Knowledge Management in a Multinational Context:
Aligning Nature of Knowledge and Technology
Cataldo Dino Ruta and Ubaldo Macchitella
Department of Management, Bocconi University
Via Roentgen 1, 20136 Milan, Italy
Abstract. Aim of this paper is to show the importance of understanding the
nature of both the technology and knowledge when promoting knowledge
sharing through knowledge management (KM) portals. This paper investigate
knowledge sharing and the “fit” between the nature of knowledge to be shared
and the nature of the technological tools that are used. Technology intended as
technical instrument could result in an empty box, and knowledge management
initiatives could not be effective and lead to a sustainable competitive
advantage. By means of an in-depth case study of a major consulting firm, the
study discusses and answer the research question. Results show that knowledge
areas with high level of codifiability can be effectively shared by using low
collaborativity and low multimodality tools. Knowledge areas with a high level
of epistemic complexity can be effectively shared by using high collaborativity
and high multimodality tools. Knowledge areas with a high level of task
dependence can be effectively shared by using low collaborativity and
intermediate multimodality tools.
1 Introduction
In the last decade knowledge management (KM) received much attention from both
practitioners and theorists. Interest in knowledge management issues was significantly
boosted by the rapid evolution of information systems [15]. The new features
introduced by innovative technological tools, like the possibility to share information
on real-time and get in touch with people around the world, has led many companies
to imagine a new world of leveraged knowledge [12].
Knowledge management systems are very often embedded in more
comprehensive technological infrastructures such as Human Resource (HR) portals
[2] and represent an invaluable instrument to foster the intellectual capital of an
organization.The goal of this paper is to investigate the relation between the nature of
knowledge and technology in order to have an effective KM system.
Ruta C. and Macchitella U. (2009).
Knowledge Management in a Multinational Context: Aligning Nature of Knowledge and Technology.
In Proceedings of the 3rd International Workshop on Human Resource Information Systems, pages 129-138
DOI: 10.5220/0002211001290138
Copyright
c
SciTePress
2 Theory
2.1 The Relevance of Knowledge Management for Sustainable Competitive
Advantage
Organizational knowledge, and therefore knowledge management, are key in
sustaining competitive advantage over time. Grant [7] develops a knowledge-based
theory of organizations. He affirms that knowledge is the most important strategic
resource for a firm. It resides in specialized form among individual organizational
members and the essence of organizations is its ability to integrate the specialized
knowledge of individuals.
However, even if originating from different fields and perspectives, these
contributions present some aspects in common. In first place, they underline the
relevance of knowledge for competitive advantage. With more or less emphasis, they
refer to knowledge management mechanisms as a key element in developing
capabilities that allow a sustainable, high performance. A second common feature
across these contributions is the strict linked between knowledge and technologies. In
some contributions technology is seen as a technical support for knowledge transfer
that favours the construction of organizational capabilities [11], [5], [18], [14].
According to other authors [7], [13], [1] technology not only concurs to the process of
developing capabilities but also embodies knowledge and capabilities in itself.
Summarizing, from the analysis of the literature emerges that: 1) knowledge and
its management are issues relevant to the construction of organizational capabilities
and that 2) the nature of both knowledge and technology should be taken into
consideration. Therefore, in the next section we present a model linking knowledge
and technology that can be applied to knowledge management projects.
2.2 The Effectiveness of HR Portals: The Match between Technology and
Knowledge
HR portals are vehicles through which HR information and applications can be
channelled effectively and efficiently [2]. HR portals have technical characteristics
that support employees contribution and participation in knowledge management
systems: employees’ personalization through information profiling, relevance of
information and customized single user interface. HR portals present several tools that
support knowledge sharing, from document repositories to more interactive tools like
forums, chat or blogs, the so called KM portals. In order to unfold their beneficial
effects on knowledge management, employees should adopt these instruments and use
the in their everyday working life. The adoption and use of technologies from the
users is an issue that has been extensively investigated within the Information System
Management literature. An established theoretical framework is the one of task-
technology Fit [4]. According to this model, a “fit” between the nature of technology
and the task to be executed should exist in order to perform the task effectively. We
apply the same idea to the knowledge management context. Considering knowledge
sharing as the “task” that should be carried out, we propose a model of “knowledge-
technology fit”, linking the nature of knowledge to be shared to the characteristics of
130
technological tools used to share it. We investigate this “fit” according to some
dimensions derived from the literature that we present in next sections. We’ll consider
the case of a world-wide consulting group. By the analysis of this case study we can
test our theoretical framework and formulate our research propositions.
Knowledge Dimensions. Knowledge has been already measured and described
according to different dimension in previous studies. Zander and Kogut [17]define
codifiability, teachability, system dependence and product observability as the four
charachterisitcs influencing the speed of transfer of organizational capabilities and,
consequently, determining the capability to imitate internal managerial practices. In
particular, these authors find out that the level of codifiability and the easiness of
teachability have a significant influence on the transfer process speed. Grandori [6]
points out three main characteristics: tacitness, computational complexity and
epistemic complexity. These knowledge dimensions influence the choice of
organization and inter-organization coordination mechanisms. Hansen [8] focuses on
codifiability and system dependence as the two main characteristics that can help to
explain difficulties in the transfer of a practice. Knowledge with a higher level of
codifiability and a lower degree of dependence will be easy to transfer. In our study,
the phemomenon of knowledge sharing, and particualrly the decision process in the
choice of KM tools, is presentend from a contingency perspective. Referring to
previous studies, we indicate four main dimensions of a knowledge area as
influencing its degree of transferability: codifiability, epistemic complexity, task
dependence and knowledge comeptitiveness.
Technologies for Knowledge Management. Drawing on the classifications of
technology suggested by the theories referring to task-technology fit, we now propose
a model for the analysis of technologies for knowledge management based on two
main dimensions. The first dimension of our model is collaborativity. Technologies
for KM enable people geographically dispersed to work jointly and to exchange
knowledge by direct interaction. Typically, these tools make possible the
collaboration among people working on the same task, or support experiences of
distribute learning [16].
The concept of “collaborativity” is also at the heart of Fulk et al. [3] studies, that
define the distinction between communality and connectivity. Tools oriented to
collaboration are the ones that present a higher level of connectivity, intended as the
ability of the tool to create connections between people.
The second characteristic by which we classify instruments for knowledge
management is multimodality. We define multimedia as “the seamless integration of
two or more media” by supporting two or more channels like text, graphic, sound and
motion, expressed in an increasingly complex order [9].
The following table presents the synthesis of the most used KM applications
(email, audio-conference, video-conference, groupware, online meeting spaces, online
discussion spaces, personal directories, text databases, intelligent data operating layer,
audio databases, video databases, multimedia simulations, multimedia encyclopaedia)
with specific levels of collaborativity and multimodality.
131
On - li n e
discussion
spaces
HIGHHIGHLOWLOW
MULTIMODALITY
COLLABORATIVITY
L
O
W
L
O
W
HIGHHIGH
Human-to-human
H u m a n- t o- n on - h um a n
On-line
me et ing sp ac es
Groupware
A u di o- co n f er en ce Video-conference
E-mail
M u lt im ed ia
simulations
Intelligent data
operatinglayer
“Mu ltim edia
en cycl op a ed ia”
Personal
Directories
Text Database A u di o D at ab ase Video Database
I
IIIII
IV
Fig. 1. A classification of instruments for Knowledge Management.
Therefore, based on these theories, we expect a relation between the nature of the
knowledge that is shared within the company and the KM tools hosted by the HR
portals.
We investigate this relation in Martinelli consulting (fantasy name). In the next
section we present the methods we used to carry out our investigation. Further, we
presents the results of our case study and discuss them in the light of the theoretical
framework we used.
3 Methods
Data and information on the KM applications of experts and users were collected,
selecting the most common applications and defining a script for each of them, in
order to have clear data on functionalities and multiple possible usages. 5 KM experts
and 5 users were asked to read these scripts and to grade the KM applications on
collaborativity and multimodality based on their characteristics from 0 to 9. Questions
about collaborativity were oriented to assess if the KM application is able to connect
and involve a great number of people, human-to-human, from one-to-one to one-to-
many. Compared to a low level of collaborativity when the KM application facilitates
interaction between human and non-human actors (i.e. databases). Questions bout
multimodality were oriented to assess if the KM application offers two or more media
(text, graphics, audio, animation, etc) in an integrated way for communicating among
people. Finally, the 10 values for each application were averaged, asking them to
decide the final value in cases of fragmented numbers (i.e. 6.5)
Knowledge dimensions could represent an important predictor in the choice of
tools (in terms of collaborativity and multimodality) for knowledge sharing. The
132
model we defined has been examinated by the analysis of a case study in order to
further investigate the ideas suggested by the theory. The research method we used is
the one of the case study as suggested by Yin [19]. This method has been selected as a
consequence of the exploratory purpose of our paper. Our analysis has been
conducted in the Italian office of Martinelli Consultants, one of the leading groups
worldwide in organization and technology consulting.
We used three techniques for data collection, so respecting the principle of
triangulation: participant-observation, qualitative interview and document analysis.
Participant-observation took place for a period of more than six months, during
which one of the authors joined the Italy Knowledge Management team of Martinelli
Consulting. In this period the researcher has been equiparated to the other members of
the team, carrying out the same activities, having the same working instruments than
his colleagues (desk, laptop, corporate e-mail, telephone), sharing the same working
spaces, and participating to all the events of the team life (meetings, work-in-
progress, training courses, presentations and so on). This helped to avoid the
“observer paradox” described by Labov [10], making the behavior of the observed
people not reactive.
A significant part of the data collection has been developed by carrying out
qualitative semi-structured interviews we made to 52 consultants in Martinelli. The
choice of the people and the groups to be interviewed was made following a
systematic approach in order to have a good representation of the entire Martinelli.
With the help of the Head of Knowledge Management Office we selected eight
groups working on the typical Martinelli business, and we intervewed people
covering all the organizational positions and different roles within the workgroups.
The contents of our interviews were related first of all to the composition of the
workgroup, to better interpret the information we obtained. A second section of the
interview protocol referred to the five or six macro tasks that the workgroup carried
out. In the same section was asked to specify the knowledge areas used to execute the
tasks that had been indicated. In the third section of the interview protocol we
investigated the four characteristics of codifiability, epistemic complexity, task
dependence and organizational competitiveness of the knowledge areas indicated by
the respondents, using the following scale: 1-3 (low), 4-6 (intermediate), 7-9 (high).
The scales were taken from Zander and Kogut’s [17] work on practices and their
transfer, and were adapted to the concept of knowledge areas that were critical in this
study. While the “codificability” construct was quite well-defined and applicable to
our context, some adaptations were necessary to measure “complexity.” We
considered “teachability” and “output observability” as part of a more expanded
knowledge complexity construct. Indeed, in this study the ease of defining cause-and-
effect relationships, and the variety of problems and solutions, are also part of the
complexity measure. Questions related to codifiability: Existing work manuals and
operating procedures describe precisely what people working in this knowledge area
actually do; most of the solutions to the problems related to this knowledge area are
described in written manuals; the outputs related to this knowledge area are well
documented. Questions related to epistemic complexity: a competitor can easily learn
how we produce outputs related to this knowledge area by analysing carefully all the
related resources used and produced for these outputs; the quality of the output
depends more on the judgment of the experts than on well defined rules; within the
practice of this knowledge area, a given action has a known outcome; the problems
133
related to this knowledge area are always different. It is not convenient to collect and
store them.
Questions related to task dependence: indicate the degree to which each
knowledge area is needed to complete each task (previously identified in the group
project). Questions related to organizational competitiveness: this knowledge area is
crucial for the success of the firm; we cannot allow that this knowledge is accessed by
external people or competitors.
The last section of the protocol was about the use of technology tools for the
working activities and the use of the corporate portal. In particular, we examined
which kinds of knowledge area were retrieved and contributed from the portal and
which ones, instead, had the project leader as an important link to external sources.
We attempted to map the habits in the acquisition and contribution of information, in
terms of “problems” and “solutions” related to a particular knowledge area.
The technique of document analysis has been adopted with the aim of
investigating the use of KM tools that consultants have at their disposal. By
classifying the documents that are on the Martinelli KM Portal it has been possible to
understand how technology supports the sharing of the different knowledge areas in
Martinelli. To operate this classification we performed a document analysis on a
sample of the documents contained in Martinelli KM portal. We analyzed 2850
documents on a total of about 8000 documents referred to the Martinelli Italian
Region. These documents have been extracted from the two most representative
sections of Martinelli KM Portal: the Global Container (fantasy name) and the
Management Section (fantasy name). The Global Container is the Martinelli general
knowledge repository, while the Management Sections is a best practice database. Of
these document we counted the frequency of appearance.
4 Findings
4.1 Knowledge Areas in use in Martinelli Consulting
From the analysis of the knowledge areas emerged by the interviews, it has been
possible to define three macro-classes of knowledge areas (KA).
A first class of knowledge area is represented by managerial knowledge areas.
This macro-class is made up of all the group management methodologies, the ability
to organize one’s work in coordination with other team members, the rules for the
interaction with other colleagues, the ability to use all the tools required by the
workgroup activities and so on.
A second macro-category of knowledge areas is represented by technological
knowledge. Seven out of the eight workgroups we examined heavily relied on this
kind of knowledge. Technological knowledge typically consists of programming
languages (ADA, C++,…), operating system source-codes, web design architectures,
technological platforms and so on.
Finally, the third knowledge macro-class coming out from the interviews is the
one of process/market knowledge. Process knowledge is intended as all the
knowledge areas that must be managed in order to implement the service to the client.
These knowledge areas can be related to the specific nature of the client or to the
134
particular kind of job delivered to it. For example, an ERP implementation will
require different knowledge areas than an Application Maintenance service.
Market knowledge, instead, is simply the information related to the specific sector
in which the client operates.
The level of organizational competitiveness for the three macro-classes was the
same. All the respondents, in fact, considered equally important the different
knowledge areas and said that not particular tensions generated when sharing any
kind of knowledge.
Far less homogeneous is the situation for codifiability. Managerial knowledge
areas, in fact, showed a high level of codifiability; technological knowledge, instead,
showed a low level of codifiability. An intermediate level of codifiability was
obtained by market/process knowledge. From the aspect of epistemic complexity, we
noticed a low level for managerial knowledge areas, an high level for technological
knowledge area and an intermediate level for market/process knowledge areas.
Finally, task dependence resulted intermediate for managerial and technological
knowledge areas, while it was very high for market/process knowledge. The results of
this assessment are reported in table 2.
Table 1. Assessment of knowledge areas.
Codifiability Epistemic complexity Task Dependence
Managerial K.A.
HIGH LOW INTERMEDIATE
Technological K.A.
LOW HIGH INTERMEDIATE
Market/process K.A.
INTERMEDIATE INTERMEDIATE HIGH
4.2 KM Tools in Martinelli and their Utilization
We analysed the main tools of the HR Portal available to the consultants for
knowledge sharing. These tools can be conducted to the general type of instruments
that we defined as repositories, that support knowledge sharing following a
distributive logic.
The document analysis conducted on the Global Container and Management
Section of Martinelli KM Portal shows the documents that more frequently appear are
the ones related to market/process knowledge areas (64%), followed by the the ones
related to managerial knowledge areas (35%). Document pertaining technological
matters, are instead totally absent, even in the three technological Boxes.
A similar composition of documents has been found in the Management Section:
also in this repository the mainly represented macro-class of knowledge is
market/process (81%), followed by managerial knowledge (19%). Documents related
to technological knowledge areas do not appear.
From the analysis of the Global Container and Management Section emerges the
complete absence of technological knowledge. This could sound quite strange in a
company that makes technology consulting its core business. This situation is
confirmed by the words of the project leader of group number five: “whenever I have
a problem related to technology I’m sure I cannot rely on the portal! Probably,
programming languages and other technological stuff are too specific to be usefully
135
shared on the portal; problems are always different and it’s not convenient to store
them. So, I usually take the telephone and make a call to a colleague expert on that
domain of knowledge”.
5 Findings
From the analysis of the Martinelli case we obtain the following findings. In first
place, we notice how the tools available on the Martinelli KM Portal can be
reconducted to only one of the four categories of KM tools we defined in our model:
the ones with low collaborativity and low multimodality. On the base of our study,
besides, we also found that, of the three knowledge areas identified, only two are
effectively shared by using the these tools. Technological knowledge, in fact, is not
available at all on Martinelli KM Portal. This suggests the presence of a relation
between knowledge dimensions and the characteristics of the technological tool used
to transfer knowledge. As a consequence of the existence of this relation, some
knowledge areas can be shared by a particular means, while others cannot. Drawing
on this we can deepen the general model of task-technology fit. In particular, within
the relation between knowledge and technology, we can observe the following
relations.
5.1 The Relation between Codifiability and KM Tools
The presence on the Martinelli KM Portal of managerial and market/process
knowledge areas reveals how a high level of codifiability requires the use of a low
collaborativity and low multimodality tools, such as the Global Container and the
Management Section.
Proposition 1
: Knowledge areas with a high level of codifiability can be effectively
shared by using low collaborativity and low multimodality tools
.
5.2 The Relation between Epistemic Complexity and KM Tools
The complete absence of technological knowledge areas on Martinelli KM Portal
shows how knowledge area. with a high level of epistemic complexity cannot be
shared by using low collaborativity and low multimodality KM tools.
On the contrary, tools with a high level of collaborativity and high level of
multimodality are indicated for this kind of knowledge.
Proposition 2
: Knowledge areas with a high level of episteimc complexity can be
effectively shared by using high collaborativity and high multimodality tools.
136
5.3 The Relation between Task Dependence and KM Tools
From the analysis developed in Martinelli Consulting we found out as knowledge
areas with a high level of task dependence can effectively be shared by using tools
with a low level of collaborativity and a low level of multimodality. Market/process
knowledge area were widely present on Martinelli KM Portal. This shows that the
“repository” logic is suitable when dealing with knowledge with a high level of task
dependence.
Proposition 3
: Knowledge areas with a high level of task dependence can be
effectively shared by using low collaborativity and intermediate multimodality tools.
6 Conclusions
Our experience shows that knowledge management initiatives can fail if they are not
included in the wider context of organizational capabilities. As Teece et al. [15] warn,
the ability to integrate, build and reconfigure internal and external competencies to
address rapidly changing environments is a matter of dynamic capabilities.
Knowledge management tools and techniques are only a partial aspect of these
mechanisms and cannot assure by themselves a sustained competitive advantage.
Our results indicate that the implementation of technologies for KM should be
accompanied by a deep understanding of the nature of the knowledge that is going to
be shared and of technology used to share it. Our paper, however, presents some
points that need to be developed. Further research could be addressed to carefully
define which is the “dominant” dimension within a knowledge area. In other words,
the three characteristics of codifiability, epistemic complexity and task dependence
could be present in the same knowledge area. It would be critical, therefore, to define
which one, of this three knowledge dimensions, has the major influence in the
decision process underlying the selection of the appropriate tool for knowledge
sharing.
Another point to be addressed by further research, could be testing these
propositions in other contexts, in order to reach a god level of statistical
generalization. What we primarily aimed in this paper has been, instead, a sound level
of theoretical generalization, consistently with the qualitative techniques we used.
References
1. Capron L. and W. Mitchell. How elephants learn new tricks: internal and external
capability sourcing in the European telecommunication industry. Academy of management
best conference paper 204 BPS: M5. (2004).
2. Firestone, J. M.. Enterprise information portals and knowledge management. Boston:
Butterworth- Heinemann. (2003).
3. Fulk, J., A. J. Flanagin, M. E. Kalman, P.R. Monge and T. Ryan. Connective and
commnunal public goods in interactive comunication systems. Communication Theory,
pp.60-87. (1996).
137
4. Goodhue, D.L. and R.L. Thompson. Task-technology fit and individual performance. MIS
Quarterly, June. (1995).
5. Gold A., A. Malhotra and A. Segars. Knowledge management, an organizational
capabilities perspective. Journal of management information systems 18:1. (2001).
6. Grandori, A. Organization Networks and Knowledge Networks. Paper presented at 16
th
Egos Colloquium, Helsinki. (1999).
7. Grant R.. Prospering in dynamically-competitive environments : organizational capability
as knowledge integration. Organization science 7:4. (1996).
8. Hansen, M. T.. The search-transfer problem: The role of weak ties in sharing knowledge
across organization subunits. Administration Science Quarterly 44 (1) 82-111. (1999).
9. Heller, R.S. and C.D. Martin,. A Media taxonomy. IEEE MultiMedia, 2, 4, 36-45. (1995).
10. Labov, W. Sociolinguistic patterns. University of Pennsylvania Press, Philadelphia. (1972).
11. Lei D., Hitt M. and Bettis R. Dynamic core competences through meta-learning and
strategic context. Journal of management 22:4. (1996).
12. McDermott R. Why information technology inspired but cannot deliver knowledge
management. California Management Review 41:40. (1999).
13. Ranft A. and M. Lord. Acquiring new technologies and capabilities: a grounded model of
acquisition implementation. Organization science 13:4. (2002).
14. Sher P. and V. Lee. Information technology as a facilitator for enhancing dynamic
capabilities through knowledge management. Information and management 41, pagg. 933-
945. (2004).
15. Teece, D.J., G. Pisano and A. Shuen. Dynamic Capabilities and Strategic Management.
Strategic Management Journal 18: 509-553. (1997).
16. Zack, M. Knowledge management and collaboration technologies. White Paper, The Lotus
Institute, Lotus Development Corporation, July. (1996).
17. Zander, U. & Kogut, B. Knowledge and the speed of the transfer and imitation of
organizational capabilities: an empirical test, Organization Science, 6: 76-92. (1995)
18. Zollo M. and S. Winter. Deliberate learning and the evolution of dynamic capabilities.
Organization science 13:3. (2002).
19. Yin, R. K. Case Study Research: Design and Methods. Vol. 5. Applied Social Research
Methods, ed. Leonard Bickman. Beverly Hills, CA: Sage. (1984).
138