Mathieu R. Westbroek Neighbouring or Distant Partners?

(1)

Neighbouring or Distant Partners?

– Geographical Proximity within Open Innovation: a consideration of the time frame –

by

Mathieu R. Westbroek

December 2014

A Master Thesis

University of Groningen (NL) Faculty of Economics and Business MSc Technology and Operations Management

Newcastle University (UK) Business School

(2)

2 |

(3)

| 3

Abstract:

This study addresses the importance of geographical proximity within the context of ‘open innovation’. More specifically, it focuses on the permanent and temporary geographical proximity within the core processes of open innovation; external exploration, external exploitation and innovative collaboration. By conducting semi-structured interviews within a multiple-case study, the importance of geographical proximity was studied within the European hub for water technology. Research findings show that most enterprises collaborate with distant innovation partners. Through a process of open- and axial-coding, it was found that especially temporary geographical proximity was important within the three core processes of open innovation. Which, more in general, suggest that the advances in ICT and transportation expanded the geographical limitations of knowledge transfer.

Keywords: permanent-; temporary-; geographical proximity; external exploration; external exploitation; innovative collaboration; knowledge transfer

Acknowledgments:

(4)

(5)

| 5 Contents 1. Introduction ... 7 2. Frame of reference ... 9 2.1. Open Innovation ... 9 2.2. Contextual Factors of OI ... 14

2.3. Geographical proximity and Open Innovation ... 16

3. Conceptual Model and Research Questions ... 19

4. Methodology ... 21 4.1. Research Method ... 21 4.2. Research Design ... 22 4.2.1. Research Propositions ... 22 4.2.2. Unit of Analysis ... 22 4.2.3. Case Selection ... 23

4.3. Case Study Protocol ... 26

4.3.1. Case Study Overview ... 26

4.3.2. Data Collection Method and Procedures ... 26

4.3.3. Data Collection Questions for the Interview ... 29

4.3.4. Case Study Report ... 30

4.3.5. Analysis ... 30

5. Findings ... 31

5.1. Web-based Questionnaire ... 31

5.2. Semi-structured interviews ... 33

6. Discussion ... 39

6.1. OI is performed due to a lack of knowledge and/or expertise ... 39

6.2. The importance of GGP for enterprises ... 40

6.3. The influence of knowledge types on the role of GGP ... 40

6.4. Reflection on the research methodology ... 41

7. Conclusion ... 43

7.5. Further research ... 45

REFERENCES ... 46

APPENDICES ... 53

Appendix A – Questionnaire ... 53

Appendix B – Semi-structured Interview ... 67

(6)

| 6

List of Tables

Table 1. Various definitions of Open Innovation between 2003 and 2014 ... 10

Table 2. Various practices of the outside-in process ... 11

Table 3. Various practices of the inside-out process ... 12

Table 4. Various practices of the coupled ... 13

Table 5. Case selection criteria ... 25

Table 6. Data Collection Questions ... 29

Table 7. Results of Web-based Questionnaire ... 32

Table 8. Interview - background information ... 73

Table 9. Example of the output of a questionnaire ... 74

Table 10. The originally proposed replication logic (30th of June, 2014) ... 75

List of Figures Figure 1. Core processes of Open Innovation (adopted from Gassmann & Enkel, 2004)…..13

Figure 2. Conceptual model………..20

(7)

| 7

I n t r o d u c t i o n

1. Introduction

Influential regional clusters like the Motor Sport Valley in the south of England (Henry & Pinch, 2000), let us believe that the geographical proximity (GGP) of enterprises is important for innovation. It is often argued that the co-location of enterprises facilitate the required face-to-face interactions for knowledge transfer. However, when considering the current information communication technologies (ICTs) and modes of transportation, is such permanent co-location still relevant? Or can temporary co-location facilitate innovation as well? Enterprises are opening up their organisational boundaries, as external knowledge and external business models are seen as equally important as internal equivalents during innovation (Mowery, 2009). Therefore, it is important for new industry hubs like the European hub for water technology to have a better understanding on the time frames of GGP.

Ever since Alfred Marshall argued the externalities of agglomeration in 1920 (see Marshall, 1920), scholars addressed the concept of GGP. Unambiguously, the literature defined GGP as “the spatial or physical distance between economic actors” (Boschma, 2005, p. 69). Note that the time frame of GGP is not addressed, which is either ‘permanent’ or ‘temporary’. Recognizing a lack of understanding, Ramírez-Pasillas (2010) explained both time frames as being ‘the permanent co-location of enterprises within a certain territorial reach’ (permanent-GGP) and ‘the co-location for a short period of time for which enterprises have to travel’ (temporary-GGP). In contradiction to GGP, the equal importance of internal and external sources of knowledge and business models is a relatively new phenomenon within innovation (Mowery, 2009). It relates to the ‘open innovation’ paradigm, which was introduced in 2003 and based on the idea “that valuable ideas can come from inside or outside the company and can go to market from inside or outside the company as well” (Chesbrough, 2003, p. 43). Within practice, this idea is applied by enterprises by focussing on both their internal and external environment for knowledge (i.e. the outside-in process), alternative business models (i.e. the inside-out process) and innovative collaborations (i.e. the coupled process) (Chesbrough & Bogers, 2014).

(8)

| 8

I n t r o d u c t i o n

second objective of this study is to contribute to the decision-making on business’ locations and innovation partners. Here, it considers the need for nearby innovation partners.

In order to address these research objectives, exploratory and explanatory research is performed at the European hub for water technology (i.e. the WaterCampus). Which is established in 2010 and covers the entire ‘water technology innovation chain’ with its managing directors; Wetsus (cluster for fundamental research), Centre of Expertise in Water technology (applied science) and the WaterAlliance (cluster for entrepreneurship). The research itself is designed as a multiple-case study that collects data through the a questionnaire and multiple semi-structured interviews.

This study contributes to the literature on economic geography, by exploring and explaining the relevance of the different durations at which GGP takes place. Something which is relevant now advanced modes of transportation enable us to cross long distances within a short amount of time. A contribution is given to the literature on OI by focussing on the importance of having nearby innovation partners. In addition, this study contributed to the decision-making in practice. More specifically, knowledge on the relevant duration of GGP will be beneficial when implementing OI processes and deciding on business’ locations and innovation partners.

(9)

| 9

F r a m e o f r e f e r e n c e : O I

2. Frame of reference

This chapter is devoted to the theoretical background and conceptualisation of this study. It provides an overview and review of relevant literature on OI and GGP. It elaborates on OI and its core processes. After that, the contextual factors that influence its adoption are addressed. Finally, it concludes with integrating the literature on GGP and OI.

2.1. Open Innovation

In 2003, Henry Chesbrough introduced the OI paradigm by arguing “that valuable ideas can come from inside or outside the company and can go to market from inside or outside the company as well. This approach places external ideas and external paths to market on the same level of importance as that reserved for internal ideas and paths…” (Chesbrough, 2003, p. 43). In essence, he argued that enterprises should open up their organisational boundaries and link with other actors in their external environment (Chesbrough, 2004; Enkel, Gassmann, & Chesbrough, 2009). By doing so, they would benefit from external innovative efforts and accelerate their internal innovation (Chesbrough & Crowther, 2006). It was not an entirely new approach to industrial innovation, as inter-institutional linkages and intellectual property (IP) markets were already used in the late 19th and early 20th century for industrial research (Mowery, 2009). Back then, enterprises mainly used their internal sources of knowledge and technology because external alternatives were related with higher costs (Mowery, 1983). Over the years, scholars began to recognize and understand that sources of innovative ideas were often located outside the enterprise (West, Salter, Vanhaverbeke, & Chesbrough, 2014) and that viable business models were missing for the technology and knowledge that ‘spilled over’ (Chesbrough & Rosenbloom, 2002). Therefore, Chesbrough emphasized an equal level of the importance for internal and external origins of ideas and paths to market. Which differentiates the OI paradigm from the use of intern-organisational linkages and IP markets in early19th and late 20th century.

(10)

| 10

2004). These are illustrated at the end of this subsection (see Figure 1) and separately addressed by the next paragraphs.

Table 1. Various definitions of Open Innovation between 2003 and 2014

Literature OI Definition

Chesbrough (2003, p. 43) “…that valuable ideas can come from inside or outside the company and can go to market from inside or outside the company as well. This approach places external ideas and external paths to market on the same level of importance as that reserved for internal ideas and paths…” Gassmann and Enkel (2004, p. 2) “Open innovation means that the company needs to open up its solid

boundaries to let valuable knowledge flow in from the outside in order to create opportunities for cooperative innovation processes with partners, customers and/or suppliers. It also includes the exploitation of ideas and IP in order to bring them to market faster than competitors can”

Chesbrough (2006)

(Adopted by e.g. van de Vrande et al. (2009); Bogers (2011); Huizingh (2011))

“open innovation is the use of purposive inflows and outflows of knowledge to accelerate internal innovation, and expand the markets for external use of innovation, respectively. This paradigm assumes that firms can and should use external ideas as well as internal ideas, and internal and external paths to market, as they look to advance their technology.”

Du Chatenier, Verstegen, Biemans, Mulder and Omta (2009)

“open innovation, in which companies develop new products, services, or markets collaboratively by using each other’s know-how,

technology, licenses, brands, or market channels.”

Lichtenthaler (2009) “open innovation processes involve multiple internal and external technology sources and multiple internal and external technology commercialization channels.”

Lichtenthaler and Lichtenthaler (2009)

“internal and external knowledge exploration, retention and exploitation.”

Lichtenthaler (2011) “Open innovation is defined as systematically performing knowledge exploration, retention and exploitation inside and outside an

organizational boundaries throughout the innovation process” Tödling, Prud’homme van Reine

and Dörhöfer (2011)

“it argues that companies should not just rely on internally developed ideas and knowledge, but increasingly also on ideas to go to market in addition to the internal path for innovation”

Chesbrough and Bogers (2014) “We define open innovation as a distributed innovation process based on purposively managed knowledge flows across organizational boundaries, using pecuniary and non-pecuniary mechanisms in line with the organization’s business model”

Outside-in

(11)

| 11

(2010) distinguished between pecuniary and non-pecuniary practices (respectively) in terms of ‘acquiring’ and ‘sourcing’. In general, acquiring involves rather tightly ‘packaged’ technologies (Mowery et al., 1997) as the involved knowledge is often ‘explicit’. Explicit knowledge means that it is articulated and captured in drawings or writings (Nonaka & von Krogh, 2009), which make it relatively easy to transfer across organisational boundaries (Foray & Lundvall, 1998). Good examples are patents and manuals, which can be transferred with only one (transactional) interaction. On the contrary, ‘tacit’ knowledge is hard to transfer because it is unarticulated and often tied to senses, movement skills, physical experience and/or intuition (Nonaka & von Krogh, 2009). Such knowledge is quite hard to ‘package’ and often context specific or ‘sticky’ (von Hippel, 1994). A good example is experience, which requires frequent and face-to-face interactions (i.e. interactions in-person) to be transferred (Gertler, 2003). Here, a sourcing practice like ‘external networking’ would be appropriate. Taken altogether, the outside-in process is directed towards the focal enterprise and needs low or high frequencies of interaction when (respectively) ‘acquiring’ or ‘sourcing’ is involved.

Table 2. Various practices of the outside-in process

Literature Outside-in Practices

(Chesbrough & Crowther, 2006; Enkel et al., 2009; Gassmann & Enkel, 2004; Lichtenthaler & Lichtenthaler, 2009)

-Buying or licensing IP and technological knowledge -Customer and supplier integration

(van de Vrande et al., 2009) -External participation -Outsourcing R&D -External networking (Enkel et al., 2009; Gassmann & Enkel,

2004)

-Global knowledge creation

-Linking to regional innovation clusters by establishing listening spots

Inside-out

(12)

| 12

(Lichtenthaler, 2007). In this way, value can be captured from knowledge that does not fit within the internal business model(s). A good example of a non-pecuniary practice is the multiplication of technologies, which is better known as ‘open sourcing’. It often transfers knowledge to the external environment for free, hereby using explicit knowledge. A good example is Linux (Henkel, 2006). In the context of the inside-out process, explicit knowledge helps to prevent unintended spill overs. When tacit knowledge is involved, one might want to interact more frequently to develop trust and a ‘socially embedded relationships’ so that knowledge spill overs can be managed when they occur.

When related with each other, the inside-out and outside-in processes can be regarded as parallels because the outbound flow of one organisations (a part of) is the inbound flow of another (Chesbrough & Crowther, 2006). However, because organisations do not have an equal ability or desire to acquire or transfer knowledge in the external environment, the implementation of both processes is not equally distributed across organizations (Huizingh, 2011). Taken altogether, the inside-out process captures additional value from internal ideas, knowledge and technologies through external paths to market.

Table 3. Various practices of the inside-out process

Literature Inside-out Practices

(Chesbrough & Crowther, 2006; Enkel et al., 2009; Gassmann & Enkel, 2004)

- Selling or licensing-out IP - Outsourcing commercialization

- Multiplying technology (e.g. transferring ideas to others)

(van de Vrande et al., 2009) -Venturing

- Involve non-R&D employees in innovation activities

(Lichtenthaler, 2007) - Stick licensing

- ‘Active approach’ to licensing

Coupled

(13)

| 13

tension between knowledge sharing and protection. Bogers (2011) called it the ‘the paradox of OI’, because two opposing goals need to be balanced. On the one hand, internal information need to be shared for the development of new and valuable knowledge. Whereas on the other hand, crucial knowledge needs to be protected from unintentionally spilling over (Du Chatenier et al., 2009). Due to this paradox, it is important that those involved act as ‘good partners’ (Du Chatenier et al., 2009; Mowery, 2009). To reduce the intentions for opportunistic behaviour (Chesbrough & Bogers, 2014), formal mechanisms (see Table 4 for a list) like strategic alliances are often used.

Table 4. Various practices of the coupled

Literature Coupled Practices

(Enkel et al., 2009) - Alliances, - Cooperation, - Joint ventures (Chesbrough & Bogers, 2014) - Strategic alliances

- Networks - Consortia - Ecosystems - Platforms

They allow for shared and easy use of developed knowledge (Kogut, 1988). In addition, through a process of mutual learning, and long lasting-, frequent- and extensive interactions (Gassmann & Enkel, 2004), trust among alliance members is developed for sharing tacit knowledge (Holste & Fields, 2010; Rallet & Torre, 1999b). With regard to the underlying linkages, it means that the coupled process requires relational and two-way linkages to support the frequent and extensive interaction. In addition, linkages often have a non-pecuniary nature because of the continuous flow of knowledge which is hard to value (i.e. tacit). Also, the formal structure already distributes the produced value to those involved. In principle, the coupled process can use any combination of the outside-in and inside-out practices. To summarize, the coupled process is often performed within a formal structure within which long lasting, frequent and extensive interactions are essential for the mutual learning and the transfer of tacit knowledge.

(14)

14 | F r a m e o f r e f e r e n c e : C o n t e x t u a l F a c t o r s

University of Groningen & Newcastle University Business School | M.R. Westbroek

2.2. Contextual Factors of OI

When researching the influence of one variable (GGP) on another (OI), it is important to understand what contextual factors moderate the relationship. Here, it is relevant to consider three factors that act as pre-conditions for the adoption of OI; enterprise characteristics, technology considerations, and environmental conditions (Gassmann & Enkel, 2004; Gianiodis et al., 2010; Perkmann & Walsh, 2007). Each affects the ‘organisational fit’ between an organisation and its external environment. It refers to “the extent to which critical firm-level characteristics (i.e. systems, process, structures, and incentives) are aligned with external environmental conditions.” (Gianiodis et al., 2010, p. 554) Because different organisational fits result in different abilities to implement OI, it is necessary to control for the pre-conditional factors of OI. In this way, the GGP variable can be studied across similar contexts of OI. Hereafter, the three pre-conditional factors of OI will be discussed separately.

Enterprise Characteristics

As the definition of OI already revealed, an organisation requires various capabilities to be effective in OI. More specifically, it requires network capabilities (Dodgson, Gann, & Salter, 2006), the absorptive capacity to recognize, assimilate and apply external knowledge (Spithoven, Clarysse, & Knockaert, 2011), and the capabilities to subsequently transfer knowledge (Gianiodis et al., 2010). Because these organisational characteristics support the processes of OI, they play a prominent role in the adoption of OI processes. Note that the size of the enterprise is not included. Within the OI literature, research findings are ambiguous on the relevancy of size for the adoption of OI. To indicate, Lichtenthaler (2008) found that large enterprises use OI more often as a strategic approach to ward off competitive pressure from smaller and new enterprises, whereas Henkel (2006) found that small enterprises were more likely to engage in OI practices than large ones’. Due to this controversy, organisational size was not regarded as a requirement for the adoption of OI within this study.

(15)

| 15

F r a m e o f R e f e r e n c e : C o n t e c t u a l F a c t o r s

related it with (Cohen & Levinthal, 1990). As such, the organisation of an enterprise requires a learning capability that stores and develops knowledge so that it creates an internal base of knowledge. By incorporating that knowledge within a innovation process, an enterprise applies it. When internal knowledge is externally exploited, an enterprise needs to be able to transfer knowledge while preventing unintended spill overs (Lichtenthaler & Ernst, 2007). In essence, this means that an enterprise should be able to articulate its knowledge and codify it in such a way that it is easy and save to transfer across organisational boundaries (Spithoven et al., 2011). Taken altogether, various capabilities are required for an effective adoption of OI.

Technological considerations

As Dodgson et al. (2006) argue in their paper, the literature is unambiguous on how technology assists with the merger of internal and external inputs for innovation. Therefore, it is relevant to consider two technological aspects of OI processes; ‘the technological distance’ between, and the ‘technological infrastructures’ of, those involved. Technological distances refer to the extent with which innovation partners have overlapping knowledge bases (van de Vrande et al., 2006). A small technological distance entails that the technology’ receiver has prior knowledge on the new technology (and/or knowledge) (van de Vrande et al., 2006). It enables enterprises to communicate with the technology sender, understand the newly acquired technology and/or knowledge, and process it successfully (Boschma & Lambooy, 1999). Such prior knowledge provides enterprises with absorptive capacity (Cohen & Levinthal, 1990), something which is required for the outside-in process. In a similar vein, a low technological distance between enterprises and their external environment provides ‘cognitive proximity (see section 2.3)’ between enterprises and their external environment. Which is required to transfer internal knowledge outward. This makes it relevant for as well the inside-out process as the coupled process.

(16)

| 16

F r a m e o f R e f e r e n c e : C o n t e c t u a l F a c t o r s

technological interfaces are required for the communication, understanding and processing of the knowledge.

Environmental conditions

Environmental conditions are beyond the control of an organisation (van de Vrande et al., 2006). However, this does not prevent them from affecting the implementation of OI. For example, when the market of an industry is highly uncertain one may want to use loosely structured (i.e. informal) collaborations (Hoskisson & Price, 2002; Steensma & Corley, 2000). Which provides flexibility if needed. On the contrary, when the uncertainty within an industry is low one may want to use structures that are more formal. Industry characteristics that can help with determining the uncertainty of an industry, are the speed (i.e. length of product life cycles, development of new markets, or frequency of changes in industry structure) technological nature (being high or low) (Chesbrough & Crowther, 2006; Gianiodis et al., 2010). For example, van de Vrande et al. (2006) state in their paper that the fast cycles and high ‘clockspeed’ of high-tech industries increase the likelihood that enterprises within it employ or will adopt an OI approach. Chesbrough approached the environmental conditions through ‘erosion factors’ (Chesbrough & Bogers, 2014). Which he used in 2003 to argue for the switch from a ‘closed’ paradigm of innovation to a ‘open’ paradigm. He argued that when these factors were less present within an industry, it would carry a smaller likelihood of employing OI. Taken altogether, the industry in which enterprises operate should be considered when studying OI on the organisational level. This is because it could influence the implementation of OI and therefore the results of the study.

2.3. Geographical proximity and Open Innovation

(17)

| 17

F r a m e o f r e f e r e n c e : G G P & O I

immobile; (tacit-)knowledge, skills, institutional- and organizational structures (Breschi & Malerba, 2001). Based on the additional advantages of proximity, five dimensions of proximity were identified; geographical-, cognitive-, organisational-, social-, and institutional proximity (Boschma, 2005). Bunnel and Coe (2001) called this the ‘de-territorialisation of closeness’, as the new dimensions may function as substitutes for GGP (Gertler, 2003). Here, GGP was defined as “the spatial or physical distance between economic actors, both in an absolute and relative meaning” (Boschma, 2005, p. 69). In which ‘absolute’ referred to the actual distance in (e.g.) kilometres and ‘relative’ to the travel time from actor A to B.

Permanent and temporary geographical proximity

In his paper, Boschma (2005) stated that GGP was not a necessary condition for the transfer of (tacit-)knowledge. It was based on the argument that other dimensions of proximity could substitute GGP. He points to the problem of coordination, which could be solved by organizational proximity and the mobility of human resources (Rallet & Torre, 1999b). In addition, he argued that tacit knowledge was actually embedded within the social relationships of employees (Gertler, 2003). If true, tacit knowledge would be a common property and a shared good within networks that are by nature geographically unbounded (Breschi & Lissoni, 2001; Gertler, 2003). However, despite that the previous arguments were well based, Boschma may omitted the notion of Rallet and Torre in his final conclusion. Who stated that, even when the combination of organizational- and cognitive proximity could substitute GGP, face-to-face interaction remained a requirement for the exchange of tacit-knowledge (Rallet & Torre, 1999a). Here, face-to-face interactions entailed the physical proximity (i.e. GGP) of those involved. Or in other words, an interaction ‘in-person’. In essence, Rallet and Torre proposed that ‘permanent-GGP’ could be substituted but that ‘temporary-GGP’ could not. Here, permanent-GGP referred to the ‘permanent co-location within a certain territorial reach’ and temporary-GGP to the ‘co-location for a short period of time for which enterprises have to travel’ (Ramírez-Pasillas, 2010). The next paragraphs will discuss these two time frames of GGP within the context of OI.

Outside-in and inside-out

(18)

| 18

of production and consumption, whereas improved codification refers to the process of converting tacit-knowledge into explicit-knowledge (Morgan, 2004). Improvements that make the inside-out and outside-in process indeed a lot easier across distances. To indicate (respectively); the increased ability to separate production from consumption enables partners to exploit the acquired ideas and/or knowledge on a location of their choice. In addition, the improved codification of external knowledge enables exploring organisations to recognize, capture and transfer external knowledge from and across long distances. From these perspectives, it can indeed be argued that the improved ICT (which underlie the advances in tradability and codification) made permanent GGP less important for the outside-in and inside-out processes of OI. This suggests that only temporary GGP is required for the face-to-face interactions in, for example, socialisation, deal-making, relationship management and evaluation (Ramírez-Pasillas, 2010; Storper & Venables, 2004).

Coupled

When related to the ‘paradox of OI’ (see section 2.1), the arguments of increased tradability and improved codification lose their strength. They do not contribute to the balance between knowledge sharing and knowledge protection, a balance that is especially important for the coupled process. As elaborated earlier, this process involves a continuous exchange of non-codified (i.e. tacit) knowledge. It results in outputs that are often only used within the formal structure of the innovative collaboration that developed the knowledge. Therefore, the arguments of increased tradability and improved knowledge codification do not apply to the coupled process of OI. However, Du Chatenier et al. (2009) made a distinction between interactive stages and individual stages within the process of collaborative knowledge creation. Some activities take place on the group level, whereas others on the individual level (Du Chatenier et al., 2009). For example, the interactive stage relates to activities that externalize, share, negotiate and revise knowledge by interaction on the group level. In contrast, the individual stage relates to the activities of knowledge interpretation, analysis, combination and creation at the individual level. This distinction suggests that GPP is not permanently required for the coupled process, but only during the stage that involves the interaction with innovation partners. Therefore, temporary GGP may be sufficient for the coupled process of OI.

(19)

| 19

of collaborative knowledge creation (Du Chatenier et al., 2009), was that between the different time frames of GGP and the processes of OI. A topic that has not been addressed before by the literature on geography, innovation or OI. Therefore, this research is focussed on filling that gap with a study on the role of GGP within the processes of OI.

3. Conceptual Model and Research Questions 3.1. Conceptual model

Based on the literature review in the previous chapter, a conceptual model was developed (see Figure 2). The theoretical concepts of permanent- and temporary-GGP were operationalized by linking interactions ‘in-person’ with the establishment locations of those involved. Which meant that those involved were on the same location and interacted ‘in the flesh’ for different time frames. OI was represented by its three core processes; outside-in, inside-out and coupled. Which were all operationalized by their practices (see Table 2, Table 3 and Table 4). However, the adoption of OI and the implementation of OI processes was influenced by contextual factors. As the study needed to control for these factors, they were incorporated within the conceptual model. Further, as explained in the literature review, GGP is especially important for complex, ambiguous and tacit interactions. Because tacit knowledge is hard to grasp, it could complicate interactions and perhaps make the transferred knowledge unclear (i.e. ambiguous). Therefore, it was expected that tacit knowledge had a moderating effect on the role of GGP within OI.

(20)

| 20

R e s e a r c h Q u e s t i o n s

3.2. Research Questions

Within this study, the RQ addresses the importance of the geographical proximity within the processes of OI;

“How important is the geographical proximity of enterprises for the processes of open innovation and why is this?”

(21)

| 21

M e t h o d o l o g y

4. Methodology

This chapter elaborates on the methodology of this study. Before the research method, – design and –protocol are discussed, it is worthwhile to state that they are based on a post positivism research philosophy. It entails that the reality is viewed as an innate truth, that can only be apprehended in a very imperfect way (Croom, 2009). Therefore, to develop an unbiased view of reality, a ‘dialectician’ researcher (Eilon, 1975) interacted with the system to subsequently combine new knowledge with existing knowledge (i.e. literature). A participative approach that was appropriate due to the large, and possibly unknown, number of contextual factors within this research topic.

4.1. Research Method

For the strength of this research, it was important that the entire study followed an underlying logic that related the ‘what’, the ‘why’ and the ‘how’ of this study (Pannirselvam, Ferguson, Ash, & Siferd, 1999). Or in other words, the research topic, the philosophical stance and the methodology. Because a social phenomenon (OI) was addressed by this study, it was important to consider a large number of known and unknown factors that could influence this research. Because it would be unmanageable to control them all in one quantitative research (Johnston, Leach, & Liu, 1999), a qualitative research approach was applied by a researcher with an interactive style. This interactive and qualitative approach meant that the main source of information would be the perceptions of those involved in OI. Together with the rather rational approach to apprehend the reality (combining old and new knowledge), Meredith, Raturi, Gyampah and Kaplan argued that multiple research methods were appropriate to use (Croom, 2009). Their ‘generic framework for research methods’ suggested field studies, field experiments, structured interviews, survey research, action research and case studies as appropriate (Meredith, Raturi, Gyampah, & Kaplan, 1989).

(22)

| 22

M e t h o d o l o g y : R e s e a r c h D e s i g n

behavioural events. By combining the previous, it was decided that a case study was the most appropriate research method for this study (Yin, 2014).

Being appropriate and feasible are two different aspects of research. Therefore, the research requirements and characteristics were combined in the framework of Corrêa (1992). Based a summary of multiple research design choices, Corrêa argued that a case study was feasible for both ‘why’ questions and data collection procedures in which the researcher would be present. Taken altogether, this study performed a case study.

4.2. Research Design

The following section elaborates on the research design by formulating the plan that was used to get from questions to answers (Rowley, 2002). It explains the logic of linking data to research questions and elaborates on the research propositions, unit of analysis and case selection. Principle decisions for data collection and analysis method are stated as well (Runeson & Höst, 2008).

4.2.1. Research Propositions

The first step in getting from questions to answers was to speculate on research findings, i.e. to develop research propositions. These propositions were based on the literature review in chapter 0. By ensuring reasonable evidence for or against them, the research propositions structured this study (Stuart, McCutcheon, Handfield, McLachlin, & Samson, 2002). One proposition was that ‘enterprises performed OI because they internally lacked knowledge and/or expertise’. In addition, it was proposed that ‘GGP was important because it facilitated the social and intensive interactions that were required for the sharing and combination of tacit-knowledge’. Lastly, it was proposed that ‘the involved type of knowledge influenced the role of GGP during the processes of OI’. When combined, they proposed a theory that ‘GGP is only required for the creation, acquirement and transfer of tacit-knowledge during OI processes’. For further elaboration on these propositions, see section 2.3 of the literature review.

4.2.2. Unit of Analysis

(23)

| 23

2009), ‘the enterprise’ was defined on both the theoretical level and empirical level. It prevented aggregation and analysis of data at one level (the empirical level), while interpreting (cross-level inference problem) or drawing conclusions (ecological fallacy problem) at another level (theoretical level) (Forza, 2009).

On the theoretical level it was defined as “the smallest combination of legal units that is producing goods or services by carrying out one or more activities as an organisational unit, at one or more locations, while benefiting from a certain degree of autonomy and responsibility in the decision-making on innovation strategy and activities” (Eurostat, 1996, p. 58; OECD et al., 2005, p. 65). In practice, the smallest combination of legal units that formed an organisation unit were representatives who had the authority to act in the name of the organisation. In addition, benefiting from a certain degree of autonomy and responsibility referred to the independency and authority required to set its own innovation strategy or – activities. Based on the previous, the enterprise was defined on the empirical level as ‘the smallest combination of autonomous enterprise representatives who were involved in the OI of their organisations and had the authority and responsibility to independently set their innovation strategy or –activities’. An operational (i.e. empirical) parallel of the theoretical concept, appropriate because temporary GGP related to the representatives of enterprises and not to concrete buildings (which refer to business’ locations).

4.2.3. Case Selection

(24)

| 24

Selection Approach

In contradiction to the random sampling approach that is often used for survey studies (Seawright & Gerring, 2008), this study used a purposive approach to decide on the appropriate number of cases, sample frame and what cases to select (Ritchie et al., 2003; Voss, Tsikriktsis, & Frohlich, 2002). Meaning that cases were chosen with a certain purpose (Eisenhardt, 1989); understanding the role of GGP within OI. It entailed that findings could only be generalized to theory (Gibbert, Ruigrok, & Wicki, 2008; Johnston et al., 1999) via a process of analytical reasoning (Yin, 2014), better known as ‘logical inference’ (Small, 2009). Understandings could only be generalized for cases that were theoretically identical to the ones used for this study (Runeson & Höst, 2008). In short, cases were purposively selected for this research.

Replication Logic

(25)

| 25

Selection Criteria

The replication logic of this study was operationalized by the actual case selection, a process that was based on selection criteria (see Table 5). Selection criteria were used to control for a similar context of OI. For example, cases were only selected if they expressed an ability to assimilate (i.e. acquire) and apply (i.e. use) external knowledge. In addition, they had to be active in the ‘extraction and distribution of water’ (e.g. the same sector). See Table 5 for a summary of all selection criteria.

Table 5. Case selection criteria

Selection Criteria Selection Criteria - Operationalized

Control variables Exogenous Uncertainty - Active in the same sector

Endogenous Uncertainty

- Acquired knowledge from external sources - Can assimilate external knowledge.

- Similar to other cases with regard to its ICT-infrastructure.

Open Innovation

- Presence of all three OI processes - Similar presence of all three OI processes

Control variables

Exogenous Uncertainty

- Active in the extraction and distribution of water

Endogenous Uncertainty

- Intensive use of multiple external knowledge sources - Innovative collaborations

- Importance of the telephone, e-mail and documentation are ‘normal’; slightly, very or extremely important.

Open Innovation

- Needs to perform outside-in, inside-out and coupled activities.

- Selected cases are similar in their 'breath' & ‘depth’ of OI processes

Theoretical replication Interactions ‘in-person’

- Be an extreme with regard to use of interactions ‘in-person’

Theoretical replication Interactions ‘in-person’

- Interactions ‘in-person ( per month) within 0-30min driving distance (car)

Interactions ‘in-person ( per month) beyond 30min driving distance (car)

(26)

| 26

or beyond the territorial reach of the case. Here, the territorial reach was set at a radius of 30 minutes travelling time. A relative distance (Boschma, 2005) which relates closely with the average 68,5 kilometres (i.e. 43 miles) that a company car covers daily (Statline & CBS, 2014). Here, it is assumed that it covers a two-way trip of 34,25km. Which is about 30 minutes by car. All interactions with innovation partners beyond this 30min radius were related with temporary-GGP. Taken all together, the selection criteria operationalized the control variables and replication logic to ensure that cases provided a better understanding on the role of GGP within OI.

4.3. Case Study Protocol

Being the core of this study, the case study protocol elaborated on how the required data had to be collected. This chapter elaborates on the used documentation; how the right focus, organisation and documentation was ensured during and after interviews. It provided this study with a trial of evidence (Stuart et al., 2002). The study overview, data collection procedures and –questions are also addressed.

4.3.1. Case Study Overview

Initiated as an graduate research at the University of Groningen (RUG) and the Newcastle University Business School (NUBS), this multiple-case study addressed the role of GGP within OI. With support of the internship company IKM (Innovation by Knowledge Mapping), this study targeted the European hub for water technology called WaterCampus. More specifically, the scientific cluster Wetsus and entrepreneurial cluster WaterAlliance. Two cases were selected from their members.

4.3.2. Data Collection Method and Procedures

(27)

| 27

M e t h o d o l o g y : C a s e S t u d y P r o t o c o l

Web-based Questionnaire

The questionnaire was used to collect information on the characteristics and innovative activities (i.e. the case selection criteria) of the research population. Here, the European guidelines and standards for innovative data collection were followed. They were acquired from the third Oslo Manual (OECD et al., 2005) and the Community Innovation Survey (CIS) (2010). This improved the comparability of data and quality of the questionnaire (Bryman & Bell, 2011; Rowley, 2014). It applied an institutional approach (see OECD et al., 2005) and covered an observation period of three years; September 2011 to September 2014. To ensure that the questions were answered with full knowledge on innovative activities, all respondents represented the empirical unit of analysis (see section 4.2.2) during the observation period. With regard to the innovative activities, a shared understanding (i.e. equal conceptualization and interpretation (Forza, 2009)) was developed by providing clear definitions (see Questionnaire in Appendix A). These, and the rest of the questionnaire, were all developed in Dutch (by a native speaker) because all respondents were Dutch (see Appendix A for a translated version).

For its distribution and administration, an efficient and convenient online survey platform named Qualtrics was used (www.qualtrics.com). It allowed for online management and completion at the convenience of the respondents, e.g. in terms of time and place (Rea & Parker, 2012). Which resulted in an enhanced response rate (Greenlaw & Brown-Welty, 2009) and an equal or improved reliability, if compared with paper based questionnaires (van Gelder, Bretveld, & Roeleveld, 2010). Taken altogether, this study used a web-based questionnaire that was developed by following the European standards.

Semi-structured Interviews

(28)

| 28

stated that all released information and interview contents were regarded as confidential. To summarize, respondents were contacted beforehand and well-informed on the background and ‘ins and outs’ of this study.

The next step within the interview procedure was the conduction of interviews. They were semi-structured and followed a ‘general interview guide approach’. Meaning that the interview was structured, but flexible in that the interviewer was free in his statement and composition of questions (Turner, 2010). Each interview lead off with a standard introduction, asking if there were any questions with regard to the provided background information or the structure of the interview. Followed by a short elaboration on the distinction between the three core processes of OI (based on Figure 1) and permanent- and temporary timeframe of GGP. In combination with the provided background information, this introduction ensured that all critical side-notes (e.g. confidentiality and purpose) were mentioned and understood. From this starting point, questions evolved from ‘easy-to-answer’ to more ‘in-depth’. By mainly asking ‘open questions’, the interviewee was invited to talk uninterrupted and elaborate his answer in his own words (Jacob & Furgerson, 2012). This provided the interview with room for unexpected topics (Turner, 2010). This gave the interviewee some control over the content of the interview, which developed a certain trust between the interviewee and interviewer (Jacob & Furgerson, 2012). However, to ensure that the required data was collected, respondents were sometimes asked to elaborate with the use of certain structure or concepts (see Table 6 in section 4.3.3). To provide the interviewer with prompts, a ‘case narrative’ (Voss et al., 2002) (see Appendix B) was developed and used as an interview guide and notebook. For each data collection question (see section 4.3.3), it showed relevant structures and key words. When each question was addressed and all the required data was gathered, the interviewer concluded with a first impression and a check of contact details for follow-up contact for the results of this study.

(29)

| 29

a case narrative for the documentation of data, the researcher prepared himself for the interviews that were at the core of this research.

4.3.3. Data Collection Questions for the Interview

This section elaborates on the core of the research protocol, the data collection questions. These questions were crucial for the link between research propositions and data collection. They provided the interviewer with prompts on what to ask. The first question (see Table 6) related with the first proposition of this study (see section 4.2.1). It encouraged the development of an understanding on ‘why’ enterprises opened up their innovation. The second question provided insight in to whom it was opened up, encouraging data collection on the role of permanent GGP in the selection of innovation partners. Therefore, the second data collection question contributed to the understanding on the importance of GGP. It related with the second proposition. The third question provided a more comprehensive picture of the OI processes and encouraged data collection on why, how and when GGP played a role in the processes of OI. Therefore, the third question related to the second and third proposition. Lastly, the fourth question developed an understanding on how the involved knowledge could influence the OI processes. Which related to the third proposition. Taken altogether, the data collection questions provided the required focus when collecting data.

Table 6. Data Collection Questions

1) Why are external parties involved in innovation?

Here, distinguish between the innovation phases; research, development and exploitation (i.e. knowledge acquirement, knowledge development and knowledge exploitation).

Keywords: knowledge, expertise, market channels, capturing value.

2) How are innovation partners selected? Why like this?

What aspects are considered and what are their weight?

Keywords: (permanent)GGP, relations, culture, organization, routines, knowledge base.

3) How do processes of ‘open’ innovation generally take place?

Here, distinguish between knowledge acquirement, knowledge development and knowledge exploitation. Elaborate by discussing the start, implementation and end of the activities.

Keywords: meetings/in-person interactions, organisation, knowledge transfer, oversight, communication.

4) Does the involvement of different knowledge types influence the role of GGP within OI processes?

(30)

| 30

4.3.4. Case Study Report

In this study, the earlier mentioned ‘case narrative’ (see section 4.3.2) was used as a case study report after the interview was conducted. It was directly expanded with first notes on e.g. first impressions, interpretations or interesting subjects. In addition, a transcript of the recorded interview was supplemented later. In the end, all data was presented by following the structure of the initial interview matrix.

4.3.5. Analysis

The analysis of data entailed an ongoing process of data articulation, arrangement, conceptualizing, organizing and interpretation. As Strauss and Corbin (1990) call it, open- and axial coding were used. Open coding entailed the continuous identification of properties and dimensions (Voss, 2009) within the collected data. Which all needed to be related with the earlier defined practices of OI (see section 2.1). Starting during the interview, the interviewer summarized the elaborations of interviewees in his own words (which were based on the literature review) and the interviewees were asked for their feedback. This was repeated after the interview; transcripts of interviews were conceptualized, categorized and structured with the use of the interview matrix. Which was send back to the interviewees for feedback. The results of case 1 (see Table 7 in chapter 5) were validated during an interview with an, in terms of the selection criteria, identical organisation (see Table 7 in chapter 5). Unfortunately, this was not possible for case 2 (see Table 7 in chapter 5) as there was no organisation within the research sample that was identical to it. The case results and ‘validation results’ were later combined by axial coding. This entailed that they were written down and combined. Here, the data collection questions that were used as a structure (see section 4.3.3).

(31)

| 31

F i n d i n g s

5. Findings

This chapter elaborates on the findings of this study by addressing the results of the questionnaire and interviews. First the results of the questionnaire are addressed to indicate why the used cases were selected. Then, based on the data collection questions (see section 4.3.3 for their elaboration), the in-depth results are presented. To conclude, this chapter reports the concluding answers of interviewees on the research question of this study.

5.1. Web-based Questionnaire

For the questionnaire of this study, 35 enterprises were approached. Resulting in contact with 25 CEO’s, directors or technology managers of which 19 completed the questionnaire. In the end, 17 of these questionnaires were useful. Their results are presented in Table 7, which are structured by following the case selection criteria (see section 4.2.3). Remarkable was the distribution of interactions ‘in-person’ across innovations partners within and beyond the territorial reach of enterprises. To highlight, 97% of the interactions ‘in-person’ involved innovation partners who were not permanently co-located with the questioned enterprise (see Graph 1).

Being underlined in Table 7, Enterprise-B and –K were the only two cases that confirmed all selection criteria. As a result, they were selected as cases for this study. Remarkable was the difference in their number of interactions ‘in-person’. Being 52 to 10, whilst they performed OI on a similar level if comparing the ‘breadths’ of their processes. The number of interactions ‘in-person’ of Enterprise-E was notable as well, however they did not want to be involved any further. Enterprise-I did and was therefore used as a validator of the case 1 results. 3% 50% 7% 25% 15%

Interactions 'in-person' (207)

0-30min drive (car) 31-60min drive (car) 61-90min drive (car) 91-120min drive(car) 120min+ drive (car)

(32)

| 32

F i n d i n g s

Table 7. Results of Web-based Questionnaire

Enterprise - A B* C D E F G H I** J K*** L M N O P Q Exogenous Uncertainty

'Water extraction and distribution' industry YES YES YES NO YES YES YES YES YES NO YES YES NO YES NO NO NO

Endogenous Uncertainty

External knowledge sources (11) extensively used 1 2 1 1 2 2 1 6 3 1 3 2 2 5 1 0 3 Innovative collaborations NO YES NO NO YES YES NO YES YES NO YES NO NO YES NO NO YES Information & communication media ‘normal’ YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES YES

Theoretical replication - Dependent variable (OI)

Performs all three OI processes (presence should be equal in cases) NO YES NO NO NO NO NO NO NO NO YES NO NO NO NO NO YES *Outside-in Breadth: Number of different outside-in practices performed 2 5 2 1 0 2 3 2 5 2 6 2 4 5 6 1 3

* Outside-in Depth: Number of highly to extremely important practices 1 4 2 0 0 2 1 2 4 2 4 2 2 3 5 0 2 *Inside-out Breadth; Number of different outside-in practices performed 1 2 1 2 1 0 1 0 4 1 3 2 1 2 3 0 1

* Inside-out Depth: Number of highly to extremely important practices 1 2 1 0 0 0 0 0 2 1 1 2 0 0 1 0 0 *Coupled Breadth; Number of innovations realized 0 15 0 0 0 5 0 1 0 0 18 0 0 0 0 0 8

Theoretical replication - Independent variable (GGP)

Interactions 'in-person' (per month) within 0-30min driving distance (car) 4 0 - 1 0 1 0 0 - 0 0 0 1 0 0 - 0 Interactions 'in-person' (per month) beyond 30min driving distance (car) 0 52 - 3 75 2,3 6 10 - 4 10 1 6 0 26,5 - 5

*Enterprise-B represents case 1.

(33)

| 33

F i n d i n g s

5.2. Semi-structured interviews

Why are external parties involved in innovation?

In the analysis of interview data, multiple reasons were recognized for the involvement of external parties during innovation. However, overarching reasons were identified; the history of the industry, the improvement of standards, the internal lack of knowledge (or expertise) and pro-active adaptation. With regard to the history of the water industry, a distinction had to be made between the ‘traditional’ water organisations and those that ‘apply the technology related with water’. Especially between traditional water organisations, who produce drinking water or are entitled as the authority for public water management, there exists a strong relationship and history when it comes to innovative collaboration. As Enterprise-B (i.e. case1) states it, they “see each other as colleagues and not as competitors” because they all serve the public interest and “…are [from origin] not commercial[ly]…” instituted. Being confirmed by Enterprise-I (i.e. validator), those who apply water technology within non-public areas, “who are closer to the market”, are more commercially oriented and see the involvement of externals more as a transactional relationship. In a similar vein, Enterprise-K (i.e. case 2) argued “pre-competitive” (i.e. not commercial) and “competitive” (i.e. commercial) collaborations. It said that organisations and people shared knowledge and ideas more easily when it is was not ‘concrete’, i.e. not in a (nearly) commercial (or applicable) form. As a result, “incubators …learn[ed] a lot from each other.” Which suggests that a long history together is not a prerequisite for the involvement of others within ones’ innovation process.

Furthermore, external parties were involved in innovation to improve standards of e.g. education and regulation. The involvement of students enabled organisations to direct the education of their future employees, thereby stimulating their innovation in the future. In addition, the involvement of other organisations was used to increase the support base for amendments to regulation. Which was sometimes necessary for a successful launch or application of a technology; “The more support a new standard has, the less likely a regulator would be to reject an application for amendment of regulation”.

(34)

| 34

F i n d i n g s

organisations accelerate their innovation process by involving the knowledge of others. This was supported by Enterprise-K, which stated that “organisations [or people] often do not know how to bring an idea further to market”. The last reason that was given for the involvement of others, was pro-active adaptation. As Enterprise-B elaborates: “Which direction is the water industry going? What do we need to do in 5 years from now? And 10 years?” By tapping into the knowledge and experience of others, the awareness of contemporary and future issues was increased so that appropriate action could be taken or planned. Both Enterprise-I and -K agreed on this, as Enterprise-K said that: “when doing in-depth R&D on an attic, you lose the connection with the outside world”. Arguing that one should stay aware of what is happening around him or her, by interacting with his or her environment. Taken altogether, enterprises often involve external organisations in their innovation because of relations, shared benefits and pro-active adaptation.

How are innovation partners selected? Why like this?

(35)

| 35

F i n d i n g s

know one another.” Here, it seemed that persons were more important than the organisation that was represented by them. Which meant that the permanent-GGP with innovation partners was less important than the temporary-GGP with their representatives. Noteworthy is that factors such as cultural-, institutional-, organisational proximity between innovation partners were considered as being (almost) irrelevant for their selection; Enterprise-B stated that “as long as they are Dutch or from the West of Europe, it is all fine with me.” In a similar vein, it was stated by Enterprise-K that “there are indeed cultural differences, but we respect those … they do not directly affect the innovation process itself … at most when it comes to keeping appointments.” In short, innovation partners are mainly selected on their knowledge and expertise.

How do processes of ‘open’ innovation generally take place?

(36)

| 36

F i n d i n g s

organisations did explore for knowledge beyond their territorial reach but often by temporarily co-locating themselves with knowledge sources.

With regard to exploitation of knowledge, Enterprise-B mainly exploited its applications at application areas of its shareholders. Using the developed knowledge for its services to others, including its life-time and noncommittal advice to application areas. It entailed a non-pecuniary exploitation of knowledge by multiplying internal knowledge in consulting activities. In this, permanent-GGP was argued as beneficial for frequent interactions required for “get[ting] to know your application areas”. It allowed them to customize their advice. Both ways of exploitation were recognized by Enterprise-I, which exploited its knowledge for ‘free’ in education and commercially by doing applied research for clients. As for Enterprise-B, it turned out that it was not involved in actual exploitation outside the organisation but that it applied its knowledge within its service delivery. For the relation between Enterprise-B and educational institutes, it was recognized that permanent-GGP was beneficial. Furthermore, Enterprise-B applied the concept of ‘Broodje Kennis’ as well to make its issues or developed knowledge approachable for others. As they said it; “if something works for you, why not share it with others?” In short, for the external exploitation of knowledge both permanent- and temporary-GGP were used.

(37)

| 37

F i n d i n g s

“… only [used] when those involved know each other well and trust has been developed”. Agreeing with that, Enterprise-K argued that the world was like a village and that Skype could substitute for permanent-GGP. Therefore, it seemed that permanent co-location of partners was not required as travelling to one another for temporary-GGP was no problem. However, interactions ‘in-person’ remained essential for the communication with new innovation partners, as no clear image was formed with regard to their experience (i.e. cognitive proximity). This was recognized by Enterprise-I and -K, both emphasizing the relevance of interactions ‘in-person’ for the development of trust and creative practices related with (open) innovation. The previous focusses more on the preluding phase of innovative collaborations, whereas Enterprise-B emphasized the importance of interactions ‘in-person’ for the evaluation of progress and results; “all research is reported in large numbers and presented to other water-enterprises”. Enterprise-I acknowledged this and stated that such evaluation moments required the physical presence of all involved to address questions like “what do we actually see? What do the results mean? What are the causes of these results? Is adjustment needed? How can we do that?” In the end, according to Enterprise-B, “technology made it easier to communicate and share knowledge [over distances]…making innovation a lot more structured and efficient.” Enterprise-K agreed. In short, GGP is important for the development of trust among partners but the need for it reduced during actual collaboration as it was mainly relevant for the evaluation of results.

Do knowledge types influence the role of GGP within OI processes?

(38)

| 38

F i n d i n g s

replace the requirement of GGP for the transfer of tacit, or unarticulated, knowledge. So in short, yes, the involved type of knowledge influences the role of GGP.

Concluding quotes of interviewees Company-B:

“Permanent geographical proximity is not necessarily required for [open] innovation…though it allows you to maintain a lasting relationship with application areas…keeping your knowledge on, and understanding of, up-to-date. With regard to [the open-] innovation [processes], if you want to stay informed on all developments you need to be ‘everywhere’…For that you should intensively use temporary GGP. However, some GGP, like within the North of The Netherlands, is beneficial.”

Company-K:

(39)

| 39

D i s c u s s i o n

6. Discussion

Recall that this study was devoted to the exploration and explanation of the importance of GGP within OI. The notion that ‘GGP is substitutable’ was questioned due to the remaining need for interactions ‘in-person’ in the transfer of (especially) tacit-knowledge. Despite that knowledge sharing is at the core of the OI, the importance of GGP within OI was not addressed by the literature until now. Therefore, this study followed an (earlier explained) research logic that collected the appropriate data for the exploration and explanation of GGP within OI. This chapter discusses that data and follows the underlying logic back to the research propositions of this study (see 4.2.1). Hence, the structure of the discussion is based on those propositions and is concluded with a discussion of the used methodology.

6.1. OI is performed due to a lack of knowledge and/or expertise

(40)

40 |D i s c u s s i o n

6.2. The importance of GGP for enterprises

It was proposed that the GGP of enterprises was considered to be important within an OI context, because it facilitates the social and intensive interactions among innovation partners. For the transfer of tacit-knowledge, it was assumed that GGP was especially relevant within the coupled process of OI. In general, the data of this study argued for this proposition. GGP was indeed important within the context of OI, as it facilitated the development of trust and evaluation of both articulated and un-articulated results. That articulated, i.e. codified, knowledge required GGP for its interpretation was somewhat remarkable. It conflicted with statements of e.g. Paavola, Lipponen and Hakkarainen (2004), who argued that explicit knowledge could be formulated in clear terms. The opposite turned out to be true. However, with regard to the transfer of explicit knowledge, the use of ICT was more efficient than GGP and therefore broadly used. This supported the distinction between permanent- and temporary-GGP, which was made at the beginning of this study. The studied cases did not see permanent-GGP as a necessity for their (with OI related) practices. However, it was beneficial for their customer relationship management and those incubators who could still learn a lot from others. In this, the lack of necessity originated from the advances in transportation. Transportation facilitated temporary interactions in-person by travelling towards each other, i.e. realizing temporary-GGP. Here, temporary-GGP was seen as a necessity for the development of trust and evaluation of results because it facilitated the transfer of tacit-knowledge (i.e. experience),‘energy’, and questions and answers. Here, trust entailed the believe in an effective and efficient collaboration whereas evaluation entailed the systematic determination of the merit, worth and significance of innovation results. Further, by frequently meeting with economic actors from beyond ones’ territorial reach, organisations remained up-to-date about the developments in the entire industry. Taken altogether, GGP was important for OI but mainly from a temporary perspective as ICT and transportation annihilated the importance of permanent-GGP.

6.3. The influence of knowledge types on the role of GGP