. They were framed as ‘conversation starters’ for a
company (see Table 2). These settings were discussed during the
KMS designer, when analyzing knowledge needs and knowledge
interviews to arrive at more generalizable framings, and to think
work in particular organizational context. Table 4 shows the
through work situations that covered all the eight different
questions and their relation to the knowledge needs-KMS fit
knowledge needs profiles. The informants agreed that
profiles. This new instrument thus complements the fit profiles
contingency variables were well chosen and that different
table and together they constitute a design model built based on
combinations account for most work contexts in organizations.
existing literatures and amended through the exploratory study in
Through the method of envisioning a KMS for each of the eight
Parsons Brinckerhoff.
contexts, we came to an agreement on updated design model (see
Table 3, compare to Table 1). Furthermore, there was an
important finding that resulted from thoroughly analyzing the
3.2.3 Evaluation of the design model
After the build phase was completed, the KMS design model was
core of the model. We realized that the fit-profiles-table might be
evaluated. An in-depth exploratory qualitative case study was
too complex to be easily understood and applicable in practice. As
performed in Samsung Electronics, which had just undergone a
noted by the interviewee:
KM-organizational change project. Samsung Electronics was
selected as the company enjoys a strong reputation of being one
“For a real world setting, I think you will have to of the most innovative and successful knowledge-based
consider a way to get to your answers more easily. I organizations in the world [51-55].
mean, the KMS designers would benefit from such
tool, as they need to analyze knowledge needs and The goal of the evaluation phase was to further validate the core
practices. They will not have the time to explain the of the model and demonstrate its usability of the model to the
entire model… all of the model’s concepts to intended users. Utility of the model was assessed with an eye for
everyone. So you need a set of efficient questions 'satisficing'. Hevner et al. [2] suggest that the model has to be
that will get the analyst the answers from ‘common’ “good enough to work” as the nature and the value of the design
 Table 4: Building a list of questions for the knowledge needs and organizational context analysis

Question Explanation
Q1: What does the knowledge worker The question is concerned with overall process- and business-orientation of both KM
need to solve and what are the business- efforts in general and KMS design as technological support for particular business
and knowledge-related goals behind problem. First, a business-related reason must exist in order to think about KMS
that? support; KM efforts need to be oriented in improving how knowledge is created and
utilized with the goal to achieve that business-related goals.
Q2: What is the shelf life of knowledge This question is concerned with the knowledge volatility in particular context; high
that is needed in everyday decision- or low. How we deal with knowledge and what is the appropriate KMS depends on
making? how much time that knowledge is current. Can the same knowledge be reused for
performing tasks/make decisions, or is it rapidly changing and needs to be created
continuously?
Q3: How do employees make decisions: The question is concerned with the way of performing the task or making decision.
on their own, by collaborating with How does he solve a particular problem/makes a decision? On his own or in
people in the same knowledge domain, or collaboration with others?
with people from other knowledge
domains?
Q4: What kind of knowledge do they This question is concerned with the type of knowledge needed to perform the task;
need to arrive at a solution? Is it a piece procedural or informational? Is it know-what that is important in particular context,
of information, a document? Or is it a or is it know-how? In other words, what kind of knowledge needs he have?
procedure, how to do something?
Q5: How do employees learn from past Based on the above argument on how knowledge needs to be created and utilized
experience before they make a decision? when employee tries to perform his activities (under Q1), an important underlying
concept is learning. So in the course of daily work, employees need to learn
Q6: How employee learns during the “before” undertaking a task, learn “during” that task, and learn “after” the task (task
particular work? being i.e. decision-making). By looking through these three lenses, KMS designer
Q7: How they instigate the double loop can answer the question of how learning and leveraging knowledge is done during
learning, e.g. learning after a task? everyday work. The argument for interest in how learning before making decisions
is done is that it is highly likely that there is somebody out there who has already
done something similar before. Thus, how can a KMS be designed in a way to
provide up-front as much as possible information necessary for fulfilling the task?
This question corresponds somehow to the early phases of knowledge work, i.e.
problem recognition and goal setting. The argument for learning after an event is that
experience and insights should be captured and transferred to similar future
occasions. This question corresponds somehow to the late phases of knowledge
work, i.e. tentative knowledge variation selection and knowledge retention. Again,
the goal is to design such a KMS that will facilitate this. In example, ways to learn
after include immediate project team meetings, codifying insights into a searchable
database, and holding retrospect meetings. The logic behind learning during is that
knowledge-related interventions can be introduced while making a decision, while
working on a project, as one can continuously learn before reaching the end of a
project, in example. Again, the goal is to design such a KMS that will take into
account how learning during is usually made.
Q8: KMS Technologies supporting these This question is concerned with existing KMS support for knowledge workers at
activities? their tasks. Should be asked with each of the Q5-Q7 questions.

science does not lie in the researching ‘why’ exactly the artifact Selection of this individual is in line with Nonaka and Takeuchi
works. [56] who argue that members of middle management, more
precisely, ‘owners’ of subunits, of processes, KM officers, or
Data collection occurred through semi-structured interviews with
KM-project managers, should be interviewed as they are in
a key informant, short informal discussions with key knowledge
possession of sufficient knowledge and are adequately involved in
workers, and through review of written materials. First, experts
the KM programs of the organization. Besides knowing the
from the company who were involved in the KM-related
business value of such projects, this person was operationally
organizational change project were queried on various aspects of
involved in design and deployment of future KM solutions,
the project. Through the course of these discussions we learned
including design and implementation of KMS. The interviewee
that the Director of Organizational Development could answer
was thus able to assess applicability and usability of the KMS
questions on all aspects of the KM project and KMS design.
design model. He was also asked to reflect on past decisions and “Let me say that I like this concept. While kept
discuss how the model could help Samsung if they had to start the uncluttered, it is also very powerful as it reminds
KM project all over again. In addition to in-depth interviews, key you of all the important boxes you need to think
knowledge practitioners [56] in the company were informally about when designing a KM environment and
observed at their daily work and were queried to evaluate mechanisms. I like it as it puts together the two
appropriateness of the existing KMS design. The purpose was to worlds of the information architect and the process
assess the fit of their KMS to the context in which it is used, and owner, and as it makes them clear for both of them.
to assess the value the proposed design model. Key knowledge Maybe these parts are already in managers’
workers know which KMS functionalities would be useful for subconscious however, I guess they are not really as
their context of work, or at least can recognize a potentially useful structured as in here. It is advantageous to have this
KMS when they are presented with its options. The primary as it shows the relationships between the elements of
sampling criterion to identify knowledge workers was opportunity successful KMS design and the role of each of the
to learn and diversity along with balancing the constraint of model’s core elements: the nature of work and the
access [57]. This approach allowed for obtaining first-hand variety of possible information technology to
experience of the challenges at various organizational levels of support it. It would be much easier to talk with the
KM-project and by interacting with those directly involved. project team members and specialists if this concept
was in everyone’s mind from the beginning.”
Utility assessment was done by picking two different contexts in
which managing knowledge is important, and analyzing them “What I see from this example that we went through
from the KMS design model standpoints: application developer is something quite unique. As the model is well
solving mobile phone software bugs, and software technology structured and as constructs are explained, analysis
manager mobilizing innovative ideas. Based on the list of seven of existing or design of new KMS has become quite
questions and the fit profiles table, interviewees were asked to easy. The model takes me from one knowledge work
discuss the suggestions (as proposed by the model) versus real life stage to another and it forces me to think from the
KMS support in their work context. knowledge worker’s view. This is really good as
KMS designer remains focused on the user and his
No alterations were proposed to the fit-profiles table (Table 5)
needs. All the designer needs to do is to think how to
even though we have asked for amendments at numerous
apply available KMS tools exactly in each stage. But
occasions during the interview. The interviewee confirmed the
the stages are here and the fitting technology is
basic tenant of the research (need for segmented approach),
here, which is most welcome.”
repeated the practical need for the design model, and confirmed
the core of the model. With regard to that, consider the following
Another goal of the study was to evaluate the set of seven
comments:
questions by assessing their wording and utility. The interviewees
had significant experiences in designing KM solution and

Table 5: The knowledge needs-KMS fit profiles – final version

WORK CONTEXT KNOWLEDGE WORK STAGES

Generation of
Selection of
Task Knowledge Type of Problem Goal tentative Retention of new Resource
tentative
domain volatility knowledge recognition setting knowledge knowledge management
knowledge
variation

1 focused low informational EKX EKX EKX,KD EKX EKX EKX

2 focused low procedural EKX EKX EKX EKX EKX,TKX EKX

3 focused high informational EKX,COL EKX,COL EKX,COL,KD EKX,COL EKX EKX,CON

Types of KMS

4 focused high procedural EKX,COL EKX,COL EKX,COL EKX,COL EKX,TKX,COL EKX,CON

5 broad low informational EKX,COL EKX EKX,COL COL EKX,COL EKX,CON

6 broad low procedural EKX,COL EKX EKX,COL,TKX COL EKX,COL,TKX EKX,CON

7 broad high informational COL,EKX EKX,COL COL,CON,EKX COL EKX,CON,COL EKX,CON

8 broad high procedural COL,EKX COL,EKX COL,EKX,TKX COL EKX,COL,TKX,CON EKX,CON

KMS Types TKX: Tacit knowledge exchange
EKX: Explicit knowledge exchange CON: Connectivity between human knowledge sources
COL: Collaborative knowledge creation KD: Knowledge discovery from explicit sources
systems. Therefore, they were knowledgeable about what and
how to ask knowledge workers when assessing their needs.
Positive assessments were received on the questions:

“I think you hit the nail here by adding these to fit

profiles table. My first thought when you presented
the model was that “this is too complex”. Well, your
model does describe very complex relationships.
Therefore it is natural to be complex on its own. The
questions that you attached very simply ask about
the constructs your model uses; I can see where
each question is pointing. This is a very good
interface to the model. Still, and this is a very
serious comment, they need to be tweaked a little. As
a whole, they are very confusing because you use
different words for same thing, in example, decision
making, work, task. KMS designers will be using
these questions when interacting with ordinary
people. Use simpler words, use one word for the
same construct, and you are fine.”

Based on the feedback, we reworded a few questions accordingly;

we also dropped conceptual words, such as ‘double loop
learning’, and ‘knowledge domain.’ The revised list of questions
is outlined in the Table 6.

Table 6: A list of questions for the knowledge needs and

organizational context analysis
Q1: What is the task that knowledge worker needs to Figure 2: 6-step process to design and deploy a KM solution
perform and what are the business- and knowledge-
related goals behind it? guidelines on how to use the model to analyze particular business
context, specify an appropriate KM system, and design other
Q2: What is the shelf life of knowledge that knowledge (organizational) parts of the KM solution. Practitioner guidelines
worker uses when performing the task in this context? on how to go about knowledge-related organizational change
projects were outlined by proposing steps for such organizational
Q3: How do knowledge workers perform the task: on
endeavors. We discovered that design and deployment of KM
their own, by collaborating with people who have the
solution happens in a 6-step process (see Figure 2).
same expertise, or by collaborating with people with
different expertise? The proposed process conceptually does not differ much from
other organizational change projects. KM-project steps follow the
Q4: What kind of knowledge does a worker need to
same logic of “understand the challenge, set the goal, model the
successfully perform the task?
as-is and to-be situations, evaluate the results.” In this sense,
Q5: How does the knowledge worker learn from past deployment of a KMS is no different from any other
experience before performing the task? What organizational change project, and KMS development is no
technologies support this activity? different from other IS development projects. This is positive both
for re-focus and for the advancement of the KM research area, as
Q6: How does the knowledge worker learn during a researchers are encouraged to take advantage of the cumulative
particular task? How can newly acquired knowledge body of research in management and IS. Earlier studies have
be transferred to other knowledge workers? What proposed various information systems development
technologies support this activity? methodologies (ISDM), which provide a consistent set of
Q7: How does the knowledge worker learn after procedures to be followed to make the process of managing and
performing the task? How can newly acquired developing information systems more efficient and effective [58].
knowledge be transferred to other knowledge In addition, ISDM imply a time-dependent sequence of action
workers? What technologies support this activity? stages [59]. Such frameworks have been developed over the
years; Jayaratna [60] estimated that there were more than 1000
3.2.4 Further results of the research available for use. Each of these ISDM has its own acknowledged
The case of Samsung Electronics also provided for a rich data strengths and weaknesses. However, one ISDM is not necessarily
source we observed their ongoing organizational change suitable for use in all projects [61].
management effort. This gave us additional insights into KM-
related organizational change projects. We investigated how to
use the proposed model for designing a KMS. We arrived at
3.3 Robustness of the overall research design rigorous data collection, analysis and synthesis, by using
The table 7 shows how design science guidelines suggested by established guidelines and techniques, common in qualitative
Hevner et al. [2] were adhered to. research. To analyze and discuss findings from the study, we
used “noting patterns”, making “contrasts / comparisons”, and
counting/listing. Counting and listing was conducted to make
Table 7: Ensuring robustness of the research design lists; in particular, it was used to identify KMS functionalities in
place in each of the eight knowledge work contexts.
G1: Design as an Artifact
Patterns and themes were noted to find support for
Description: Design-science research must produce a viable conceptualization of the constructs.
artifact in the form of a construct, a model, a method, or an
instantiation.
Addressed: Result of this research is a KMS-design model. G6: Design as a Search Process
Description: The search for an effective artifact requires
G2: Problem Relevance utilizing available means to reach desired ends while satisfying
Description: The objective of design-science research is to laws in the problem environment.
develop technology-based solutions to important and relevant Addressed: The study employed iterative approach to building
business problems. the model as the first version of the theory-based-model was
Addressed: The criticality of knowledge creation and utilization first amended in exploratory empirical study. The final model
issues and the role of technologies to foster this has been called was proposed after the evaluation stage.
out in the literature. In the final stage of the build phase,
exploratory multiple case studies were performed to establish G7: Communication of Research
the face validity of the model;, including an assessment of Description: Design-science research must be presented
problem relevance from the practitioner’ perspective. Research effectively both to technology-oriented as well as management-
questions ignited numerous discussions with executives in oriented audiences, while satisfying rigorous academic
companies. standards.
Addressed: Findings and conclusions were presented from both
G3: Design Evaluation research and practical aspects throughout the research report.
Description: The utility, quality, and efficacy of a design artifact Several academic contributions have been disseminated from
must be rigorously demonstrated via well-executed evaluation the study to communicate the results of the study (Baloh, 2007a,
methods 2007b, 2008a, 2008b, 2008c, 2009). For practitioners, research
Addressed: Case study methodology was rigorously employed results have been communicated in a manner of executive
for model development and evaluation. reports and presentations. Both types of communication were
prepared in a manner for technology-oriented audience to
include sufficient detail to be able to construct an instantiation
G4: Research Contributions (computer-based KMS) in particular organizational setting, and
Description: Effective design-science research must provide for management-oriented audience to find guidelines on how to
clear and verifiable contributions in the areas of the design analyze a business context in the KM sense, and how this
artifact, design foundations, and/or design methodologies. influences the choice of KMS functionalities.
Addressed: Contributions to research and practice were
discussed in relation to the disciplines in which this research is
3.4 Contributions of the research
This paper makes several contributions to the KM literature. First,
situated (IS-design science and management and organizational
contrary to mainstream research, it confirmed that developing one
studies).
company-wide KM solution is of limited value. Not only do
different knowledge challenges exist in organizations, but people
G5: Research Rigor also perform different tasks in course of their daily work. This
Description: Design-science research relies upon the application calls for a portfolio of KM solutions that, which are tailored to
of rigorous methods in both the construction and evaluation of satisfy the knowledge needs of an individual. Second, the paper
the design artifact. has introduced a model to guide the design of KMS based on
Addressed: In terms of artifact construction, existing theories knowledge needs. The model consists of “ideal” combinations of
such as Evolutionary Information Processing Theory of knowledge needs and characteristics of KMS, which should result
Knowledge Creation (Li & Kettinger, 2006), Task Technology in improved knowledge utilization and creation. The design
Fit Theory (Goodhue & Thompson, 1995; Zigurs & Buckland, model developed enables the KM community to critically
1998), and Process Value of IT Theory (Barua et al., 1995; evaluate efforts underway to leverage organizational knowledge
Sambamurthy, 2001; Tallon et al., 2000) were used. These with KMSs. The proposed model can also be applied to analyze
theories were used to inform the theoretical part of artifact- successful and unsuccessful KMS implementations retroactively.
building. Exploratory multiple case studies were performed in Third, practitioner guidelines on how to use the model to build
order to complete the build of the KMS design model. In terms KMSs as a part of knowledge-related organizational change
of artifact evaluation, this research used case study research as projects were proposed.
the research strategy. The goal was to evaluate the usability of
the design model for the intended users – managers and KMS
developers. In both phases, special attention was paid to
4. DISCUSSION AND CONCLUSION [2] Hevner, A. R., March, S. T., Park, J. and Ram, S. Design
March, Smith, Hevner and others [1, 2] have documented Science Research in Information Systems. MIS Quarterly,
guidelines and steps for conducting design science research. A 28, 1 2004), 75-105.
contribution of our work is to suggest a methodological [3] Simon, H. A. The Sciences of the Artificial. MIT Press, 1996.
enhancement to the current design science guidelines. In design [4] Denning, P. J. A New Social Contract for Research.
science guidelines as put forward by Hevner et al. [2], the Communications of the ACM, 40, 2 1997), 132-134.
mechanics of the “build” and the “evaluate” phases are [5] Orlikowski, W. J. and Iacono, C. S. Research Commentary:
straightforward. Purposeful artifacts are first built to address Desperately Seeking the “IT” in IT Research—A Call to
unsolved problems and then they must be evaluated with respect Theorizing the IT Artifact. Inf. Syst. Res., 12, 2 2001), 121-
to the utility provided in solving those problems. Rigor in the 134.
build phase is derived from the effective use of the knowledge [6] Schön, D. A. The reflective practitioner. Basic Books, 1983.
base: artifacts should be built upon theoretical foundations [7] Pries-Heje, J. and Baskerville, R. THE DESIGN THEORY
provided in existing cumulative body of knowledge. This model NEXUS. MIS Quarterly, 32, 4 2008), 731-755.
should then be evaluated through a sturdy empirical study, where [8] Butler, T., Heavin, C. and O’Donovan, F. A Theoretical
methodological decisions are justified and appropriate for the Model and Framework for Understanding Knowledge
research context [2]. Management System Implementation. Journal of
However, such approach is dominantly positivistic and presumes Organizational & End User Computing, 19, 4 2007), 1-21.
that existing body of knowledge is sufficient for construction of [9] Economist Intelligence Unit. CEO Briefing: Corporate
an artifact. Moreover, this approach is contradictory when priorities for 2006 and beyond. The Economist, London,
compared with another guideline that strictly requires novelty of 2006.
an artifact. As per Hevner et al. [2] guidelines, the only artifacts [10] Economist Intelligence Unit. Know how - Managing
that design science should be concerned with are those that are knowledge for competitive advantage. The Economist,
novel, solving previously unresolved problems. Then, is existing London, 2005.
body of knowledge on its own enough to construct a novel artifact [11] Economist Intelligence Unit. CEO Briefing: Corporate
that can go into the evaluation phase of a design science research? priorities for 2007 and beyond. The Economist, London,
2007.
To reduce the risk of building a “wrong” initial model, rooted
[12] Desouza, K. C. Barriers to Effective Use of Knowledge
solely in existing literature, we decided to add an exploratory
Management Systems in Software Engineering.
empirical study to the build phase. This, we argue, added to
Communications of the ACM, 46, 1 2003), 99-101.
higher level of practical relevance and novelty of the model. It
allowed us to test the basic propositions of the study, justify its [13] Desouza, K. C., Awazu, Y. and Tiwana, A. Four Dynamics
relevance and check if there are any additional constraints or for Bringing Use Back into Software Reuse.
constructs that are important. A set of in-depth interviews were Communications of the ACM, 49, 1 2006), 97-100.
performed with KM executives who represented a ‘well-informed [14] Wing, L. and Chua, A. Knowledge Management Project
informants’ to 1) gain insights about relevance and potential Abandonment: An Exploratory Examination of Root Causes.
applicability of the model; 2) receive feedback on the model Communications of AIS, 2005, 16 2005), 723-743.
which will be used to amend and validate the tentative model [15] Davenport, T. H. and Glaser, J. Just-in-Time Delivery
suggestions. The a priori (theory based) proposed model was thus Comes to Knowledge Management. Harvard Business
improved through acquired practical insights. Without this step, Review, 80, 7 (July 2002, 2002), 5-9.
we believe, the evaluation phase would surely not have yielded [16] Desouza, K. C. and Awazu, Y. Engaged knowledge
the results expected. Moreover, the research instrument (interview management : engagement with new realities. Palgrave
protocol and the definitions document) were significantly Macmillan, Basingstoke, 2005.
amended during this additional step. [17] Lucier, C. and Torsiliera, J. Why Knowledge Programs Fail.
Strategy and Business, 4th quarter 1997), 14-28.
For the evaluation phase, the goal is to demonstrate usability of
[18] Maier, R. and Remus, U. Implementing process-oriented
the model to the users for whom it is intended. KM managers that
knowledge management strategies Journal of Knowledge
are responsible for design and deployment of KMS need to find it
Management, 7, 4 2003), 62-74.
useful for the purpose intended. Here we are seeking input on the
utility of the model, how it can be used thus, and if it is helpful [19] Von Krogh, G. Care in Knowledge Creation. California
for them. The interview protocol designed and used was thus Management Review, 40, 3 1998), 133-153.
different from the one used in the first, exploratory, study in the [20] Malone, D. Knowledge Management A Model for
build phase. We believe that adding an exploratory study in the Organizational Learning. International Journal of
build stage is critical. Only if the artifact is constructed effectively Accounting Information Systems, 3, 2 2002), 111-123.
will it be found useful by the intended audience. [21] Tiwana, A. The Knowledge Management Toolkit:
Orchestrating IT, Strategy, and Knowledge Platforms.
Prentice Hall PTR, Upper Saddle River, NJ, 2002.
5. REFERENCES [22] Goh, S. C. Improving organizational learning capability:
[1] March, S. T. and Smith, G. F. Design and natural science
lessons from two case studies. The Learning Organization,
research on information technology. Decision Support
10, 4 2003), 216-227.
Systems, 15, 4 1995), 251-266.
[23] McDermott, R. and O’Dell, C. Overcoming cultural barriers [40] Li, Y. and Kettinger, W. J. An Evolutionary Information-
to sharing culture. Journal of Knowledge Management, 5, 1 Processing Theory of Knowledge Creation. Journal of the
2001), 76-85. Association for Information Systems, 7, 9 (September 2006
[24] De Long, D. W. and Fahey, L. Diagnosing cultural barriers 2006), 593-617.
to knowledge management. Academy of Management [41] Barua, A., Kriebel, C. H. and Mukhopadhyay, T.
Executive, 14, 4 2000), 113-127. Information Technologies and Business Value - an Analytic
[25] Markus, M. L. Toward a Theory of Knowledge Reuse: Types and Empirical-Investigation. Inf. Syst. Res., 6, 1 (Mar 1995),
of Knowledge Reuse Situations and Factors in Reuse 3-23.
Success. Journal of Management Information Systems, 18, 1 [42] Sambamurthy, V. Editor’s Comments-Research in
(Summer 2001), 57-93. Information Systems: What We Haven’t Learned. MIS
[26] Wenger, E. C., McDermott, R. and Snyder, W. M. Quarterly, 21, 4 2001), v-xv.
Cultivating Communities of Practice. Harvard Business [43] Tallon, P. P., Kraemer, K. L. and Gurbaxani, V. Executives'
School Press, Boston, MA, 2002. perceptions of the business value of information technology:
[27] Storey, J. and Barnett, E. Knowledge management A process-oriented approach. Journal of Management
Initiatives: Learning from Failure. Journal of Knowledge Information Systems, 16, 4 (Spr 2000), 145-173.
Management, 4, 2 2000), 145-156. [44] Balmisse, G., Meingan, D. and Passerini, K. Technology
[28] Stenmark, D. Knowledge creation and the web: factors Trends in Knowledge Management Tools. International
indicating why some intranets succeed where others fail. Journal of Knowledge Management, 3, 2 2007), 118-131.
Knowledge and Process Management, 10, 3 2003), 207-216. [45] Barron, T. M., Chiang, R. H. L. and Storey, V. C. A
[29] Bock, G. W. and Kim, Y.-G. Breaking the Myths of semiotics framework for information systems classification
Rewards: An Exploratory Study of Attitudes About and development. Decision Support Systems, 25, 1 1999), 1-
Knowledge Sharing. Information Resources Management 17.
Journal, 15, 2 2002), 14. [46] Garfield, S. Initiating & Running a Successful Worldwide
[30] Kankanhalli, A., Tan, B. C. Y. and Wei, K.-K. Contributing KM Program: Examples from HP. Hewlett Packard Services
Knowledge to Electronic Knowledge Repositories: An Consulting & Integration, City, 2006.
Empirical Investigation. MIS Quarterly, 29, 1 (Mar 2005), [47] Janev, V. and Vraneš, S. Comparative Analysis of
113-143. Commercial Knowledge Management Solutions and their
[31] Forcadell, F. J. and Guadamillas, F. A Case Study on the Role in Enterprises. Journal of Information & Knowledge
Implementation of a Knowledge Management Strategy Management, 4, 2 2005), 71-81.
Oriented to Innovation. Knowledge and Process [48] Liao, S. H. Technology management methodologies and
Management, 9, 3 2003), 162-171. applications - A literature review from 1995 to 2003.
[32] Butler, T. and Murphy, C. Understanding the design of Technovation, 25, 4 (Apr 2005), 381-393.
information technologies for knowledge management in [49] Roads & Bridges Roads & Bridges identify the top design
organizations: a pragmatic perspective. Information Systems firms. Scranton Gilette Communications Inc., City, 2007.
Journal, 17, 2 2007), 143-163. [50] Balogun, J. Managing Change: Steering a Course between
[33] Kankanhalli, A., Tanudidjaja, F., Sutanto, J. and Tan, B. C. Intended Strategies and Unanticipated Outcomes. Long
Y. The role of IT in successful knowledge management Range Plan., 39, 1 2006), 29-49.
initiatives. Communications of the Acm, 46, 9 (Sep 2003), [51] Desouza, K. C. and Dombrowski, C. Six Case Studies on
69-73. Organizational Innovation. Technical Report #I4I-I3M-
[34] Tsui, E. The role of IT in KM: where are we now and where InnovCases-1, Six Case Studies on Organizational
are we heading? Journal of Knowledge Management, 9, 1 Innovation, 2006.
2005), 3-6. [52] Edwards, C. Sony and Samsung's Big HDTV Bet. Business
[35] Davenport, T. H. and Prusak, L. Working knowledge : how Week Online, 4/11/20062006).
organizations manage what they know. Harvard Business [53] Hesseldahl, A. Samsung's Fuel-Cell Gambit. Business Week
School Press, Boston, Mass., 1998. Online, 5/18/20062006).
[36] Dilnutt, R. Knowledge Management in Practice, Three [54] Kim, S., Park, Y. and Seol, H. On the design of process-
Contemporary Case Studies. International Journal of oriented knowledge management system for fourth
Accounting Information Systems, 32002), 75-81. generation R&D. Knowledge and Process Management, 14,
[37] Desouza, K. C. Knowledge Management Maturity Model 4 2007), 287-302.
Theoretical Development and Preliminary Empirical Testing. [55] Moon, I., Edwards, C. and Port, O. Samsung Inside? , City,
Doctoral Thesis, University of Illinois at Chicago, Chicago, 2004.
IL, 2006. [56] Nonaka, I. and Takeuchi, H. The knowledge-creating
[38] Goodhue, D. L. and Thompson, R. L. Task-Technology Fit company : how Japanese companies create the dynamics of
and Individual-Performance. Mis Quarterly, 19, 2 (Jun innovation. Oxford University Press, New York ; Oxford,
1995), 213-236. 1995.
[39] Zigurs, I. and Buckland, B. K. A theory of task/technology [57] Stake, R. E. Qualitative case studies. Sage, City, 2005.
fit and group support systems effectiveness. MIS Quarterly, [58] Yadav, S. B., Shaw, N. G., Webb, L. and Sutcu, C.
22, 3 (Sep 1998), 313-334. Comments on "Factors that Impact Implementing a System
 Development Methodology". IEEE TRANSACTIONS ON [61] Meso, P., Madey, G., Troutt, M. D. and Liegle, J. The
SOFTWARE ENGINEERING2001), 279-281. knowledge management efficacy of matching information
[59] Walters, S. A., Broady, J. E. and Hartley, R. J. A Review of systems development methodologies with application
Information Systems Development Methodologies. Library characteristics—an experimental study. The Journal of
Management, 15, 6 1994), 5-19. Systems & Software, 79, 1 2006), 15-28.
[60] Jayaratna, N. Understanding and Evaluating Methodologies: [62] Simon, H. A. The New Science of Management Decision.
NIMSAD, a Systematic Framework. McGraw-Hill, Inc. New Harper & Bros.: New York, 1960.
York, NY, USA, 1994.

View publication stats