Open Innovation and Public–Private Collaboration in Manufacturing: A Case Study from Piedmont, Northern Italy

Abstract

This study explores the dynamics of Open Innovation (OI) in manufacturing firms, with particular attention to collaboration with public research institutions. The research is performed in the Piedmont region, Northern Italy, which represents one of Italy’s leading innovation regions, with a strong manufacturing heritage and an active strategy to foster industrial transition through innovation clusters and partnerships. The research analyzes survey responses from 82 managers and decision-makers in manufacturing firms belonging to the local manufacturing ecosystem. The questionnaire investigated how company size, organizational structure for research and development (R&D), perceived importance of collaboration, innovation drivers and barriers, and trust in research institutions affect four types of innovation: product, process, marketing, and organizational. Results indicate that collaboration with other private companies is significantly associated with product innovation, while collaboration with public research institutions is associated to both product and process innovation. The level of R&D structuring in the management of innovative projects and trust in the expertise of public research organizations are also positively associated with product innovation. In addition, key drivers—such as the availability of dedicated financial resources, staff creativity, and openness to external partnerships—are significantly related to process innovation. The findings suggest that regional policymakers and industry stakeholders should promote targeted measures to strengthen OI adoption, particularly by improving the perceived competence and transparency of public research organizations.

1. Introduction

In recent years, the challenge of sustainable development has been recognized worldwide [1], since the estimated growth of the world population [2] requires development pathways that balance human needs with environmental protection [3]. Sustainability and sustainable development are systemic and multi-objective issues: they require the simultaneous improvement of environmental, economic, and social performance, often with trade-offs (e.g., among cost, quality, energy, health, safety, and lead times) and indirect effects across the entire product and process life cycle [4,5]. Sustainable development that meets people’s needs also requires radical improvements in eco-efficiency [1]. For instance, the transition toward more sustainable models involves technological and managerial changes to reduce impacts on energy and resource consumption, and emissions, particularly in the context of manufacturing firms [6,7].

In the sustainability transition literature, innovation is understood as a mechanism that enables systemic change toward more sustainable socio-technical configurations [4,5,8]. Innovations contribute to the development, diffusion, and utilization of new ways to fulfil societal needs that are aligned with sustainability objectives, thereby supporting transitions in socio-technical systems at multiple levels. Within this framework, the Eco-Innovation Action Plan (EcoAP) launched in 2011 as key priority to Europe’s future competitiveness, defined eco-innovation as “any innovation resulting in significant progress towards the goal of sustainable development, by reducing the impacts of our production modes on the environment, enhancing nature’s resilience to environmental pressures, or achieving a more efficient and responsible use of natural resources” [9].

Approaches to innovation paradigms are grounded on numerous factors, including integrated collaboration, co-creation, and value sharing, which emerge from networked knowledge connections [10,11,12]. These appear stronger when grounded in interactive engagement between universities that provide cutting-edge knowledge and competitive firms [13]. The development of innovation requires a specific ecosystem that emphasizes the position and role of local and public actors in collaborative and co-creative activities [12]. The challenge for public policy is thus to provide the instruments to transform traditional environments into innovative ones [10].

The most fitting example is that of innovation ecosystems, as a network of relationships that enhances sustainability by bringing together actors that establish connections, reinforcing the importance of institutions, the environment and enabling information and knowledge flows through value co-creation systems [14]. An innovation ecosystem will provide firms with an innovative environment in which they can share value to create new products with a community of stakeholders who share interests, ranging from governments to end users [8]. The innovation ecosystem will consist of a dynamic, interactive network embedded in an innovation mindset and an interactive environment focused on knowledge creation and diffusion [12].

In recent years, the manufacturing industry has faced rapid acceleration of innovation trajectories driven by digitalization and integration of the physical and digital worlds. In particular, the Industry 4.0 vision describes the transformation of production systems through technologies such as the Internet of Things, cyber-physical systems, and advanced automation within smart manufacturing, enabling more connected, adaptive, and intelligent processes [15,16].

Unlike previous industrial transformations, Industry 4.0 is grounded in interdisciplinarity and in deep interconnection among diverse competences [17]. Building on innovation and collaboration, the pursuit of shared value creation, sustainable growth, and development will primarily concern the economic and social spheres, as companies shift from maximizing short-term financial performance to assuming long-term economic and social responsibility [18]. Only through joint research within an innovation ecosystem will it be possible to meet these requirements, accelerate the process, and raise the standards of outcomes [12].

It is important to observe that innovation in the manufacturing sector is a multidimensional phenomenon that includes the introduction of new or improved products and processes, as well as changes in how a firm organizes and carries out its activities [19]. Next to technological innovation, organizational innovation is essential to translate digital potential into tangible outcomes [20]. Another increasingly relevant component of manufacturing innovation concerns the transition toward the circular economy framework, in which innovation also includes experimenting with new product, process and business models to reduce waste and valorize resources across the life cycle [7,21].

1.1. The Open Innovation Paradigm

The business environment in the 20th century was mostly defined by the Closed Innovation (CI) paradigm, defined as “[…] an internal paradigm of industrial R&D, in which playing chess is critical to success in extending the current businesses”, or “[…] an approach that is fundamentally inwardly focused […]” [22] (p. 18). As in the game of chess, where the players know the pieces in advance and know what moves they can and cannot make, in CI, the companies know all the information in advance, what the competitors will do and what the customer expects. Therefore, they can plan moves in advance, and often this is key to winning. The paradigm of CI is increasingly at odds with today’s knowledge landscape. In the first half of the 21st century, companies are increasingly moving beyond their internal agendas, moving towards building mechanisms to access and tap into the wealth of external knowledge that surrounds them.

Open Innovation (OI) is a term coined by Henry Chesbrough in 2003 that “[…] means that valuable ideas can come from inside or outside the company and can go to market from inside or outside the company as well” [22] (p. 43). This approach placed external market ideas and paths at the same level of importance as internal ones. The boundaries of the organization are thus opened to source in complementary external resources for innovation and source out internal resources to be externally commercialized [23]. This made it possible to move toward a more comprehensive approach to external collaboration [24]. In this way, innovation can be conceived as the result of distributed inter-organizational networks rather than individual companies [25,26]. OI thinking changed the role of the research function, expanding the role of internal researchers to include not only knowledge generation, but also knowledge brokering [22].

In the private sector, OI mainly involves other private companies [27] as external stakeholders such as suppliers, customers, competitors, and partners [28]. In this sector, innovations are mainly focused on gaining a competitive advantage [29], allowing the company to benefit from the innovation itself [30] through access to external expertise, shorter time to market, and reduced innovation failure rates [31]. The focus of OI in the private sector is thus on the development of new products and services [32].

Universities and other Public Research Organizations (PROs) constitute a significant external source of knowledge for innovation development. PROs occupy a central position in R&D and innovation activities across numerous industries [33], and their role as providers of external knowledge has received growing attention in the literature [34]. Although the importance of university–industry collaboration for innovation has been widely examined, the organizational mechanisms that shape these relationships remain insufficiently understood [35]. Despite the considerable potential of universities and PROs as generators of novel knowledge, firms often encounter substantial difficulties in absorbing such knowledge [36], as illustrated by the frequent failures of university–industry knowledge transfer initiatives [37].

A key set of obstacles stems from firms’ uncertainty regarding the benefits that collaborative arrangements with PROs may yield, as well as from perceptions that PROs may lack the competencies required for effective cooperation [38,39,40,41,42]. These concerns underscore the critical role of trust in shaping collaborative partner selection. Firms are more inclined to share resources and engage in joint activities when they trust their counterparts [43], and strong trust-based relationships facilitate both the transfer of knowledge [44] and the willingness of partners to exchange information [45]. Sustained collaboration is more likely to emerge when interactions yield fair and reliable outcomes, reinforcing long-term relational commitments [46]. Trust, together with mutual commitment, is therefore essential to maintain strong interorganizational ties [47]. High levels of trust also reduce opportunistic behavior, knowledge monitoring costs [48], and promote more meaningful communication between partners [49].

In collaborative innovation projects, differences in institutional logics, procedural requirements, and performance expectations often generate delays and misalignments [50]. Trust can mitigate these tensions by lowering the need for excessive monitoring and formal control, facilitating open information exchange, and supporting flexible problem-solving practices [51]. Moreover, trust contributes to the alignment of expectations concerning intellectual property management, resource commitments, and project timelines, thereby enhancing the efficiency and stability of industry–research partnerships [48]. In this sense, trust acts as an organizational mechanism that sustains collaborative innovation processes within regional ecosystems.

Innovation is seen as an evolving process composed of four main dimensions [52]. Product innovation is represented by the introduction of new or improved goods or services, and involves changes in technical specifications, components and materials, ease of use, and integration of software and other functional features [53], thus potentially leading to competitive advantages [54]. Process innovation can be defined as the development of a new or improved production method, involving changes in techniques and equipment, aiming to reduce production or distribution costs, improve the quality and distribution of products themselves [53], and also reduce resource and energy consumption and lower emissions across the production process life cycle [5,6]. Marketing innovation can generate improvements in the product by enhancing its uniqueness, addressing the implementation of new methods with changes in product development, packaging, promotion, distribution, and pricing [55]. Its purpose is to meet consumer needs by opening new markets and repositioning a company’s products within those markets, thereby increasing sales [53].

Finally, organizational innovation is essential for companies seeking to address strategic challenges, as it results in improvements in organizational management [55]. Thus, this component of innovation involves the implementation of a new organizational method in business attitudes, such as workplace organization and external relations. New methods thus contribute to business routines and procedures, as well as direct work and practices that facilitate learning and knowledge sharing within the company [53].

Alongside the different forms that innovation may take, firms’ ability to innovate is shaped by a combination of internal and external drivers and barriers. The main barriers to innovation are represented by the lack of internal or external financial resources [38,56], qualified personnel [57], good ideas to innovate [58], a scarcity of partners to collaborate with [38], and finally, market demand variability [59]. The literature reports several drivers of innovation, such as the high number of employees [60], encouraging individual staff proposals on topics related to innovation [61], investing time and resources in creativity [62] and experimentation [63], having dedicated financial support for innovation [64], receiving training from external personnel and professionals [65], collaborating with external organizations and/or companies [66] and acquiring new technologies in a timely manner [67].

1.2. Rationale and Aim of the Present Study

Despite the widespread adoption of the OI paradigm, the literature makes little distinction between types of partners and types of innovative outputs [22,28]. Even though extensive research on the antecedents of firm-level innovation exists and collaborations between companies and PROs are analyzed, the knowledge remains limited on how such collaborations differentially influence the various dimensions of product, process, marketing, and organizational innovation [35].

From a theoretical standpoint, previous literature seems to follow two partially disconnected approaches in the study of OI and private-public collaboration. Knowledge-based and resource-oriented perspectives emphasize complementarities, absorptive capacity, and external knowledge acquisition as primary drivers of innovation performance [21,27]. In contrast, institutional and relational approaches highlight governance mechanisms, trust, and coordination barriers as critical factors shaping collaboration outcomes [36,38]. However, these perspectives rarely engage in an integrated empirical assessment, particularly within public–private OI contexts. As a result, the field lacks a comprehensive understanding of how structural characteristics, relational mechanisms, and perceived drivers and barriers jointly influence differentiated innovation outputs.

Furthermore, although the literature on university–industry collaboration has extensively classified barriers, highlighting cultural, institutional, operational, governance and transactional obstacles such as divergent goals, lack of incentives, bureaucratic burdens and misalignment between academic and industrial worlds, most studies remain descriptive, focusing on typologies rather than empirically testing how these barriers affect specific innovation outcomes. Similarly, drivers of collaboration are frequently examined independently of barriers, without assessing their simultaneous operation within the same analytical framework.

Also, how these barriers and drivers affect different types of innovation outcomes within manufacturing ecosystems remains limited [39,40,68]. Another underexplored dimension concerns perceptual and relational factors [28,37]. The perceived importance attributed to collaboration represents a cognitive and motivational component that may shape firms’ engagement in OI activities, yet it has received limited empirical attention as a direct antecedent of innovation. Likewise, the impact of trust in PROs—widely acknowledged as a key relational governance mechanism—has been overlooked in its differentiated impact across multiple innovation types.

These gaps are particularly relevant in manufacturing-centered regional ecosystems characterized by strong institutional support and dense public–private interaction, where structural conditions, perceptions, and relational quality are likely to interact in shaping innovation dynamics.

Addressing these limitations, the present study proposes an integrated empirical framework that simultaneously considers structural and relational variables related to firms’ characteristics, the importance of collaboration with private firms and PROs, perceived drivers and barriers to innovation and trust in research institutions’ expertise. The analysis was conducted within the manufacturing ecosystem of the Piedmont region (Northern Italy), characterized by high R&D intensity, active innovation clusters, and well-established public–private partnerships [69].

The study intended to answer three research questions:

• Do structural (i.e., size and level of R&D) and relational (i.e., the importance of collaboration with private firms and PROs) factors have different effects on different types of innovation (product, process, marketing, and organizational)?
• How do perceived innovation drivers and barriers shape different types of innovation performance?
• Does trust in PROs differentially impact innovation outputs within a regional manufacturing ecosystem?

The study focuses on four distinct innovation outcomes, i.e., product, process, marketing, and organizational innovation, to capture the multidimensional nature of OI and to assess whether collaboration mechanisms operate differently across innovation domains. Private–private and public–private collaborations within the same regional ecosystem were considered, to provide a more controlled comparison of how distinct governance logics and relational dynamics operate under comparable systemic conditions.

The originality of this study lies in the combined analysis of differentiated innovation types, comparative assessment of collaborations between private and public partners, and explicit integration of perceptual and relational mechanisms within a unified empirical model. The final aim is to contribute to the literature on Open Innovation by providing empirical evidence on public–private collaboration within a manufacturing-centered regional ecosystem, while offering policy-relevant insights to support the design of targeted interventions that strengthen collaboration, trust, and innovation performance at the regional level.

2. Materials and Methods

2.1. Context of the Present Study

In 2021, there were 163,550 small and medium-sized enterprises in Italy, 97,100 of which were in the North of the country [70]. In the following year, there were 358,500 manufacturing companies in Italy out of 2.2 million active companies in the same sector in the European Union (EU) [71], which is approximately 17% of the total. In 2022 these companies created about 3.9 million jobs in Italy compared to 30 million created by companies in the same sector in the EU. Eurostat statistics [71] reported that among the 5 largest EU countries, Italy had the second-largest share (12.7%) of the value added generated in the EU manufacturing sector in 2022 and the third-largest share in the business economy (10.5%). Eurostat [71] estimates that in the same year the Italian manufacturing sector generated EUR 1.3 billion in Gross Value Added (GVA) out of EUR 9.8 billion generated by companies in the same sector in the EU, representing nearly 15% of Gross Domestic Product (GDP). While the largest contribution in the EU came from all those companies involved in the manufacture of coke and refined petroleum products, in Italy it came from those involved in the production of rubber, plastic and metal, with EUR 76 billion, or 26% of the total.

Already during 2020, Italian manufacturing companies had 55% of the total tax credit for investment in Research & Development (R&D) [72]. This is a tax incentive allocated by the State since 2015 to support companies’ R&D activities and strengthen the Country’s production capacity. Italian companies in general (and specifically manufacturing companies) can thus invest in fundamental, industrial research and experimental development, such the training of highly qualified employees; purchasing instruments, laboratory equipment, and industrial patents; and financing research contracts signed with public research institutions [72].

Based on the regional attractiveness indicators according to The European House—Ambrosetti (TEHA), compiled by the Piedmont Region in collaboration with TEHA Group, Piedmont is confirmed as the Italian leader in attracting new investments [69]. TEHA identified five dimensions of Piedmont’s attractiveness for businesses and investors (economic and industrial competitiveness; innovative capacity and talent; infrastructure, networks, and connectivity; inclusion, social services, and environmental sustainability; culture, tourism, and lifestyle) according to the Global Attractiveness Index (GAI) model. To this end, it reconstructed historical trends and developed a database of 50 Key Performance Indicators (KPIs), for a total of approximately 23,000 data points. The work also included individual interviews with investors and entrepreneurs on the most relevant indicators for attractiveness. Overall, Piedmont has grown or consolidated its position in 73.9% of the KPIs considered compared to the pre-pandemic period. The comparison between the pre- and post-COVID periods shows that performance was equal to or well above compared to that of the pre-pandemic period in the five areas of analysis.

It is important to emphasize that in the macro-area of economic and industrial competitiveness, Piedmont has the fourth-highest manufacturing added value per company (EUR 919,000) and a growing incidence of manufacturing added value (fourth place for increase over the last 15 years compared to a national average decline). In addition, Piedmont is a leader in terms of openness of the economic system: 4th in Italy for the incidence of manufacturing exports on the national total (10.5%), 1st in Central-Northern Italy for manufacturing export growth (+7% in 2023 compared to 2022) and 3rd in Italy for the number of foreign-controlled industrial companies (943 in 2021, equal to 10% of the national total). As a further important aspect, in the macro-area of “Innovative capacity and talent,” Piedmont is a benchmark in Italy for R&D (2nd Italian region with an incidence of 2.09% on regional GDP compared to the national average of 1.48% in 2022) thanks to a strong focus on innovation among businesses—second-highest propensity at national level to introduce innovative activities.

Finally, Piedmont is an investment-oriented region, with the second-highest incidence of gross fixed investments on GDP (24.6% vs. the Italian average of 20.5%) and the third-highest growth rate for the same (between 2011 and 2021, the incidence increased by +2.7 p.p., compared to an increase of 1.1 p.p. at the national level). In 2024, Piedmont ranked among the most virtuous European regions for attracting Foreign Direct Investment (FDI) (2nd in Italy) in the “Large European Regions of the Future 2024—FDI Strategy” ranking compiled by the Financial Times.

In addition to being a region well-positioned in Italy in the manufacturing sector and having a strong propensity for innovation, as evidenced by its second-place ranking in Italy, Piedmont has the distinctive feature of being the first Italian region to establish a structured system of Innovation Hubs in 2009 [73]. The mission of the hubs, from their foundation to the present day, has remained unchanged and consists, in addition to technology transfer, of promoting innovation, bringing together businesses and public and private research players, and sharing knowledge and skills.

The Piedmont region represents a particularly relevant case within the Italian innovation landscape. According to the Regional Innovation Scoreboard (RIS) published by the European Commission (2023), Piedmont is classified as a “Moderate Innovator +”, showing strengths in indicators such as the introduction of process innovations by Small and Medium Enterprises (SMEs) and public–private scientific co-publications [74]. However, it shows relatively weaker performance in some dimensions related to the knowledge base and R&D. This mixed innovation profile could highlight the presence of structural resources along with systemic coordination challenges.

Further findings from the Organization for Economic Co-operation and Development (OECD) [75] pointed out that, despite relatively high levels of private R&D spending and patenting activity compared to other Italian regions, Piedmont’s companies, particularly manufacturing SMEs, face difficulties in fully exploiting collaborative innovation networks and in achieving innovation outcomes [75]. The OECD analysis highlights the need to improve coordination between ecosystem actors and to further strengthen links between science and industry.

This combination of innovation achievements alongside persistent challenges and critical areas for improvement—such as ecosystem coordination and science-industry linkages—makes Piedmont a highly relevant context for investigating the dynamics of open innovation (OI) and the factors affecting it, including collaboration patterns, trust in public research organizations (PROs), and perceived drivers/barriers.

In the Italian context, prior studies on Open Innovation in manufacturing have examined the effects of external knowledge acquisition on innovation performance in relation to human resource management practices [76] and the impact of different R&D collaborations on product innovation and innovation performance in SMEs [77]. However, these studies have not systematically explored the combined role of structural characteristics, relational mechanisms, and perceived drivers and barriers within a territorially bounded innovation ecosystem, nor have they differentiated across multiple innovation types and partnerships (private–private vs private–PROs). By focusing on a region characterized by institutionalized collaboration mechanisms and a consolidated manufacturing base, also facing challenges in ecosystem coordination and science industry linkages, this study provides a contextually grounded yet analytically transferable framework for examining Open Innovation dynamics. While the findings should be interpreted within the scope of manufacturing-centered ecosystems with comparable institutional density, the Piedmont case offers a relevant benchmark for future comparative analyses across innovation systems with different levels of maturity.

2.2. Participants and Procedure

Data was collected through a cross-sectional online survey administered via Google Forms. To contact the potential participants, the authors reached out to MESAP, the Piedmont innovation hub for Smart Systems and Smart Products. MESAP brings together 240 companies, including SMEs, research centers and large groups, with initiatives aimed at fostering connections and synergies within its ecosystem and beyond. The survey link was distributed to the full list of firms associated with MESAP, targeting key informants holding managerial or senior positions and therefore considered knowledgeable about the firm’s innovation activities and external collaborations. The invitation email explained the academic purpose of the research, ensured confidentiality and anonymity of responses, and clarified that participation was voluntary. To improve participation, one reminder was sent two weeks after the initial invitation.

Eighty-two participants agreed to be involved in the survey and completed the online questionnaire. Among them, 73.2% were men, 25.6% were women, and 1.2% indicated other gender. Their age ranged from 25 to 71 years (M = 43.878, SD = 11.57). A total of 6.1% of participants had a middle school diploma, 20.70% had a high school diploma, 57.30% had a university degree, and 15.9% had postgraduate education. In addition, 32.9% held managerial roles, 29.3% were area managers, 29.3% held executive roles, 4.9% held other roles, and 3.7% did not specify their role. The main characteristics of the firms involved in the study are reported in

Table 1. Main characteristics of the 82 companies that took part in the study.

Company Characteristic		N	%
Number of employees	Up to 10	8	9.8
	Up to 50	34	41.5
	Up to 250	12	14.6
	More than 250	28	34.1
Level of R&D organization to manage innovation	The company does not have any form of R&D organization	13	15.9
	The company chooses R&D organizational solutions for innovation project management from time to time	16	19.5
	The innovation projects are entrusted to employees who already hold other positions within the company	29	35.4
	There is an R&D department within the company, and a staff dedicated to innovation	24	29.3
		Mean (SD)
Years of activity		40.96 (4.66)

.

2.3. Instruments

An 18-item questionnaire was developed based on a previous survey on innovation among Italian companies performed by the Italian National Institute of Statistics (ISTAT) [78] and other previous surveys related to firms’ innovation in the Italian and European context [79,80]. The questionnaire was pre-tested with two academic experts in innovation and regional development, and two practitioners with managerial experience within the MESAP cluster, to assess clarity, relevance, and contextual appropriateness of the items.

In the introductory section of the questionnaire, participants were informed that the study investigated innovation initiatives within the regional manufacturing ecosystem with particular attention to their potential contribution to firms’ sustainability and long-term competitiveness. Then, the questionnaire included three main sections.

The first section collected background characteristics of the firms, such as the number of employees (company size), the level of R&D organization to manage innovative projects [80,81], and the number of years the company has been in business. Questions were used to gather information about the characteristics of the company, and a multiple-choice question with four alternative answers was used to gauge the level of R&D in the organization. The questions were based on the Eurostat Community Innovation Survey (CIS) [80].

In the second section, participants were asked to rate on a four-point scale (from 1 = not at all to 4 = very much) the extent to which each of the four innovations has been implemented within the company over the past three years. Using the same four-point scale, they were asked to rate the importance of seven drivers that may push the company to innovate [80,82], the importance of six barriers for the implementation of innovation [80,83] and their level of trust in the expertise of research institutions [84].

The third section refers to collaboration with Italian companies and public entities for innovation development. Thus, participants were asked to rate on a four-point scale (from 1= not at all to 4 = very much) how important they considered collaborating with other Italian companies for innovation development and how important they considered collaborating with Italian public institutions for innovation development. In addition, participants were asked to indicate (no = 0, yes = 1) the types of contributions provided during collaboration with both companies and public institutions: 1. Writing a funded research proposal, 2. Providing suggestions and ideas for the development of innovative products, processes, or strategies, 3. Providing administrative management, 4. Giving technical support, and 5. Co-creating innovative products, processes, or strategies. Furthermore, a question addressed the extent to which specific public institutions (European Commission [EC], University, National Research Council of Italy [CNR] and other research institutes, Italian Ministry of University and Research [MUR], and the Regional system) have contributed to innovation in their company through specific strategies, such as 1. Training, 2. Development of growth strategies, 3. Experimental strategies, and 4. Sharing of obtained results.

Finally, a standard socio-demographic section (including gender, age, education, and role in the company) for the respondents closed the questionnaire. The questionnaire took 10–15 min to fill out.

2.4. Data Analysis

First, the dataset was examined for missing values and inconsistencies prior to conducting statistical analyses. Descriptive statistics were computed for all variables considered. Mean scores were calculated for all items with a four-point scale. Cronbach’s alphas were computed to measure the reliability of the aggregated scores of drivers and barriers. Thus, driver and barrier variables were calculated by the sum of the scores of each item.

To identify influential regression outliers, standardized DFFIT (SDF) was computed [85] $SDF = 2\sqrt{{\displaystyle \frac{K +1}{n}}}$ where n is the sample size (equal to 82), and k is the number of predictor variables (equal to 7). As such, four cases with SDF >0.6 were excluded. Sample-size–adjusted thresholds, rather than absolute values, were used to detect outliers, as it has been previously explained that this approach provides a more reliable exclusion of outliers when dealing with small sample sizes [86,87].

As a result, the final sample consisted of 78 respondents. Multicollinearity was assessed using variance inflation factors (VIF). Heteroscedasticity was evaluated through standard statistical tests, and robust standard errors were applied to account for potential heteroscedasticity (Breusch–Pagan test) and ensure reliable inference. When evidence of heteroscedasticity emerged, heteroscedasticity-robust standard errors were used to correct the standard errors and maintain valid statistical inference.

Furthermore, Harman’s single-factor test was conducted for detecting common method bias (CMB) [88]. Correlation analysis was conducted to examine the relationships among the study variables and assess the items’ validity (Pearson coefficient > 0.30).

The structural characteristics of the company (i.e., number of employees, level of R&D) were then used together with importance of collaborating with other companies, importance of collaborating with PROs, drivers, barriers, and trust in the expertise of PROs as the independent variables in four regression models to investigate the effects of these variables on the four type of innovation (i.e., product, process, marketing and organizational) that companies have implemented in the last three years. Analyses were performed using IBM SPSS Statistical Package for Social Science v29.

3. Results

Regarding the investigated four type of innovation that firms have mainly implemented in the last three years, both cooperating with other private companies and public organization and institutions, participants reported a higher mean score for product innovation (M = 2.48, SD = 0.835), followed by process innovation (M = 2.141, SD = 0.764), and then marketing (M = 1.698, SD = 0.671) and organizational (M = 1.666, SD = 0.606) innovation.

Regarding the items exploring the determinants of companies’ attitudes towards innovation,

Table 2. Mean (M), Standard Deviation (SD), and reliability coefficients (α) of the different items and scales representing the firms’ perceived innovation determinants.

Scales	Item	Mean (SD)	Cronbach’s α	Pearson’s Coefficient
Drivers	Encouraging individual staff proposals on topics related to innovation	3.18 (0.818)	0.788	0.738
	Investing time and resources in creativity	3.08 (0.864)		0.664
	Investing time and resources in experimentation	3.42 (0.730)		0.641
	Having dedicated financial support for innovation	3.35 (0.753)		0.565
	Receiving training from external personnel and professionals	3.00 (0.883)		0.771
	Collaborating with external organizations and/or companies	3.08 (0.849)		0.696
	Acquiring new technologies in a timely manner	3.18 (0.769)		0.619
Barriers	Lack of internal financial resources within the company	2.73 (1.002)	0.718	0.663
	Lack of external funding sources for the company	2.69 (0.916)		0.679
	Lack of qualified personnel	2.77 (0.992)		0.678
	Lack of partners to collaborate with	2.41 (0.932)		0.737
	Market demand variability	2.56 (0.877)		0.481
	Lack of good ideas for innovation	2.41 (1.037)		0.625

Note: Cronbach’s α coefficients indicate the reliability of the drivers and barriers scales. Pearson’s coefficients represent item–total correlations, providing evidence of the internal validity.

reports the mean values for all items related to drivers and barriers, as well as the aggregated scores considered and the reliability coefficient. Regarding trust in the expertise of PROs, our participants provided a mean rating of 3.14 (SD = 0.768) on the four-point scale.

The mean rating score attributed to the importance of collaborating with other Italian companies to develop innovation was 3.32 (SD = 0.712). Whereas the mean rating attributed to the importance of collaborating with Italian public institutions to develop innovation was 2.71 (SD = 0.941). Detailed information regarding the type of contribution provided by other companies or public institutions during the collaboration is reported in

Table 3. Contributions provided by the other company/companies or public institutions during the collaboration.

Item	Other Private Companies	Public Institutions
	N (%)	N (%)
Writing a funded research proposal	28 (35.9)	22 (28.8)
Providing suggestions and ideas for the development of innovative products, processes, or strategies	58 (74.4)	31 (39.7)
Administrative management	20 (25.6)	14 (17.9)
Technical support	63 (80.8)	41 (52.6)
Co-creation of innovative products, processes, or strategies	62 (79.5)	26 (33.3)

. As can be seen, the main contribution from both private companies and private institutions was providing technical support. This was followed by co-creation activities for private companies and the provision of ideas from the public institutions.

Furthermore,

Table 4. Participants’ perception of public institutions in contributing to innovation within the company.

	The Extent to Which This Institution Contributes to Innovation in Your Company Through:
	Training	The Development of Growth Strategies	Experimental Research	The Sharing of Obtained Results
Public Institutions	Mean (SD)
European Commission	1.83 (0.918)	2.03 (1.006)	1.68 (0.933)	1.99 (1.013)
University	2.26 (1.050)	1.83 (0.903)	2.27 (1.136)	2.22 (1.065)
CNR and other institutes	1.74 (0.889)	1.63 (0.854)	1.86 (1.053)	1.86 (0.977)
MUR	1.71 (0.839)	1.68 (0.890)	1.60 (0.858)	1.68 (0.890)
Regional ecosystem	2.01 (0.830)	1.92 (0.905)	1.76 (0.840)	1.63 (0.740)

reports the mean values for participant perceptions of the role of different public institutions in contributing to innovation. The participants perceived that the university mainly contributed to experimental research, training, and sharing of results followed by the EC in the development of growth strategies and in the dissemination of results.

Correlation analysis was conducted to examine the relationships among the study variables, and the results are reported in

Table 5. Correlation analysis.

	1	2	3	4	5	6	7	8	9	10	11
1. Product innovation	1
2. Process innovation	0.381	1
3. Marketing innovation	0.227 *	0.496 ***	1
4. Organizational innovation	0.359 **	0.586 ***	0.620	1
5. N of employees	0.094	0.215	−0.096	0.102	1
6. Level of R&D	0.414 ***	0.047	−0.017	0.003	0.384 **	1
7. Importance of collaborating with other companies	0.600 ***	0.274 *	0.137	0.251 *	0.007	0.270 *	1
8. Importance of collaborating with public institutions	0.430 ***	0.384 **	0.145	0.269 *	0.029	−0.011	0.434 ***	1
9. Drivers	0.266 **	0.291 *	−0.030	0.212	0.213	0.261 *	0.363 *	0.123	1
10. Barriers	−0.009	−0.020	−0.054	−0.009	−0.116	0.090	−0.012	0.086	0.311 **	1
11. Trust in the expertise of PROs	0.450 ***	0.143	0.046	0.158	−0.153	0.162	0.486 ***	0.112	0.211	0.089	1

* p < 0.05, ** p < 0.01, *** p < 0.001.

.

Table 6. Regression analysis for the four types of innovation investigated.

	Product Innovation		Process Innovation		Marketing Innovation		Organizational Innovation
	B	t	B	t	B	t	B	t	VIF
N of employees	−0.001	−0.012	0.153	1.713	−0.0688	−0.766	0.069	−0.875	1.324
Level of R&D organization to manage innovation	0.244	3.111 **	−0.093	−1.045	0.001	0.012	−0.077	0.945	1.353
Importance of collaborating with other companies	0.324	2.402 *	0.017	0.111	0.108	0.708	0.070	−0.520	1.931
Importance of collaborating with public institutions	0.256	3.059 **	0.274	2.876 **	0.079	0.834	0.131	1.416	1.308
Drivers	0.003	0.120	0.051	2.013 *	−0.007	−0.286	0.028	1.530	1.450
Barriers	−0.006	−0.311	−0.025	−1.092	−0.011	−0.462	−0.014	−0.555	1.219
Trust in the expertise of PROs	0.255	2.392 **	0.106	0.876	−0.022	−0.180	0.083	0.672	1.406

* p < 0.05, ** p < 0.01, *** p < 0.001.

reports the results from the regression analyses. Level of R&D organization had a significant positive effect on product innovation (R² = 0.522, F(7,70) = 10.931 p = 0.003). The importance of collaborating with other companies and trust in the expertise of PROs had a significant positive effect also on product innovation (R² = 0.522, F(7,70) = 10.931, p = 0.019 and R² = 0.522, F(7,70) = 10.931 p = 0.019). The importance of collaborating with public institutions had a significant positive effect on two out of four types of innovation implemented (product R² = 0.522, F(7,70) = 10.931, p = 0.003; process R² = 0.262, F(7,70) = 3.545, p = 0.005). Drivers had a significant positive effect on process innovation (R² = 0.262, F(7,70) = 3.545, p = 0.048). The other variables investigated did not report any significant effect. The variance inflation factor (VIF) values ranged from 1.116 to 1.466, indicating that multicollinearity was not a concern. Also, the Breusch–Pagan test was not significant at p = 0.226 (χ² = 1.404), p = 0.960 (χ² = 0.002), and at p = 0.458 (χ² = 0.552), for product, process and marketing innovations, respectively, but significant p = 0.021 (χ² = 5.298), for organizational innovations.

To assess common method bias, Harman’s single-factor test was conducted by performing an exploratory factor analysis (EFA) without rotation, forcing to one factor. The results indicated that the single factor accounted for 29.133% of the variance, which is below the 50% threshold, suggesting that common method bias is not a major concern.

4. Discussion

This study examined how structural characteristics of companies, perceived importance of collaboration, internal drivers and barriers, and trust in PROs relate to four types of innovation, i.e., product, process, marketing, and organizational, in manufacturing companies embedded in a highly developed regional ecosystem in Piedmont, Northern Italy. Overall, the findings indicate that Open Innovation practices are present and differentiated, with collaboration patterns and internal conditions playing distinct roles across innovation outcomes.

Consistent with previous research [89], a positive effect of the level of R&D organization on product innovation was found, suggesting that a formal structure may foster the development of such innovations. This result may be interpreted considering that a formal R&D organization provides firms with structured routines and dedicated resources that enhance their ability to acquire, integrate, and exploit knowledge, which is particularly critical for product innovation due to its higher technological complexity and resource requirements. Collaboration with other companies emerged as a positive predictor of product innovation [29,30], confirming that firms rely strongly on peer organizations and value-chain partners to introduce new products. Firms often express uncertainty regarding the benefits of collaborative arrangements with PROs, along with perceptions that PROs may lack the specific competencies required for effective industry cooperation [38,39,40,41,42]. However, in our sample, collaboration with public institutions was also significantly associated not only with product but also with process innovation, suggesting that PROs are primarily involved in projects requiring technical knowledge, experimentation, and production-related improvements rather than market- or organization-oriented change. This is probably because public research institutions are mainly perceived as creators of services and processes, rather than producers of a physical artifact [32]. Furthermore, companies may be interested in collaborating with PROs because they are committed to improving service performance and generating public value [29,30] through extensive citizen input and enhanced government trust [90]. This pattern is explained by the virtuous Piedmont ecosystem, where innovation hubs like MESAP foster structurally embedded public-private ties centered on technology transfer and experimental process innovations, aligning PROs’ expertise with the region’s manufacturing focus on sustainable production enhancements, despite science-industry coordination challenges. These findings point out the importance of creating and supporting such innovation clusters to strengthen public-private collaborations, particularly for advancing process innovations that drive firms’ sustainable development at the regional level.

Internal drivers, such as investments in creativity, experimentation, and timely acquisition of new technologies, were particularly relevant for process innovation. These proactive practices enhance firms’ absorptive capacity by fostering an internal culture of exploration, knowledge assimilation, and rapid adaptation, which are essential for implementing new production methods, equipment upgrades, and workflow optimizations typical of process innovation [61,62,63,64,65,66,67]. For instance, investments in creativity and experimentation create psychological safety and structured opportunities for employees to test process improvements [91], while timely technology acquisition enables the integration of Industry 4.0 tools like automation and IoT into manufacturing operations [92]. In contrast to previous evidence, where financial constraints, lack of skilled personnel and of suitable partners and market uncertainty were identified as major barriers to innovation [56,57,58,59], the perceived barriers did not significantly influence any of the four innovation outcomes considered in this study.

These findings suggest that firms operating in mature innovation ecosystems may be less constrained by perceived barriers, such as established routines, collaborative networks, and accumulated experience, which enable them to adapt to resource limitations and market uncertainty. In such contexts, innovation appears to depend more on the presence of enabling internal capabilities—such as experimentation, creativity, and openness to external knowledge—than on the absence of obstacles [34,36]. This interpretation aligns with prior research showing that, in highly developed innovative environments such as Piedmont, proactive organizational capabilities can mitigate structural constraints, whereas barriers tend to play a more limiting role in less developed or less coordinated systems [38,59,93,94].

Another important factor facilitating collaboration between private companies and PROs is the construction of social relationships and the development of trust and social connections [37,38,42]. In the present results, trust in PRO expertise emerged as a significant predictor specifically for product innovation. This is particularly promising, as product innovation typically involves market-oriented outputs where firms predominantly turn to other private companies for rapid commercialization and customer insights [38,39,40,41,42], whereas our findings demonstrated PROs’ potential to contribute novel technical knowledge when trust overcomes institutional barriers. This likely occurs because trust reduces perceived risks in sharing sensitive product development data, enabling firms in Piedmont’s innovation ecosystem to access PRO-generated innovations that enhance firms’ sustainable competitiveness. These results support the need to promote and strengthen innovation hubs and collaboration networks, which institutionalize trust-building interactions and structurally embed public-private ties to unlock PRO contributions beyond traditional process domains.

Interestingly and unexpectedly, our results did not show any significant relationships between the independent variables and both marketing and organizational innovations. This lack of correlation can be explained by a combination of structural, organizational, and institutional factors that characterize the regional innovation system and the local industrial context [95].

A first explanatory factor may concern the fragmentation of the regional innovation ecosystem. Piedmont hosts a rich set of actors involved in innovation activities, including firms, universities, research centers, clusters, and innovation hubs. However, interactions among these actors, in spite of the presence of the innovation hubs, are still often project-based [75]. As a result, OI initiatives tend to remain isolated episodes rather than being integrated into firms’ long-term organizational and marketing strategies. Consequently, collaboration with external partners does not always translate, for example, into internal organizational change, such as the restructuring of R&D processes, decision-making routines, or governance mechanisms, nor into sustained marketing innovations like new distribution channels, product repositioning, or customer engagement strategies that could amplify market reach and competitiveness. Second, the manufacturing sector in Piedmont is largely dominated by small and medium-sized enterprises (SMEs) [95]. While SMEs may engage in forms of OI, especially through informal collaborations with suppliers, customers, or local research institutions, they frequently lack the managerial capabilities, financial resources, and organizational slack required to implement systematic organizational innovations [74]. In this context, external openness often remains opportunistic and reactive, without leading to formal changes in organizational structures, roles, or internal processes. Therefore, OI practices may exist in the absence of deliberate organizational transformation. Finally, regional industrial and innovation policies have historically emphasized technological and product innovation more than non-technological forms of innovation, such as organizational and marketing change. This technology-oriented focus may encourage firms to adopt advanced technologies or collaborate externally without simultaneously investing in organizational redesign, change management practices, and new marketing strategies. The result is an asymmetric innovation pattern, in which OI initiatives are not accompanied by corresponding organizational and marketing innovation activities.

These factors suggest that fostering a stronger alignment between OI and both organizational and marketing innovation in the investigated regional ecosystem requires not only promoting external collaboration but also supporting firms in developing internal organizational capabilities, change management skills, and integrative governance structures that enable sustained marketing strategies [74].

Limitations of the Present Study and Future Research Development

This study has several limitations that should be acknowledged. First, the relatively small sample size (82 respondents) appeared to be sensitive to even a few extreme responses, and this may limit the generalizability of the findings beyond the firms included in the study. Therefore, the results should be interpreted with caution, and future research should replicate these analyses on larger samples to assess the stability of the observed relationships.

Second, the focus on a single region, i.e. Piedmont, while theoretically justified by its strong manufacturing base and innovation ecosystem, constrains the external validity of the results, as collaboration dynamics and OI practices may differ in other regional or national contexts. Third, participants were recruited through an innovation cluster (MESAP), whose member firms are likely to be more exposed to and engaged in OI activities, potentially introducing a systematic selection bias as cluster-affiliated firms may be structurally more collaboration-oriented than manufacturing firms outside such ecosystems. It is important to clarify that the primary objective of this study was not to assess whether firms innovate but to examine how firms operating within an innovation-supportive environment perceive and engage in collaborative OI practices, particularly with different partner types, including other firms and public research institutions. Therefore, since findings from non-representative samples typically affect the absolute level of the variables, not their relationships [96], and the latter were the main focus of this study, we are confident in the overall validity of our results. However, larger and more diverse samples, including firms from multiple regions and innovation ecosystems with different levels of maturity, would improve the generalizability of findings and enable comparative analyses across different institutional and innovation contexts.

A further limitation may concern the use of single-item self-report measures to assess product, process, marketing, and organizational innovation (one for each innovation). Although this approach is consistent with survey-based innovation research and has been previously used in studies on SMEs [97], single-item measures may not fully capture the complexity or intensity of innovation activities. Future studies could use multi-item scales or objective indicators to provide a more comprehensive assessment of firm innovation.

Finally, the reliance on self-reported data may introduce subjective bias, including social desirability and perceptual distortions regarding innovation activities and collaboration effectiveness. Moreover, the cross-sectional design of the study does not allow for causal inferences or the examination of how collaboration and innovation dynamics evolve over time. Although perceptual measures are widely used in innovation research [98], future research could strengthen empirical evidence by integrating survey-based data with objective innovation outputs and longitudinal performance indicators (e.g., patents, sales from innovation, productivity gains), thus enabling a more comprehensive and causally grounded understanding of innovation dynamics within regional ecosystems.

5. Conclusions

By clarifying how collaboration patterns and relational mechanisms relate to differentiated innovation outcomes, this study contributes to a better understanding of the conditions that may support sustainability-oriented transformation processes in a regional manufacturing innovation ecosystem. The results of the present study suggest that collaboration is not a uniform innovation input, but a differentiated strategic resource whose effects depend on the partner type and the innovation domain. The collaboration patterns that emerged in the studied manufacturing ecosystem support a view of OI in which manufacturing firms leverage private partners more strongly for market-facing and internally transformative changes, while they turn to public institutions when innovation requires technical depth, experimentation, and process-oriented improvements. Since the perceived importance of collaboration is a consistent predictor of innovation, ecosystem intermediaries and policy actors should invest in mechanisms that raise the perceived value and clarity of collaboration, particularly with public institutions, through transparent engagement pathways, brokerage services, and more visible evidence of outcomes (e.g., demonstrators, pilot projects, and diffusion initiatives). Furthermore, the relatively high trust in research institutions’ expertise reported by respondents suggests that trust may function as an enabling condition, but also that converting trust into concrete co-creation and implementation may require reducing coordination and administrative frictions and supporting the translation of research results into deployable solutions. Finally, while barriers were perceived at moderate levels, they did not emerge as significant predictors of innovation outcomes in the regression models. Instead, internal drivers, such as experimentation, dedicated financial support, timely technology acquisition, staff creativity, external training, and openness to partnerships, showed a significant positive relationship with process innovation. This indicates that, at least in this regional manufacturing hub, proactive internal conditions may matter more than perceived obstacles, suggesting the need for managers to strategically combine partnerships with other firms with targeted engagement of public research institutions while consolidating internal drivers that enable the effective absorption and exploitation of external knowledge.

Theoretical and Practical Implications of the Study

From a theoretical perspective, this study contributes to the OI literature by offering a differentiated analysis of innovation outcomes within a regional manufacturing ecosystem. By integrating structural, relational, and perceptual dimensions within a single empirical framework, the research advances beyond descriptive accounts of collaboration and provides evidence on how these factors relate to specific innovation domains. Importantly, the findings reveal heterogeneous patterns across innovation types. While certain relational and structural mechanisms are associated with product and process innovation, no significant effects emerged for marketing and organizational innovation. Rather than weakening the contribution, these results suggest that collaborative dynamics within manufacturing-centered ecosystems may primarily influence technological innovation domains, whereas non-technological innovations may depend on internal organizational capabilities or alternative strategic processes. This highlights the importance of avoiding assumptions of uniform effects of OI practices across innovation types.

From a practical standpoint, the study provides nuanced insights for managers and policymakers. The evidence suggests that strengthening collaboration with external partners and fostering trust may be particularly relevant for technological innovation activities. At the same time, the absence of significant associations for marketing and organizational innovation indicates that additional organizational levers—such as internal culture, managerial practices, or strategic alignment—may require targeted attention beyond inter-organizational collaboration alone.