Developing a competency model for open innovation: From the individual to the organisational level

From innovation to OI

Innovation research embraces a broad array of topics, including types, dimensions and determinants of innovation, which are studied at the macro, organisational and micro levels, using a range of theoretical lenses that include institutional and evolutionary economics, networks, the resource-based view, learning and change theories (Crossan and Apaydin, 2010). As a dynamic triptych of process, outcome and context (Crossan and Apaydin, 2010; Chesbrough and Bogers, 2014), the term “innovation” is broadly used by scholars and practitioners to denote new products and artefacts, new services, new manufacturing and business processes and new organisational structures, procedures, business models and strategies (Keupp et al., 2012; Keely et al., 2013). This broadening of conceptualisation and approach took root during the 1990s with the emergence of the innovation systems (IS) approach (e.g. Freeman, 1987; Lundvall, 1992; Nelson, 1993). Drawing primarily on evolutionary economics and a macro perspective, the IS approach has, however, been criticised for treating the firm as a black box, offering little guidance as to how companies, the innovation engines of an economy, might act in order to leverage the dynamics of such systems (Walrave and Raven, 2016). To address this issue, complementary approaches were developed, including innovation clusters (Porter, 2000), triple helix (Etzkowitz and Leydesdorff, 2000) and entrepreneurial ecosystems (Feld, 2012), emphasising the interactions and interplay among firms, institutions and market conditions in IS. Over the last decade, this has resulted in a shift of focus in firm-level innovation research from the product to the business model (Visnjic et al., 2016).

In this context, OI has emerged as a model that bridges the macro and micro levels in innovation studies (Carayannis and Campbell, 2011; Huizingh, 2011), where companies strive for innovation partly by tapping into knowledge residing outside their boundaries and partly by allowing their own internally developed knowledge to flow outward for external use (Chesbrough, 2003). Environmental dynamism enhances the development of OI, but it is a complex process, requiring effective management of both external and internal knowledge (Martinez-Conesa et al., 2017). The theoretical foundations of OI can therefore be seen as a firm-level mirroring of the IS literature. OI conceptualises innovation as occurring between companies and other relevant actors, who exchange knowledge and co-develop products and services in loosely coupled networks, including technology-enabled social networks (Palacios-Marqués et al., 2015), where business models are dynamically created, reshaped, dissolved and recreated to continuously enhance innovation competency, outcomes and performance (Chesbrough, 2003; Chesbrough, 2006; Chesbrough and Bogers, 2014).

OI encompasses a wide range of forms and degrees of openness in the innovation process (Laursen and Salter, 2006). Companies can engage in outbound OI (by revealing or selling ideas, knowledge, or technologies) or in inbound OI (by sourcing or acquiring innovation assets from outside) (Dahlander and Gann, 2010), or in both (by coupling external knowledge sources and outbound commercialisation activities) (Chesbrough and Bogers, 2014). A growing body of research on OI adoption shows that firms in various sectors use a range of different organisational modes (e.g. licensing agreements, alliances, purchase and supply of technical and scientific services) to enter into relationships with different types of partners (e.g. indirect and even direct competitors, suppliers, service and platform providers, technology brokers, universities, research organisations, crowds, lead users) with the aim of acquiring inbound OI and/or commercially exploiting outbound OI technologies and knowledge (e.g. Chiaroni et al., 2011; Bianchi et al., 2011; Spithoven et al., 2013; Virlée et al., 2015; Kortmann and Piller, 2016). The essential contribution of this body of research is to describe the content of change towards OI in terms of different OI activities and the context-specific industries/sectors in which it happens while less emphasis is placed on the process of change (Bianchi et al., 2011; Virlée et al., 2015). This corresponds to the call by Bogers et al. (2017) for research at the intra-organisational level of analysis “that help[s] to explain how individual-level attributes and behaviours as well as design elements of the organisation need to adapt as the organisation transitions to OI” (p. 22).

To link the “what” of OI activities to the “how” of individual attributes and behaviours in our analyses and competency model development, we compiled a list of OI activities based on Chesbrough and Brunswicker (2013, 2014). As in our own approach, that research relied on an original cross-country survey, anchored in innovation surveys such as the community innovation survey, which (among other things) provides systematic empirical evidence of OI activities. Their research confirmed that OI is a persistent and widespread phenomenon across industries and countries (Chesbrough and Brunswicker, 2014).

Professional competencies in support of OI

As noted above, the challenges of implementing and managing OI activities place new demands on the skills and abilities of the individuals involved. However, it is difficult for managers to define and source these skills and abilities in new recruits (Dabrowska and Podmetina, 2014). To identify the specific skills and behaviours for the successful execution and management of OI activities that must be found, developed and cultivated in the workforce, researchers have recently turned to the concept of professional competencies (Du Chatenier et al., 2010; Mortara et al., 2009; Hafkesbrink and Schroll, 2014). Before reviewing competency models for OI, it seems important to resolve the confusion around terms relating to individual competence and competency and to organisational capability.

Competence can be defined as a combination of the nature of the work and the characteristics of the worker performing it (Sandberg, 2000). It is a professional’s generic capability (Mulder, 2015), consisting of the integrated set of knowledge, skills and attitudes of a person (Mulder, 2007) or that an employee possesses (Rowe, 1995; Garavan and McGuire, 2001). It follows that a person or employee cannot have two or more sets of competences – in other words, a professional cannot have several competences (Mulder, 2015). A competency (plural competencies) is a part of competence (Mulder, 2015) – that is, the individual’s array of discrete knowledge, skills and attitudes/abilities/behaviours that enable them to cope with the demands and responsibilities of their job (Boyatzis, 1982; Lathi, 1999). Competencies should be formulated as “can” expressions (Mulder, 2015); they are observable and measurable, and their application results in effective and/or superior job performance, enabling a distinction between superior and average performance (Boyatzis, 1982; Catano, 1998). As the notion of organisational capability typically refers to distinctive strategic strengths at the organisational level (Athey and Orth, 1999; Luoma, 2000), organisational capabilities are distinct from competence and competency, both of which relate to the individual level and are therefore a human resource management issue.

Drawing on the above, the present research focuses on reviewing, identifying, measuring and empirically analysing competencies as the primary unit of analysis and as the building blocks of competences. Based on their review of the extant literature, Soderquist et al. (2010) synthesised three analytical perspectives that clarify the meaning and applicability of the various types of competency encountered in organisations. First, generic vs organisation-specific competencies, which are competencies characterizing a specific job, “either generically, i.e. common to all individuals occupying a specific job, or specific to the job in a particular organisation” (Soderquist et al., 2010, p. 328). Second, managerial vs operational competencies, which refer to competencies needed to successfully perform in a managerial or operational role, respectively. Third, competencies as skills vs competencies as abilities/behaviours, “which refer to characteristics of individuals that are either learned and describe what an individual does, or fundamentally inherent and describe how people do their job” (Soderquist et al., 2010, p. 328, emphasis added).

This typology of combinations of the three identified pairs has general applicability for competency management, as it suggests an analytical “scale” of competencies, ranging from the most specific “organisation-specific operational skills” to the most general “generic management abilities”. For the purposes of the present research, we focus on the analysis of generic managerial competencies for OI across industries and countries while distinguishing between skills and abilities in the review and further operationalisation of competencies in the research instrument and subsequent data analysis.

As skills, competencies specify the sufficiency of skills needed to execute a specific job (Soderquist et al., 2010), which is particularly important for the knowledge-intensive work (Lawler, 2005) that characterises innovation. Skills are developed through training, experience and knowledge transfer (Lauby, 2013). As behaviours, competencies are often referred to as abilities that are more innate than skills (which are more acquired) (Lauby, 2013), defining the desirable attributes for a specific job (e.g. creativity, initiative, persistence in problem solving, discipline, assertiveness, empathy and the ability to communicate and cooperate with others) (Nordhaug, 1998; Rowe, 1995). Abilities are particularly important in organisational contexts that are characterised by discontinuous change (Harvey et al., 2000), as well as by innovation and, in particular, by OI. Rather than sufficiency, abilities can only be evaluated in terms of performance expectations, embracing the notion of excellence (Rowe, 1995).

Competency models for OI

Turning to studies that explicitly analyse OI competencies, Mortara et al. (2009) proposed four categories of relevant skills:

1. introspective skills related to the organisation’s assessment of opportunities from inside;

2. extrospective skills related to the organisation’s assessment of opportunities from outside;

3. interactive skills that convey the value, internally and externally, of relations with the world outside the organisation; and

4. technical skills, encompassing all the technical, management and business skills needed to support the first three categories.

These skill categories are complemented by a set of desirable personal attributes (i.e. abilities as defined above), including (among others) motivation, sociability, a techno-business mindset, systems thinking, adaptability and flexibility (Mortara et al., 2009). Table I summarises indicative skills and abilities from the competency models reviewed here, which we use as inputs to the research instrument. The final complete list of variables retained and analysed is shown in Table III (see methodology section), which also details their operationalisation.

Based on an extensive literature review and qualitative validation through interviews and focus groups, Du Chatenier et al. (2010) identified four categories of OI competencies:

1. self-management (seen as the basis for achieving the central tasks related to OI);

2. interpersonal management (essential for managing inter-organisational collaboration);

3. project management (essential for managing the overall innovation process); and

4. content management (essential for creating new knowledge collaboratively).

Although it represents a great step forward in understanding the specificities of OI competencies, the work of Du Chatenier et al. suffers from generality (Mulder, 2015), as it fails to distinguish analytically between OI skills and abilities and further lacks clarity by conflating the concepts of competence and competency (Mulder, 2015). While the present study corrects for these imperfections, we nevertheless employed several of the competencies proposed by Du Chatenier et al. (2010) (indicatively listed in Table I) in building our research instrument.

Hafkesbrink and Schroll (2014) drew on the theoretical underpinnings of exploration, exploitation (March, 1991) and their simultaneous integration in ambidexterity (Raisch et al., 2009; Blindenbach-Driessen and van den Ende, 2014) to elaborate a comprehensive catalogue of organisational and individual OI competencies. In innovation studies, exploration – associated, for example, with search, variation, risk taking, experimentation and discovery (March, 1991) – has become synonymous with radical innovation, while exploitation – associated, for example, with refinement, efficiency, selection, implementation and execution (March, 1991) – has become synonymous with incremental innovation (Jansen et al., 2009). Ambidextrous organisations are “capable of simultaneously exploiting existing competencies and exploring new opportunities” (Raisch et al., 2009, p. 685). In this way, exploration and exploitation can work together towards innovation that is simultaneously incremental and radical, referred to as innovation ambidexterity (Li et al., 2008). External knowledge sourcing, which is central to inbound OI, was recently found to be positively associated with organisational ambidexterity, mediating the relationship between organisational ambidexterity and firm performance (Vrontis et al., 2017). Based on empirical research in 189 Italian knowledge-intensive firms, Vrontis et al. proposed that greater use of external knowledge in pursuit of innovation helps in managing the internal tensions entailed by joint exploitation and exploration activities, identified by March (1991) as one of the main difficulties in developing ambidexterity. Firms that adopt OI can obtain greater benefits from ambidexterity; in turn, ambidexterity is central to innovation performance, especially in the OI context, which is characterised by intensive knowledge transfer and learning (Ferraris et al., 2017).

In relation to the ambidextrous nature of OI, Hafkesbrink and Schroll (2014) identified four interdependent categories of key individual competencies for managers and employees: professional competencies, methodic competencies, social competencies and personal competencies. It should be noted that, such as Du Chatenier et al. (2010), Hafkesbrink and Schroll’s model makes no distinction between skills and abilities but uses the words competencies, skills and abilities interchangeably for what we have defined here as either competencies as skills, or competencies as abilities. Hafkesbrink and Schroll emphasise the importance of knowledge and knowledge management for the effective implementation and execution of OI, as further analysed and confirmed in recent empirical studies (Ferraris et al., 2017; Scuotto et al., 2017), and indicative competencies from Hafkesbrink and Schroll (2014) used in our instrument-building and analysis are listed in Table I.

Finally, Dabrowska and Podmetina (2014) conducted an extensive literature review (keywords: “open innovation” and “capability*” or “competence*” or “skill*”) and collected data on job offers (keyword: “OI” (in the job title or job description)). In addition to the competencies identified in the above studies, we also used items from Dabrowska and Podmetina (2014) as indicatively listed in Table I.

Questionnaire development and validation by experts

With a view to conducting the first empirical study of requisite competencies for OI in a European industrial context, we developed a structured questionnaire covering “the following broad topic areas: 1) the current state of open innovation adoption in industry; 2) the perceived importance of open innovation at present and in the near future; and 3) the employee competencies currently considered important for open innovation” (Podmetina, Equey, Kleer, Lopez Vega, Dabrowska, Albats, Petraite, Soderquist, Rethi and Hafkesbrink, 2017; Podmetina, Soderquist, Dabrowska, Hafkesbrink and Lopez-Vega, 2017, p. 45). To assess the level of OI adoption, we applied the list of OI and collaborative activities adopted from Chesbrough and Brunswicker (2013, 2014) for inbound and outbound OI, indicatively comprising customer/consumer co-creation, crowdsourcing, collaborative innovation with external partners (suppliers, universities, competitors, etc.), participation in standardisation and IP in- and out-licensing (see Table AI).

The initial list of questionnaire items was reduced the following evaluation by 15 business and academic experts in OI. Next, the questionnaire was piloted in 52 organisations (both business and academic) from 24 European countries. Based on that feedback, the original questionnaire was edited, and the large-scale online survey was subsequently launched. For the purposes of this paper, we addressed only those variables related to OI competencies: skills, abilities and OI activities. In addition, we controlled for firm size, industry and location. The final list of variables and their operationalisation is presented in Table AI.

Data collection process and sample description

The online survey was launched in September 2014 on Webropol. While the main language was English, the survey was also translated into 12 other languages. In total, 528 respondents answered the survey, providing data from 38 countries across Northern, Southern, Eastern and Western Europe. Using stratified sampling, companies were selected on the basis of their industry’s economic significance (five to ten leading industries per country). To collect the data, we contacted one respondent per company by e-mail; those targeted included top management, innovation and R&D managers and HR specialists. The cover letter described the survey objective, supplying Chesbrough’s (2003, p. xxiv) definition of OI to avoid possible bias due to different understandings of the concept. The average response rate was 10 per cent, but this varied between countries. After cleaning the sample and removing incomplete questionnaires, the final number of responses accepted for further analysis was 264.

The sample consisted mainly of SMEs and large firms (41 and 38 per cent, respectively). Micro enterprises (i.e. those with fewer than ten members of staff) were relatively under-represented (21 per cent). The majority of respondents were located in Southern and Eastern Europe, with slightly fewer from Northern and Western Europe (see Table II). The industrial classification was based on the Global Industry Classification System (www.msci.com/gics). A majority of respondents (28.4 per cent) came from industries that included capital goods, professional and commercial services and transportation (see Table II). Information technology, grouping software and services, technology hardware and equipment, and semiconductors and semiconductor equipment, accounted for 17 per cent of the sample, and consumer discretionary firms such as automobile and components, consumer durables and apparel, hotels, restaurants and leisure, media and retail accounted for 14.4 per cent, with other industries achieving a lower share. Overall, 37 per cent of our respondents represented the manufacturing companies. The proportion of high-tech firms was small at just 18 per cent; the high-tech classification was based on Kile and Phillips (2009).

To control for possible common method bias, we implemented Harman’s single factor test (Podsakoff and Organ, 1986). We conducted a principal component analysis for all the studied variables, which resulted in 14 factors with eigenvalues greater than 1. While the overall percentage of explained variance was 66.5 per cent, the variance accounted for by a single factor was only 19.8 per cent, indicating that bias due to cross-sectional data is unlikely.

OI activities and competency measurement scales

Because the measurement scales used for data collection were new in part (see Table AI), we checked for scale reliability in addition to the expert evaluation and pilot test (Tables III and IV) to ensure the quality of the instrument and data. Cronbach’s α was high for OI activities (i.e. higher than 0.8), and all items fit well, so validating the scale (see Table III).

Cronbach’s α for OI competencies was also high (i.e. higher than 0.8). All items on the scale fit well, with the exception of IP management skills, indicating a higher α for the whole scale if these items were removed (Table IV). However, given the study’s exploratory nature, we decided to retain the item at this stage.

OI competency descriptive statistics

OI competencies (divided into skills and abilities) were evaluated using a Likert-type attitude scale, ranging from 1 (not important) to 7 (very important) (see Table AI). Respondents identified networking, communication, team-working and problem-solving skills as the most important for OI specialists (Figure 1). All the skills were considered important, with scores ranging from 5.261 to 6.163 and an average importance (for the whole sample) of 5.676. The abilities of OI specialists were also mainly considered important or very important, with average importance ranging from 4.992 to 6.023 and an average for the whole sample of 5.684 (Figure 2). Respondents emphasised that technology and business mindset, the ability to share knowledge within the organisation, creativity, adaptability and strategic thinking is most important for OI adoption.

To develop the OI competency model, we selected only those questionnaire responses indicating high importance of an OI skill or ability – that is, scores of 5, 6 or 7 on the Likert scale. The percentage of respondents assigning this level of importance to OI skills and abilities (5, 6 and 7) is shown in Figure 3.

The importance of OI competencies does not vary significantly for companies of different sizes (Table V) and from different industries. A Welch’s ANOVA test revealed no significant differences between most competencies in relation to company size; the notable exceptions were negotiation skills, the ability to manage inter-organisational collaboration processes and cultural awareness. Similarly, t-tests comparing service and manufacturing firms and high-tech companies and others (low and medium tech) showed that OI competencies were considered equally important by companies from different industries. It is notable, however, that entrepreneurial skills seem more important for service firms than for those from manufacturing. This suggests that the proposed scale is generic for companies of different sizes and from different industries.

Conversely, companies located in different European regions were more heterogeneous in their perceptions of the importance of various competencies. In particular, significant differences were found for entrepreneurial and networking skills, technology and business mindset, ability to work in internal cross-functional teams, new media literacy, cultural awareness and failure tolerance. Therefore, although it appears that industry and company size exert no particular influence on the set of required competencies, regional features (such as cultural characteristics) may play a significant role in shaping the profile of OI professional competencies.

Factor analysis of OI activities and competencies

Factor analysis of open and collaborative innovation activities confirmed the possibility of reducing the number of activities to three groups. The rotated component matrix (rotation converged in five iterations) is presented in Table VI. The selected extraction method was principal component analysis, and varimax with Kaiser normalisation was chosen as the rotation method. The Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy returned a value of 0.867, which is much higher than the threshold value of 0.5. Bartlett’s test of sphericity was significant (p<0.001), indicating that the data were suitable for factor analysis. To include only those factors that explained more variance than a single variable, we limited the minimum eigenvalue to 1. The factor analysis explained around 55 per cent of the sample variance. It is notable that all items other than idea and start-up competitions exhibited good loadings without sufficient cross-loadings while that item (idea and start-up competitions) had significant cross-loadings, with commonality below 0.5, leading to the exclusion of that variable from further analysis.

Interestingly, all “traditional” inbound and outbound monetary (pecuniary) OI activities (such as external technology acquisition, technology commercialisation, in- and out-licensing and participation in industry standardisation) appeared in the first group (see Table VI, Factor 1). We subsequently refer to this group as open technology (in and out) sourcing. These monetary activities represent technology acquisition and commercialisation across organisational borders, which are well-established OI practices. These practices differed in average intensity of adoption, while companies reported high intensity of adoption of external technology acquisition (mean 4.311) and participation in standardisation (mean 3.947), in- and out-licensing (mean 2.909 and 2.660 respectively) and technology commercialisation (mean 2.693) were not very intensively adopted. Average intensity of adoption for this group was 3.303.

The second group (Table VI, Factor 2) includes activities that involve non-monetary (non-pecuniary) inbound and outbound activities that open the innovation process to a wide range of partners, such as free revealing (ideas, IP, etc.) to external partners, crowdsourcing and use of external networks (e.g. associations, intermediaries, knowledge brokers). We describe these practices as open mass innovation, characterised by less formal, non-contractual collaboration and involving a broader range of partners as compared to the OI practices in the first group. Most companies use external networks intensively (mean 4.167), but crowdsourcing (mean 2.898) and free revealing (mean 2.943) are very specific activities that are intensively adopted in a smaller number of firms. Average intensity of adoption for this group was 3.336.

Finally, the group we characterised as open collaborative innovation included practices most intensively (group mean 4.955) adopted by companies (Table VI, Factor 3): collaborative innovation with external partners (i.e. suppliers, universities, competitors), R&D subcontracting, scanning for external ideas and customer co-creation in R&D projects. Even the least adopted activity in this group – subcontracting R&D – exhibited a high level of adoption (mean: 4.053), and the most adopted activity – collaborative innovation with external partners – was the most adopted of all OI activities in this study (mean: 5.570). In the academic literature, these inbound collaborative practices are often analysed as OI practices, but they are in effect a source of critique, as many companies adopted them before the emergence of the OI paradigm.

Next, we performed a factor analysis for OI competencies, which yielded seven factors. The rotated component matrix (rotation converged in 8 iterations) is presented in Table VII. Principal component analysis was applied as an extraction method, and varimax with Kaiser normalisation was chosen as a rotation method. KMO was high at 0.886, and Bartlett’s test for sphericity was significant at p<0.001, indicating that the data were suitable for factor analysis. The number of factors was based on eigenvalues (where the factor eigenvalue should exceed 1). The factor analysis explained around 59 per cent of the sample variance. Although all variables demonstrated quite good commonalities (i.e. above 0.5), we had to remove several items because of significant cross-loadings between several factors. In particular, we removed virtual collaboration skills from Factor 3 and strategic thinking ability from Factor 4.

This factor analysis served as the background for development of our competency model, which is discussed in the next section. OI management competency, which defines the distinct core OI management competency, included ten items (Table VII, Factor 1): networking skills, communication skills, ability to work with different professional communities, ability to share knowledge and ideas internally, ability to share knowledge and ideas externally, ability to work in internal cross-functional teams, ability to work in an interdisciplinary environment, managing inter-organisational collaboration processes, adaptability and flexibility and cultural awareness. This competency reflects the specialist’s ability to manage both the inter-organisational collaboration process (collaborating with different stakeholders, sharing knowledge externally) and intra-organisational collaboration (sharing knowledge internally, working in internal cross-functional teams), as well as using communication and networking skills. Working in a holistic collaboration process (internal and external) requires adaptability and flexibility, along with networking and communication skills and the ability to work in an interdisciplinary, cross-cultural, cross-functional environment.

Other groups of competencies comprised smaller sets of skills and abilities reflecting the specific competencies of OI management. In fact, respondents evaluated all items included in the identified competencies as important, indicating that although the competencies clearly relate to different tasks, they all play an important role in the OI process. For example, entrepreneurial leadership competency (Table VII, Factor 2) combines entrepreneurial, leadership and trust skills. Although entrepreneurial and leadership skills are essential for any successful innovation project manager, the addition of trust becomes especially critical in the context of external collaboration. In contrast to how the first group brought together the broader set of skills required for OI specialists at various levels, this competency is essential for taking on a high-level management role in an OI context.

Innovative team work competency includes problem solving, multi-tasking and team-working skills (Table VII, Factor 3), contributing to effective and efficient teamwork in an OI context. In isolation, this competency may be considered important for any project team member and for in-house (internal) innovation. However, in combination with other competencies, it becomes an important factor in the overall competence of OI specialists and team members in such projects.

Creative work competency (Table VII, Factor 4) combines creativity, new media literacy and risk awareness, which are critical for transforming creative ideas into innovations that are likely to succeed in a dynamic and uncertain environment. Indeed, while this competency per se should not be considered as a unique requirement for OI specialists, it contributes to search, evaluation and integration of external knowledge in innovation projects and to external exploitation.

Innovation process competency consists of negotiation and IP management skills, technology and business mindset, and project management ability (Table VII, Factor 5). This competency is critical for those in an OI project management role, combining technical skills with the ability to manage trade-offs between technical concerns and business imperatives.

Inter- and intra-organisational collaboration competency combines internal and external collaboration skills (Table VII, Factor 6). This competency is important for all OI professionals, but it is especially crucial for the OI manager, where this role is defined and exists within the organisation. Internal and external collaboration skills are the essence of any boundary-spanning activity.

Failure tolerance competency (Table VII, Factor 7) is of equal importance across all professions and employee levels in an innovation-intensive environment. Given the high levels of uncertainty in innovation activities and the high failure rates of such projects, this competency is essential for innovation managers. Although OI is thought to have the potential to reduce the costs and associated risks of internal R&D activities, it creates additional challenges for relationships with external partners, making this an essential competency for OI managers.

Descriptive statistics, correlation matrix and reliability assessment are presented in Table VIII. While some factors (2, 3, 4 and 5) have Cronbach’s α values below the commonly accepted cut-off point of 0.7, the coefficients are still relatively high (greater than 0.6), making them suitable for exploratory study. At the same time, all the factors correlate with each other, indicating the appropriateness of joint application for the evaluation of OI competencies.

To bring together OI activities and competencies in the competency model, a final step analysed the relationship between the three identified OI activities modes – open technology (in and out) sourcing, open mass innovation and open collaborative innovation – and the seven identified OI competencies (see Table IX).

This leads to several interesting observations. First, although Welch’s ANOVA revealed significant differences in the perceived importance of several competencies for different degrees of engagement in open technology sourcing and open mass innovation activities, it revealed no such differences with regard to degrees of engagement in open collaborative innovation activities. In other words, companies tend to evaluate the importance of OI competencies evenly, irrespective of their current degree of engagement in collaborative innovation activities. Second, companies that are intensively involved in open technology sourcing – that is, the most traditional OI activities – consider all competencies (except Factor 4 creative work competence, Factor 6 inter- and intra-organisational collaboration competency and Factor 7 failure tolerance) as more important than do those who adopt this OI activity mode less intensively. As the same can be said of companies that intensively adopt (non-monetary) open mass innovation, the only exceptions in this case are Factors 4 and 7. Notably, in the majority of cases, companies with a low degree of engagement in OI activities differ more from those with medium and high degrees of engagement while the difference between those with medium and high degrees of engagement is often negligible.

The most important competencies for all three groups of OI activities are OI management competency (Factor 1), innovative teamwork competency (Factor 3) and inter- and intra-organisational collaborative competency. Interestingly, the importance of these competencies is higher for companies that adopt open mass innovation more intensively than for other corresponding groups. Notably, companies with a high degree of engagement in open collaborative innovation activities place less importance on OI competencies than firms with a high degree of engagement in other groups of activities. At the same time, low adoption of collaborative innovation indicates higher importance of competencies than in the case of low adoption of open technology sourcing and open mass innovation. For that reason, the gap in competencies evaluation between active adopters of open collaborative innovation and those who are not actively involved in such activities is narrower than for other types of OI activity.

Theoretical contributions

OI is a developing concept that still lacks theoretical grounding, widely accepted measurement instruments and quantitative empirical studies (Bogers et al., 2017). The present study contributes to theory building by proposing and validating measurement scales for OI activities and professional competencies. This measurement tool is generic and can be used for OI competencies analysis in firms that differ in size, industry and geographical location. Using these validated measurement scales, constructs related to open and collaborative innovation activities and their relation to competencies were validated and analysed, contributing to the growing though still limited body of research on the interconnections between HRM and OI (Vanhaverbeke et al., 2014; West et al., 2014).

The proposed CMOI builds on the existing literature (e.g. Du Chatenier et al., 2010; Hafkesbrink and Schroll, 2014; Podmetina et al., 2015) while moving beyond the conceptual, case-based or discrete empirical insights of those studies by proposing an integrated view of critical OI management competencies across all areas of OI expertise. This, in turn, facilitates construction of a dynamic model and links competency categories to particular managerial roles in OI. The core of the model is the distinct OI management competency, supported by professional, interpersonal and intrapersonal competencies, so reinforcing the conceptual links between competencies and innovation management (Mulder, 2015; Petroni et al., 2012).

We found empirical evidence that, rather than focusing separately on inbound/outbound or internal/external collaboration processes, many companies instead consider OI as a holistic process and treat the organisation as part of an ecosystem, involving collaboration at different levels. This view informs recent calls for a more integrated approach to the different areas and levels of analysis in OI studies (Bogers et al., 2017).

Practical implications

The results of this study build linkages between HR, competency and innovation management and provide insights for managers of different levels and functions. These results show how individual competencies can help to realise the potential of OI by articulating a model of the requisite competencies for executing OI and collaborative activities and constructing the organisational competencies for OI (Bogers et al., 2017).

For practitioners, the following implications support the implementation, development and management of OI in organisations. Based on the CMOI model, we can trace the OI professional’s profile, which includes specific knowledge of the ecosystem, a holistic approach to company operations and knowledge of inbound, outbound, internal and external collaboration processes. Additionally, the OI professional should possess a number of intrapersonal competencies, including creativity, leadership and entrepreneurial skills, as well as risk awareness and failure tolerance.

The linking of individual OI competencies and organisational level expertise, processes and managerial roles contributes to the discussion about how companies can build and structure an OI function. To date, the cross-functionality of OI has been neglected, and our results support managers in organising OI and clarifying linkages between OI and internal, cross-functional collaboration. The inventory of internal processes and competencies can help companies to establish distinct competitive advantages through innovation. The competency model for OI described here provides a general framework for OI competency analysis and development in organisations of different sizes across industries and countries. We believe that successful development of an organisational competency in OI is highly dependent on general characteristics such as company aims and ambition, management capability and maturity rather than just on the choice of OI mode. However, further success within each of the OI modes will depend on acquiring a set of skills to pursue particular activities. For example, while expertise in open technology in and out sourcing requires competencies in industry standardisation, the success of open mass innovation will rely heavily on the ability to apply crowdsourcing to OI, and improvement in open collaborative innovation requires a focus on collaboration for innovation with external partners. On the other hand, each of the identified OI processes requires the full set of skills for OI management competencies within organisations. It is also important to note that although OI activity modes were extracted for analytical purposes, firms in reality combine various OI modes, resulting in coupled OI. In this regard, skill and competency combinations remain to be further elaborated in relation to OI management maturity, engagement level and other variables.

Limitations and future research

The present study has some limitations that should be addressed in future research. First, the data were collected from just one key informant per company. Although we sought to ensure that the respondent was a senior manager responsible for innovation, we cannot rule out the possibility of some bias in terms of how OI activities and related competencies are perceived. Future studies should therefore be designed to capture more opinions from each company, especially those of HR managers, CEOs and project managers in addition to innovation managers. This leads to a second limitation, as the exploratory nature of the research invites more systematic investigation of OI activity modes and competencies that inform the resulting competency model. In particular, competencies should be evaluated by an inter-professional sample of employees who are involved in and/or responsible for innovation, development and HR management, along with leaders of innovating companies.

Third, although the sample was of sufficient size for the analyses undertaken, it was too small to enable more fine-tuned analysis of regional differences across Europe in terms of how OI is managed through the development and implementation of competencies. Further analysis of these differences (which we assume reflect cultural variations) would be an interesting research direction that could be realised by introducing variables measuring cultural characteristics such as leadership style, strength of hierarchy, team dynamics and relational trust, and linking these to the most important OI competencies. Fourth, the present study did not control for familiarity with competency management and the possible use (or not) of a competency management system in the responding organisations. As our aim was to include companies of all sizes across most European countries, no such control was introduced because of the potential for definitional confusion and possible misinterpretation. In more focused samples, future research should introduce such controls and should analyse the interplay between more or less advanced HR practices and OI implementation and outcomes.

In conclusion, this paper makes a significant contribution to the study of OI by proposing measurement tools for OI activities and competencies, and by developing the OI management competency model. This research invites discussion of required and desired employee skills in firms implementing OI. It also proposes an interdisciplinary approach by seeking to integrate OI and HRM research streams, which can contribute to the development of practices related to OI human resource management, including training, reward systems, recruitment and an understanding of the role of OI managers. Additionally, the questionnaire and validated scales provide a unique tool for evaluating current and desired OI-related skills. We believe that the present findings will be of great value to a wide range of innovation professionals in academia, industry and consulting and will stimulate debate around the crucial common competencies for OI, as well as competencies related to local industrial needs.