1
Partner selection in the Fuzzy-Front End:
Know why your customer wants you.
A Case study at Pezy Group Name: Gerben Driest Student-number: S2048566 University of Groningen
Faculty of Economics and Business Program: MSC Business Administration Specialization: Strategy and Innovation Supervisor: R.A. van der Eijk
Co- Assessor: C. Carrol
Abstract
This research attempts to enhance the understanding of how manufacturers select suppliers for collaboration in the Fuzzy-Front End (FFE) of product innovation. The results based on data collected from 11 interviews with manufacturers and 7 interviews with suppliers, show that knowledge, trust, competencies, pro-active attitude, commitment and agreement in motivation and goals are the most important selection-criteria. Furthermore, the results of this research allow the suppliers to better understand their customers in the Business to Business (B2B) market, which could enhance their possibility to be chosen as development partner in the customers’ FFE.
Keywords:
2
Executive summary
This study has been written as the final piece of work of my period as student. Together with the Pezy Group (PG), we have searched for a research question that was both relevant to them and related to my study and interests. Especially my interest in the subject made it possible to stay motivated over the whole period. The PG is a group of firms that find their core competencces in the construction part of New Product Development (NPD), and they want to increase their grip on the first phase of NPD in order to gain more assignments in the latter phases of NPD. The first phase is also known as the Fuzzy Front End (FFE) which can be defined as the period between when an opportunity is first considered and when an idea is judged ready for development.
The purpose of this research is to answer the question how innovation suppliers like PG can increase the probabililty to be chosen as partners in the FFE of a manufacturer. Besides the information that this research provides for the suppliers, it also gives more insight in the decision making process of manufacturers in the FFE. Several researches have been dedicated to this subject, and built on the insights from these papers the current research is conducted. In addition to the previous research, this cross case study investigates in an explorative way the mutual difference between selection criteria and different kinds of manufacturers. Also a comparison is made between the suppliers’ assumptions about the selection criteria of manufacturers and the actual ones. The results are collected using a semi-structured interview which included twenty pre-selected criteria which had to be graded on importance by the participant. This interviews provided both qualitative and
quantitative information, which allowed to make assumptions although the number of cases is low (N=11 manufacturers, N= 7 Suppliers). The results of the collected data are compared using several statistical tests.
The results of the research show that suppliers need to be aware of the different requirements of their customers in the Business to Business market (B2B) which is caused by the high mutual difference between the demands of the manufactures. By knowing the customers, suppliers can deliver optimal value which increases the probability to be chosen as a partner. The quantitative results show that knowledge and competencies are valued as very important criteria. Knowledge is mentioned as difficult to sell but customers are searching for authorities and experts in their specific area of interest. Also the trust, commitment, pro-active attitude, goal correspondence and
motivation correspondence are mentioned as decisive criteria. Especially in the FFE, these partner-related criteria are of great importance due to the lack of boundaries. The unpredictability of the FFE also requires a business model that allows risk taking, which is considered by several respondents as a business model which is not based on short term revenue but on intrinsic interest.
3
Table of contents:
1. Introduction: ... 5
1.1 Research question ... 5
1.2 Scope of the research ... 6
1.3 Reading guide ... 6
2. Theoretical reflection ... 6
2.2 Fuzzy-Front End ... 7
2.3 Innovation... 9
2.4 Need for collaboration ... 10
2.5 Partnership ... 11
2.6 Partner selection ... 13
2.7 Summary ... 15
3. Methodology ... 17
3.1 Research strategy ... 17
3.2 Case study design ... 17
3.2.1. Unit of analysis ... 17
3.2.2 Selection criteria ... 18
3.2.3 Validity & Reliability ... 18
3.3. Type of case study ... 19
3.4 Independent grouping variables ... 19
3.5 Data collection ... 20
3.5.1 Expert interviews with suppliers and manufacturers ... 20
3.5.2 Measurements ... 21
3.5.3 Limitations ... 21
3.6 Summary ... 21
4. Analysis and results ... 23
4.1 Introduction ... 23
4.2 Descriptive statistics ... 23
4.3 Dimension reduction ... 27
4.4 Partner- and task-related criteria ... 27
4.5 Cluster analysis ... 28
4.6 Qualitative data ... 28
4.6.1 Reasons for collaboration ... 28
4
4.6.3 Selection criteria ... 29
4.6.4 Qualitative data matrix ... 30
4.7 Cross case-analysis ... 36 4.7.1 Supplier/ Manufacturer ... 36 4.7.2 Size ... 36 4.7.3 Startup ... 37 4.7.4 Market ... 37 4.7.5 Collaborative history ... 38 4.8 Summary ... 38 5. Discussion ... 38 6. Conclusion: ... 39
7. Limitations and further research: ... 40
Literature ... 41
Appendixes: ... 45
Appendix A: Interview ... 45
Appendix B: Statistical answers Manufacturers ... 48
Appendix C: Statistical answers supplier ... 50
Appendix D: Answers open questions manufacturers ... 50
Appendix E: Answers open questions suppliers ... 54
Appendix F: Test statistics comparison supplier and manufacturer ... 57
Appendix H: Test statistics comparison size manufacturers ... 59
Appendix I: Test statistics comparison startup manufacturers. ... 62
Appendix J: Test statistics comparison manufacturer market ... 65
Appendix K: Comparison manufacturer’s customer focal supplier... 68
Appendix L: Factor analysis results... 71
5
1. Introduction:
The demand for collaboration in New Product Development (NPD) has grown in recent years, which has resulted in a higher amount of integration of suppliers, customers and other organizations (Chesbrough, 2003). This co-development is a result of a loss of internal industrial research effectiveness, the growing mobility of skilled people, the increasing turbulence of the external environment, the increasing complexity of technology that is needed to innovate and the rising costs of R&D (Emden et al., 2006), all of which leads to a paradigm shift from a closed to an open
innovation model (Chesbrough, 2003). This study will focus on co-development in the first phase of NPD (figure 1), that is also known as the “Fuzzy Front End”(FFE) which can be defined as the period between when an opportunity is first considered and when an idea is judged ready for development (Kim & Wilemon, 2002). The focal group of firms in this case-study, Pezy Group (PG) find their core competences in the pipeline of the NPD (figure, 1), and want to increase their grip on the FFE. This should allow them to guide the direction of NPD which could result in more work for the PG firms in the latter phases of NPD. PG positions themselves as innovation accelerators for companies whose businesses are related to NPD. To increase the probability to be chosen as a partner in the FFE, it is important to know which selection criteria their (potential) customers are using. This knowledge may be a basis for the future positioning of PG companies as partner in the Fuzzy Front-End of New Product Development.
Figure 1: NPD phases
Source: (Russel & Tippett, 2008)
1.1 Research question
There are several studies on NPD collaboration and the results on innovation and profits (Perks, 2000; Deck & Strom, 2002; Chesbrough, 2003) but the partner selection process in collaborative NPD is a neglected topic (Emden et al., 2006). For innovation partners like PG, it is important to know how companies select their partners in the FFE. Anexploratory research that is done by Emden,
6 by several researchers to further investigate the selection process of companies in NPD (Feng, Fan, & Ma, 2010; Zolghadri et al., 2011; Feng, Fan, & Li, 2011;), but they never wrote from the perspective of the innovation partner/ supplier. In this study the focus will be on how and where innovation suppliers have to position themselves as a partner in the FFE.
This results in the following research question:
How can a supplier increase the probability to be chosen as partner in the customers Fuzzy Front End of the new product development process?
The research question is divided in several sub-questions:
Which NPD partner selection criteria can be identified in scientific research? Which criteria are relevant for the partner selection in the Fuzzy Front End? Which selection criteria do Pezy Group firms expect from their customers? To what extent do the prior criteria meet the criteria of Pezy Group’s customers?
1.2 Scope of the research
The field of innovation and collaboration offer an immense research environment. In this research only the selection procedure of suppliers, by manufacturers in the FFE of product innovation will be investigated. More specific; the criteria that manufacturers are using in the selection of suppliers as partners in their FFE.
1.3 Reading guide
The structure of the paper will be the following; in the theoretical reflection the topics of FFE and partner selection will be discussed. This reflection will be based on peer-reviewed papers that are found in electronic journals, and topic related books. The methodology part will contain the development of a theoretical framework on the topic of partner selection, which will be tested on different cases at the companies that are members of the PG. This will result in a knowledge base on which PG can build their positioning in the FFE of their customers. Based on the gathered knowledge some managerial implications and destinations for further research will be provided.
2. Theoretical reflection
Organizations are constantly searching for the most efficient way to maximize their innovation management efforts, in a changing global environment (Christiansen, 2000). Innovation is the responsibility of all the different parts of an organization and their involvement needs to be
determined accordingly (Tucker, 2002; van de Ven, 1986). In this context Brem & Voigt (2009) posit that the ability of an organization to identify, acquire, and utilize (external) ideas can be seen as a critical factor with regard to market success. The Business to Business (B2B) market will be the focal market in this research; therefore B2B relationships will be deliberated. Based on the studies on B2B marketing by Anderson and Narus (1995 and 1998) it can be assumed that the buying decisions are based on tangible parameters in comparison to the Business to Consumer (B2C) market where a higher degree of intangible buying incentives is found. The need for risk reduction should be answered by the supplier with customized solutions (Anderson and Narus 1995), which require a good understanding of the specific needs of the customers, and the added (monetary) value of the service or product. Not only in the buying decision but also in selection of partners, firms are searching for ways to reduce the risk. There are several potentials that can be selected as partner, and the selection is dependent on the needs of the manufacturer. Examples of potential partners are suppliers, consumers, universities or knowledge institutions that can be involved in product
7 areas of the production process and van Echtelt., et al (2008) give a clear definition of supplier involvement which will be used in this study:
“Supplier involvement refers to the resources (capabilities, investments, information, knowledge,
ideas) that suppliers provide, the tasks they carry out and the responsibilities they assume regarding the development of a part, process or service for the benefit of a buyer’s current or future product development projects” (van Echtelt et al., 2008, p. 182).
Suppliers can provide additional resources which could increase the end-consumers satisfaction and can also reduce the time and money spent of unnecessary research and development on features and extra services (Anderson & Narus, 1995; 1998; Haffmans et al., 2011). The reason that many manufacturers do not choose to involve suppliers in the early stages of product development is twofold: first, most of them want to develop internally and don’t see the necessity of collaboration while second; many suppliers do not have the competencies to engage in this phase (Haffmans et al., 2011).
2.2 Fuzzy-Front End
In the literature there are different theories on the different phases in New Product Development (NPD). The first phase is best known as the fuzzy-front end (FFE) which can be defined as “the period
between when an opportunity is first considered and when an idea is judged ready for development (Kim and Wilemon, 2002, p. 269)”. Kim & Wilemon (2002) mention that in the FFE an organization is
busy formulating a product concept and determining if the idea is worth the investment of resources for development, in addition they refer to Khurana & Rosenthal (1998) who state that the FFE includes the formulation of the product strategy, the identification of opportunities and ideas that precede a further development of the concept. Koen et al., (2001) mention that the front end is one of the greatest areas of weakness of the innovation process and that it fundamentally determines the later innovation success.
This FFE is not always completely covered by the different theories on NPD, e.g. the Stage-Gate™ model of Cooper (2001) that shows a broad overview of process from a broad spectrum of ideas towards a final selection. A downside of this model is the lack of in-depth information on the FFE. A theory that offers more detail on this specific part of the NPD was made by Khurana & Rosenthal (1997).
Figure 2 NPD Stages Khurana & Rosenthal
8 The model of Khurana & Rosenthal (1997) (Figure 1) shows that the FFE contains preliminary
opportunity identification, and market and technology analysis, followed by the Front End where the product/ concept is defined. When an idea has made it through the front-end the management will make the Go/ No-Go decision. The model of Khurana & Rosenthal (1997) is in line with the NPD model of Wheelwright & Clarck (1992) that divide the process in similar phases. Thom (1980) divides the NPD process into the stages of idea-generation, idea-acceptance and idea-realization.
The FFE phase is characterized by its fuzziness, broad focus with a low degree of formalization, small teams and budgets and a low presence or complete absence of management commitment (Kim & Wilemon, 2002). In addition to this enumeration Koen et al., (2001) state that is an unpredictable stage with a high degree of chaos because it is difficult to plan ‘Eureka’ moments. The
unpredictability in this phase has as advantage that there are lower costs involved with technical changes, and the fact that there are more possibilities to reduce the manufacturing price (van Weele, 2005). These advantages are shown in figure 2. In addition to the information that is shown in figure 2, Haffmans et al., (2011) mention that many manufacturers concentrate on research and sales, and outsource the development to partners, while the participation of suppliers in the research part could reduce the cost
Figure 2: NPD Specifications 1
Source: Adapted from (Haffmans et al., 2011)
Another advantage is the possibility to reduce the lead time in the beginning of the process, when the time pressure is low (figure 3). Unfortunately, managers pay most attention to the process when most of the early advantages are gone (Wheelwright & Clark, 1992). In addition, Murphy & Kumar (1997) argue that the appropriate management of the FFE in intra-firm settings is essential, and that unsuccessful management of this phase could have considerable consequences.
9
Figure 3: NPD specification 2
Source: Adapted from (Haffmans et al., 2011)
2.3 Innovation
The FFE is a process of producing ideas for incremental and radical product or service concepts (Jörgensen et al., 2011). The degree of newness does increase the unpredictability of the concept. For the definition of newness the theory of Garcia & Calantone (2002) will be used, which holds that new product innovations can contain incremental changes that create a marginal impact on technical and market level while radical innovations are characterized by their changes in both existing
technologies and marketing infrastructure. Between these extremes are the really new innovations (those that involve changes on either market or technological level). Von Hippel (1988) mentions that both internal and external sources are a source for new ideas, which is obvious according to
Boeddrich (2004) because every innovation is based on an idea from inside or outside the company. In general there are two ways of innovation incentives: market pull and technology push (Bullinger, 1994). Market pull is an idea source that results from an inadequate satisfaction of customer needs, while technology push is caused by the application push of a technical capability without the
presence of a demand. According to Brem & Voigt (2009), the technology push can be described as a process of creative destruction with new improvements, while the market pull is better characterized as replacement or substitution for current products or services. Technology is in general more relevant in the first phase of a product life cycle, while market shows to be more important for further distribution (Pavitt, 1984). However the two ways of innovations incentives can’t be separated from each other. A full focus on R&D could lead to a “lab in the wood approach” which means that the research is disconnected from the rest of the company, while market and technology both determine the success of a company’s research and have to be connected constantly (Brem & Voigt, 2009). Hausschildt (2004) mentioned that a targeted combination of market pull and
10 The role of information collectors (boundary spanners) in the network is mentioned to be very important in gathering information for potential innovations that can be combined with internal knowledge. Most information for radical innovations arises from outside the usual network which holds that there is a need for active networking to overcome structural challenges, and stay ahead of the competition (Brentani & Reid, 2012). The effectiveness of the boundary spanners is dependent on the position of the person, the characteristics of the innovation and the ability of the individual to recognize and interpret the information (Brentani & Reid, 2012).
Brentani and Reid also mention that companies might lack the ability to gain the needed information from their network, or from outside their network. To cope with this problem they mention the position of a broker to solve the information problem.
In summary, innovations differ in newness between radical and incremental. This difference has an impact on the information collection and diffusion throughout the company. Ideas for innovations could occur from technology push or market pull. More in detail they can be found internal or external, where the gatekeeper plays an important role. Collaboration with customers and suppliers could also provide an important source of ideas. The influential factors are summarized in table 1.
Table 1: Influential factors
Influential factors
Long or short term collaboration (Gadde & Hakansson, 2001) (Axelsson & Wynstra, 2002) (Uzzi, 1997) (van Echtelt et al., 2008) Radical or incremental innovations (Crawford & De Bendetto, 2003) Garcia & Calantone (2002) (Baum et al., 2010) (Brentani & Reid, 2012) Technology push or market pull (Bullinger, 1994) (Walsch et al., 2002) (Pavitt, 1984) (Munro & Noori, 1988)
2.4 Need for collaboration
Product development is stated as one of the most risky investments of firms, the chance that a product does become a success is approximately 33%, were another 33% is doubtful and the last 33% is a failure (Haffmans et al., 2011). The reason that so many projects fail can in general be brought back to three critical success factors: Marketing, technical and project management, while
environmental factors of failure were only mentioned a couple of times (Haffmans et al., 2011). The selection of one or more partners for a strategic alliance is primary based on a lack of competences and resources at the requiring firm (Hitt et al., 2000), Not only the ability to gain the right knowledge or complementary assets (Teece, 1986) in NPD process is a problem, but also the internal research has become less effective (Chesbrough, 2003). Emden et al., (2006) appoint the increasing global competition, the shortage of product life cycles, the growing mobility of skilled people, the
turbulence of the external environment and the increased complexity of technology and R&D costs as incentives for collaboration. These developments resulted in an increased outsourcing of
components and subsystems (Wynstra., 2010), which led to a shift in paradigm from a closed to open innovation (Chesbrough, 2003). In addition Dyer & Singh (1998) mention that critical resources may extend beyond firm boundaries.
The research is focused on the first stage of NPD and it is stated that an early involvement of suppliers or intermediaries could result in a faster development process (Gupta & Wilemon, 1990). Early involvement can contribute knowledge of design, technology, costs, manufacturing lead times, product definition and project planning (Kim & Wilemon, 2002; van Echtelt, et al., 2008). Timesaving through joint technologies and earlier problem recognition in early cooperation can result in long-term relationships (Kim & Wilemon, 2002; Dyer & Singh, 1998) and could lead to a competitive advantage, through new insights and targeted solutions (van Echtelt, et al., 2008; Dyer and Singh, 1998, Hitt et al., 2000). Partnership is not per definition a competitive advantage because
11 production networks could out-innovate other networks by improving their knowledge-sharing competences. He also mentiones that in some industries (e.g. wire termination equipment) the majority of innovations could be traced back to suppliers.
In summary, there are several reasons for companies to engage in a collaboration with their suppliers, these motivations are gathered in table 2. By involving suppliers in an early stage of NPD, companies are better able to obtain the benefits.
Table 2 Motivations for collaboration
Motivations for collaboration:
Risk share (Haffmans et al., 2011) (Feng et al., 2010) (Hartley et al., 1997) Access to complementary assets, (Hitt et al., 2000) (Teece, 1986) (Luo, 1997)
knowledge and markets Barney, 1991)
Increasing costs of R&D (Rindfleisch & Moorman, 2001) (Wynstra, van Weele & Weggemann, 2001) Increasing competition (Blackwell & Eilon, 1991) (Johnson et al., 2000)
Shortage of product life cycles (Chen & Li, 1999) Increasing complexity (McCann & Selsky, 1984) (Mintzberg, 1983)
Collecting new, innovative solutions (Wheelwright & Clark, 1992) (Dyer and Singh, 1998) (Powell et al., 1996) and ideas (van Echtelt, et al., 2008) (Brentani & Reid, 2012)
Timesaving (Gupta & Wilemon, 1990) (Feng et al., 2010)
Outperform others (Kim & Wilemon, 2002) (Dyer and Singh, 1998) (Von Hippel, 1988) Cost saving Khurana & Rosenthal, 1997) (van Echtelt et al., 2008)
2.5 Partnership
The engagement in a partnership with suppliers is not without risks, because it is costly and takes time and effort (van Echtelt et al., 2008). For that reason supplier engagement should be applied selectively (Wynstra et al., 2010). The extent of supplier involvement in NPD is dependent on the supplier’s ability to accomplish the job (Corswant & Tunälv, 2002). The extent of involvement has an impact on the supplier’s role in NPD. These roles are described by Kamath and Liker (1994) in table 3. As can be seen in the table, the different stages contain different responsibilities for the suppliers. This research will focus on the partner role of the supplier, because in other stages, the provided specifications are too specific for the FFE, and partners are allowed to get involved in the pre-concept stage. Many suppliers have an explicit strategy to become involved in product development
(Wynstra et al., 2010), but need to have a clear incentive to join. The needs of suppliers could lead to a paradox that on the one hand suppliers want an open and unstructured environment to discuss in, while on the other hand they prefer a clear outcome (Jörgensen et al., 2011).
12
Table 3: Supplier roles according to Kamath and Liker (1994)
Source: Adapted from (Corswant & Tunälv, 1994, p. 251)
The neoclassical view suggests that transactions can take place trough loose arm-length connections which are impersonal and continuously shifting that self-interest motivates the actions and that firms are avoiding dependence (Uzzi, 1997). On the other hand the social view represents a network that exists of stable and close relationships with embedded exchanges (Uzzi, 1997). These views can be compared with the transaction oriented purchasing and relation-oriented purchasing views (Axelsson & Wynstra, 2002), where transaction-oriented purchasing is focused on getting the most profitable deal from one of the several suppliers that are kept at arm’s length, while the relation-oriented is focused on a sustainable collaboration with a limited number of partners (Wynstra et al., 2010). van Echtelt et al., (2008) apply these views on the topic of partnering in NPD where they distinguish the long-term strategic processes from the short-term operational processes. The long term collaboration is focused on the creation of a supplier-network that can engage in the product development processes, which should result in more efficient/ effective future collaboration, access to supplier’s technology, technology roadmap alignment and the transfer of earlier developed solutions to other projects. The short-term collaboration processes are concerned with the collaboration in specific development projects, which should result in shared technology and development costs and a reduction of the lead-time. Both views have their advantages and disadvantages, and according to Uzzi (1997) firms should search for a balance between close and distant partners. Wynstra et al., (2010) name the share and development of tacit knowledge as results of collaborative arrangements and access to internal, which might over time enable partners to develop a common understanding of processes which could lead to a reduction of coordination expenses (van Echtelt et al., 2008; Dyer & Singh, 1998). A downside of strong ties is the fact that they are more vulnerable to abuse than the weaker ties, due to the trust in the other party that it would not harm a partner; making firms less alert to signs of abuse (Anderson & Jap, 2005). Another problem with strong ties, or ties with partners within the network is that it might create a resistance to change (inertia), and a reduction of new information flow (Baum, Cowan, & Jonard, 2010). This phenomenon might also occur by selecting partners that have too much overlap in competences. Granovetter (1983) states that weak ties with parties outside or at the borders of the network, can provide insights that are needed for innovation. In addition (Powell et al., 1996, p. 118) mention that: “Sources of innovation […] are commonly found in the interstices between firms, universities,
research laboratories, suppliers, and customers”. This is in line with Wheelwright & Clark (1992) who mention that organizations should stimulate initiatives and procedures that encourage innovation and input from all parts of the organization as well as from customers, competitors and suppliers. In summary: partnering is not without risks, but it could provide some advantages. The transaction and relation oriented partnerships embody two different streams in partnerships. The strong
13 of tacit knowledge. The more distant transaction partnerships are more short- term focused on getting the best deal. A company should search for a good balance between these two kinds of partnerships between their different relations. Together with the earlier mentioned disruptive character of innovations and the source of innovations, the way partners interact is an influential factor on collaboration in the FFE (table 3).
2.6 Partner selection
When a company has decided to cooperate with other firms in their NPD process, the problem of selecting the right one arises. Traditional criteria like costs and quality are not sufficient anymore, and more specific criteria are brought to the table in selecting the right partner (Araz & Ozkarahan, 2007). Due to the imperfect information about the trustworthiness, motives and qualities of the potential partners, collaboration contains a certain risk (Baum et al, 2010). Choosing the wrong partner may entail high costs and disappointment, so to reduce the uncertainty around the partner decision, many firms rely on past ties or references from third parties (Gulati & Gargiulo, 1999). Familiarity with partners or good referrals can increase trust, which could result in an assumption that the collaboration partner will not take advantage of you (Gulati, 1995). Morgan & Hunt (1994) define trust as one party’s confidence in an exchange partner’s reliability and integrity, and appoint
trust and commitment as the two key factors in a successful collaboration. In addition to the
definition of trust by Morgan & Hunt (1994), Uzzi’s (1997) define trust in partnerships as: ‘(a) A social
process in which (b) partners would not act in self-interest at another’s expense and that (c) develops when extra effort is voluntarily given and reciprocated’.
A high degree of trust in exchange relationships is mentioned to be important because it enriches the firm’s opportunities, access to resources and flexibility in ways that that are difficult to emulate using arm’s-length ties (Uzzi, 1997).
Companies are more likely to invest in relation-specific assets when these are protected by effective safe guards (Dyer & Singh, 1998). Several studies have argued that informal safeguards like goodwill and trust are the best manner to protect the investments and exchanges that are made during a complex transaction (Uzzi, 1997; Hill, 1995). The search for partners is based on complementary knowledge but when the knowledge becomes too similar, there is no incentive to collaborate and when the knowledge is too different it becomes very hard to communicate (Baum et al., 2010). In addition Hitt et al., (2000) state that the need for additional resources is a primary reason for strategic alliances and for selecting specific alliance partners, they also mentioned that the reputation of a partner weights heavily in selection. When two firms have been engaged in an alliance they are likely to do so again, although the probability that they will learn again has decreased (Baum et al., 2010). The effect degree of disruption of innovations on the network is mentioned by Baum et al., (2010). They state that radical innovations will increase the distance of disclosure but also infect less companies due to the small amount of embedded companies in the network while an incremental innovation is more likely to create a disclosure at more levels due to the increased size of the network.
The research of Wynstra et al., (2010) showed that the position of the supplier in the supply chain and the extent of focus on innovation did impact the probability to be engaged in NPD. They divided the supply chain in five stages from raw material supplier (downstream) to system supplier
(upstream), where the chance to be chosen as development partner was higher when a supplier was positioned upstream in the supply chain.
pro-14 active suppliers must be able to evaluate their products critically, especially in regard to their
products interplay with other components. Geringer (1988) divided partner selection criteria in 2 groups of factors: (1) task related criteria that refer to the variables that are closely related to the viability of a partnership and the resources and capabilities of a potential partner to compete
effectively. This is regardless of the amount of partners available. The criteria can vary in nature from tangible to intangible and from human to nonhuman. (2) Partner Related criteria are, in contrast, more dependent on the availability of multiple partners with more focus on the way in which companies can work together effectively (Roy & Oliver, 2009). In table 4 several criteria are shown according to the definition of Geringer (1988).Roy and Oliver (2009) found in their studies on partner selection in international joint ventures that partner-related criteria had a bigger impact than the task-related ones, but this cannot be generalized because there are major differences in the criteria that are used for selection (Geringer, 1991).
Emden et al., (2006) provide a theory on partner selection in NPD, which is based on three phases: (1) Technological alignment, (2) Strategic alignment and (3) Relational alignment, which has to result in an optimal relationship (Figure 4).
Figure 4: Partner selection criteria
Source: (Emden, Calantone, & Droge, 2006, p. 336)
In the first phase of technological alignment, firms make decisions based on the unique competences of the potential partners. This view is supported by the resource based view (RBV) of Barney (1991) that states that firms should find partners with unique technological resources. A partner’s
complementary resources and technological knowledge can add value and create new business opportunities, which might result in a competitive advantage. But in order to realize the technological potential and to discover complementarities that are communicated inter-organizational there needs to be a common ground based on overlapping knowledge.
15 be different incentives for firms to enter alliance relationships (Emden et al., 2006). This could result in opportunistic behavior when e.g. a company uses the knowledge of a partner to enter the
domestic market of the partner. In addition to the motivation Emden et al., (2006) found that a high goal correspondence improved the consistency of expectations and assured mutual gains. Goal correspondence does not mean that partners pursue exactly the same goals, but the goals must not be contradicting.
In the third phase culture plays an important role in the relational alignment. Emden et al., (2006) use the definition of culture by O'Reilly et al., (1991) who state that “culture is the collection of cognitions, expectations, mindset, norms and values within an organization”. The findings of Emden et al., (2006) show that partners with compatible cultures are better able to overcome conflicts, and that cultural and procedural differences may have a negative impact on the quality of partnership interactions. It might be a very though job to find a partner with a compatible culture and in these cases openness and consistency may create the necessary ground for collaboration (Emden et al., 2006). To overcome the absence of a compatible culture, Hitt et al., (2000) pointed out that the commitment of a partner provides the necessary basis for fertile collaboration.
Besides the culture there needs to be a preparedness to adapt, which according to Sivadas &Dwyer, (2000) forms the root for NPD. Finally Emden et al., (2006) mention that there need to be a long term focus in NPD because it supports the willingness to continue in uncertain periods. This is in line with the research of van de Ven (1986) who states that a short-term focus could lead to the abandonment of ideas.
Innovation requires collective action to create shared understandings from heterogeneous
perspectives (Brem & Voigt, 2009), in addition Hamel & Prahalad (1994) state that the future will be bright when technicians are gifted with marketing imagination and marketers know the technical possibilities. But the conversion process from market to technology orientation and the other way around requires a change in mindset (Ulijn et al., 2001). To increase the value for the manufacturer, suppliers should invest in the competences of marketing, technical knowledge and capabilities and project management (Haffmans et al., 2011; Corswant & Tunälv, 2002). The problem with the theory is that managers do not always make rational decisions, and do not always have the right information which forces them to choose known or recommended partners for their NPD (Baum et al., 2010). The case study will give more information about the reasons to select a certain partner over the other, which has to provide information for the positioning of partners like PG.
2.7 Summary
Partnering is not without risks, but it could provide advantages in cost, risk and time reduction. Collaboration is in many cases a result of a company’s strategy to cope with external forces or market complexity. With collaboration, companies are better able to collect complementary assets and knowledge that may provide new innovative solutions and ideas and give them a competitive advantage. Co-operating with suppliers could have a very possitive impact on the NPD process, especially when they are engaged from the start of the project, because in this phase the
improvements and adjustments can be applied more easily and at lower costs than later on. The FFE is mentioned as the first phase of NPD, and is characterized by its open structure. This phase offers a lot of possibilities, but also a high degree of uncertainty. Despite the importance of this phase, the management attention is very low. When companies have chosen to engage in a partnership there are several factors that influence the collaboration. The transaction and relation-oriented
16 and uncertain radical innovations. The source of the innovation is roughly divided between market pull and technology push. These sources could also influence the relationship due to the fact that technology is in most cases more important in the early stages of the product life cycle, while market aspects increase in importance later on. These factors could influence the partnership and when a manufacturer decides to collaborate with a supplier the selection needs to be done very secure, to reduce the risks. Companies are more likely to choose partners from their network than from a larger distance. Besides the location of the partner, his competencies, strategy and relational aspects need to be in line with the focal company. When the partnership covers the competences of marketing, technical knowledge and project management, the NPD process is more likely to succeed. The selection criteria are gathered in table 4 and table 5. The selection criteria that are found in literature will be used in the latter part of the research.
Table 1: Selection criteria descriptions
Selection criteria
SC 1 Pro-active supplier Suppliers take initiative without a direct demand of the customer SC 2 Reputation The reputation of the supplier
SC3 Portfolio The product/service portfolio of the supplier
SC4 Tariffs The tariffs in the suppliers service compared to other potential suppliers SC5 Level of trust Confidence in an exchange partner’s reliability and integrity
SC6 Unique (technical) competencies The supplier has additional competencies SC7 Unique Knowledge The supplier has additional knowledge SC8 Market knowledge The supplier knows the market of the customer SC9 (repeat) Past collaborations The supplier is known from prior collaborations
SC10 Geographical location The geographical location of the supplier with respect to the location of the customer SC11 Position in the network The suppliers are in each other's direct network
SC12 Referrals Suppliers are recommended by third parties
SC13 Overlapping knowledge bases The exchange partners have overlapping knowledge on certain technologies SC14 Position in the supply chain The supplier s’ position in the supply chain of the customer SC15 (3Kkey) Knowledge of Marketing, The manufacturer is aware of the 3 elements; Marketing, Project management and
Project management and Technology Technology, during the selection of suppliers SC16 Motivation correspondence The motivation of exchange partners are in correspondence
SC17 Goal correspondence The goals of the exchange partners are in correspondence SC18 Compatible culture The cultures of the exchange partners are compatible SC19 Commitment of partner The supplier is committed to the project, and willing to adapt SC20 Long term orientation The supplier will not be guided by short term goals
Table 2 Criteria in Partnerships
Selection criteria
Pro-active supplier (Corswant & Tunälv, 2002)
Reputation (Hitt et al., 2001) (Dacin et al., 1997)
Portfolio No relevant literature found, but this criterion showed up from the supplier interviews. Tariffs (Huang & Keskar, 2007) (Roodhooft & Konings, 1996)
Level of trust (Uzzi, 1997) (Hill, 1995) (Hitt et al., 2000) Access to complementary assets and (Emden et al ., 2006) (Hitt et al., 2000), Barney, 1991)
knowledge (Luo, 1997) (Teece, 1986)
(Repeat) Previous collaborations (Gulati & Gargiulo, 1999) Geographical location (Weber et al., 1991)
Position in the network (Brem & Voigt, 2009) (Gulati & Gargiulo, 1999) Referrals (Baum et al., 2010) (Burt & Knez, 1995) (Coleman, 1988) Overlapping knowledge bases (Emden et al ., 2006) (Dacin et al., 1997) Position in the supply chain (Wynstra et al., 2010)
(3Key) Knowledge of Marketing, (Haffmans et al., 2011) (Hamel & Prahalad, 1994) (Hausschildt, 2004) (Ulijn et al., 2001) Project management and Technology (Freeman, 1982) (Munro & Noori, 1988) (Lee, 2003) (Corswant & Tunälv, 2002) Motivation correspondence (Doz & Hamel, 1996) (Emden et al ., 2006) (Perks, 2000) (Corswant & Tunälv, 2002) Goal correspondence (Emden et al ., 2006) (Hamel et al., 1989) (Wu et al., 2009)
Compatible culture (Parkhe, 1991) (O'Reilly III et al., 1991) (Emden et al ., 2006)
17
3. Methodology
In order to answer the research questions of this research, the selection criteria that are found in the previous section will be investigated. This section will first describe the research strategy followed by the research design and methods of data collection that will be used.
3.1 Research strategy
This methodology is built on the literature that is found in the literature research. In the literature review, several reasons for collaboration are found, along with selection criteria for partners in the FFE. By conducting an explorative research, on the mutual importance of the partner selection criteria in the FFE, new insights can be provided on the topic. The inductive approach of this research will provide results that can be used in further theory development. A case study model will be used to investigate the selection procedure in a real-life context. The case study model is most suitable in this situation for two reasons (1) It allows the researcher to observe and analyze a phenomenon in its context, without influencing the behavioral context and (2) it allows research a contemporary event that lacks sufficient empirical evidence (Yin, Case Study Research: Design and Methods, 1984)
3.2 Case study design
According to Yin (2012)there are three important criteria that need to be mentioned in relation to the case study design (1) unit of analysis, (selection criteria and (3) validity/ reliability.
3.2.1. Unit of analysis
This component is according to Yin (1984) a description of what should be considered in a case. For this research the cases can be separated in two different groups. The first group contains a person that is involved in the decision-making process regarding partner selection form a supplier
(dependent) perspective and the second group can be described as a person that is part of an organizations decision-making unit towards partner selection for product innovation from a
manufacturer’s perspective. Tables 6 and 7 give an overview of the persons that are involved in the case study, and the function within their company.
Table 6: Supplier cases
Company Description Name Function
Studio Pezy Industrial Design A.Veendijk Senior designer We Are Perspective Branddrive sustainable design N. van Marle New business manager Pezy product innovation
(mechanical) Product
innovation support H. van Es Project manager Pezy product innovation
(mechanical) Product
innovation support J. Hoekstra Manager engineering and Tooling Pezy product innovation
(mechanical) Product
innovation support A. Fennema Sales manager and business developer Promise Project management M. Rijken managing director
18
Table 7: Manufacturer cases
Company Description Number of Employees business Experience in FFE Startup Customer of Pezy Group name Function A Victron Hybrid power solutions
SME B2B Yes No Yes M.Vader Product manager
B Limis Institute for Surgery innovation
SME B2B Yes No Yes H.Bouma Managing director
C Philips Electronics Large B2B/B2C Yes No Yes C.
Kraaijenhagen
Former New Venture integration leader D Xsens Sensor Technology SME B2B No No Yes M. Timmermans & B. Kooi
Product designer and engineering manager E Bywin Innovative
Toys
SME B2C Yes Yes Yes W. Schreuder van Pijkeren
Owner/founder F Eessy Professional
tools
SME B2B Yes Yes Yes E. van Benthem Owner/founder G Neopost Mailroom
equipment
SME B2B Yes No Yes B. Nieuwenhuis Manager Mechanical design engineering H ART.bv Radar
Technology
SME B2C Yes Yes Yes T. Punt Founder/CFO
I Inalfa Roof systems
Vehicle roof systems
Large B2B Yes No No J. Sanders Vice President Global Advanced Technology J Stork/Marel Poultry Food processing equipment
Large B2B Yes No Yes R. Deckers Coach Mechanical Design Engineering K Pro/
Shimano
Bicycle equipment
Large B2C Yes No No M. Gerritsen Team leader Product development
3.2.2 Selection criteria
The selection of the participants was dependent on the market their company was in, and second their function in the specific company.
For the first selection-criterion all firms had to be involved with product development as
manufacturer or supplier, regardless their kind of product. Due to the explorative character of the research, a mutually different set of firms is composed, to have a broader view on the topic.
The second criterion contained the specific function of the interviewee. This person needed to have a sufficient influence in the decision-making process, and needed to know the procedure concerning the selection of partners.
3.2.3 Validity & Reliability
Yin (1984) defines three important factors that determine the validity of a case study. These factors concern (1) the validity of the construct which is controlling for correct operational measures, (2) the external validity, which is related to the domain to which the findings of a study can be generalized and (3) the reliability which is controlling for repeatability of the study with similar results.
- Construct validity; the independent variables that will be used in the cross-case analysis are derived from multiple empirical sources, what benefits the validity (Yin, 1984). Also the selection criteria that are derived for the literature study in the previous section are based on multiple sources and afterwards checked and supplemented by seven professionals that are working in the field of product innovation.
19 company (Anderson & Narus, 1998). Therefore generalization only applicable on several dominant selection-criteria that are of importance for the fast majority of the cases. - Reliability; the case studies can be reproduced due to a fixed interview that is made to
increase the reliability of the results. The interview can be found in appendix A. A replication of this study is due to the mutual differences of the participants not likely to occur.
3.3. Type of case study
In order to get a good overview of the selection procedure and the selection criteria involved, multiple cases will be conducted. Due to the strong diversity between firms in the B2B market the results of the different case studies will be used in a cross-case analysis where multiple cases are grouped based on company or market specifications. A single case study could provide a very subjective view in contrast to an aggregation of several cases.
3.4 Independent grouping variables
Based on literature research, five independent grouping variables are selected for the cross-case analysis (1) supplier versus manufacturer, (2) the firm size, (3) Start-up, (4) market and (5) collaborative history.
1. Supplier versus Manufacturer
The first measurement that will be taken is the comparison between the internal- and the external view. The research of Tuli et al., (2007) showed that there was a difference between the customers and suppliers perception of value in the B2B market. In the current research there might also occur a difference in the expectation of the customers’ selection criteria, and the estimated selection criteria.
2. Firm size
According to Roy & Oliver (2009), and Dacin et al., (1997), the firms’ size is a potential influencer of partner selection criteria, due to market power and its ability to dominate the partner. In this study the firm size was measured using the absolute number of employees. The number of employees was used as a measure because of its often high correlation with the sales and capital of a firm
(Contractor & Kunu, 1998). de Faria et al., (2010) mentioned that the relationship between the firm size and the decision to cooperate in innovation activities is not clear, because large firms are on the one hand expected to assign necessary resources to partner search, but they are also able to carry out the whole innovation process by themselves due to their internal capabilities (Cassiman & Veugelers, 2002). In this study there will be made a distinction between Small and Medium Entrepreneurs (SME’s) and the larger companies. The SME’s contain all companies with <250
employees (European Commission, 2005). Some companies were sub brands of a larger company; in these cases the company was still classified as large because the umbrella organization can have impact on their decisions for certain suppliers
3. Collaborative history
Selecting solely companies in the research that have a history in collaboration with one of the brands in the group could give a biased view on the topic. To ensure that the current customers of the group are a representative of the wider population, a small control variable is integrated with the
20 4. Start up
According to Baum et al., (2000) start up firms have different needs than more mature ones, and the rate of failure is higher for startup’s in relation to their grown up counterparts. There are several reasons mentioned for the differences; one is unfamiliarity with the new roles and work
relationships, at a time that the resources of an organization are heavily taxed; secondly startups lack a broad basis of influence, endorsement quality, reliability and legitimacy and stable exchange relationships that have to grow during years of experience (Baum et al., 2000). In addition, Boeker (1989) mentioned that startups lack stability, social approval and adequate resources. Stuart et al., (1999) found out that the technology startups ‘perform better when they have exchange partners or a prominent alliance. In addition, Baum et al.(2000) mention that startups should invest in
establishing alliances and networks, because this could give them access to technical, social and commercial competitive resources in a fraction of the time it would take to develop them
individually. The incentive to select a partner might differ between startup and mature firms due to their different position.
5. Market
The differences between players in the B2B market and the B2C market are extensively mentioned by Anderson & Narus (1998). Their research showed that there were clear differences between preferences in the B2B and B2C market according to their customer preferences and expectations. This could be reflected in the way they organize their NPD process and the criteria they value. They also mentioned that in a B2B market every customer is a new segment with totally different expectations and preferences.
3.5 Data collection
3.5.1 Expert interviews with suppliers and manufacturers
Primary data is collecting by means of interviews with experts from suppliers and manufacturers. In order to gain both qualitative and quantitative information the interview is conducted in a semi-structured way with both open and closed questions. This allowed asking in depth questions which is seen as a robust manner to do qualitative research (Mc Cracken, 1988). The open questions about incentives for collaboration and partner selection criteria are formulated in a way that reduces potential bias, as the respondent can answer questions without guidance from pre-selected criteria. The qualitative information will be used to explain the results from the quantitative research. The interview can be found in appendix A.
The closed questions contained the 20 selection criteria that were derived from prior literature research. The criteria contain 13 partner-related and 7 task-related criteria according to the definition of Geringer (1991) (table 7) For each item the respondent was asked to indicate the importance of this criteria in their company’s partner selection decision in a FFE collaboration, using a 5 point Likert-scale that ranged from 1 ( strongly disagree) to 5 (strongly agree). The selection criteria were introduced by a fictive case where the manufacturer was asked to select suppliers as partners in his FFE. An early analysis of the suppliers’ results showed that a major part of the criteria were indicated as important. To have a better insight in the mutual difference between the most important criteria, external respondents were asked to rank the 5 most important pre-selected criteria and divide 10 points over them, giving the most important criteria more points than the ones with lower importance. This ranking is only conducted at the interviews with participants from the suppliers.
The interview setting allowed the researcher to distill the motivation behind the response, which provided a more qualitative view on the quantitative information, in contrast to a regular
21
3.5.2 Measurements
For the cross-case analysis of the quantitative data of the Likert-scale measurements will be treated as interval by assuming that the mutual differences between the 5 options is equal. By assuming the data to be interval, more analysis can be conducted on the data. Together with the interval data from the top 5 measurement, independent sample T-test will be conducted to measure the impact of the group variable on the selection criteria. The mutual correlation of the different criteria will be measured using a bivariate correlation (Spearman) test. Also a factor analysis will be used in order to measure if there are groups of criteria which constitute a common factor.
In addition to the independent grouping variables, the different manufacturer cases are subjected to a cluster analysis, to measure if there is a grouping variable between mutual cases.
The qualitative data will be processed using a qualitative data matrix.
3.5.3 Limitations
Due to the small sample, significant differences between groups will not be expected, but the results could give some indications which are useful in this explorative research. For this reason a confidence level of 0,8 will be used, and only approximately significant criteria with a p-value of < 0,1 (two tailed) will be deliberated as possible differences.
3.6 Summary
The methodology that is used in this research is built on eighteen case studies that are conducted at both suppliers and manufacturers. All the participants are active in the field of product development and have sufficient power in the decision making process regarding to collaboration. The case-studies contain a collection of both qualitative data and quantitative data. The quantitative data will be interpreted as interval to support more statistical tests. The quantitative test results will be
22
Table 8: Dependent and independent variables
Dependent variables
Dependent 1 Pro-active supplier Suppliers take initiative without a direct demand of the customer. Dependent 2 Reputation The reputation of the supplier.
Dependent 3 Portfolio The product/service portfolio of the supplier.
Dependent 4 Tariffs The tariffs of the suppliers’ service compared to other potential suppliers. Dependent 5 Level of trust Confidence in an exchange partner’s reliability and integrity.
Dependent 6 Unique (technical) competencies The supplier has additional competencies Dependent 7 Unique Knowledge The supplier has additional knowledge Dependent 8 Market knowledge The supplier knows the market of the customer Dependent 9 (repeat) Past collaborations The supplier is known from prior collaborations
Dependent 10 Geographical location The geographical location of the supplier with respect to the location of the customer Dependent 11 Position in the network The suppliers are in each other's direct network
Dependent 12 Referrals Suppliers are recommended by third parties
Dependent 13 Overlapping knowledge bases The exchange partners have overlapping knowledge on certain technologies
Dependent 14 Position in the supply chain The supplier s’ position in the supply chain of the customer. Dependent 15
(3Kkey) Knowledge of Marketing, technology and project management
The manufacturer is aware of the 3 elements; Marketing, Project management and Technology, during the selection of suppliers.
Dependent 16 Motivation correspondence The motivation of exchange partners are in correspondence. Dependent 17 Goal correspondence The goals of the exchange partners are in correspondence. Dependent 18 Compatible culture The cultures of the exchange partners are compatible. Dependent 19 Commitment of partner The supplier is committed to the project, and willing to adapt. Dependent 20 Long term orientation The supplier will not be guided by short term goals.
Independent variables
Independent 1 Supplier and manufacturer
The difference between the assumptions of suppliers and the actual scores of the manufacturers.
Independent 2 Size Comparison based on size SME (< 250 employees) vs. Large (> 250). Independent 3 Startup Comparison between startup firms and more mature ones. Independent 4 Market Comparison between firms in the B2B and B2C market. Independent 5 previous collaboration
23
4. Analysis and results
4.1 Introduction
This research is interested in the way manufacturers select their suppliers as partners in their FFE. The results that are shown in this section will provide a better insight in this process. First the descriptive statistics of the data will be discussed and afterwards the results of the different cross case-study analysis.
4.2 Descriptive statistics
The descriptive statistics in table 8 show the outcomes of the different case-studies with manufacturers. These results show that knowledge, trust, competencies, a pro-active attitude, commitment, common motivation and common goals are the highest graded criteria. These results are supported by the data in table 9.
Table 9: Descriptive Statistics Likert-scale Manufacturers
Descriptive statistics Likert-scale measurement (Manufacturers)
proactive reputation portfolio Tariffs trust competencies knowledge
N Valid 11 11 11 11 11 11 11 Missing 0 0 0 0 0 0 0 Mean 4,6364 3,4545 4,0000 3,0909 4,8182 4,3636 4,8182 Std. Deviation ,67420 1,12815 1,00000 ,70065 ,40452 ,80904 ,40452 market repeat location network references overlap threekey
N Valid 11 11 11 11 11 11 11 Missing 0 0 0 0 0 0 0 Mean 3,5455 4,0000 3,2727 2,3636 3,6364 3,4545 3,3636 Std. Deviation 1,57249 ,44721 1,10371 1,02691 ,92442 ,93420 1,20605 supplychain motivation goal culture commitment longterm
24
Table 10: Descriptive statistics top 5 manufacturers
Descriptive Statistics Top 5 measurement (manufacturers) Criteria N Sum Mean
Std.
Deviation Criteria N Sum Mean Std. Deviation proactive 11 14 1,273 1,1037 networkpostion 11 0 0 0 reputation 11 3 0,273 0,6467 references 11 0 0 0 Portfolio 11 0 0 0 overlap 11 0 0 0 Costs 11 3 0,273 0,6467 threekey 11 0 0 0 Trust 11 17 1,545 1,3685 supplychain 11 2 0,182 0,603 competencies 11 17 1,545 1,4397 motivation 11 9 0,818 1,328 knowledge 11 18 1,636 1,5667 goal 11 6,5 0,591 0,97 marketknowledge 11 2 0,182 0,603 culture 11 3 0,273 0,9045 repeat 11 2 0,182 0,603 commitment 11 10,5 0,955 1,0596 location 11 1 0,091 0,3015 longterm 11 1 0,091 0,3015
The descriptive statistics of the case studies of the suppliers show a slightly different result (table 10). The scores are in general lower than the scores of the manufacturers, but this topic will be discussed in more detail in paragraph 4.7.1.
Table 11: Descriptive statistics likert-scale Suppliers
Descriptive statistics Likert-scale measurement (Suppliers)
proactive reputation Portfolio Tariffs Trust competencies knowledge
N Valid 7 7 7 7 7 7 7
Missing 0 0 0 0 0 0 0
Mean 3,8571 4,2857 4,4286 3,7143 4,1429 4,0000 4,0000
Std. Deviation ,89974 ,48795 ,78680 ,95119 1,06904 ,57735 ,57735 market repeat location network references overlap threekey
N Valid 7 7 7 7 7 7 7
Missing 0 0 0 0 0 0 0
Mean 3,1429 4,4286 3,1429 3,1429 4,0000 3,5714 3,2857
Std. Deviation ,69007 ,53452 1,06904 ,89974 ,57735 ,97590 1,11270 supplychain motivation goal culture commitment longterm
N Valid 7 7 7 7 7 7
Missing 0 0 0 0 0 0
Mean 3,1429 3,1429 3,1429 2,8571 4,2857 2,5714
Std. Deviation ,89974 1,46385 1,34519 1,21499 ,75593 ,78680
25
Table 12: Correlation between the factors (likert-scale)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 proactive Pearson cor. 1 -,024 -,297 ,077 -,267 ,267 -,267 -,549 ,000 ,147 -,368 -,233 -,188 -,313 ,092 -,052 -,281 ,232 -,029 ,083 Sig. (2-tailed) ,944 ,376 ,822 ,428 ,428 ,428 ,080 1,000 ,667 ,266 ,490 ,581 ,349 ,787 ,880 ,403 ,492 ,933 ,808 2 reputation Pearson cor. 1 ,177 -,311 -,020 -,090 ,199 ,015 ,396 ,613* -,071 ,654* ,164 ,234 ,538 ,123 ,144 ,123 -,121 -,309 Sig. (2-tailed) ,602 ,353 ,954 ,793 ,557 ,964 ,227 ,045 ,837 ,029 ,630 ,489 ,088 ,718 ,673 ,718 ,724 ,356 3 portfolio Pearson cor. 1 ,143 ,494 ,124 ,000 -,191 ,224 ,544 ,292 ,325 ,642* ,332 ,000 ,383 ,297 ,096 ,000 ,371 Sig. (2-tailed) ,675 ,122 ,717 1,000 ,574 ,509 ,084 ,383 ,330 ,033 ,319 1,000 ,245 ,375 ,779 1,000 ,262 4 Tariffs Pearson cor. 1 ,417 ,112 -,289 -,231 -,319 ,094 -,329 -,407 ,389 -,635* -,022 ,149 -,039 -,261 -,528 -,064 Sig. (2-tailed) ,202 ,742 ,389 ,494 ,339 ,783 ,324 ,214 ,237 ,036 ,948 ,662 ,910 ,438 ,095 ,851 5 trust Pearson cor. 1 ,528 -,222 -,143 -,553 ,122 -,306 -,194 ,241 ,149 -,346 -,043 -,111 ,194 -,289 -,083 Sig. (2-tailed) ,095 ,511 ,675 ,078 ,720 ,359 ,567 ,476 ,662 ,297 ,900 ,744 ,568 ,389 ,808 6
competencies
Pearson cor. 1 -,083 -,093 -,276 -,010 -,055 ,061 -,241 -,149 -,500 -,194 -,134 ,043 ,024 ,236 Sig. (2-tailed) ,808 ,786 ,411 ,976 ,873 ,859 ,476 ,662 ,117 ,568 ,695 ,900 ,944 ,485 7 knowledge Pearson cor. 1 ,486 ,000 -,326 -,066 ,608* -,289 -,056 ,500 ,430 ,869** -,753** ,241 ,222 Sig. (2-tailed) ,130 1,000 ,328 ,848 ,047 ,389 ,870 ,117 ,186 ,001 ,007 ,476 ,511 8 market Pearson cor. 1 -,142 -,440 ,236 ,494 -,390 ,043 -,059 ,155 ,527 -,454 ,223 ,300 Sig. (2-tailed) ,677 ,176 ,484 ,122 ,236 ,900 ,862 ,649 ,096 ,161 ,510 ,370 9 repeat Pearson cor. 1 ,608* ,653* ,484 ,239 ,000 ,191 ,000 ,000 ,000 ,000 ,000 Sig. (2-tailed) ,047 ,029 ,132 ,478 1,000 ,573 1,000 1,000 1,000 1,000 1,000
26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
10 location Pearson cor. 1 ,168 ,303 ,450 ,068 ,190 -,063 -,163 ,284 -,423 -,122 Sig. (2-tailed) ,621 ,365 ,165 ,842 ,575 ,854 ,631 ,398 ,195 ,720 11 network Pearson cor. 1 ,364 ,227 ,286 -,227 -,153 -,009 ,034 ,436 ,427 Sig. (2-tailed) ,271 ,501 ,393 ,501 ,654 ,980 ,921 ,180 ,191 12 references Pearson cor. 1 -,021 ,130 ,345 ,377 ,653* -,348 ,211 ,328 Sig. (2-tailed) ,951 ,702 ,298 ,254 ,029 ,294 ,534 ,324 13 overlap Pearson cor. 1 ,194 ,283 ,354 -,039 ,252 ,083 ,024 Sig. (2-tailed) ,568 ,399 ,285 ,910 ,456 ,808 ,944 14 threekey Pearson cor. 1 ,019 -,130 -,090 ,585 ,549 ,158
Sig. (2-tailed) ,955 ,703 ,793 ,059 ,080 ,642 15 supplychain Pearson cor. 1 ,343 ,463 -,313 ,100 -,077 Sig. (2-tailed) ,302 ,152 ,349 ,770 ,822
16 motivation Pearson cor. 1 ,690* -,283 ,149 ,280
Sig. (2-tailed) ,019 ,399 ,662 ,405
17 goal Pearson cor. 1 -,733* ,174 ,479
Sig. (2-tailed) ,010 ,610 ,136
18 culture Pearson cor. 1 ,149 -,194
Sig. (2-tailed) ,662 ,568
19 commitment
Pearson cor. 1 ,553
Sig. (2-tailed) ,077
20 longterm Pearson cor. 1
Sig. (2-tailed)
27
4.3 Dimension reduction
In order to see if the amount of criteria can be reduced, a factor analysis is conducted. To do this a Varimax rotated factor analysis is performed on the seven most important selection criteria. The reason for this selection is the inapplicability of the results by using all the criteria. Two components were extracted from the factor analysis (table 12).
Table 13: Factor analysis
Rotated Component Matrix
Component 1 1 2 Goal ,944 Knowledge ,848 Motivation ,715 Proactive -,478 Trust ,946 competencies ,690 Commitment -,395
The two components contain respectively four and three criteria, which are in a certain way related to each other. An explanation for this relation is not found in the literature, and comparisons of the different independent variable on these components show only approximately significant results with the comparison of suppliers’ assumptions and manufacturers’ actual scores. The manufacturers scored higher on both components, which can be explained by the fact that all these seven criteria are graded higher by manufacturers. The other comparisons did not show any significant results (appendix L). For these reasons the factor analysis will not be used in the further analysis.
4.4 Partner- and task-related criteria
In addition to the factor analysis another criteria grouping variable is reviewed. Based on the definitions of Geringer (1991) two types of criteria are mentioned; (1) task-related and (2) partner-related. In this research 7 task-related and 13 partner-related criteria are involved (table 13). The results show that knowledge and competencies are the most important task-related criteria, whereas trust, pro-active attitude and commitment are graded as the most important partner-criteria. The research of Geringer (1991) and Roy & Oliver (2009) argue that partner-related criteria are of greater importance than task-related ones, and this confirms the results of this research where the partner related criteria are in general higher graded than the task-related ones.
28
Table 14: Task- and Partner related scores
(mean is measured on a scale from 1 (very unimportant) to 5 (very important)
4.5 Cluster analysis
A cluster analysis is conducted to investigate if there are other case-grouping variables in addition to the current five independent variables. In addition to the cluster analysis, only the seven most important criteria are included. The results of the analysis show that two of the manufacturers’ cases are slightly different from the nine others (appendix M). The two firms grade the motivation
approximately significant and goal significantly lower than the others but besides that the
commonality between these two firms (LIMIS and Neopost) stops with their geographical location in the Dutch province of Friesland. The usage of different cluster methods in addition to the Ward linkage method showed similar results and for this reason there will not be added another independent grouping variable.
4.6 Qualitative data
In order to have a better insight in the motivation behind the numbers, the respondents were asked to support their grades. This resulted in a database of qualitative knowledge which will be discussed in this section. First the manufacturers’ were asked why they should collaborate with suppliers in the product innovation, and afterwards what their experience was with collaboration in the FFE and how they should select suppliers.
4.6.1 Reasons for collaboration
The reasons to start collaborating in product innovation differ between companies. For most
respondents the main reasons for collaboration were collecting additional knowledge, competences, capacity and new insights to stay ahead of the competition. One of the respondents summarized the goal of collaboration as ‘finding the needed professions that can make what is inside our head’. In addition, speed is also mentioned as one of the advantages of collaboration due to the more
Task-related Partner-related Task-related mean score
