Prospector orientation and innovation performance : the mediating role of customer involvement and the impact of connectedness

(1)

Prospector Orientation and Innovation Performance:

The mediating role of Customer involvement and the impact of

Connectedness

Student: Iris Blokland (11418699) Supervisor: Michiel Tempelaar Date: 23-06-2017

Final version

University of Amsterdam - ABS Faculty of Economics and Business MSc BA – Strategy Track

(2)

Statement of originality

This document is written by Student Iris Blokland who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

While prior research shows that firms with a prospector strategic orientation tend to have the highest values of innovation performance, this study extends this research by seeing prospector orientation as a continuum; by examining the mediating role of customer involvement in this relationship; and by further exploring the relationship between customer involvement and innovation performance through examining the moderating effect of connectedness. To test the hypotheses, an online survey was sent to managers from firms in different industries. Overall, the findings show that there is a direct relationship between being prospector to a higher extent and having higher values of innovation performance, indicating that firms can differ in their degree of prospector orientation and that this results in different outcomes. However, the expected mediating effect of customer involvement, and the expected moderating effect of connectedness are not confirmed.

(5)

1 Introduction

Recently, the environment of organizations worldwide has changed to a more dynamic, complex and unpredictable one. This environment is created by increased international interdependence, blurring industry boundaries, and more demanding markets (Alegre, Lapiedra & Chiva, 2006; Özsomer, Calantone & Benedetto, 1997). Therefore, the need for organizations to ensure their future viability by exploring new competences becomes increasingly important, and has even become a necessary condition for organizations to grow and survive (Georgellis, Joyce & Woods, 2000; Han, Kim & Srivastava, 1998; Laforet, 2008). This results in organizations searching for new ways of conducting their business through investing in innovations, leading to a higher innovation performance. This, in turn, creates the ability for a firm to gain higher profits and leads to the ability to adapt and manage their changing environment (Aragón-Sánchez & Sánchez-Marín, 2005; Kumar et al., 2012; Laforet, 2008).

The way in which organizations respond to this changing environment, is mainly determined by their strategic orientation, shaping its internal policies, processes and procedures (Hambrick, 1983: McDaniel & Kolari, 1987). This, in turn, determines the core strategy of the firm, meaning that it, in the end, determines the organization’s level of innovativeness. Hence, the strategic orientation can be seen as a factor by which organizations are constrained in their responses to changes in the environment (Hambrick, 1983; Özsomer et al., 1997).

While strategic orientation is conceptualized in many different ways throughout the literature, this research’s focus is on the extent to which organizations do have a prospector orientation. This type of orientation is identified and explained by Miles and Snow (1978) as being the opposite of a defender orientation. The underlying difference is the way organizations respond to and act within their environment (Hambrick, 1982; Hambrick, 1983; McDaniel & Kolari, 1987). As firms become more prospector oriented, they become more proactive (Slater

(6)

& Narver, 1993), and more entrepreneurial and market oriented (Cheng & Huizingh, 2014; Han et al., 1998).

Whereas internal features of a prospector orientation of a firm are extensively studied (i.e. Ren & Guo, 2011; March, 1991; McDaniel & Kolari, 1987), there is, however, still a lack of research on external focus, meaning that far less is known about how those firms strategize for cross-boundary activities. What makes this important, though, is that two trends, identified by Lakhani and Tushman (2012), which are digitization and the rise of numerous people that can actively participate in knowledge production at very low costs, show that customers are gaining more and more power. This results in the need for organizations to interact with them, instead of merely be the seller to them, and therefore, in letting customers actively participate in organizational processes (Schreier, Fuchs & Dahl, 2012; Seltzer & Mahmoudi, 2013).

As firms become more prospector oriented in their strategy, they become more focused on exploring new ideas, developing new knowledge and selecting new markets and customer bases (Slater & Narver, 1993). As a result, it becomes more and more important for them to shift their focus towards external knowledge sources originating from one of the most important stakeholder groups, that is, their customers. Through involving customers in, for example, innovation processes, a prospector firm is able to actually fulfill their strategic intentions and to meet the specific demands of evolving customers. In turn, in this changing environment, as firms become more prospector oriented, actively involving customers in innovation processes becomes a necessary condition for them to develop the right products and services to meet the demands of the customer, to ultimately guarantee the success of new innovations and therefore to increase their innovation performance (Abdul Adis & Jublee, 2010). This results in the first research question: How does customer involvement mediate the relationship between a prospector orientation and the innovation performance of a firm?

(7)

Since involving customers in the innovation process is an activity that goes beyond the boundaries of a firm, the knowledge obtained has to be brought inside. Although organizations actively and consciously choose for involving customers, they do have to find a way in which they are able to integrate this external knowledge effectively and actually leverage it. Therefore, there is a need for effective knowledge sharing within the organization. Since customer involvement is about integrating exploratory, that is new and divergent knowledge, it seems that the most effective way of doing this is by stimulating informal social interaction, or connectedness, since this allows for non-regulated interaction and a certain openness to something else than regulated communication related to exploitation (Sheremata, 2000).

The question that arises is to what extent connectedness helps organizations to turn the external knowledge obtained from involving customers into actual integration of this knowledge, which in turn results in new product development, and thus in an increase in innovation performance. This implies a moderating effect of this social integration mechanism, resulting in the second research question: How does connectedness moderate the relationship between customer involvement and innovation performance of the firm?

This paper tends to contribute to both practice and literature in several ways. First, it tends to contribute to research on strategic orientation by focusing on prospectors only and stating that different degrees of being prospector can lead to different outcomes in terms of i.e. innovation performance. Second, as customer involvement becomes more and more important in this changing world for organizations, this research tends to gain more insight in how customer involvement is a necessary condition nowadays for prospectors to gain a higher innovation performance. By doing this, this paper is an extension on research about cross-boundary activities, since most research on external modes of operations focuses on knowledge originating from acquisitions or alliances, that is, formal interorganizational agreements (Stettner & Lavie, 2013; Russo & Vurro, 2009). Further, although prior research about customer

(8)

involvement does focus on different forms of it and the contribution to innovation performance (i.e. Cui & Wu, 2016), there is a call for more research on strategic and organizational factors that have the potential to facilitate customer involvement, since most research is about the customer’s perspective and lacks the firm’s perspective (Cui & Wu, 2016). The firm’s perspective, underlined as the focus of this research, is on the extent to which firms are prospector and how this influences the ability to increase the innovation performance through engaging in customer involvement.

Third, through combining the mediating effect of customer involvement with the moderating role of connectedness, this paper tends to offer a complete picture on how knowledge deriving from engaging customers leads to actual innovation within an organization, and thus a higher innovation performance, given a certain extent of being prospector oriented. This is also underlined by other researchers, who state that there is a need for more specific answers on how customer involvement contributes to innovation and new product performance, since the results on this are still inconclusive (Cui & Wu, 2016; Foss, Laursen & Pedersen, 2011; Hoyer et al., 2010). This research tends to extent this understanding by including an internal socialization process, that is, connectedness, to find out how this affects this relationship.

Fourth, this research tends to contribute to practice as well, by offering insights in how organizations could be better able to absorb this external knowledge: whether stimulating informal interaction would make this process easier and more effective in terms of innovation performance. These insights could help managers to determine whether it is beneficial to involve customers depending on their strategic orientation. Moreover, it could help managers locate the inertial pressures that cause the inability of integrating new external knowledge and thus the inability to actually innovate, and ultimately leverage those achievements in terms of

(9)

better meeting the demands of customers and therefore gaining a higher innovation performance.

The next section covers the existing literature on innovation performance, prospector orientation, customer involvement and connectedness. Here, the hypotheses, based on this review, are presented. Subsequently, the methodology of this study is discussed in which the research design, the procedure and the measurements are specified. Next, the results section shows the findings of this study, after which the discussion concludes with the interpretation of the results as well as with the theoretical and practical implications of this study and the limitations and suggestions for future research. Finally, this paper ends with a small revision in the form of a conclusion.

2 Theoretical framework and hypotheses

Strategic orientation and innovation performance

One of the most important factors for a firm to be able to innovate successfully, is the ability to create a clear strategy (Georgellis et al., 2000; Laforet, 2008). A clear strategy identifies itself as being effective, efficient, valuable, rare and inimitable (Barney, 1991). In turn, formulating this strategy results in a competitive advantage for a given firm. Further, in order for this strategy formation process to be successful, it has to match the characteristics and configuration of a firm’s strategic orientation, which defines the broad outline while leaving out the details of the content and implementation of a firm’s strategy (Slater, Olson & Hult, 2006). Those broad outlines define in which direction a firm tends to go in creating behaviors in such a way that they can achieve superior performance (Slater et al., 2006).

The most frequently used classification of strategic orientation is the typology of Miles and Snow (Miles et al., 1978). This typology makes a distinction in strategic orientations based on the organization’s response to changes in the environment by taking into account market,

(10)

innovation and technological orientation (Hambrick, 1982; Hambrick, 1983; McDaniel & Kolari, 1987; Miles et al., 1978). As organizations engage in an ongoing process of restructuring and evaluating their strategies, processes and structures, they tend to move through an adaptive cycle in different ways (Miles et al., 1978). Although this adaptive cycle is mainly determined by environmental conditions, it is stated that top-managers’ choices are important as well for shaping the internal processes and structures (Miles et al., 1978). Managers have to define their organizational domain; they have to create a technological system for production and distribution; and they have to develop administrative systems that allow for monitoring and directing current and future activities (Conant, Mokwa & Varadarajan, 1990; Slater & Narver, 1993). Making those choices results in the identification of four different strategic types, that is: defenders, prospectors, analyzers, and reactors.

Whereas the results on overall firm performance given a specific strategic orientation are still inconclusive (Aragón-Sánchez & Sánchez-Marín, 2005), results on the impact of strategic orientation on new product development are consistent throughout the literature. It seems that for prospectors, the development of new products is the most important source for competitive advantage, meaning that this can result in substantial growth (Hambrick, 1983; Laforet, 2008; McDaniel & Kolari, 1987; Slater & Narver, 1993). This results in prospectors having the highest values of innovation performance as opposed to the other strategic orientations included in the typology of Miles and Snow (Aragón-Sánchez & Sánchez-Marín, 2005; Slater & Narver, 1993). This claim is supported by research on innovation performance of firms. It seems that more innovative firms are characterized by being R&D oriented and by being proactive in developing new technologies and products, which in turn could lead to more radical innovations with the potential for a competitive advantage (Gatignon & Xuereb, 1997).

Hence, prospectors can be identified as being more innovative based on their proactiveness, as being the ‘aggressiveness with which businesses pursue growth opportunities

(11)

in their product-markets’ (Slater & Narver, 1993, p. 35), and based on their focus on identifying and exploiting new product-market opportunities by focusing on exploration as their core strategy (McDaniel & Kolari, 1987; Ren & Guo, 2011; Slater & Narver, 1993). This proactiveness results in prospectors tending to distinguish themselves by differentiating their product lines (Laforet, 2008; Mustafa et al., 2015). It is about finding new opportunities and diversifying their activities (Hambrick, 1983; McDaniel & Kolari, 1987). Furthermore, since a product differentiation strategy is about valuing specific customer needs (Dirisu, Iyiola & Ibidunni, 2013), prospectors tend to be active in keeping up-to-date with how customers evolve (Laforet, 2008). When scanning the environment of their customers, and in turn, focusing on developing product lines that fits this environment, prospectors are able to increase the willingness to pay of those customers (Awa, 2010; Dirisu et al., 2013).

However, while Miles and Snow seem to make a definite distinction between four different strategic orientations, there can be differences between firms with the same strategic orientation based on the different dimensions of which it consists (Conant et al., 1990). Besides, the fact that the world of organizations is changing to a more complex, and dynamic one, results in a call for more research on examining organizational processes as more dynamic entities: they do not fit into categorical ideas anymore, meaning that they rather exist on a continuum.

As a result, being a prospector does not guarantee high values for innovation performance per se. In contrast, it could depend on the extent to which firms are actually prospector oriented in their strategy. Hence, as prospectors tend to be more innovative as opposed to the other strategic orientations identified by Miles and Snow, it is unclear how the extent to which firms are actually prospector oriented influences their innovation performance. The arguments mentioned above lead to the expectation that being more prospector is associated with higher amounts of the above mentioned characteristics. Therefore, the following hypothesis is formulated based on this argument:

(12)

H1: The more a firm is prospector oriented in its strategy, the higher its values of innovation

performance are.

Moreover, research on the relationship between a prospector orientation and innovation performance tends to be mostly about internally oriented determinants explaining this relationship. Although those internal factors are still considered to be the main determinants of innovation by many researchers (i.e. Romijn & Albaladejo, 2002; Vega-Jurado et al., 2008), changes in both the internal and external environment, as explained earlier, call for more research on the impact of external factors that could explain how a more prospector orientation could lead to a certain innovation performance.

Co-creation with customers

One of the most important external success factors for innovation is the interaction with customers, since they can provide missing input for which the organization itself cannot provide (Romijn & Albaladeo, 2002; Xu, Ribeiro-Soriano & Gonzalez-Garcia, 2015). Moreover, the role of customers is becoming more and more important, especially within industries where customer needs become more heterogeneous and where transparency, dialogue, and understanding of risk-benefits becomes essential (Sarkar & Costa, 2008; Sawhney, Verona & Prandelli, 2005). This, together with the rise of information technologies, results in customers being a key part of the external environment of organizations and the interaction between organization and customer becoming the locus of value creation (Humphreys & Grayson, 2008; Prahalad & Ramaswamy, 2004). Therefore, the market is changing from a place where customers merely play a role in product exchange at the end of the supply chain, to a market that is becoming a forum for conversation with customers, meaning that customers can be

(13)

involved in earlier stages of product development (Prahalad & Ramaswamy, 2004; Sawhney et al., 2005). This means that there is a shift of organizations creating value for customers, to creating value with customers, also called co-creation or customer involvement (Cui & Wu, 2016; Humphreys & Grayson, 2008).

In the process of co-creation, organizations acknowledge the important role of communities outside the boundaries of the firm in inventing, creating and developing innovative ideas. Those communities are defined as ‘a voluntary association of actors, typically lacking in a priori common organizational affiliation (i.e. not working for the same firm) but united by a shared instrumental goal, (…) [that is,] creating, adapting, adopting or disseminating innovations’ (West & Lakhani, 2008, p. 224). The Internet seems to be the most important source for tapping into those communities, since it offers an open and cost-effective network (Sawhney et al., 2005). It allows organizations to efficiently gain information about larger groups of people with shared interests and use the knowledge that is created by those groups.

However, using virtual communities is just one of the many ways in which organizations can strategize for co-creation. Especially the Internet offers more possibilities for organizations in how to involve customers in the new product development process. To make it more insightful, Cui and Wu (2016) offer a categorization perspective of customer involvement in innovation, by dividing the concept of customer involvement into three different forms. Those forms are distinguished from each other by the extent to which customers are actually involved in innovation processes and how active their role is.

The first form comprises ‘customer involvement as an information source’, meaning that the role of customers is passive rather than active: they only provide the information and are not actively involved in the actual innovation process. The second form, which is ‘customer involvement as co-developers’, is instead about actively engaging and collaborating with customers. Customers do not only provide information, but also solutions to problems and

(14)

specific inputs for i.e. new products. During the innovation process, there is a close interaction between customer and organization.

Finally, the third form comprises ‘customer involvement as innovators’, in which the locus of product development shifts outside the organization. In this form of customer involvement, the roles of organization and customer are switched, with the organization providing the knowledge and information and customers being the actual innovators, without any interaction with the organization during the innovation process (Cui & Wu, 2016). This research’s focus is on the second form of customer involvement, being ‘customer involvement as co-developers’, since it is about actively involving customers in the innovation process, without them being merely information sources or independent innovators.

Prospector orientation and customer involvement

When a firm decides to actively involve customers into their innovation processes, there is a need for them to have both the willingness and the ability to actually use this knowledge and combine it with their own knowledge. In this process, the strategic orientation of a firm starts to play an important role, since this constant factor determines what strategic initiatives are noticed, how firms respond to changes in their environment and, in turn, which decisions are made (McDaniel & Kolari, 1987; Ren & Guo, 2011).

First, the more a firm is prospector oriented in its strategy, the more its focus is on exploration. Since having a more exploratory focus tends to result in noticing exploratory initiatives more easily (Ren & Guo, 2011), those firms have a tendency towards finding new and divergent opportunities in order to expand, for example, their product lines, to ultimately achieve a higher performance. Those new and divergent opportunities, in turn, can mainly be found in external knowledge sources, since this allows organizations to bring in unknown knowledge from outside their own bubble (Romijn & Albaladeo, 2002). Especially involving

(15)

customers into the innovation processes is a way for organizations to enlarge and improve their knowledge base, since there is a certain distance between the organization and customers (Jespersen, 2010). They do not tend to have a common language at first, meaning that customers could come up with all kinds of different ideas diverging from the ideas that are brought up internally.

Second, as firms become more prospector oriented in their strategy, they tend to be more customer-centric in their approach, meaning that their highest priority is on creating customer value by actively keeping up-to-date with how customer evolve (Laforet, 2008; Olson, Slater & Hult, 2005). Since knowing your customers as well as their needs is of great importance nowadays, involving customers in the innovation process is the ultimate way for prospectors to increase the willingness to pay of their customers and therefore to meet the specific demands of existing and new customers (Awa, 2010; Dirisu et al., 2013).

Third, as firms become more prospector, they tend to have higher values of strategic flexibility due to their focus on exploration, that is, finding new opportunities and diversifying their activities (Hambrick, 1983; McDaniel & Kolari, 1987). This results in a stronger ability to deal with uncertainty and changes in the internal business as well as in the environment. They tend to have more experience and more capabilities to anticipate those changes. In turn, this indicates that firms with higher values of strategic flexibility can be identified by having a dynamic capability that allows them to ‘reallocate and reconfigure its organizational resources, processes, and strategies to deal with environment changes’ (Zhou & Wu, 2010, p. 549). Hence, the more a firm represents a prospector orientation, the higher their values of strategic flexibility, the stronger their willingness and ability to engage in cross-boundary activities like customer involvement that could result in more uncertainty and changes in their internal business, that is, their resources, processes and strategies.

(16)

In line with the arguments given, it is expected that the more firms are prospector, the stronger would be both their willingness and their ability to search for opportunities outside their organizational boundaries, leading to the following hypothesis:

H2A: The more a firm is prospector oriented in its strategy, the more it engages in customer involvement.

Knowledge management and absorptive capacity

Involving customers in the innovation process is an activity that goes beyond the boundaries of a firm, meaning that firms gather external knowledge through a process of outsourcing. This is associated with what is called ‘knowledge acquisition’ or ‘external knowledge exploration’, meaning that an organization deliberately chooses to actively listen to, in this case, their customers (Liao, Welsch & Stocia, 2003; Lichtenthaler & Lichtenthaler, 2009). By means of knowledge acquisition, organizations are able to access knowledge that is not available internally (Grimpe & Kaiser, 2010), which in turn enables them to focus more on differentiation (Zahra & Nielsen, 2002). Furthermore, using external knowledge sources for innovation can reduce investments in internal R&D, and therefore spare money, time and energy (Zahra & Nielsen, 2002).

However, the fact that this external knowledge exists outside the organization, does not mean that this knowledge is absorbed instantly (Escribano, Fosfuri & Tríbo, 2009; Matusik, 2000). Instead, engaging in knowledge acquisition implies the need of an integration process in order to use the external knowledge most effectively. When it comes to integrating knowledge, firms have a certain extent of absorptive capacity which enables them to recognize, assimilate, transform, and exploit new external knowledge, which ultimately results in the integration of this new knowledge with existing knowledge (Zahra & George, 2002).

(17)

When it comes to customer involvement, new knowledge is created outside the organizational boundaries after which it can be internally applied and retained over time (Lichtenthaler & Lichtenthaler, 2009). The use of those external knowledge sources creates new opportunities for organizations, in addition to the opportunities created via their internal R&D investments and innovations (Cheng & Huizingh, 2014). This is explained by the fact that customer involvement can result in the ability to actually develop products and services that closely meet the demands of the customer. Those developed products and services, in turn, can provide greater returns in terms of innovation performance (Abdul Adis & Jublee, 2010). Furthermore, due to growth of IT and communication technology, it has become easier to search for new information, which in turn results in acquiring and handling external knowledge more efficiently (Kang & Kang, 2009). Therefore, organizations can both access new knowledge more easily and more thoroughly, and, as a result, innovate more successfully (Kang & Kang, 2009).

Taken together, customer involvement can be considered as an important source of innovation for organizations, since it allows for obtaining tacit and explicit knowledge from customers to improve internal processes and to create new products (Xu et al., 2015). This results in customer involvement being identified as a potential external determinant through which innovation performance can be predicted, leading to the following hypothesis:

H2B: The greater the extent to which a firm engages in customer involvement, the higher is the

innovation performance of this particular firm.

The mediating effect of customer involvement

Prospector firms tend to have a more market and entrepreneurial orientation, meaning that those firms can be characterized by proactiveness, entrepreneurial processes, and an

(18)

outward orientation (Cheng & Huizingh, 2014; Han et al., 1998; Huizingh, 2010). This, in combination with the changing environment of the organization, results in the need for them to gain a more external focus to be able to maintain and broaden customer bases, and meet the specific demands of existing and new customers.

Hence, nowadays, merely monitoring the customer base, as indicated by the market orientation, and focusing on internal exploration, as indicated by the entrepreneurial orientation, is not enough anymore to reach successful innovations. In contrast, customers have become more demanding than ever before, with their demands being more complex and volatile. This results in customer involvement being the eminent way for firms to maintain their strategy, and execute the strategic intentions resulting from this strategy, to ultimately actually being able to innovate (Awa, 2010; Dirisu et al., 2013). Therefore, instead of only monitoring customer demands, firms need to open up their organizational boundaries and actively engage and involve the most important stakeholder group into their organizational processes. Only through this process, prospectors are able to understand evolving customer needs, by which they can offer specific products and services meeting the demands. This, in turn, determines the success of innovations: firms are able to increase the willingness to pay of customers (Abdul Adis & Jublee, 2010).

Thus, the choice to either involve customers or not, eventually determines whether prospectors will come to innovations and thus a certain innovation performance: it is nowadays the necessary road to walk for prospectors. This leads to customer involvement mediating the relationship between prospector orientation and innovation performance, by which hypothesis 2A and 2B are combined in the following hypothesis:

H2: Customer involvement mediates the relationship between a prospector orientation and the

(19)

The moderating effect of connectedness

Zahra and George (2002) were the first that actually opened the black box of absorptive capacity by stating that it is a multidimensional construct, by which they build upon the dynamic capabilities view of the firm. This results in their distinction between two different forms of absorptive capacity, that is, potential absorptive capacity on the one hand, and realized absorptive capacity on the other. The first is defined as the ability of an organization to acquire and assimilate external knowledge (Zahra & George, 2002). The latter encompasses the ability to internally transform and exploit the acquired knowledge (Lichtenthaler & Lichtenthaler, 2009). This distinction is important, since some organizations will be able to acquire and assimilate external knowledge, but not have the ability to actually leverage the knowledge that has been absorbed (Zahra & George, 2002).

Since involving customers in the product development process can be seen as an effort of identifying, acquiring and selecting new external knowledge, this directly relates to an organization having a potential absorptive capacity, as defined by Zahra and George (2002). This potential absorptive capacity is the ability of an organization to acquire and assimilate external knowledge (Zahra & George, 2002). Hence, in choosing to involve customers in the innovation process, organizations acknowledge the opportunity of using the knowledge of the customers in order for them to, ultimately, be able to innovate successfully.

However, the question that arises here, is whether outsourcing innovation by means of customer involvement always results in successful innovation, and thus a higher innovation performance (Grimpe & Kaiser, 2010). The fact is, namely, that in order for an organization to retain the knowledge acquired and assimilated, it needs to be integrated and institutionalized throughout all organizational levels (Grant, 1996). This means that having a potential absorptive capacity does not automatically result in what is called a realized absorptive capacity

(20)

(Zahra & George, 2002). Realized absorptive capacity encompasses the ability to internally transform and exploit the acquired knowledge (Lichtenthaler & Lichtenthaler, 2009).

In order for an organization to integrate knowledge effectively, it needs to be structured in such a way that it enables knowledge flows to reach all the relevant departments and individuals (Liao et al., 2003). This means that an organization’s absorptive capacity and, in turn, the way an organization manages its knowledge flows, strongly depend on its employees’ knowledge sharing activities (Cohen & Levinthal, 1990; Grant, 1996; Lichtenthaler & Lichtenthaler, 2009). Moreover, the extent to which an organization can, in fact, integrate specialized knowledge effectively, determines whether it will gain innovative capabilities, and, in the end, a competitive advantage (Escribano et al., 2009; Grant, 1996b).

Grant (1996b) further states that the knowledge integration process can be viewed as a hierarchy, meaning that different degrees of knowledge integration are needed for different types of activities. At the highest level of this hierarchy, new product development arises, since this activity requires wide-ranging cross-functional integration and is characterized by a high degree of novelty and being exploratory (Grant, 1996b; Sheremata, 2000). This implies that in order for an organization to effectively integrate external knowledge to eventually engage in new product development, they have to develop certain organizational capabilities, named combinative capabilities, that enable them to achieve this highest level of hierarchy (Jansen, Van den Bosch & Volberda, 2005).

Since integration is defined ‘as the process of achieving unity of effort among the various subsystems’ (Lawrence & Lorsch, 1967, p.4), effective organizational structures can be created by the use of so-called socialization mechanisms or social integration mechanisms (Grant, 1996b; Lawson et al., 2009; Zahra & George, 2002). By use of those mechanisms, social interaction is stimulated. More social interaction means that people within the organization get more opportunities to share their ideas and new information. This, in turn, facilitates the

(21)

exploitation of the generated knowledge throughout the organization and therefore in higher levels of knowledge sharing (Tsai, 2002; Tsai, Liao & Hsu, 2015). Intensive knowledge sharing, in turn, has shown to have a positive effect on product innovation performance (De Luca & Atuahene-Gima, 2007).

The most crucial way for organizations to stimulate social interaction is by focusing on the more non-regulated, informal, voluntary and personal mode of interaction (Tsai, 2002), since this form of interaction stimulates problem solving and social trust (Atuahene-Gima, 2003; Chow & Chan, 2008); blurs boundaries between organizational units (Tsai, 2002); and allows a specific unit to use firm-wide knowledge for creating new ideas (Atuahene-Gima, 2003). This form of informal social interaction, working as an integration mechanism, is called connectedness, or strength of social ties (Jansen, Van den Bosch & Volberda, 2006; Jansen et al., 2009; Sheremata, 2000). Connectedness is defined as ‘the extent to which individuals in a department [are] networked to various levels of hierarchy in other departments’ (Jaworksi & Kohli, 1993, p. 59). Furthermore, higher levels of connectedness refer to the fact that individuals within an organization have more opportunities to informally connect with each other (Jansen et al., 2006). This, in turn, results in enhancing the assimilation of external knowledge, as well as the transformation and exploitation of this knowledge, meaning that it is both positively related to potential and realized absorptive capacity (Jansen et al., 2005).

Hence, it is expected that connectedness works as a moderating factor in the relationship between customer involvement and innovation performance, resulting in the following hypothesis:

H3: The higher the extent to which a firm engages in connectedness, the stronger the

relationship is between the use of customer involvement and the innovation performance of this particular firm.

(22)

Conceptual model

The theoretical review results in the following conceptual model, as shown in Figure 1, presenting all variables and relations between those variables in the form of the formulated hypotheses.

3 Methodology

3.1 Research method

This research is conducted by using a web survey, since this allows for collecting a large amount of data in a short period of time. Besides, a web survey can easily provide respondents with anonymity, which in turn is important for increasing the response rate (Rogelberg et al., 2006). In addition, by using a survey there is no researcher subjectivity involved. Therefore, it provides so-called standardized stimulus, meaning that each participant will be exposed to the same questions. In turn, this results in the possibility to compare the scores with each other more easily. Finally, it seems that surveys are especially helpful when the subject of matter is not too sensitive, since subject sensibility is negatively related to response rate (Fan & Yan, 2009). In this case, there could be said to be hardly any sensibility included, since the organizational processes examined in this research do not ask for providing personal or

Figure 1. Hypothetical model: mediating effect of customer involvement and moderating effect of connectedness

(23)

confidential information. Further, the constructs to be measured in this research are what is called behavioral in nature, meaning that they are subject to different interpretations and that they rely on individual’s self-perceptions. It namely seems that survey research is especially useful when the topics to be examined comprise complex questions like what, how, or why (Pinsonneault & Kraemer, 1993). For example, a construct like connectedness depends on the human interpretation, since it encompasses a certain judgment, and therefore, only a survey can capture those judgments in answering complex questions.

3.2 Sample and data collection

This quantitative, non-experimental research is conducted at firms from several different industries. The data were collected from senior executives, being CEOs, managing directors, senior managers or the like. Each manager counted as representative of their firm, meaning that the amount of managers matched the amount of firms included in this research.

Before approaching firms, a selection of firms was made by using the website of ‘EuroPages’, where companies from all European countries and all industries can be found. Here, the database was selected by country, being the Netherlands, and industry. Further, the database of ORBIS was used to select even more companies, selecting the database by the same characteristics. In selecting by industry it was important to only include the industries in which customer involvement could play a role, since this is the focus of this study. In the end, this resulted in five industries included in the sample, being the food industry, the software industry, the travel and leisure industry, luxury goods and services, and the fashion industry.

Since personal information about senior executives is rarely provided, the general mail addresses stated on the websites were used to contact all firms. If a personal mail address of a senior executive was stated as well, however, a personal e-mail was send to him/her instead of using the general mail address. Firms were kindly asked to participate by including a letter

(24)

(Appendix 2). This letter briefly explained the purpose of the research, without giving away the exact research question, since this would push the respondents in a certain direction too much. In the e-mail sent to the general mail addresses, the receiver was kindly requested to forward this mail to someone in the function of CEO, CFO or the like. In addition, respondents were promised anonymity, as well as the possibility to receive a management summary with the results of the research. All to increase the possibility that they would be incentivized enough to participate. The link to the web survey was immediately included in the mail to decrease the amount of mail contact needed. In this way, it was made as convenient as possible for the senior executives, sparing them time and energy.

The sample consisted of 453 firms. In the end, 82 senior executives representing one firm participated in the survey. Before conducting the analyses, the data file was checked for missing values, meaning that only completely finished surveys were included for running the analyses. The others were deleted from the file. In the end, 50 respondents completely finished the survey. 28 respondents represented firms in the food industry (56%), 13 respondents represented firms in the software industry (26%), and 9 respondents represented firms in ‘other industries’ (18%). Overall, this reflects a response rate of 11%.

3.3 Measures and reliability

The survey used was conducted in Dutch (Appendix 3). In order to translate the items, direct translation was used. First, in order to measure the consistency of the measurements, reliability tests have been conducted for each variable scale. Further, for each construct, an exploratory factor analysis was done, by which the underlying structure of each construct could be examined. Appendix 1 shows all the original measurement items including the translation.

Dependent variable: innovation performance Innovation performance is operationalized in several different ways throughout the literature. Some studies use objective

(25)

measures like patents (Rothaermel & Alexandre, 2009) or share of sales of new products (Grimpe & Kaiser, 2010). However, some other researchers prefer measuring innovation performance by using a survey, since this allows for including the multidimensionality of this kind of performance (Carmona-Lavado, Cuevas-Rodríguez & Medina, 2010; Cabello-Medina, López-Cabrales & Valle-Cabrera, 2011). Alegre, Lapiedra and Chiva (2006) make a distinction between two dimensions of innovation performance, that is, efficacy and efficiency, meaning that both the success of innovations (efficacy) and the effort carried out to achieve this success (efficiency) are important, since the innovation performance is not only determined by the success of the innovations, but also by the costs and time put in those projects (Alegre et al., 2006). As a result, this research measured innovation performance by using the three-item measure of Carmona-Lavado et al. (2010), including ‘proportion of technologically new or improved products in the turnover of the company’ (α = .80). The Likert-scale ranged from (1) for less than competitors to (7) for more than competitors. Exploratory factor analysis showed that the construct loaded on one factor, explaining 72.8% of the variance with an Eigenvalue of 2.2. Besides, all factor loadings showed a score above 0.67.

Independent variable: prospector orientation The extent to which a firm has a prospector oriented strategy is measured by using a survey, since prior research shows that strategic awareness is positively related to hierarchical level, meaning that top management is able to identify their strategy quite accurately (Hambrick, 1981; Snow & Hrebiniak, 1980). It even seems that different managers within one organization all tend to agree on the strategy that is identified (Snow & Hambrick, 1980). In the survey, prospector orientation was measured by using the twelve-item measure of Avci, Madanoglu and Okumus (2011), including ‘our organization adopts a growth oriented strategy’ (α = .78). The Likert-scale ranged from (1) for strongly disagree to (7) for strongly agree. Exploratory factor analysis showed that the construct loaded on four factors. The first factor explained 30.3% of the variance with an Eigenvalue of

(26)

3.6. However, not all items had factor loadings higher than 0.50, which resulted in the deletion of four items. After deleting those items, the construct loaded on three factors instead, of which the first factor explained 41.8% of the variance, with an Eigenvalue of 3.3. The second factor explained 16.5% of the variance, with an Eigenvalue of 1.3. The third factor explained 14.7% of the variance, with an Eigenvalue of 1.2. Since the first factor accounted for the highest variance and the other two factors accounted for substantially less variance, prospector orientation was measured as one factor. After deletion, all items had factor loadings higher than 0.58. Further, one item was counter-indicative, which resulted in recoding this item into a different variable.

Mediator: customer involvement The extent to which a firm engages in customer involvement was measured by using the five-item scale for ‘customer involvement as co-developers’ of Cui and Wu (2016). Although Cui and Wu (2016) use three different scales for different types of customer involvement, this research made use of the scale for ‘customer involvement as co-developers’, since the focus of this study is on an active role of customers in the innovation process, without them being innovators independent from the organization (Cui & Wu, 2016). The scale was changed from past simple to present simple, including ‘our customers’ involvement as co-developers is significant’ (α = .93). The Likert-scale ranged from (1) for strongly disagree to (7) for strongly agree. Exploratory factor analysis showed that the construct loaded on one factor, explaining 78% of the variance with an Eigenvalue of 3.9. Further, all loadings showed a score higher than 0.78 on this factor.

Moderator: connectedness Connectedness was measured by using the seven-item scale of Jaworski and Kohli (1993), including ‘people around here are quite accessible to those in other departments’ (α = .85). The Likert-scale ranged from (1) for strongly disagree to (7) for strongly agree. Exploratory factor analysis showed that the construct loaded on two factors. The first factor explained 51.3% of the variance with an Eigenvalue of 3.6. The second factor

(27)

explained 15.7% of the variance with an Eigenvalue of 1.1. Further, one item’s factor loading was lower than 0.50, which resulted in deleting this item from the scale. After deletion, the construct loaded on but on factor, explaining 59.8% of the variance with an Eigenvalue of 3.6. Further, all factor loadings scored higher than 0.56 on this factor.

Control variables In testing the hypotheses, this study first controlled for industry, size (employees) and age (years), since larger and older companies have shown to have more tendency towards innovation as well as being more exploratory (Bantel & Jackson, 1989). For the controls firm age and firm size, natural logarithms were created of the number of employees within the organization and the number of years from the firm’s founding to correct for non-normality. For the control variable industry three dummies were created, accounting for the food industry, the software industry, and other industries.

Second, this study controlled for competitive intensity (Jaworski & Kohli, 1993), slack (De Luca & Atuahene-Gima, 2007), and environmental turbulence (Citrin, Lee & McCullough, 2007; Han, Kim & Srivastava, 1998), since integration mechanisms tend to become more critical when the environment is less stable (Menon, Jaworski & Kohli, 1997). Competitive intensity was measured by using the six-item scale of Jaworksi and Kohli (1993), including ‘price competition is a hallmark in our industry’ (α = .83). Exploratory factor analysis showed that the construct loaded on one factor, explaining 52.7% of the variance with an Eigenvalue of 3.2. However, one item had a factor loading lower than 0.50, resulting in the deletion of this item. After deletion, the construct still loaded on one factor, explaining 60.7% of the variance, with an Eigenvalue of 3.0. Further, all factor loading scored above 0.53. Further, one item was counter-indicative, meaning that this one was recoded into a different variable.

Slack was measured by using the four-item scale of De Luca and Atuahene-Gima (2007), including ‘we have uncommitted resources that can be used to fund strategic initiatives on short notice’ (α = .73). Exploratory factor analysis showed that the construct loaded on one

(28)

factor, explaining 65% of the variance with an Eigenvalue of 2.0. Further, all factor loadings were above 0.50. The Likert-scale for competitive intensity and slack ranged from (1) for strongly disagree to (7) for strongly agree.

Environmental turbulence was split apart into both market and technological turbulence. Market turbulence was measured by using the four-item scale of Han et al. (1998), including ‘[there are] frequent changes in customer preferences’. However, market turbulence showed a Cronbach’s alpha of .04, which resulted in excluding this variable from any further analyses. Technological turbulence was measured by using the three-item scale of Citrin et al. (2007), including ‘the technology in this industry is changing rapidly’ (α = .86). Exploratory factor analysis showed that the construct loaded on one factor, explaining 78.3% of the variance with an Eigenvalue of 2.3. Besides, all factor loadings were higher than 0.68. Respondents were asked to rate the extent to which the items for environmental turbulence were applicable to their industry on a scale from (1) for to a low extent to (7) for high extent.

Finally, this study controlled for the extent to which organizations engage in formal socialization mechanisms, that is, cross-functional boundary spanning (Jansen et al., 2009), since companies which do invest more in these kind of integration mechanisms tend to put more effort in creating an environment where knowledge can be exchanged and integrated more easily. Formal socialization mechanisms were measured by using the five-item scale of Jansen et al. (2009), including ‘we have cross-functional teams to exchange knowledge between departments’ (α = .72). The Likert-scale ranged from (1) for strongly disagree to (7) for strongly agree. Exploratory factor analysis showed that the construct loaded on one factor, explaining 52.3% of the variance with an Eigenvalue of 2.6. Further, all factor loadings scored above 0.50.

(29)

4 Results

4.1 Correlation analysis

In conducting the analysis several control variables were included. The choice for those controls was based on the change in R2_{compared to the model without any controls, and on the} correlation matrix. This resulted in selecting the control variables cross-functionality, competitive intensity, technological turbulence, industry, and firm size, since those showed a substantial change in the R2 compared to the base model and since those indicated a significant correlation with the variables included in the models. As a result, the correlation matrix only shows the controls that were included into the models.

In Table 1, the means, standard deviations, and correlations coefficients of the study variables are provided. First, it shows that a prospector orientation was a strong predictor for innovation performance, meaning that the more prospector firms are, the higher their innovation performance is (r = 0.43, p < .01). Second, competitive intensity and technological turbulence did positively relate to innovation performance as well, with competitive intensity being negatively correlated (r = -0.29, p < .05) and technological turbulence being positively correlated (r = 0.36, p < .05). Further, software industry firms had higher values of innovation performance than firms in the food industry or in other industries.

Third, being more prospector correlated with engaging more in customer involvement (r = 0.29, p < .05), and with having to deal with a more technological unstable environment (r = 0.33. p < .05). Fourth, both connectedness and cross-functionality positively correlated with customer involvement (r = 0.33, p < .05), meaning that as firms have more informal or formal socialization mechanisms, they engage in customer involvement more. Lastly, connectedness positively correlated with cross-functionality (r = 0.36, p < .05).

(30)

Table 1. Means, Standard Deviations, and Correlationsa M s.d. (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (1) Innovation performance 4.95 1.00 (0.80) (2) Prospector orientation 5.42 0.65 0.43** (0.78) (3) Customer involvement 4.92 1.36 0.21 0.29* (0.93) (4) Connectedness 6.05 0.78 0.04 -0.04 0.33* (0.85) (5) Cross-functionality 4.58 1.19 0.24 0.21 0.33* 0.36* (0.77) (6) Competitive intensity 4.48 1.18 -0.29* -0.24 -0.26 -0.17 0.06 (0.83) (7) Technological turbulence 4.79 1.38 0.36* 0.33* 0.21 0.14 0.22 -0.10 (0.86) (8) Firm sizeb 4.77 1.81 -0.10 0.06 -0.13 -0.08 0.18 0.44** 0.06 (-) (9) Food 0.56 0.50 -0.20 -0.07 0.05 0.02 -0.02 0.47** -0.42** 0.37** (-) (10) Software 0.26 0.44 0.34* 0.01 0.24 0.17 0.07 -0.37** 0.54** -0.30* -0.67** (-) (11) Other 0.18 0.39 -0.13 0.09 -0.33* -0.22 -0.06 -0.18 -0.07 -0.13 -0.53** -0.28 (-)

Notes. Numbers in the parentheses on the diagonal are Cronbach’s alphas of the composite scales a _{n = 50}

b_Log

** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).

(31)

4.2 Hypothesis testing: direct effect prospector orientation and innovation performance

First, to examine the direct relationship between prospector orientation and innovation performance (H1), a hierarchical regression was conducted (Table 2). The model with innovation performance as dependent variable was significant (F (7, 42) = 3.06, p < .05). Prospector orientation explained an additional variance in innovation performance of 9%. This change in R2 was significant, F (1, 42) = 5.63, p < .05. As Table 2 shows, a prospector orientation was significantly related to innovation performance (b = 0.35, t = 2.37, p < .05). In other words, a prospector orientation positively related to innovation performance, meaning that two firms that differed by one unit in their prospector orientation, were estimated to differ by 0.35 units in their reported innovation performance. As a result, H1 was supported.

Table 2. Results of the hierarchical regression for prospector orientation explaining innovation performance Innovation performance Model 1 Model 2 Control variables Cross-functionality 0.19 0.14 Competitive intensity -0.27 -0.14 Technological turbulence 0.26 0.10 Firm size -0.04 -0.06 Food industry 0.19 0.17 Software industry 0.20 0.33 Independent variable Prospector orientation 0.35* Adjusted R2 0.14 0.23 D adjusted R2 _0.25 _0.09

Note. N = 50. Standardized regression coefficients are reported. * p < .05

(32)

4.3 Hypothesis testing: mediating effect of customer involvement

Second, to examine the mediating effect of customer involvement in the relationship between prospector orientation and the innovation performance of a firm, Model 4 of the PROCESS macro written by Andrew F. Hayes for SPSS was used (Figure 2).

Table 3. Results of PROCESS analysis (mediation) with dependent variable innovation performance Consequent

CustInv (M) InnovPerf (Y)

Antecedent Coeff. SE p Coeff. SE P

Prospector (X) a1 .33 .26 .21 c1’ .49 .20 p < .05 CustInv (M) --- --- --- b1 -.10 .12 .38 Cross-functionality .34 .15 p < .05 .15 .12 .21 CompIntens -.28 .17 .09 -.13 .13 .31 TechnTurb .09 .17 .60 .08 .13 .54 Food industry 1.48 .48 p < .01 .49 .41 .23 Software industry 1.19 .58 p < .05 .86 .49 .07 Firm size -.12 .11 .27 -.05 .08 .60 constant i1 1.89 1.59 .24 i2 1.58 1.10 .16 R2 = .38 R2 = .35 F (7, 42) = 3.62, p < .01 F (8, 41) = 2.76, p < .05

First, the results in Table 3 indicate that the model as a whole with customer involvement as mediating factor was significant (F = 3.62, p < .01) and explained 38% of the variance in customer involvement. Further, it seemed that a prospector orientation did not significantly predict customer involvement (b = .33, t = 1.27, p = .21). This means that the effect was not statistically different from zero: with including the control variables, firms did not significantly

(33)

engage in more or less customer involvement as they became more prospector oriented. Therefore, H2A was not supported.

Second, the model as a whole with innovation performance as dependent variable was significant as well (F = 2.76, p < .05). Together the variables explained 35% of the variance in innovation performance. Further, it can be seen that, with customer involvement and the aforementioned control variables included in the model, a prospector orientation did significantly predict innovation performance (b = .49, t = 2.49, p < .05). The positive b for the prospector orientation means that as firms are more prospector oriented, their innovation performance increases as well. However, the effect b1 = -.10 (t = -.88, p = .38) indicates that customer involvement did not significantly predict the innovation performance. Hence, there was no significant increase of decrease in innovation performance as firms engaged in more or less customer involvement. Therefore, based on those results, H2b was not supported either.

Finally, interesting to see here is the significant coefficient between cross-functionality and customer involvement (b = .34, t = 2.30, p < .05), meaning that as firms were more cross-functional organized, they significantly engaged more in customer involvement. In addition, firms from the food industry as well as from the software industry were engaging in significantly more customer involvement compared to firms from ‘other industries’ (b = 1.52, t = 3.17, p < .01).

Table 4. Direct, total and indirect effect with dependent variable innovation performance

Effect SE p

Direct effect c1’ .49 .20 p < .05

Total effect c1 .46 .19 p < .05

Boot SE Boot LLCI Boot ULCI

(34)

In Table 4 the difference between the direct and total effect of prospector orientation on innovation performance is shown. The direct effect of prospector orientation was c1’ = 0.49, indicating that the firm that was more prospector oriented but equally engaging in customer involvement, was estimated to be 0.49 units higher in its reported innovation performance. This direct effect was statistically different from zero, t = 2.49, p < .05. The total effect of prospector orientation was c1 = 0.46, meaning that two firms that differed by one unit in their prospector oriented strategy, were estimated to differ by 0.46 units in their reported innovation performance. The positive sign means that firms scoring higher on having a prospector orientation had a higher innovation performance. This effect was as well statistically different from zero, t = 2.37, p < .05.

Further, Table 4 shows that the indirect effect was not statistically different from zero, as indicated by the 95% confidence interval from -0.27 to 0.06. Since zero is part of this interval, it cannot be stated that an indirect effect of customer involvement on the relationship between prospector orientation and innovation performance existed. Therefore, based on this analysis, the mediation hypothesis (H2) was not supported: there was no significant mediating effect of customer involvement.

4.4 Hypothesis testing: moderating effect of connectedness

Second, to examine the moderating effect of connectedness in the relationship between customer involvement and the innovation performance of a firm, Model 1 of the PROCESS macro written by Andrew F. Hayes for SPSS was used (Figure 3).

(35)

Table 5. Results of PROCESS analysis (moderation) with independent variable customer involvement and dependent variable innovation performancea_.

b SE t p Intercept 3.64 [1.53, 5.76] 1.05 3.48 p < .01 Connectedness (M) -0.11 [-0.71, 0.49] 0.30 -0.37 .71 Customer involvement -0.02 [-0.39, 0.35] 0.18 -0.10 .92 CONN x CUST 0.22 [-0.28, 0.71] 0.25 0.89 .38 Cross-functionality 0.20 [-0.10, 0.50] 0.15 1.38 .18 CompIntens -0.22 [-0.55, 0.11] 0.16 -1.32 .19 TechnTurb 0.17 [-0.13, 0.47] 0.15 1.14 .26 Food industry 0.56 [-0.55, 1.68] 0.55 1.02 .31 Software industry 0.60 [-0.52, 1.73] 0.56 1.08 .29 Firm size -0.01 [-0.21, 0.20] 0.10 -0.06 .96

Notes. 95% bias corrected and accelerated confidence intervals reported in parentheses. a_R2_{= 0.33, F (9, 40) = 1.48, p = .19}

For a moderation effect to take place, it is important to first look at the direct relationship between customer involvement and innovation performance. As is shown by previous analyses, this relationship was not significant. This can be confirmed by the results in Table 5, indicating that customer involvement was no significant predictor of a lower or higher innovation performance (b = -0.02, t = -0.10, p = .92). Besides, the 95% confidence interval, [-0.39, 0.35], shows that zero is part of the interval, meaning that the correlation could be either negative, positive or zero (no effect). In addition, there was neither a significant effect for connectedness on innovation performance (b = -.11, t = -0.37, p = .71). Both those insignificant relationships resulted in an insignificant interaction effect (b = .22, t = .89, p = .38). Moreover, the model as a whole also showed no significance (F = 1.48, p = .19). Therefore, H3 was not supported based on this analysis.

Finally, though all measurement scales have been tested in prior research thoroughly, an additional statistical procedure was taken to test the effect of common method bias. In this, a Harman’s one-factor test on items was conducted. Since the factor analysis showed that the

(36)

first factor did not account for the majority of the variance (30.2%), there were no reasons to be concerned about problems associated with common method bias.

4.5 Post-hoc analyses

Since both the mediation and moderation relationships were not significant, additional post-hoc analyses were conducted.

First, as was indicated, firms from different industries participated in the survey, being the food industry, the software industry and ‘other industries’. To find if there were any differences between those industries, independent t-tests were conducted for all variables. The results showed that, on average, firms from the software industry reported a higher innovation performance (M = 5.51, SE = 0.23), than firms from the food industry (M = 4.77, SE = 0.18). This difference, -0.74, BCa 95% CI [-1.38, -0.11], was significant t (39) = -2.39, p < .05. There was 0.82 of a standard deviation difference between those two industries in terms of their innovation performance, representing a large effect. The results showed that there was no significant difference between the innovation performance of firms from the food industry and ‘other industries’ (t (35) = 0.28, p = .78), and firms from the software industry and ‘other industries’ (t (20) = 2.04, p = .06). Further, firms from ‘other industries’ scored lower on customer involvement (M = 3.96, SE = 0.52) compared to firms not from the ‘other industries’ (M = 5.13, SE = 0.19). This difference -1.17, BCa 95% CI [-2.13, -0.21], was significant t (48) = -2.45, p < .05. There was 0.83 of a standard deviation difference in terms of their customer involvement, representing a large effect.

Moreover, the results from the independent t-test showed that competitive intensity was substantial higher for firms in the food industry (M = 5.01, SE = 0.21), compared to firms not from the food industry (M = 3.81, SE = 0.26). This difference 1.20, BCa 95% CI [0.54, 1.86], was significant t (48) = 3.66, p < 01. There was 1.04 of a standard deviation difference in terms

(37)

of their experienced competitive intensity, representing a large effect. In addition, technological turbulence was substantial higher for firms in the software industry (M = 6.03, SE = 0.31), than for firms not from the software industry (M = 4.35, SE = 0.20). This difference 1.67, BCa 95% CI [0.91, 2.44], was significant as well t (48) = 4.40, p < .001.

Second, a split sample between food industry, software industry, and ‘other industries’ was conducted to see whether there were any differences between those industries in the relationships between the variables included in the model. The results of the linear regression showed that, for firms from ‘other industries’, there was no relationship found between a prospector orientation and innovation (F (1,7) = 1.76, p = .23). However, some interesting results were found for firms in the software industry. The regression showed that, for those firms, a prospector orientation explained 57% of the variance in customer involvement (F (1, 11) = 14.54, p < .01). It was found that the more prospector a firm in the software industry was, the higher was its engagement in customer involvement (b = 1.16, p < .01). For the food industry and firms in ‘other industries’, there was no significant relationship found between a prospector orientation and customer involvement (p > .05). Further, whereas for firms in the food industry and ‘other industries’ no relationship between customer involvement and innovation performance existed, for firms in the software industry, customer involvement explained 47% of the variance in innovation performance (F (1,11) = 9.59, p < .05). It was found that the higher customer involvement was for firms in the software industry, the higher its reported innovation performance was (b = 0.48, p < .05).

Moreover, there was found a significant mediation effect of customer involvement on the relationship between prospector orientation and innovation performance for the software industry. The results showed that there was a significant effect of prospector orientation on customer involvement (b = 1.16, t = 3.81, p < .01), as well as a significant effect of customer involvement on innovation performance (b = 0.55, t = 2.42, p < .05). The indirect effect showed

(38)

significance (b = 0.49), as indicated by the 95% confidence interval from 0.09 to 1.40. In addition, the total effect was as well significant (b = 0.61, t = 2.26, p < .05). This resulted in a significant partial mediation effect for software firms of customer involvement on the relationship between prospector orientation and innovation performance.

Finally, interestingly, based on the correlation analysis, there might be a mediation effect of prospector orientation on the relationship between customer involvement and innovation performance. The results showed that a direct effect of customer involvement on innovation performance was not significant (b = 0.07, t = 0.68, p = .50), as was already indicated by the analyses conducted earlier. However, there was in fact a significant effect of customer involvement on prospector orientation (b = 0.16, t = 2.07, p < .05), as well as a significant effect of prospector orientation on innovation performance (b = 0.52, t = 2.91, p < .01). The indirect effect showed significance as well (b = 0.08), as was indicated by the 95% confidence interval from 0.01 to 0.21. Hence, although there was no significant mediation effect of customer involvement on the relationship between prospector orientation and innovation performance for all firms, there was in fact found a significant full mediation effect of a prospector orientation on the relationship between customer involvement and innovation performance.

5 Discussion

5.1 Discussion of the results

This paper tends to contribute to research on cross-boundary activities by including an external factor through which more prospector oriented firms are able to come to innovation performance. Moreover, it was expected that connectedness would positively affect the relationship between customer involvement and innovation performance.

Prospector orientation and innovation performance : the mediating role of customer involvement and the impact of connectedness