The relation of embeddedness
characteristics and knowledge integration;
case studies of the Technical Expert Group
of Philips
Public version
2
The relation of embeddedness characteristics and
knowledge integration; case studies of the
Technical Expert Group of Philips
Master Thesis
Public version*
Groningen, October 17th 2013 Author:
Rachelle Jade Cartigny
rachellecartigny@hotmail.com
MSc Strategic Innovation Management University of Groningen
Supervisors at the University of Groningen: Dr. Ir. M. W. Hillen
Prof. Dr. W. A. Dolfsma Supervisors at Philips: Ing. A. de Vet
Ir. S. Woldberg
3
I.
Management summary
A firm can only retain its competitive advantage when they continuously develop and update their competencies in accordance with their environment (Collins, 1987). This research looked into how the competence or knowledge development of the Technical Expert Group (TEG) of Philips Drachten should be aligned with the requirements of their main businesses. TEG’s main task is to support the New Product Development (NPD) efforts of several innovation sites or businesses of Philips by providing their expert technical knowledge. As the business problem of TEG was that they wanted to know how to develop themselves for future demands, this research focuses on how TEG should align itself with their businesses such that the TEG will properly develop its knowledge as required by the businesses. From the knowledge that lack of familiarity with pre-existing knowledge and practices, and troublesome communications channels are factors hampering sharing of knowledge and aligned knowledge development (Adenfelt & Lagerstrom, 2008), this research will incorporate the embeddedness concept in researching the knowledge integration. The research question set out to answer by this paper is:
“How does the embeddedness of the TEG with its businesses facilitate knowledge integration?”
Embeddedness describes that all behaviors are embedded within a network of relations (Granovetter, 1985) and functions as an exchange system which can offer unique opportunities with higher success rates (Uzzi, 1996). It is proposed in this paper that to get a proper insight of the embeddedness and how it influences knowledge integration, one should consider the strength of ties (Granovetter, 1973; Ruef, 2002), the scope of ties (Burt, 1992), and the formal/informal alignment of ties (Nee & Ingram, 1998). It is also proposed that for knowledge integration for NPD best results can be expected when having strong, diverse ties, with formal/informal alignment. To look into this a network analysis was performed which mapped the relations of TEG with its main businesses. From this network analysis a selection of four businesses was selected because of their distinct embeddedness patterns.
4
II.
Preface
For the final part of my study I got the great opportunity to go and write my master thesis at the Technical Expert Group (TEG) of Philips. As a student of Strategic Innovation
Management this internship provided me with very interesting insight into the front end of the product development of Philips and into the role the technical support institution delivered in supporting this difficult task.
This challenging and practical working environment within which I was able to write my report greatly motivated me during the ups and downs of writing the report and helped me keep my goal and visions clear for myself. Therefore I consider myself very lucky to have worked in such an inspiring and motivating environment. Therefore I would like to thank Dr. Ir. Hillen for setting up the initial contact. I also like to thank Dr. Ir. Hillen for consulting me on the academic side of the report as well as helping me translate academic insights into practical insights and vice versa. Also, I would like to thank Prof. Dr. Dolfsma for sharing his extensive experience on some of the difficult topics as network analysis and guiding me through the correct usage of these methods and concepts.
Of course as is promoted throughout this report my actions and thoughts are well embedded and therefore directed within the network of key figures within both Philips as well as the University of Groningen. Therefore the results of this report should be credited to them. I would like to thank my two supervisors at Philips, Arno de Vet and Sven Woldberg for guiding me throughout the process; for sharing their ideas and visions, for giving me the chance to pitch my ideas and visions; and for guiding the process of aligning all this into the report you currently have in front of you. Also, I like to thank all colleagues of TEG for cooperating in the network analysis and for providing me with extensive and honest opinions on all kinds of issues concerning the organization of TEG. Then I would like to thank
5
III. List of figures and tables
Figure 1: The conceptual model………15
Figure 2: Distinction of strength and scope of ties………...……….17
Figure 3: Network overview of the TEG with their businesses………...……….22
Figure 4: Network overview of the TEG; with filter only showing contact persons at the businesses with at least three contact persons………….………..24
Table 1: Network analysis factors and its scales……….………..17
Table 2: Constructs used to measure embeddedness and the operational variables used in the network analysis and the interviews………..20
Table 3: Correlations of relational characteristics……….………...21
Table 4: Average characteristics for all businesses……….…..23
Table 5: Average characteristics per TEG specialization………..24
Table 6: Overview on number of TEG specializations with whom the categories have contact and nr of relations with the TEG management………..………..25
Table 7: Overview on embeddedness scores of selected businesses from the network analysis………...………27
6
Inhoud
I. Manamgement summary……….3
II. Preface……….………4
III. List of figures and Tables………5
1. Introduction ... 7 2. Problem definition ... 8 3. Research question ... 9 4. Literature review ... 9 4.1 Embeddedness ... 9 4.2 Knowledge Integration ... 12 5. Conceptual framework ... 15 6. Method ... 16
6.1 Network analysis method ... 16
6.2 Interviews method ... 18
6.3 Overview on network analysis and interviews ... 19
7. Analysis and results ... 21
7.1 Network analysis ... 21
7.1.1 Analysis strong and weak ties ... 23
7.1.2 Analysis scope of ties ... 24
7.2 Interviews analysis ... 27
7.3 Linking network analysis and the interviews ... 31
8. Results ... 34
9. Discussion ... 36
Bibliografie ... 37
Appendix A: Interview format ... 43
7
1. Introduction
“It is not just about information sharing, it is about information sharing and inspiring each other.” (Interviewee, 2013)
8
2. Problem definition
Today’s working conditions require ongoing adaptation of employees and organizations to their environments new demands and competitive opportunities (Ericsson, et al., 2009). It is not enough to have competent employees as other firms can also hire competent people and since competence is not a static concept. A firm can only retain its competitive advantage when they continuously develop and update their competencies in accordance with their environment (Collins, 1987). The TEG department has some of the most skilled employees of Philips but has to keep working on the employees competences. The Technical Expert Group can be recognized as an centre of excellence as it conforms to the concept’s two conditions: (1) It is ‘excellent,’ in the sense that the department has some distinct competences, being their unique knowledge. And also, (2) the TEG is a ‘centre,’ in the sense of using these competences for other corporate units (Holm & Pederson, 2000). The managers of the TEG were struggling with the question:
“We are doing quite well currently, but how can we ensure we will keep doing well in the future and what competences should we develop for future demands?”
Defining however what competences need to be worked on involves many actors, uncertainties and choices. Another issue this question imposes is discussed by many authors discussing ambidexterity, and concerns how to both efficiently work on current matters while being flexible and be prepared for future changes and demands (Eisenhardt, Furr, & Bingham, 2010). As the TEG is a technical support institution, their main responsibility is to provide technical support to the different innovation sites, also called businesses, of Philips. Therefore to be prepared for future demands they have to align their competences/knowledge to what is required by the businesses. The direction of the New Product Development (NPD) of these businesses is defined by their department roadmaps, which describes the plans of their future for two up to five years. These roadmaps are composed from different technical and consumer market insights. These insights are either required by themselves from their previous experience with products, from external institutions such as universities, or through the marketing department or any other department of Philips. For the proper execution of these plans the businesses need to integrate competences and knowledge from different knowledge areas, such as technical and managerial/market, and from different actors either within or outside their own department/company. The TEG is an institution that can provide certain technical competences to the different business groups of Philips to help them in the execution of their NPD, and can also have a role in signalizing new technical developments either from outside the company or by leveraging developments of other business groups within Philips.
9
3. Research question
The request from the TEG department for this research was to look into the competence development of the department and the positioning of the TEG pertaining to their main businesses. From a preliminary literature study it is seen that the context in which a person/department should function prescribes the required competence development (Boyatzis, 1982; Collins, 1987). In the problem definition was described that the different innovation sites of Philips developed roadmaps based on a selection of market developments and that support of TEG should help to execute certain steps of the NPD. What competence or knowledge should be developed by the TEG thus depends on the relevant trends registered by the different innovation sites. It is therefore very important for TEG to be able to align the knowledge requirements of their businesses with their knowledge bases. Furthermore, as it is TEG’s main task to support the innovation sites in their NPD and sometimes to point out interesting technological developments, the success of the TEG depends on the extent to which they are able to transfer and integrate the relevant knowledge to the businesses. From these two considerations; TEG needs to incorporate the trends marked by their main businesses and TEG needs to be able to transfer the relevant knowledge related to these new developments, we can see the importance of effective knowledge integration in alignment with collaborative relations.
An important notion of von Krogh et al. (2000) stated that the ultimate success of knowledge creation depends on how organizational members relate through the different steps of the process. From the problem definition it could also be recognized that the alignment of the knowledge bases of the TEG and the businesses is from upmost importance for a successful cooperation on the short as well as the long term. This paper will incorporate these notions by including the embeddedness, which describes that all behavior is embedded within a network of relations (Granovetter, 1985). The paper will research how the TEG is positioned within a network of relations with its businesses. For the second part, it will be researched how this embeddedness influences knowledge integration between the TEG and its businesses. The research question answered by this paper will be:
“How does the embeddedness of the TEG with its businesses facilitate knowledge integration?”
4. Literature review
This chapter provides an overview of the main literature regarding the two main concepts used in this paper, namely; embeddedness and knowledge integration. As the first part of the research will map the characteristics of the relations between TEG and the businesses, we will first discuss embeddedness. For the second part of the research we will research how the embeddedness influences knowledge integration, hence, the second concept discussed here will be knowledge integration.
4.1 Embeddedness
10
The Problem of Embeddedness, 1985, p. 504). According to Granovetter (1985) this ubiquity of embeddedness accounts for much of the order that is found in both markets and in firms. A comprehensive study by Uzzi (1996) on the unique features, functions and sources of embeddedness revealed the concept of an exchange system with unique opportunities offering higher success rates to firms organized in networks. It has been found by many authors that the embeddedness of the company within its network of interfirm relationships matters for its innovative as well as its economic performance (i.a. Hagedoorn, 2002, Ahuja, 2000). A multinational corporations (MNC) can be observed to be a network of product, capital and knowledge transactions (Calvet, 1981; Ghoshal & Bartlett, 1990). This view implicates that the embeddedness concept also upholds within a MNC. There is also growing consensus on the potential of intrafirm relationship embeddedness on aspects like innovative performance (Nooteboom, 1992) and competence development (Andersson, Forsgren, & Holm, 2001). When considering the concept embeddedness in more detail, literature provides two specified forms of embeddedness which together form the concept of social capital: structural embeddedness and relational embeddedness (Moran, Structural vs. Relational Embeddedness: Social Capital and Managerial Performance, 2005). Structural embeddedness is concerned with the aggregate configuration of relationships in some structural form, while relational embeddedness is concerned with the concrete personal relationships themselves (Granovetter, Economic Action and Social Structure: The Problem of Embeddedness, 1985). When considering the former, issues like structural holes and the configuration of the relationships are discussed. Relational embeddedness is characterized in terms of the strength of their social ties, their level of trust, and their shared common processes and values (Kale, Singh, & Perlmutter, 2000). While both forms of embeddedness contribute to the performance of an organization, structural embeddedness has more explanatory power considering execution-oriented tasks, and relational embeddedness has more explanatory power considering oriented tasks (Moran, 2005). Because TEG is mainly involved in innovation-oriented tasks, it is most important to consider relational embeddedness. Because this paper considers the embeddedness of intra-organizational groups, the difference in common processes and values are expected to be less, and therefore also be less influential, then they were in studies which considered interorganizational relations.
When considering whether an organization is well embedded within a certain context we can distinguish a couple of characteristics. The most used characteristic is the strength of the tie, which is conceptualized by Granovetter (1973:1361) as: ‘a probably linear combination of the amount of time, emotional intensity, intimacy, and the reciprocal services’. This definition would assume that the amount of time invested is linearly related to the amount of information exchanged, an important notion for this paper to consider as this paper looks into knowledge integration. However, as we can all recognize from our own lives, the amount of time spend with persons is not always linear related to the strength of the tie. Depending on whether there is a ‘click’ or whether you have more or less in common interests or knowledge influences how quickly you are able to form a relation with another actor. Therefore we also need to take other attributes into account when mapping the embeddedness.
11
a willingness to work on mutual problem-solving (Uzzi, 1997). Lane et al. (2004, p.1141) add that trust ‘influences both the extent of knowledge exchanged and the efficiency with which it is exchanged.’
On the influence of the strength of ties authors differ. A view that is well represented in literature is characterized by the following: as stronger ties may provide a better means to get more conformity (Ruef, 2002), weaker ties have a greater propensity to link disconnected groups, thereby are more likely to result in more unique and therefore more valuable information (Burt, 1992). A study of Tiwana (2008) finds that strong ties complement weak ties in enhancing the knowledge integration and indirectly therefore enhance alliance ambidexterity. This study hints that configuring an efficient balance of strong and weak ties may contribute to an organization’s ability to plan for short-term as well as long term successes. Intrinsically held by the concept of weak and strong ties used by these authors is the assumption that the strength of the tie is connected with the scope of the ties For this paper we will consider the strength of the tie and the scope of the tie, as being two separate characteristics. As it is recognized that these are indeed separate characteristics, but linked because of a greater propensity of stronger ties to be convergent, and a greater propensity of weak ties to be more divergent. However the best combination would be to have strong, divergent ties (Jacobs, 1965). Jacobs (1965) recognized the importance of crosscutting, strong personal ties developed over time, providing a basis for cooperation, trust and collective action. Also, as this research will focus on the portfolio level ties versus the dyadic ties used in most other studies, the likelihood of strong and divergent ties is better possible through the relations of multiple individuals. An important contribution furthermore from applying embeddedness is the notion that informal constraints stemming from personal relationships may be critical to enforce the rules of the game (Nee & Ingram, 1998). This suggests that the informal networks constituted within the organization must be aligned with the formal rules of the organization in order for them to enforce each other’s effectiveness.
Previous literature on the topic of embeddedness was inconsistent concerning what characteristics should be taken into account when defining the embeddedness. Literature was also inconsistent as to what measure of a characteristic, for example weak or strong ties or a combination, would provide the best results concerning the innovative oriented tasks and execution oriented tasks. For the cases discussed in this paper we will describe the portfolio of ties rather than the specific dyadic ties, and therefore agree with Jacobs (1965) that the strength of the tie is not automatically related to the scope of the tie and that the best results for the NPD would be accomplished by having strong and diverse ties. Added to this is the insight from the study of Nee & Ingram (1998) that we furthermore also have to take into account the formal and informal alignment as this may enforce or diminish the positive/negative effects from the previous two characteristics. Considering all this, it is suggested that the best results for the NPD would be accomplished by having strong and diverse ties with an formal and informal alignment that would reinforce this. The more the relations between two departments are according to these measures, the higher we consider the degree of embeddedness to be. From the literature discussed in this paragraph we come to the following propositions concerning embeddedness:
Proposition 1: The degree of embeddedness of two intra-organizational groups is considered higher when the relation’s characteristics consist of:
- Strong ties (versus weak ties) - Diverse ties (versus convers ties)
12
Applying the embeddedness concept in a network of relationships is suggested to be able to explain why some organizations achieve higher results than others, both in terms of market performance and their competence development within the MNC (Andersson, Forsgren, & Holm, 2001). As there is little doubt that embeddedness influences the communication patterns between different organizational groups (Granovetter, 1973; Kerckhoff, Back, & Miller, 1965) the following paragraph will discuss knowledge integration.
4.2 Knowledge Integration
‘Knowledge integration is a vital part of the dynamics of the innovation process’ (Berggren, Bergek, Bengtsson, Hobday, & Soderlund, 2011, p. 1). When considering complex innovation problems, integration and synthesis of diverse complementary knowledge is often needed to come to a solution (Obstfeld, 2005). The concept has also been described as realized absorptive capacity (Zahra & George, 2002). Knowledge integration is defined by Tiwana (2008) as ‘the process of jointly applying specialized knowledge held by various alliance partners’. For this purpose we will use the definition of knowledge integration formulated by Huang & Newell (2000) as being: ‘an ongoing collective process of constructing, articulating and redefining shared beliefs through the social interaction of organizational members’. This definition, which will be used in this paper, highlights that knowledge integration is embedded in a network of relations and has a dynamic character.
Like in this paper, much of the research regarding integration of different kinds of specialized knowledge is done in the context of NPD (Clark & Fujimoto, 1991; Carlile, 2002; Becker & Zirpoli, 2003). As noticed by Leenders et al. (2003), cross-fertilization and interaction of ideas beyond the boundaries of groups can be essential for NPD. Clark & Fujimoto (1991) state on this subject that NPD involves particularly wide-ranging integration as it requires a wide range of cross-functional integration. While some of the innovations stem from applying new knowledge, others stem from reconfiguring existing knowledge, a process that leads to ‘architectural innovations’ (Henderson & Clark, 1990). These architectural innovations provide the opportunity to further explore the potential of the technology and the capability to do this is defined as ‘combinative capabilities’ (Kogut & Zander, 1992).
Grant’s (1996) theory on knowledge integration synthesizes two streams of literature on knowledge development. The first stream is concerned with activities such as knowledge creation, sharing, acquirement and appliance, while the second literature stream is concerned with the need to externalize tacit knowledge to make sure it is easy to access (Huang & Newell, 2003). One important contributor to this second literature stream is Nonaka (1996), who recognized the transformation of tacit knowledge to explicit knowledge and vice versa; and the distribution of knowledge among the individuals, groups, and on organizational and inter-organizational level. Grant emphasizes the importance of knowledge integration by stating that ‘if the strategically most important resource of the firm is knowledge, and if knowledge resides in specialized form among individual organizational members, the essence of organizational capability is the integration of individuals’ specialized knowledge’ (Grant, 1996). The importance of knowledge is directly related to the positive relation of firm-level knowledge assets with level innovation, and the subsequent positive relation of firm-level innovation with firm-firm-level performance (Thornhill, 2006).
13
determining features to whether or not relationships are likely or unlikely to contribute to the knowledge sharing are (1) knowing what the other knows, (2) the likelihood of getting access to that person, (3) the willingness of the other actor to engage in problem solving rather than information dumping, (4) a certain amount of safety that promotes learning and creativity (Cross, Parker, Prusak, & Borgatti, 2001). From these features, ‘knowing what the other knows’, seems most important for this study as this might be greatly differ from one business to another. Getting access and a certain amount of safety will be less of an issue as this paper discusses intra-organizational relations. Furthermore as the businesses ask for support from the TEG and as it is TEG’s main task to support the businesses, it is expected that both parties are willing to participate in problem solving. From a wide range of literature it can be found that the capability of the organization for creating and sharing knowledge stems from a wide scope of factors, among which is the facilitation for the creation, translation and transfer of tacit knowledge (Nonaka & Takeuchi, 1995; Kogut & Zander, 1993); the structure of the organization, through which individuals communicate, get coordinated and cooperate (Connor & Prahalad, 1996); and the nature of the social community of the organization (Kogut & Zander, 1992).
The Social Capital Theory provides a paradigm that articulates that the network of relationships in which an individual is positioned and the resources which are embedded in their network have a strong influence on the degree to which interpersonal knowledge sharing occurs (Nahapiet & Ghoshal, 1998). Nahapiet and Ghoshal (1998) furthermore find that: ‘(1) social capital facilitates the creation of new intellectual capital; (2) organizations, as institutional settings, are conducive to the development of high levels of social capital; and (3) it is because of their more dense social capital that firms, within certain limits, have an advantage over markets in creating and sharing intellectual capital. According to these statements on social capital it can be argued that the primary goal of any firm is to create the essential social capital needed to be able to leverage its resources into the creation of new intellectual property.
A publication of Lang (2004) argues that social capital and social context enable knowledge integration and proposes a distinction as to what different social contexts and what social capital enables what kind of knowledge integration. She proposes that a combination of high and low social embeddedness, previously referred to as weak and strong ties by a string of authors (Ruef, 2002; Marsden & Campbell, 1984; Burt, 1992; Tiwana, 2008), together with loose or tight coupling, which reports on the interdependence of the organizations on each other, resulted in four kinds of knowledge integration, namely: frontier, incremental, instrumental and combinative knowledge integration (Lang, 2004). Dhanaraj et al. (2004) also found that an increase in tie strength, trust, and a shared system all significantly influence the tacit knowledge transfer, while it has an insignificant influence on explicit knowledge transfer. As the degree of codification increases with the maturity of a research (Saviotti, 1998), this notion can be linked with a result found by Haahtela et al. (2003) that enhanced social interaction is needed for the forming of a shared vision which is needed in the early stages of the NPD.
14
is risk-seeking, variation-seeking and experimentation oriented, exploitation is variety reducing and experimentation oriented (March, 1991). Exploitation activities require search for proximate, familiar, mature or current knowledge; and exploration activities require search for distant or remote and unfamiliar knowledge activities (Ahuja & Katila, 2004; Benner & Tushman, 2002; Katila & Ahuja, 2002). An important study of Miller et al. (2006) linked tacit knowledge and interpersonal learning to March’s exploration-exploitation model and found that local search associated with converse ties is more consistent with organization-level exploration, while distant search associated with divers ties search would aggregate to system wide exploitation.
Thus considering that distant knowledge from a diversity of ties is associated with exploration and that tacit knowledge is more related to exploration (Miller, Zhao, & Calantone, 2006), it can be expected that a high degree of embeddedness associated with more tacit and divers knowledge integration will lead to more explorative activities and learning. In a similar way, considering that proximate or familiar and explicit knowledge is associated with exploitation, it can be expected that a low degree of embeddedness associated with explicit and convergent knowledge integration will lead to more exploitative activities and learning. From the literature previously discussed it could been perceived that both exploration and exploitation are needed for a sustained performance, and therefore we should consider contributions of both forms of knowledge integration and their associated degree of embeddedness.
In a similar way as Lang (2004) proposes, this paper recognizes that different forms of embeddedness with different characteristics as to the strength of the ties, the scope of the ties, complemented with the formal and informal alignment of the organizations would facilitate different forms of knowledge integration. Divergent ties have a greater propensity to enable the integration of very dissimilar knowledge and could therefore avoid lock-in while providing the opportunity to access, assess and appropriate emergent knowledge (Lang, 2004). Convergent ties will be less likely to enable the transfer of dissimilar or new knowledge but do increase the chance of transferring the information because the information held by similar sources then transfers through different channels. Also, local search enables to choose the best from similar alternatives (Miller, Zhao, & Calantone, 2006). Strong ties would also increase the chances of transferring the information, and as found by the research of Dhanaraj et al. (2004) it would certainly increase the level of tacit knowledge transfer as the level of trust, the amount of time spend together, and the amount of information exchanged on certain topics increases. It is also found by Danaraj et al. that the amount of explicit knowledge transferred does not significantly change when there are only weak ties.
Taking all this in consideration, the following propositions are developed:
Proposition 2a: A high degree of embeddedness, characterized by strong and divergent ties, enforced by a formal/informal alignment, would better facilitate the knowledge integration of tacit and divergent knowledge than a low degree of embeddedness would.
15
5. Conceptual framework
From considering all literature and the working process of TEG the following conceptual framework was configured.
Figure 1: The conceptual model Figure 1: Conceptual model
The conceptual model used for his paper consists of two main areas, represented by the two grey squares; embeddedness and knowledge integration. The area embeddedness can be divided among the degrees of embeddedness, from low to high, but to make differentiation more recognizable and easier to understand this paper just distinguishes ‘high degree of embeddedness’ and ‘low degree of embeddedness’ which are both characterized by the scope of ties, the strength of the ties and the formal-informal alignment as discussed in paragraph 4.1. In the second area, knowledge integration, it can be seen that ‘high degree of embeddedness will more likely lead to more tacit and divers knowledge integration, while ‘low degree of embeddedness will more likely lead to explicit and convergent knowledge integration, as also discussed in paragraph 4.2. Because tacit and divers knowledge integration is more associated with explorative activities, this paper grouped this as ‘explorative knowledge integration’. Because explicit and proximate/familiar knowledge integration is more associated with exploitative activities, this is grouped as ‘exploitative knowledge integration’ (Ahuja & Katila, 2004; Miller, Zhao, & Calantone, 2006). As described by Eisenhardt et al. (2010), there should be a balance of efficiency and flexibility, managed through a balance of exploitative and explorative activities and learning. Therefore both types of knowledge integration are considered in this paper. Like will be discussed in the methods chapter, the embeddedness will be mapped by using a network analysis and the knowledge integration will be researched having interviews.
High degree of embeddedness
- strong ties
- divergent ties
- formal/informal alignment
Low degree of embeddedness
16
6. Method
The research for this paper is done at the Technology Expert Group of Philips Drachten over the course of five months. It is an explorative research requiring quantitative and qualitative data, which is well available as the TEG department has provided me access to their working place, enabling me to observe how the employees interact and work. Also, because of limited previous research on the exact relation of the main concepts and because of the suggestion of Lang (2004) for future research to differentiate several types of knowledge integration, an exploratory research seems best fit. Information is required through attending meetings, through the possibility to observe them work with each other, a network analysis survey, and through interviews.
As mentioned before, the first part of the research will map the embeddedness characteristics of the relations of the TEG. The following paragraph will cover the method used for network analysis. The second part of the research will look into the knowledge integration between TEG and its main businesses and their experiences when working with each other. Paragraph 6.2 will cover the method for the interviews used for this part. Finally, paragraph 6.3 provides an overview on the two methods used and their connections.
6.1 Network analysis method
The Social Network Analysis (SNA) used in this research is a egocentric network approach which was introduced by a group of authors among which was B.A. Radcliff-Brown. The approach comprises the people that an individual or a certain group knows (Friedman & Aral, 2001). For the definition of networks this paper uses the concept proposed by Mitchell (1979): “a specific set of linkages among a defined set of persons, with the additional property that the characteristics of these linkages as a whole may be used to interpret the social behavior of the persons involved”. Network analysis is concerned with the structure and patterning of the relationships within a defined set of people/groups and tries to identify both their causes and consequences (Tichy, Tushman, & Fombrun, 1979)
17 Rated factors for each
provided contact
Scales
Length of contact 1: contact for less than a year 2: contact for 1-2 years
3: contact for more than 3 years
Frequency of contact 1: contact once every few months 2: contact every few weeks
3: contact every other week or more often
Short term information importance/long term information
importance/effectiveness
1-5, from low to high
Table 1: Network analysis factors and its scales
The data collected thus provides data on each relation between a TEG employee and a business contact person. But more interesting is to combine all these individual relations and consider the network of relations between the TEG department and each of the businesses. From this it is possible to see the amount of individual relations between the TEG and the businesses. At this point it is important to make a clear distinction between the concepts of ‘scope’ and ‘strength’ of ties as used for this research. Strength of the ties is concerned with the relational embeddedness (Kale, Singh, & Perlmutter, 2000) and thus with the concrete relationships themselves (Granovetter, 1985). All relations were rated on ‘length’ and frequency’ of contact and ‘short and long term information importance’, which together provide information on the strength of the tie, or the relational embeddedness (Granovetter, 1985). The scope of ties is concerned with the structural embeddedness of the network (Moran, 2005), which is depicted by the network of ties in the figure below. In this figure the blue dots represent individuals working at TEG, the red dots represent individuals within a certain business, the purple dots represent individuals in another business, and the lines represent the relations. From this figure it can be seen that there are seven relations depicted in total between the TEG and the business. The individual relations will be described by the strength of the ties, and in this figure a thick line depicts a strong tie, characterized by regular contact over a longer period of time and important for the knowledge exchange. The scope of ties describes the configuration, depicted by the patterns of ties. Then what can be seen in this example is that the TEG has four relations with two individuals, with the ‘red’ business, and three relations with three individuals with the ‘purple’ business.
18
Thus when we consider the example depicted in figure 2, it can be recognized that the scope of ties, from the perspective of the TEG, is wider for the ‘purple’ business than for the ‘red’ business, which means that TEG can provide and receive their information through three persons at that business. For the ‘red’ business we see that one contact person was mentioned three times, this is then probably an important contact person to the TEG. You can see that this person has a wide scope of possible channels through which this person may provide or receive his/her information. This may be important to consider when the three individuals at the TEG considered would be from different specializations, which will mean they have dissimilar knowledge bases. What furthermore should be considered is that some businesses, are much bigger than others and may have more contacts, but the percentage of employees with whom there is contact may be much lower than for smaller, newer businesses.
Finally, from databases of Philips we know all the functional positions of the persons in the network analysis. This enables observation on the amount of contact the formal management of TEG has with the different businesses, as well as the amount of contact the management of the businesses has with the TEG employees and management. Like mentioned before the ratings on long term information importance for the managers and the employees is also interesting to connect to the functional positions of the individuals to gain insight into the formal/informal alignment; thus whether long term information is mainly distributed through managers or through employees. And finally, it would be interesting to consider whether managers rated their relations in a similar way as employees did. If managers have a different perspective on their relations, the knowledge exchange and the results of the collaboration, then this should be accounted for during the next stage of the research: the interviews.
6.2 Interviews method
From the results of the network analysis we choose only to go further with the businesses four businesses with interesting embeddedness patterns. What these results were and what reasoning for proceeding with these departments are behind this will be discussed in paragraph 7.1. From each of the departments one manager and one employee were interviewed to get information from multiple perspectives. The manager interviewed was usually the most mentioned person who scored high on long term information importance. The employees have comparable functions, usually system architects or project leaders, and were mentioned by multiple TEG employees and usually scored somewhat higher on the short term information importance. The interviews covered two general questions after which we discussed four topics related to this research. The purpose of this qualitative part of the research is to see how the knowledge integration is linked to the embeddedness scores we found in the network analysis.
19
in literature furthermore is the level of trust between two parties. As TEG and the businesses are part of the same company, this paper does not actively consider this aspect. One of the main aspects of knowledge integration considered for this paper is whether or not the knowledge exchange is mainly explicit or implicit. Thus does the information exchange mainly occur through written material or through verbal communication?
The first general question discussed for what issues the businesses would contact a technical support institution, and the second question concerned with which specializations and/or TEG employees they have contact. Then moving on to the first topic, we discussed what factors they took into account when considering collaborating with a technical support institution and what distinguished the TEG from external technical support institutes. This topic was discussed to get a better understanding of what was expected from the relation and to provide a clearer scope on what knowledge integration would be required. The second topic concerned the information sharing. The amount of information, the process through which it was shared from the business to TEG, the information known by TEG, the competences of their employees and how proactively the TEG employees were in approaching them with technical developments and other information, were all covered during this part of the interview. This part provides information on the process of knowledge integration. The third part discusses how the relation between the interviewee and its represented business evolved, both on an informal as well as an formal level. This part was included to see what role formal and informal mechanisms have throughout the relation. Then finally the last part comprised the experienced results by the interviewee from collaborating with the TEG. The interview questions can be found in appendix A.
6.3 Overview on network analysis and interviews
This paper links the concepts of embeddedness with knowledge integration by doing a network analysis and by using interviews. It can therefore also be considered that this paper pursues mixed method, an increasingly popular research method which combines the quantitative and the qualitative approach in an attempt to use both these approaches their strengths and minimize their weaknesses (Johnson & Onwuegbuzie, 2004). A schematic overview on all the constructs used and its operational variables in the network analysis and the interviews can be found in table 2.
20
Constructs used to measure embeddedness
Operational variable in network analysis
Operational variable in interviews
Strong/weak ties - length of contact - frequency of contact - importance of contact for
short/long term information
- importance of contact for long term information - Effectiveness of the contact
- experience working together - experienced expert level of
TEG employees
- experienced knowledge level on their business of TEG employee
- Amount and detail of
information shared both short term as long term
- Proactive attitude TEG employee (maybe result) - Experienced effectiveness
collaboration Divergent/convergent ties - Contact scope considered
from different levels - relative amount of contacts
considering business size/budget spend on TEG
- Applying solutions from other businesses to their business - Difference in working with
different TEG specializations Formal/informal ties - Contact with formal
management
- importance of contact for short term information - importance of contact for
long term information
- Contact with formal management
- Discuss what issues with formal management
- Sharing of planning - Involvement in planning
21
7. Analysis and results
7.1 Network analysis
All employees except for a new employee, the secretary, and some flex workers were included in the network analysis. Therefore the network analysis includes the responses of 28 TEG employees. These reported a total of 478 relations with 255 individuals from outside the TEG. This difference comes from the fact that a lot of these individuals outside the TEG were mentioned by several employees of the TEG. An overall overview on the network of the TEG can be seen in figure 3.
CENSORED*
Figure 3: Network overview of the TEG with their businesses.
*Censored public version, please contact author for more information.
In the middle of the figure the red dots indicate the TEG employees. The lines indicate the relationships they have with individuals from other businesses within Philips. All these businesses are also grouped, this is indicated by the different colors and squares in which they are placed. Also abbreviations on their department names are depicted. The light blue group in the upper right corner is the only exception as this group depicts some very small businesses and some cooperation partners of TEG, such as the research department of Philips. This group will be left out of the analysis for the rest of this research as I chose to focus on the analysis of the relationships TEG has with its main businesses.
As mentioned in the previous paragraph every relation was also rated on the ‘frequency’ and ‘length’ of the contact, ‘short term information importance’, ‘long term information importance’ and ‘effectiveness’. This data enables us to filter on the relations of these characteristics. This and other analyzes will be discussed in the next paragraph.
22 Correlations TEG employees TEG MAN Difference in correlation (z)
correlation short info importance with effectiveness 0.613 0.238 3.45 correlation long info importance with effectiveness 0.447 0.193 2.09 correlation frequency with effectiveness 0.343 0.579 -2.22 correlation length with effectiveness 0.154 0.263 Not significant correlation short with long term information
importance 0.53 0.356 1.59
correlation frequency with short term information
importance 0.283 -0.065 2.61
correlation frequency with long term information
importance 0.13 0.034 Not significant correlation length with short term information
importance 0.023 0.111 Not significant correlation length with long term information
importance 0.061 -0.064 Not significant
Table 3: Correlations of relational characteristics
As a score from 0 till 0.3 is a weak positive relation, a score of 0.3 till 0.7 indicates a moderate positive relation and a score of 0.7 up to 1.0 indicates a strong relation this first analysis could not find a strong relation. The strongest correlation was found for the ‘short term information importance’ with ‘effectiveness’ of the collaboration for the TEG employees. This may indicate that according to the employees the effectiveness of single relations would be greatly reflected by the short term information importance.
Important to consider when doing the interviews and regarding the formal/informal alignment, are the great differences of correlation results for the TEG employees and the TEG managers. This would suggest that managers and employees experience a very different result from working together for a longer time or with a higher frequency, and that they have a different view on the influences of for example ‘short term information importance on ‘effectiveness’. The biggest difference in correlation for TEG management and the TEG employees corresponds to the correlation of ‘short term information importance’ with ‘effectiveness’, with a significant z-score of 3.45 (>98%).
23 7.1.1 Analysis strong and weak ties
For the analysis concerning the strength of the ties we can look at the characteristics ‘frequency’, ‘length’, ‘short term information importance’ and ‘long term information importance’. Table 4 provides an overview on the averages scores of all categories on these characteristics. Business Frequency (1-3) Length (1-3) Short (1-5) Long (1-5) Effectiveness (1-5) Business A 1.59 2.95 4.09 4.18 4.45 Business B 2.02 2.76 3.86 3.40 4.18 Business C 2.10 2.44 3.84 3.22 4.05 Business D 1.50 2.32 3.21 3.05 3.47 Business E 1.68 2.23 4.14 3.91 3.95 Business F 1.72 2.89 3.78 3.67 3.67 Business G 2.14 2.38 3.76 3.52 3.95 Business H 1.80 2.75 3.60 2.75 3.75 Business I 1.58 2.88 3.58 3.35 3.98 Business J 1.60 2.80 3.30 3.30 3.30 Business K 1.60 2.50 3.50 4.00 2.83 Average 1.76 2.63 3.70 3.49 3.78
Table 4: Average characteristics for all businesses
From table 4 it can be observed that Business A scores highest on almost all characteristics and Business B also scores pretty high on everything. Business C stands out as it does score high on effectiveness and short term information importance, but has one of the lowest scores on long term information importance. This might suggest a vulnerability in the long term. The business with the lowest scores from these characteristics is Business D. According this data the Business A business thus has the strongest relation and the Business D business has the weakest relation with TEG from the businesses considered. The same can be done for the different TEG specializations that TEG offers. The table below provides an insight into the average scores of the characteristics for all TEG specializations. However, while the businesses where all scored by at least several TEG specializations and by quite a proportion of the whole TEG crew, the scores the specialization get are only scored by an average of four employees per category. Also, the scores provided not represent what the TEG specializations were rated but what they on average rate their own relations. Therefore these scores are less reliable as the scores for the specializations are.
Specialization Frequency Length Short Long Effectiveness
24
Specialization F 1.84 2.29 3.89 3.31 3.5 Specialization G 1.45 2.38 3.23 2.57 3.29 TEG MAN 1.77 2.65 3.16 3.88 3.51 P-value normal test 0.912 0.922 0.176 0.53 0.155
Table 5: Average characteristics per TEG specialization
From the table it can be seen that Specialization B and E score quite high on all the characteristics, while for example Specialization G scores lowest on almost all characteristics. However this may very well be biased because of one or two persons within a specialization which provide more positive or negative ratings overall.
7.1.2 Analysis scope of ties
The figure below portraits all contact persons at the different categories with at least three contact persons at the TEG.
CENSORED*
Figure 4: Network overview of the TEG; with filter only showing contact persons at the businesses with at least three contact persons.
*Censored public version, please contact author for more information.
25 Business Nr of relations Nr of contact persons Average number of relations per contact person Nr of specializations of TEG with whom they have contacts
Number of relations with TEG management Business A 36 11 3.27 7 5 Business B 67 29 2.28 6 8 Business C 55 33 1.67 7 6 Business D 22 12 1.83 3 6 Business E 46 27 1.7 7 6 Business F 28 18 1.56 6 4 Business G 21 4 5.25 5 3 Business H 18 13 1.38 4 1 Business I 97 49 1.98 7 9 Business J 16 9 1.78 4 2 Business K 11 5 2 3 4
Table 6: Overview on number of TEG specializations with whom the categories have contact and nr of relations with the TEG management.
From table 6 it can be seen that some businesses have considerably more relations and contact persons than others. These differences have several explanations; some businesses are just bigger than others, others businesses are more involved in projects in need of the specializations TEG offers, and finally, some businesses worked with TEG for many years while collaboration with other businesses is still young. For example, Business D is still quite a new business and the relation with this business is still developing. The ‘average number of relations per contact person’ is interesting to consider as the higher this number this, the relatively more broad the scope of ties from the business perspective is, but the relatively narrower the scope is from TEG’s perspective. Especially Business G and A have high averages which both can be explained. Business G is a very small business with whom TEG started working very intensively. So their overall portfolio of relations through which cooperation evolves is still developing; TEG employees do not yet know many Business G employees, of which there also are not many, and the manager of Business G was still very involved. In the case of Business A it could be seen from the data that there was one very influential manager who had no less than fifteen mentions from the TEG personnel.
It can also be seen that the number of specializations of TEG with whom the businesses have contact differ. This difference is partly explained by the overall difference of relations/contact persons with similar reasoning behind it, but can also come from the fact that some specializations are just less interesting for some businesses. The last column on the number of relations with the TEG management will be discussed in the next subparagraph.
Analysis divergent and convergent ties and formal and informal ties
26
new and developing, which requires more involvement of the managers to initiate first contacts.
Although managers generally are the most mentioned persons, the Business B manager has substantially more relations with the TEG than other managers have. This high number of mentions is partly because this manager is very involved in the work TEG does and since he leaves a very positive impression as could been seen from his high scores on the long and short term information importance and effectiveness. His management style is partly just good management but is also necessary as the TEG delivers some crucial specializations for Business A and since, unlike many other businesses who are also situated in Drachten, their location is mainly at Singapore and partly in Amsterdam, making more informal and spontaneous contact more difficult. It is therefore assumed that a more formal approach is more suitable for Business A as the informal way of communication is less effective because of the distance. This assumption will be further looked into during the interviews. For the other businesses the managers had more mentions and generally scored higher on the long term information importance, but there were no outstanding cases. The only business scoring slightly higher on the ‘average number of relations per contact person’ without such obvious explanation is Business B as this business their circumstances could very well be compared with the one of especially Business C, but also Business F and E. We thus recognized two businesses with slightly more formal approaches; Business D, as it is a developing business relation; and Business A, as it is a very important client, which wants some control, and is not located at Drachten.
Overall conclusions network analysis
27
The table below provides an overview on the main characteristics of embeddedness for the businesses considered.
Businesses Business A Business B Business C Business D
Nr. of contact persons 11 29 33 12
Nr. of relations 36 66 55 22
Average number of relations per contact person
3.27 2.28 1.67 1.83
Frequency 1.59 2.02 2.10 1.50
Length 2.95 2.76 2.44 2.32
Average score on short term information importance
4.09 3.86 3.84 3.21
Average score on long term information importance 4.18 3.40 3.22 3.05 Average score on effectiveness 4.45 4.18 4.05 3.47 Number of businesses with whom they have contact
7 6 7 3
Number of relations with TEG management
5 8 6 6
Table 7: Overview on embeddedness scores of selected businesses from the network analysis
7.2 Interviews analysis
The interviews with one employee and one manager of Business A, Business C, Business B and Business D were all recorded and later on written down. To analyze this information, a big overview on how each of the employees reacted on the questions is used. This paragraph will go through the main things discussed for each part of the interview. After that, this paragraph will give a brief overview on the main differences of answers given by the different businesses.
General questions
The businesses indicate that they ask support from technical support institutions for quite a broad range of issues, but the general trend is that they only ask support when dealing with very complex issues which requires a level of competence they cannot held up themselves because of the discontinuous basis to which these issues occur.
Part one: what distinguishes TEG?
The most mentioned factor considered when choosing to cooperate with a technical support institution is the competence the technical support institute has. Also mentioned a lot is the availability of time of the employees holding these competences and the accessibility of the employees and the organization.
This accessibility of the TEG is also mentioned as a factor that really distinguishes TEG from other technical support institutions and can be attributed to several causes: the proximity of the TEG as most important businesses are located at the same site which allows for
28
face and unscheduled spontaneous contact; the fact that TEG is part of Philips, which allows for open information sharing and allows for a more informal working processes in which NDA’s etc do not have to be predefined; and the history of contact the businesses have when working with TEG, as most of them know both some employees as well as some management of TEG from previous collaborations. All these factors keep the threshold of contacting TEG at a minimum. Furthermore, the businesses experienced the TEG to deliver quality. Factors contributing to this are: the level of competences TEG has, the attitude of the employees to go the extra mile; and the business understanding they have on their businesses. And finally some mention the benefits from TEG as having lots of experience in domestic appliances in general, and having an overview on what developments are going on in other businesses of Philips that may also be of interest to them.
A reason for collaborating with other technical support institutions would mostly come forth out of TEG not having certain competences. Another possible cause would be that a business would like to have a project executed independently of them. In this case a business might consider this option as they have too little time to carry out the project themselves and as they have enough resources to outsource the activity. As TEG’s role is more supportive and they are less suited for carrying out whole projects by themselves, these projects would more likely go to other institutions.
Part two: Information sharing
When asking on the amount of information and the detail to which it gets shared, the businesses generally answer that all information required can be made available to the TEG employees as they are all Philips employees and they do not have to worry about confidentiality. Some interviewees do mention that the constraint to information sharing is
time related as both their time as well as the TEG employees time is valuable and sparse. Some mention that the information shared should be of direct value to the problem they encounter while others try to bring across a vision and discuss issues not directly related to the problems they are now facing. Some of the employees interviewed, also mention that they feel that the TEG employees have less interest in things like planning and trends.
When considering whether or not TEG employees are proactive in sharing new developments and projects they are working on with them, the answers differ a little. The experience on this is quite different from one interviewee to another as they feel that this proactive attitude differs from one TEG employee to another.
This difference can be explained from the different technical fields they are in, but could also be explained because of the time available to different employees to take on such activities, and finally, it can also just be because of the different working attitudes of employees. For example, one particular TEG specialization is experienced to be more proactive, namely Specialization A. This could be explained because of the many advancements there are seen
“First of all there are no boundaries or anything. We can tell them everything, what we don’t share is mostly time related. We have little time and they also are really busy.” - Interviewee Business C
“If you are looking for something you have to announce it, if you don’t tell, people will
not just come with an solution.” - Interviewee Business D
“The process of getting to a project is much more streamlined, we don’t need nda's etc. TEG can just participate in a project. When working with other agencies this is much more official.”
29
in this technical field, especially from the automobile industry, or it could be explained from the fact that advancements in this field can easily be applied to various businesses, providing the possibility to leverage previous experiences. Another explanation could be that unlike for some other TEG specializations, the employees in this specializations all have quite similar levels of competences which makes them more interchangeable and therefore more flexible. In for example Specialization C, we can distinguish one employee who clearly has the highest level of competences and who gets approached a lot by different businesses after which he will try to manage his time by handing part of this work of to his colleagues. This difference in working process and number of assignments he gets without pursuing them, may make this highly specified employee at Specialization C less eager to go out and find his own projects. The other employees at Specializaion C which are still developing their knowledge may encounter less potentially interesting projects as their competences are still being developed. The final explanation could be that the different relations of the TEG with the businesses provide less or more possibilities or/and incentives to proactively share information, this will further be discussed in paragraph 7.3.
In general interviewees experience that TEG, except from maybe some individuals, is not proactively enough in sharing what technical developments are going on. They do have experienced that TEG applies technologies they encountered at different categories to their business and experience this to be one of the strengths of TEG. However they share too little on what technological developments are available developed outside Philips and are also not sure whether TEG knows enough on the latest technological developments.
Concerning the level to which interviewees think that the TEG employees are really experts in their technical fields, they almost all rate the employees with an 6 on a scale from 1 to 7. Interviewees seem to be content with the level of knowledge of the TEG employees. They do not rate them a 7 because they do not have the time to do a lot of research. One interviewee also states that their business does not challenge TEG enough as they mostly only require incremental innovations. Concerning the amount of knowledge the TEG employees have on their business the opinions differ greatly from one interviewee to another. What does stand out in is that Business D grades the TEG sufficiently lower on the amount of knowledge TEG employees have on their business.
Part three: Evolvement of the relation
What gets mentioned a lot during this part of the interview is the ease of contacting TEG as they already knew them from previous experiences. Business C and Business B can both really go to the TEG building and discuss issues when they need to. Business A already cooperates with TEG for a long time and a the manager visits the TEG very regularly. The TEG employees also get exchanged with Business A for a couple of months to deepen their knowledge on Business A. These mechanisms make sure that the TEG and Business A are well acquainted with each other. On the other hand, one of the interviewees at Business D says it is generally really hard to access TEG as he does not know how TEG is structured and how to formally contact them. He also explains that many colleagues of Business D do not know TEG or what TEG can do for them. This might be explained from the fact that many people are either new or temporary employees hired from agencies. These employees do not know about TEG but also rather seem to figure out themselves how to do certain things.
30
These efforts costs them very much time and are a lost effort to Philips as many of these newly developed competences will leave the company with these temporary employees. Most of the interviewees are connected with somebody of the TEG management. TEG managers get contacted for account issues or when they do not know which employee or competence they require for certain projects.
All businesses interviewed share their roadmaps with TEG, however some businesses are better able to define their plans than others. Another difference concerning the issue is how actively they share their roadmaps, as an interviewee of Business D says TEG can just look into their plans as they are shared on the digital working space while some other departments actively share their plans through meetings etc.
Part four: Experienced results
Almost all interviewees answer that they get the results they expected from collaborating with TEG. Some mention that this comes from keeping their interests aligned. Interesting enough only one interviewee from Business D answers that they get more from TEG than what they expected. This may also have to do with the expectations of this business.
Regarding what factors are considered to contribute to an effective collaboration, many answer this question again with the level of competences or skills, the convenient location and the planning or availability of the
employees/organization. What gets added here by both Business C as well as by Business A is that the communication, open mindedness and trust/confidence also contributes to the effectiveness of the collaboration. Then finally an interviewee of Business A thinks it is
important to be able to inspire each other. As availability seems to be an important factor to the businesses I then also asked if the availability of employees was a problem to them. The general answer was that they may experienced some employees having little time for them. Although they have not encountered any real problems with it. One interviewee does points out that because a certain TEG specialization has too little time for him he feels like he is always fire fighting. This would suggest that the availability of the employees can inhibit the business from moving from a problem driven approach to an opportunity driven approach. When trying to distinguish whether they experience a difference of the effectiveness for the collaboration through different stages or with different specializations, the respondents generally had a lot of difficulty making a distinction. Concerning the intensity of information shared and the amount of experience with working with each other they either say they cannot really make a distinction, or they experience a more intensive collaboration and previous experience to have a positive effect on the collaboration. The same goes for the amount of information shared. However it has to be noted that most of them share an amount of information that seems appropriate to them for the kind of project or problem they are collaborating on. Therefore this would be an indirect effect.
“It is not just about information
