Innovation Ecosystem Paradox: How Strong External Support Weakens Project Management—Sustainability Innovation Link

Abstract

This study examines the impact of structured internal innovation project management (IPM) practices and external innovation ecosystem (IE) characteristics on sustainable and responsible innovation (SRI) in EU widening countries. Using a two-stage Delphi-informed survey of 100 firms across Bosnia and Herzegovina, North Macedonia, Albania, and Serbia, the research applies moderated multiple regression analysis to examine the interplay between internal processes and external ecosystem maturity. Results show that both structured innovation phases and tools have a positive impact on SRI. However, while innovation phases consistently enhance SRI regardless of ecosystem conditions, the effect of innovation tools weakens in stronger ecosystems, suggesting a resource substitution dynamic. These findings challenge the assumption that greater ecosystem support uniformly improves innovation outcomes. The study contributes to the theoretical integration of the Resource-Based View and Innovation Ecosystem Theory, highlighting context-specific conditions in transitional economies. Practical implications are offered for managers and policymakers; firms in weaker ecosystems should prioritize building internal innovation capabilities, while those in mature ecosystems may gain more from leveraging external collaborations. The research advances debates on sustainable innovation strategies by showing how the effectiveness of internal management practices depends on ecosystem maturity, offering insights for both policy interventions and strategic innovation management in developing economies.

1. Introduction

In recent decades, patterns of production and consumption have changed substantially [1], resulting in problems like climate change, resource overexploitation, and biodiversity decline, all of which threaten life on Earth [2]. Under pressure from the public, non-governmental organizations, and institutional stakeholders, as well as driven by their own interest in becoming and remaining socially responsible, companies are shifting their paradigm from a linear to a circular economy, embracing sustainable and responsible innovation. Nowadays, organizations are more interested in the sustainable aspects of their operations for they know that, to remain competitive, they have to invest in innovations that include and promote sustainable development [1]. The path to innovation is not straightforward, especially for small and medium-sized enterprises (SMEs), which, due to limited resources, generally cannot rely on their own research and development (R&D) activities and are compelled to shift their innovation management paradigm from closed to open innovation (OI) [3], a new paradigm of collaboration involving a large number of stakeholders. Within these ecosystems, startups often act as critical knowledge brokers, enhancing knowledge flows and fostering innovative collaborations with incumbent firms [4]. Therefore, knowledge spillovers [5] and collaboration within the entrepreneurial ecosystem in all directions (inbound, outbound, and coupled innovation) within an open innovation ecosystem (IEs) [6,7,8,9] become a vital opportunity for SMEs to maintain their competitive position. Firms and other actors collaborate by combining their products and services into an integrated solution that better meets customer needs. Such collaboration creates more value than any partner could generate alone by harnessing complementarities in resources, capabilities, and know-how [10]. Meanwhile, frameworks such as the natural resource-based view emphasize dynamic capabilities, including green entrepreneurial orientation and corporate social responsibility, as drivers of green innovation and sustainable performance [11,12,13]. These patterns are especially noticeable in EU-candidate (widening) countries. The potential for green innovation is clear, yet these regions still lag on core sustainability metrics, most notably R&D spending, renewable-energy uptake, and waste-management capacity [14]. Innovation finance remains largely public, and ecosystem structures are only slowly maturing. In the Western Balkans, for example, average annual PM2.5 concentrations exceed EU and WHO limits, signaling both environmental and institutional weaknesses and underscoring the need for stronger innovation strategies [15]. Embedding sustainability-focused practices into Innovation Project Management (IPM) can help close this gap by enabling firms to deliver environmentally and socially responsible solutions without sacrificing business performance [16]. While a small but growing literature links IPM to stronger innovation results, we know far less about the specific pathways through which internal practices translate into sustainability-oriented outcomes across different organizational contexts. Innovation ecosystems (IEs) can substantially enhance sustainable innovation by moderating firm-level efforts; their impact, however, depends on ecosystem maturity, collaboration density, and supporting infrastructure. This points to the importance of policies that expand inclusiveness, resource availability, and connectivity in less developed settings [17,18,19]. Well-designed IEs, offering collaborative networks, supportive regulation, knowledge-sharing platforms, and accessible finance and mentoring, can markedly strengthen firms’ capacity for sustainable innovation [20,21].

On the other hand, resource substitution theory suggests that abundant external support may reduce the marginal value of internal capabilities such as IPM tools [21]. Huang & Li [22] found that aligning with strong external resources (e.g., well-matched external partners and networks) significantly enhanced green innovation outcomes. It implies that external partners can improve performance, suggesting that internal resources alone are less decisive when alignment is strong. Zheng & Cai [22] argued that in innovation systems that evolved into mature ecosystems (with ample government support, infrastructure, and partnerships), firms relied less on in-house R&D, as external institutions provided many innovation functions. Audretsch et al. [5] state that when industry-level knowledge ecosystems are strong, even firms with modest internal R&D can innovate successfully by leveraging external spillovers, illustrating a substitution effect between internal R&D and external knowledge spillovers. This prompts a deeper inquiry: “Is the path to sustainable innovation always paved by structured internal processes, or does a strong ecosystem change the rules? And if it does, does it amplify or diminish the role of internal project management?”.

This question is especially relevant in EU-widening countries, where businesses often operate in transitional economies with restricted access to ecosystem support [23]. In such settings, innovation is usually led by SMEs and necessity-driven entrepreneurs who have limited R&D resources. Such settings bring into focus the interplay between internal project management practices and external ecosystem maturity that remains underexplored in the empirical literature. Although previous research has examined IEs and sustainable innovation separately, few studies explore how ecosystem maturity interacts with IPM to influence sustainable and responsible innovation (SRI), and empirical evidence of this moderating role remains especially limited in EU widening countries with transitional economies.

In line with the European Green Deal and the EU’s 2030 Agenda, sustainable and responsible innovation has become a strategic imperative for firms seeking long-term competitiveness and legitimacy [1,16]. However, empirical evidence shows persistent disparities between EU core and widening countries in innovation inputs and outputs [14,15]. Carayannis et al. [17] argue that effective innovation ecosystems depend on interactions among government, academia, industry, and society—a model often underdeveloped in transitional economies. Recent OECD reports [14,15] confirm that R&D intensity, digital readiness, and collaboration density in the Western Balkans remain below EU averages, limiting firms’ access to ecosystem-level resources. Consequently, many SMEs rely on internal project management capabilities to integrate sustainability goals into innovation practice. This dual dependence on internal structure and external support justifies our focus on the moderating role of ecosystem maturity in shaping sustainable and responsible innovation outcomes.

This study addresses this gap by empirically testing whether ecosystem maturity weakens or strengthens the effectiveness of IPM in driving SRI, i.e., whether firms can “make up” for external deficits through internal structures, or, conversely, whether strong innovation ecosystem support can compensate for weak internal innovation capabilities. Although sustainability is widely recognized as a strategic priority, its formal adoption among SMEs remains uneven [24,25]. This underscores the need to examine the internal and external conditions that shape how sustainability principles are implemented at the project level. Our analysis centers on two core dimensions of Innovation Process Management (IPM): (1) how firms structure the phases of innovation and (2) how they deploy specific innovation tools.

We found that although previous studies examined the link between innovation management and sustainability, very few addressed how ecosystem maturity moderates this relationship, particularly in EU widening and transitional economies. For example, Khiewngamdee & Chanaim [26] examined how institutional and innovation-ecosystem factors moderate the effects of green practices on export performance in European countries. Shawesh et al. [27] examined how non-core firms’ eco-embeddedness influences breakthrough innovation, finding that ecological legitimacy and technology turbulence moderate this effect. Wang et al. [28] used green-technology turbulence as a moderator in the green-innovation strategy-performance link. Despite these contributions, studies that explicitly model ecosystem maturity as a moderating variable remain scarce, particularly for transitional and EU-widening economies. Our study addresses this gap by examining how ecosystem maturity moderates the relationship between innovation project management tools and sustainable innovation outcomes across four Southeast European countries.

Grounded in the Resource-Based View (RBV) [29,30] and Innovation Ecosystem theory [31,32], we develop and test a conceptual model using a Delphi-informed survey of 100 firms spanning diverse sectors.

The aim of this research is to explore the interaction between internal IPM practices and external ecosystem features in the context of EU-widening countries.

In line with EU research-policy terminology, the term ‘widening countries’ refers to Serbia, Bosnia and Herzegovina, Albania, and North Macedonia, which represent emerging innovation ecosystems within the EU enlargement context. While it’s believed that strong ecosystems support SRI, the study finds a surprising moderation effect: internal tools are less effective in more developed ecosystems. These findings are important for SMEs managing the green transition amid institutional gaps. The paper makes several contributions: first, it empirically confirms that structured internal capabilities, especially innovation phases, significantly support SRI results; second, it reveals a moderation effect where the impact of tools lessens in mature ecosystems, aligning with a “resource substitution” effect, thus deepening understanding of resource interaction during institutional change; third, it offers practical guidance for managers and policymakers, SMEs in weaker ecosystems should focus on internal processes, while those in mature ecosystems should emphasize external partnerships; finally, it stresses the importance of ecosystem-level programs (e.g., incubators, accelerators) prioritizing long-term, inclusive innovation over short-term gains. The paper is organized as follows. The next section reviews literature on IPM phases, tools, and IEs’ role in sustainable, responsible innovation. The subsequent section describes the methodology, including the sample, model design, and empirical strategy. It then presents empirical results and discussion, followed by conclusions, limitations, and ideas for future research.

2. Literature Review and Hypothesis Development

The proposed research paper is grounded in the IEs perspective [31,32] and the resource-based view (RBV) of the firm [29,30], which highlight that innovation outcomes are not only a function of internal project management capabilities but also of external ecosystem support.

The Innovation Ecosystem (IE) perspective views innovation as the result of dynamic interactions among a diverse network of actors, institutions, and resources that together foster knowledge sharing, technological progress, and value creation [21,33]. From this standpoint, a firm’s ability to innovate does not rely solely on its internal strengths, but also on how effectively it collaborates with its environment—including universities, research institutes, government bodies, and other firms—to exchange knowledge, leverage complementary resources, and co-create value. This perspective underscores that innovation is inherently systemic, depending as much on the quality of relationships and coordination among ecosystem members as on the firm’s own efforts.

In contrast, the Resource-Based View (RBV) emphasizes the importance of what lies within the firm. It argues that lasting competitive advantage stems from unique resources and capabilities that are valuable, rare, difficult to imitate, and non-substitutable [29,30]. Within the sphere of innovation management, RBV highlights project management skills, human capital, and absorptive capacity as crucial enablers of innovation success. Yet, when considered alongside the IE perspective, RBV suggests that these internal capabilities achieve full potential only when strategically aligned and integrated with the broader ecosystem in which the firm operates.

2.1. Innovation Project Management Phases and Tools

IPM is a systematic way organizations arrange people, activities, and resources to turn new ideas into offerings that create market value. In practice, the innovation journey typically moves through a series of phases, idea generation, selection, development, testing, and commercialization, supported by established tools and methods such as the Stage-Gate model, Design Thinking (DT), and the Business Model Canvas (BMC), all of which help improve efficiency and raise the likelihood of successful outcomes [34,35]. Among these, the Stage-Gate model, developed by Cooper, remains one of the most widely used frameworks [34]. It segments innovation into discrete stages from ideation to launch, separated by decision “gates” where projects are assessed and either advanced, redirected, or stopped [34]. Drawing on evidence from more than 300 firms, Cooper reports that organizations applying a rigorous Stage-Gate approach achieve up to 30% higher success rates in new product development, largely because the model structures uncertainty, guides resource allocation, and reduces the risk of late-stage failures [34]. Complementing such structured approaches, creative tools like DT and the BMC have gained traction across startups and established firms alike [35,36,37]. DT keeps the process grounded in user needs through iterative cycles of empathy, problem definition, ideation, prototyping, and testing [36,37], while the BMC helps teams articulate how an idea creates and captures value, ensuring technological promise is matched by a viable business model [35].

Baldassarre et al. [38] propose a novel framework for what they call “responsible design thinking,” linking specific DT practices with conceptual dimensions of responsible innovation. Baldassarre et al. [38] substantiate ethical thinking in innovation management by drawing on responsibility literature [39,40], and reposition DT to bring back more of its original purpose of doing innovation for the benefit of society and the natural environment, rather than just for competitive advantage. The capacity of firms to implement these phases and tools systematically reflects their dynamic capabilities, that is, their ability to adapt, integrate, and reconfigure internal and external competencies to meet rapidly evolving environments [41]. While economic impact is indeed an important aspect, it is becoming increasingly clear that innovation management must consider environmental and social impacts as well [40,42].

2.2. Moderating Role of Innovation Ecosystem Characteristics in Producing Sustainable Innovation

Businesses find it challenging to achieve novel innovation by relying exclusively on their R&D and innovation capabilities and resources [43]. This is mainly due to the complexity of new technologies and the necessary complementary resources to develop novel and sustainable innovation [44,45]. In recent years, the OI notion has started to be explored in the context of sustainability [46]. There is no consensus among authors and three main concepts (green innovation, sustainable innovation, and eco-friendly innovation) are used interchangeably [24]. Yet, despite this momentum, many studies continue to frame sustainability through a narrow environmental lens, neglecting its economic and social dimensions [47]. Emerging perspectives like “outnovation”, the intentional simplification of products in favor of authenticity and sustainability, suggest a reframing of innovation priorities [48]. Digital technologies play a pivotal role in enhancing these collaborative processes by providing the necessary infrastructure for real-time communication, data sharing, and collaborative development [49]. Recent empirical studies highlight that advanced digital technologies—such as artificial intelligence, big data analytics, and digital twins—reshape ecosystem interactions, innovation processes, managerial decision-making, and customer-centric value creation [50,51,52]. A supportive IEs, marked by enabling policies, collaboration networks, knowledge-sharing mechanisms, and accessible financial and mentoring resources, can enhance a firm’s capacity for sustainable innovation by fostering adaptive learning and reducing barriers to experimentation [20,21]. Understanding the processes by which business networks adapt to growing environmental pressures and collaborate to attain sustainability in their strategies and operations are of growing interest [53], given the rapidity of the expected change as well as the multitude of challenges its implementation entails [54,55]. Much of this rests with the challenges of integrating environmental sustainability into product and process innovation [56], a process resulting in eco-innovation [57].

IEs are increasingly viewed as moderating variables that shape the relationship between IPM practices and sustainable innovation outcomes [20,41]. Sustainable innovation (SI), innovation that pursues environmental and social goals alongside economic performance, demands substantial coordination and long-term commitment, often beyond the capacity of any single firm [58,59]. In this setting, supportive innovation ecosystems (IEs) act as critical enablers. Evidence backs this view; Janik and Ryszko [60], in a content analysis of 324 studies, show that firms’ adoption of open and environmentally sustainable innovation (OESI) is shaped by internal factors (e.g., cooperative culture, absorptive capacity) and by external, inter-organizational dynamics such as network position, trust, and shared cognitive frames. These results underscore the moderating role of the broader ecosystem, policy frameworks, financial incentives, and research linkages, in influencing firm-level SI behaviors. Academy-to-Business (A2B) or University–Industry Collaborations (UIC) further strengthen these dynamics, delivering benefits that range from enhanced national competitiveness to shared value creation across partners [61,62]. Many authors have also highlighted UIC’s importance in improving innovation capabilities [63,64]. Universities and industries work together with the principal objective of generating new knowledge [65], but even though collaborative R&D projects have become more frequent, many still fail [66]. While supportive ecosystems can significantly enhance IPM outcomes, particularly when aligned with sustainable goals [56], barriers rooted in structural gaps, institutional inertia, or collaboration deficiencies may undermine this potential [67]. Comparable studies in Central and Eastern Europe (e.g., Poland, Croatia, and Hungary) reveal similar patterns of ecosystem asymmetry, where institutional maturity moderates the role of internal project management tools in fostering sustainable innovation [68,69,70,71,72]. Therefore, future innovation policy and practice must focus on ecosystem-level interventions that promote inclusiveness, resource accessibility, and network connectivity, especially in regions where ecosystem development is still emerging [19,73].

2.3. Sustainable and Responsible Innovation as an Outcome of the Innovation Project Management

Sustainable and Responsible Innovation (SRI) has become a central aim of Innovation Process Management (IPM). Society now expects innovation to drive growth while also protecting the environment and improving social well-being. In practice, responsible innovation means tackling “wicked” problems through experimentation, collaboration, and continuous learning, with clear attention to economic, social, and environmental effects [40]. This view aligns with the Triple Bottom Line (TBL); profit, planet, and people should be advanced together rather than traded off [42]. Projects that apply TBL principles tend to create value that lasts and to avoid unintended harms. For firms, the implication is practical and direct: build these principles into every stage of the innovation process, i.e., governance that sets the right guardrails, meaningful engagement with stakeholders, and transparent evaluation of outcomes, so that sustainable results are not an afterthought but the norm [74]. Green innovation allows manufacturers to create and develop items that do not cause harm to the environment [75]. Companies leading in green innovation may gain a competitive edge and charge premium prices for their creative eco-friendly products, thereby enhancing their environmental performance [76]. In according to Hanaysha et al. [77] research, the use of green innovation will yield enhanced sustainable performance, including cost effectiveness, increased profitability, heightened customer satisfaction, reduced environmental pollution, enhanced product quality, and an elevated reputation. Consequently, the owners/managers of SMEs must enhance their comprehension of the importance of ongoing innovation in green product and process improvements to attain improved business success [77]. Green innovation is shifting the central question for firms from “How do we make things more efficiently?” to “How do we use cleaner materials and cut pollution while we make them?” [78]. Many large companies are moving in this direction under policy and strategy pressures, but uptake among SMEs remains uneven [24]. On the ground, teams often lack practical, standardized tools to embed Sustainable and Responsible Innovation (SRI) into projects, and sustainability goals can slip when cost, time, and scope pressures mount. Internal limits, such as low absorptive capacity and weak cross-functional coordination, also slow progress [79]. These hurdles are amplified by external barriers, including scarce sustainability-oriented finance, regulatory uncertainty, and fragmented innovation ecosystems.

2.4. Conceptual Development and Hypothesis

2.4.1. Direct Effect Hypotheses

Empirical studies show that structured innovation phases and tools matter. Carlgren et al. [37], for instance, find that design thinking (DT) helps teams work across functions and align solutions more closely with user needs. Likewise, the Business Model Canvas (BMC), introduced by Osterwalder and Pigneur [35], offers a practical way to map how a firm creates and captures value. Joyce and Paquin [80] report that the BMC is particularly useful in early-stage settings, where teams must iterate rapidly in fast-moving markets. In practice, canvases like the BMC have helped both startups and large companies (e.g., GE, Nestlé) move beyond a product-only lens to a fuller, business-model view [81]. These tools, however, are not cure-alls. Their iterative logic can clash with rigid project governance or tight budgets, and the BMC can oversimplify complex environments, especially in multinationals or multi-sided markets where relationships are more fluid than static boxes suggest [82]. This is where organizational capabilities come in. From a dynamic capabilities perspective [41], firms need not just access to tools but the skill to use them strategically and adapt them over time. Barreto [83] reinforces this point, showing that dynamic capabilities significantly moderate the link between innovation practices and firm performance. Without such capabilities, such as knowledge integration, organizational learning, and strategic flexibility, tools like design thinking and the BMC may remain underutilized or misapplied. The concept of Responsible research and innovation (RRI) underpins the theoretical foundation of SRI, emphasizing anticipatory, inclusive, and reflexive innovation processes. According to Stilgoe et al. [39], empirical research supports the assertion that adopting sustainability and responsibility-oriented practices within IPM can lead to better innovation outcomes. Bocken et al. [16], using a multiple–case study design, identify several sustainable business model archetypes, such as closing resource loops, turning waste into value, and prioritizing stakeholder well-being, that directly shape how projects are executed. Firms that embed these principles systematically into their innovation projects are more likely to deliver offerings that meet societal and environmental standards without sacrificing commercial viability. Yet implementation remains uneven. Silvius and Schipper [25], in a cross-industry study, document a persistent gap between project managers’ stated recognition of sustainability’s importance and the formal integration of sustainability metrics into project management methods.

H1. 

The extent to which companies go through innovation project management phases has a significant positive effect on sustainable and responsible innovation outcomes.

H2. 

The extent to which companies use innovation project management tools has a significant positive effect on sustainable and responsible innovation outcomes.

2.4.2. Moderating Effect Hypotheses

Empirical work shows that well-developed ecosystem support can substantially improve innovation outcomes. Ritala et al. [84] find that ecosystems fostering cognitive proximity, mutual trust, and shared goals among stakeholders deliver more effective sustainable innovation, particularly in high-tech settings. Likewise, Cillo et al. [85] argue that ecosystem-level capabilities, such as mechanisms for integrating knowledge and making sense of emerging trends together, are central to sustainability-oriented innovation. Taken together, these studies suggest that for innovation ecosystems (IEs) to genuinely moderate the link between innovation management and sustainability, they must do more than supply resources; they must enable the circulation of knowledge, build trust, and cultivate a collaborative intent among actors. In this sense, IEs can play a pivotal moderating role, though their effectiveness hinges on institutional maturity, cross-organizational collaboration, and supportive infrastructures [17,18]. At the same time, research highlights significant ecosystem barriers to sustainable innovation (SI). Janik and Ryszko [60] point to internal obstacles, resistance to externally developed ideas and limited absorptive capacity, as well as external constraints such as scarce funding and weak stakeholder engagement, especially early in the innovation process. Hueske and Guenther [86] similarly emphasize the drag of organizational rigidity, limited learning mechanisms, and low openness to external knowledge. Cooperation-related frictions, misaligned incentives, cultural mismatches, and difficulty identifying suitable partners, are also common and can blunt the benefits of ecosystem participation [87]. Overall, while IEs can enhance SI, their real impact depends on the alignment and maturity of institutional, financial, and social infrastructures [17,18,73].

On the other hand, some studies show substitutional effects of external support and mature innovation ecosystems on sustainability outcomes. Huang & Li [22]. analyzing Chinese firms’ environmental innovation strategies and their green innovation performance, found that aligning with strong external resources, such as external partners and networks, significantly enhanced green innovation outcomes. It indicates that strong collaboration with external partners enhances performance, implying that internal resources become relatively less critical when such alignment is high. Zheng & Cai [88], focusing on the role of public policy, through case evidence and analysis, show that in innovation systems that evolved into mature ecosystems (with plentiful government support, infrastructure, and partnerships), firms relied less on in-house R&D processes, as external institutions provided many innovation functions. Audretsch et al. [5] examine how firms benefit from external knowledge spillovers across different industries. Their study finds that the impact of knowledge-rich environments on innovation depends on a firm’s internal R&D capabilities. In some sectors, smaller or less R&D-intensive firms reap greater innovation gains from external knowledge networks than their more capability-rich counterparts. In other words, when industry-level knowledge ecosystems are strong, even firms with modest internal R&D can innovate successfully by leveraging those external spillovers.

Viewed through a systems lens, innovation ecosystems (IEs) can either amplify or constrain firm-level practices; their features strengthen or weaken internal innovation depending on how well they align with organizational goals and capabilities [18,89]. Our results on tool use fit resource-substitution logic. When external support is plentiful, the incremental value of internal tools falls; in weaker ecosystems, firms must lean more on their own toolkits, which then become essential substitutes for missing external capacity and are critical for achieving sustainable innovation (SI) outcomes. In mature innovation ecosystems, abundant external resources, policy incentives, and networked collaborations can substitute for internal project management tools, thereby reducing their marginal contribution to sustainable innovation outcomes. This resource-substitution logic explains the negative moderation effect observed between ecosystem maturity and tool-based innovation performance [33,90,91]. These dynamics are especially pronounced in emerging and transition economies. In the Western Balkans, for example, business–academia (B2A) collaboration remains limited, and funding dedicated to cooperative R&D is scarce [23]. Although North Macedonia and Serbia have begun to build institutions that support innovation linkages, the system remains fragmented and too often neglects cross-sector collaboration [23]. Such structural weaknesses curb firms’ ability to pursue open and sustainable innovation, reducing the moderating role that IEs might otherwise play in these settings.

H3. 

The positive relationship between the extent to which companies go through innovation project management phases and sustainable and responsible innovation outcomes is weakened in the presence of stronger innovation ecosystem characteristics.

H4. 

The positive relationship between the extent to which companies use innovation project management tools and sustainable and responsible innovation outcomes is weakened in the presence of stronger innovation ecosystem characteristics.

2.5. Empirical Model

Building on the conceptual development and grounded hypotheses, the empirical model (Figure 1) illustrates the direct and moderating relationships proposed in this study. Specifically, the model posits that both structured innovation project management phases (H1) and the use of innovation tools (H2) positively influence sustainable and responsible innovation (SRI) outcomes. In addition, the model integrates the moderating role of innovation ecosystem characteristics (IES), reflecting the idea that the strength of an external ecosystem can either reinforce or attenuate the effectiveness of internal innovation efforts. Hypotheses H3 and H4 test whether a mature and resource-rich ecosystem weakens the influence of IPM phases and tools, respectively, on sustainability outcomes. The model is designed to capture these interactions in the context of transitional economies, offering a nuanced perspective on the interplay between firm-level capabilities and ecosystem conditions in driving sustainable and responsible innovation.

Figure 1. Empirical model.

• IP = the extent to which companies go through innovation project management phases (e.g., opportunity identification, idea generation, prototyping, etc.);
• IT = the extent to which companies use innovation project management tools (e.g., design thinking, MVP testing, business model canvas, etc.);
• IES = innovation ecosystem characteristics (moderating variable);
• SRI = sustainable and responsible innovation outcomes.

2.5.1. Independent Variables

In the empirical model, the independent variables are Innovation project management phases (IP) and Innovation project management tools (IT). We include eight items that measure firms’ use in the innovation process management phases. Each item is measured on a five-item Likert scale that asks firms to evaluate how significant the various phases of the IPM are.

2.5.2. Dependent Variable

Dependent variable in the empirical model is Sustainable and responsible innovation (SRI) outcomes measured by using a set of twelve items derived from two subscales: sustainable innovation and responsible innovation. The survey question is as follows: How significant are the following characteristics of sustainable innovation?

2.5.3. Moderating Variable

Moderating variable in our conceptual model is Innovation ecosystem characteristics (IES) measured by using a set of seven items. The survey question is as follows: How significant are the following characteristics for sustainable innovation ecosystem?

3. Methodology of Empirical Analysis

3.1. Research Design

This study employed a mixed-method research design, integrating a preparatory Delphi method [92] with a subsequent cross-sectional quantitative survey. The research aimed to explore the impact of innovation management practices, specifically, IPM phases and tools, on SRI outcomes. Additionally, the study examined how characteristics of the IEs may moderate these relationships.

3.2. Delphi Study and Instrument Development

The Delphi study took place in November 2024 within the USE IPM Horizon Europe project, which supports the transfer of knowledge and skills from EU universities to researchers in widening countries, Serbia, Bosnia and Herzegovina, Albania, and North Macedonia. The Delphi method is well suited to building expert consensus through successive questionnaires; its key features are respondent anonymity, iterative rounds with controlled feedback, and statistical aggregation of group judgments [93]. It is especially useful when hard data are limited and expert input is essential, such as in forecasting, policy design, priority setting, and best-practice identification [94], and is widely used to clarify complex issues, refine theoretical frameworks, and guide decisions in uncertain or emerging fields [95]. In this study’s first phase, a two-round Delphi was used to validate and refine both the conceptual framework and the survey instrument. Experts from academia, industry, and policy were purposefully selected for their experience with innovation ecosystems in transitional economies. The panel comprised 20 stakeholders (10 women, 10 men) across the four widening countries, who anonymously rated and commented on proposed constructs and indicators related to innovation management and sustainable innovation. Fourteen participants were managers (e.g., R&D, innovation, CEOs, incubator centers), three were entrepreneurs (including one start-up founder), and three were innovation specialists (a national technical inspector, a machine-learning expert, and an innovation researcher) (Figure 2).

Figure 2. Methodological framework.

The expert questionnaire contained 13 items, each with more than five response options. Across the two rounds, five options were collectively discarded, indicating notable convergence despite the panel’s cross-country diversity. This alignment suggests shared views on priorities and needs and strengthens the relevance and validity of the Delphi results. Guided by the panel’s feedback, the questionnaire items were finalized to ensure content validity, contextual fit, and clarity (see Appendix A). The finalized instrument was then used for the main survey.

3.3. Sample and Data Collection

Following the Delphi phase, we fielded a structured survey as part of the needs analysis in four widening countries—Serbia, Bosnia and Herzegovina, Albania, and North Macedonia. The aim was to quantify how firms perceive and practice innovation and sustainability. We used purposive sampling to reach founders, managers, and R&D staff actively engaged in innovation within emerging ecosystems. The final sample included 100 respondents from varied business contexts. By firm size, 32% were from micro enterprises (fewer than 10 employees), 37% from small (10–49), 21% from medium (50–249), and 10% from large firms (250+). By sector, 38% operated in services, 29% in manufacturing, 19% in trade, and 14% in other activities. The sample was geographically balanced, with 25 participants from each country. Data were collected from December 2024 through late January 2025.

3.3.1. Target Population, Sampling Frame, and Sample Justification

The target population comprised innovation-active firms operating in Serbia, Bosnia and Herzegovina, Albania, and North Macedonia, specifically decision-makers (founders, senior managers, and R&D/innovation staff) with direct responsibility for innovation practices. Access to this population was operationalized through partner networks (e.g., university centers, incubators/accelerators), professional associations, and direct outreach (email/LinkedIn) to firms known to be engaged in innovation activities.

This approach focuses inference on organizations where the constructs of interest are observable in day-to-day practice, thereby aligning the sample with the study aims. Such purposive sampling is appropriate when inclusion requires domain-specific expertise and role-based knowledge central to the research question [96]. We also employed purposive, quota-guided sampling to ensure geographic balance (25 respondents per country) and heterogeneity by firm size and sector. The achieved sample (n = 100) included 32% micro, 37% small, 21% medium, and 10% large firms; 38% services, 29% manufacturing, 19% trade, and 14% other activities. Data were collected online between December 2024 and late January 2025. For the moderated regression models, our sample size meets established guidance for testing multiple correlation with several predictors (e.g., N ≥ 50 + 8 m) [97]. While stricter criteria for testing individual predictors (e.g., N ≥ 104 + m) are often difficult for organizational field studies, we benchmark model complexity and effect sizes against standard references and conduct robustness checks to guard against overfitting [97]. Consistent with AAPOR (American Association for Public Opinion Research) guidance on non-probability samples, we are transparent that findings generalize to innovation-active firms in the four countries rather than to the full population of all firms. The quota design prioritizes cross-country comparability and construct validity over population representativeness [98].

3.3.2. Assessing and Addressing Self-Selection Bias

As participation was voluntary and recruitment relied on professional networks, coverage and self-selection bias are possible; those most engaged with innovation and sustainability may have been more inclined to respond. We follow survey methodology guidance that such risks should be explicitly reported and empirically probed where feasible [99]. To evaluate whether sample composition due to self-selection could bias estimates, we included firm size, sector, and country as covariates (factors) in the GLM for both model specifications (the Innovation phases model and the Innovation tools model). Across models, these controls were not jointly associated with the outcome (size: p = 0.671/0.661; sector: p = 0.155/0.195; country: p = 0.142/0.052), and effect sizes were small to modest.

Crucially, focal relationships remained stable; in the Innovation phases model, In-novation phases remained positive and significant while Innovation ecosystem characteristics and the Innovation phases × Innovation ecosystem characteristics interaction were non-significant; in the Innovation tools model, Innovation tools and Innovation ecosystem characteristics were positive and significant and the Innovation tools × Innovation ecosystem characteristics interaction was significant. Taken together, these results indicate that our inferences are not driven by observable composition differences, thereby mitigating selection on observables linked to self-selection. Residual bias from unmeasured factors cannot be ruled out and is noted as a limitation.

3.4. Measurement Instrument

The questionnaire instrument was developed based on literature and refined through the Delphi method. It consisted of four primary constructs measured via multi-item Likert-type scales ranging from 1 (Completely insignificant) to 5 (Very significant):

• Sustainable and Responsible Innovation (SRI) was assessed with twelve items organized into two six-item subscales: sustainable innovation and responsible innovation. The sustainable-innovation subscale captured firms’ environmental practices, use of resource-efficient and renewable inputs, circular-economy integration, environmental stewardship, adaptability, emission-reduction technologies, and associated economic benefits [100]. The responsible-innovation subscale reflected ethical and societal dimensions, adherence to ethical principles, transparency and inclusiveness, social accountability, long-term societal value, community engagement, and mitigation of negative impacts [101]. Taken together, these twelve items provide a comprehensive measure of SRI that spans both ecological and social aspects of innovation performance.
• Innovation project management phases (IP). Six items measuring identification of innovation opportunities: evaluation of innovation ideas, testing and prototyping, user involvement and feedback loops, iterative development process, and implementation and scaling [39,102].
• Innovation Tools (IT). Six items measuring use of Design thinking methodology [36], application of the Lean Startup principles [103], use of Business model canvas [35], development of MVPs (Minimum Viable Products) [103], use of SWOT and PEST analyses in planning [104], and monitoring innovation Key Performance Indicators (KPIs) and dashboards.
• Innovation ecosystem characteristics (IEC). Seven items measuring availability of support institutions and services: presence of collaborative networks, access to public policy incentives, strength of university-business cooperation, digital infrastructure readiness, flexibility of regulatory frameworks, and innovation culture and mindset in the region [17,18,105].

A total of 100 valid responses were obtained. Data were collected online, with informed consent obtained from all participants. Anonymity and confidentiality were guaranteed, and participation was voluntary.

3.5. Data Analysis

All analyses were run in IBM SPSS Statistics 21. We first computed descriptive statistics and Pearson correlations to examine variable distributions and pairwise relationships. To test the hypotheses, we estimated two moderated multiple regression models. Model 1 tested the interaction between Innovation Phases (IP) and Innovation Ecosystem Characteristics (IEC) on SRI; Model 2 tested the interaction between Innovation Tools (IT) and IEC on SRI. Continuous predictors were mean-centered prior to creating interaction terms to limit multicollinearity [106]. Assumptions of linearity, normality, homoscedasticity, absence of multicollinearity, and limited outlier influence were checked and satisfied [107]. We report standardized coefficients, adjusted R², F-statistics, and p-values. This approach balances conceptual clarity with empirical rigor, enabling a reliable assessment of how innovation practices relate to sustainability outcomes in emerging ecosystems.

4. Empirical Results and Discussion

4.1. Descriptive Statistics and Correlations

Table 1 presents the means and standard deviations for the key variables in the study. Participants reported relatively high levels on all constructs: phases in innovation process management (M = 4.45, SD = 0.56), tools in innovation process management (M = 4.23, SD = 0.62), innovation ecosystem characteristics (M = 4.41, SD = 0.61), and sustainable and responsible innovation (M = 4.16, SD = 0.70).

Table 1. Descriptive statistics.

Variable	Mean	Std. Deviation	N
Phases in innovation process management	4.447500	0.5573714	100
Tools in innovation process management	4.227143	0.6172451	100
Innovation ecosystem	4.410000	0.6127401	100
Sustainable and responsible innovation	4.159167	0.6987337	100
Valid N (listwise)			100

Note: Authors’ calculation.

Table 2 presents the Pearson correlation coefficients among the key study variables. All correlations were statistically significant and positive. Notably, phases of innovation process management were strongly correlated with the use of innovation tools (r = 0.72, p < 0.001) and with the innovation ecosystem characteristics (r = 0.63, p < 0.001). Sustainable and responsible innovation was also positively associated with phases (r = 0.64, p < 0.001), tools (r = 0.57, p < 0.001), and ecosystem characteristics (r = 0.49, p < 0.001).

Table 2. Pearson correlations among study variables.

Variable	Phases in Innovation Process Management	Tools in Innovation Process Management	Innovation Ecosystem	Sustainable and Responsible Innovation
Phases in innovation process management	1	0.716 **	0.625 **	0.642 **
Tools in innovation process management	0.716 **	1	0.472 **	0.566 **
Innovation ecosystem	0.625 **	0.472 **	1	0.485 **
Sustainable and responsible innovation	0.642 **	0.566 **	0.485 **	1

Note: Authors’ calculation. ** Correlation is significant at the 0.01 level (2-tailed).

These findings indicate strong positive associations among the studied constructs, suggesting that structured innovation processes and supportive ecosystem characteristics are linked to enhanced sustainability-oriented innovation outcomes. Nonetheless, further analyses are necessary to examine the hypothesized moderation effects and to establish the robustness of these relationships.

4.2. Reliability Analysis of Measurement Scales

To evaluate the internal consistency of the measurement instruments used in this study, Cronbach’s alpha coefficients were computed for each construct. The results demonstrate that all four scales achieved good to excellent levels of reliability, thereby confirming that the items within each construct consistently measure the same underlying theoretical concept.

• Innovation project management phases (IP): The scale consisted of 8 items and achieved a Cronbach’s alpha of 0.889, which is well above the acceptable threshold of 0.70 [108]. This high value suggests a strong internal consistency, indicating that the items reliably capture different phases of innovation project management within firms.
• Innovation project management tools (IT): The 7-item scale for the use of innovation tools yielded a Cronbach’s alpha of 0.851, reflecting good internal consistency. This supports the reliability of the instrument in capturing how systematically companies apply tools in managing innovation activities.
• Innovation ecosystem characteristics (IES): The scale for assessing the perceived supportiveness of the external innovation ecosystem included 6 items and produced a Cronbach’s alpha of 0.827. The result confirms that the items collectively represent the perceived supportiveness of the external innovation ecosystem as experienced by firms.
• Sustainable and responsible innovation (SRI) The composite scale consisting of 13 items achieved an excellent Cronbach’s alpha of 0.932, signifying outstanding internal consistency. This exceptionally high value indicates outstanding internal consistency and suggests that the scale is highly reliable in capturing the construct of sustainable and responsible innovation outcomes.

4.3. Test of Assumptions

Before running the moderated regression, we checked the standard assumptions. Linearity was evaluated by plotting standardized residuals against standardized predicted values. The points were randomly dispersed with no clear curvature or pattern, indicating that the relationships between predictors and the outcome were adequately linear [109]. Normality of residuals was examined using a histogram and Normal Q–Q plot of the standardized residuals. The histogram was roughly bell-shaped, and the Q–Q points tracked the diagonal closely, supporting approximate normality [110]. Homoscedasticity (constant residual variance) was assessed with the same residuals-versus-predicted scatterplot. Residuals were evenly scattered around zero, with no funneling or systematic structure, consistent with homoscedasticity [111]. Multicollinearity was evaluated using Tolerance and VIF values from the coefficients output, as well as Condition Indices and Variance Proportions from the collinearity diagnostics table. For both models, multicollinearity was assessed using Tolerance and VIF values from the coefficients tables. All Tolerance values > 0.2 (lowest: 0.522 for Model 1 and 0.590 for Model 2) and All VIF values < 2 (highest: 1.917 in Model 1 and 1.694 in Model 2) [110,112]. Condition indices and variance proportions from the collinearity diagnostics: No condition index exceeded the conservative threshold of 30; there were no high variance proportions > 0.90 across multiple variables in the same dimension. These findings indicate that no serious multicollinearity concerns are present in either model [113]. Outliers and influence: Standardized residual statistics show for Model 1, the standardized residual range was approx. −4.14 to +1.79, for Model 2, approx. −3.00 to +1.58. Though a few residuals in Model 1 exceed ±3.0 (a common threshold), they do not cluster or distort the model based on the plots and residual distribution [114]. Hence, there is no clear evidence of influential outliers affecting model estimates. In conclusion, all critical assumptions for moderated multiple regression were tested and met reasonably well in both Model 1 (Innovation Phases) and Model 2 (Innovation Tools). This validates the interpretation of the regression results as robust and reliable.

4.4. Regression Model Summary

A hierarchical multiple regression was conducted to examine the predictive effects of innovation process management (phases and tools) and the moderating role of the innovation ecosystem on sustainable and responsible innovation (SRI). Two separate models were tested and are summarized below. In Model 1 (Table 3), the predictors included Phases in Innovation process management (IP_C), Innovation ecosystem characteristics (IES_C), and their interaction term (IP × IES). The first block (main effects only) was statistically significant, F (2, 97) = 35.66, p < 0.001, and explained 42.4% of the variance in Sustainable and responsible innovation, R = 0.651, R² = 0.424, Adjusted R² = 0.412. The addition of the interaction term in Block 2 increased the explained variance by 1.3% (ΔR² = 0.013), which was not statistically significant, The Block 2 remained significant, F (3, 96) = 24.81, p < 0.001, explaining 43.7% of the variance, R = 0.661, R² = 0.437, Adjusted R² = 0.419. The Durbin–Watson statistic was 2.03 [115], indicating no concern with autocorrelation [116].

Table 3. Model summary and ANOVA for moderated regression with innovation phases (Model 1).

Block	R	R2	Adjusted R2	ΔR2	F	df	p	Durbin–Watson
Block 1	0.651	0.424	0.412	0.424	35.66	(2.97)	<0.001
Block 2	0.661	0.437	0.419	0.013	24.81	(3.96)	<0.001	2.03

Note: Authors’ calculation.

In Model 2 (Table 4), the predictors included Tools in Innovation process management (IT_C), Innovation ecosystem characteristics (IES_C), and their interaction term (ITxIE). The first block (main effects only) was statistically significant, F (2, 97) = 29.92, p < 0.001, and explained 38.2% of the variance in Sustainable and responsible innovation, R = 0.618, R² = 0.382, Adjusted R² = 0.369. When the interaction term was added in Block 2, the explained variance increased by 2.5% (ΔR² = 0.025), and this change was statistically significant. The final model (Block 2) was also significant, F (3, 96) = 21.95, p < 0.001, explaining 40.7% of the variance, R = 0.638, R² = 0.407, Adjusted R² = 0.388. The Durbin–Watson statistic was 2.04, indicating no concern with autocorrelation.

Table 4. Model Summary and ANOVA for moderated regression with innovation phases (Model 2).

Block	R	R2	Adjusted R2	ΔR2	F	df	p	Durbin–Watson
Block 1	0.618	0.382	0.369	0.382	29.92	(2.97)	<0.001	—
Block 2	0.638	0.407	0.388	0.025	21.95	(3.96)	<0.001	2.04

Note: Authors’ calculation.

Together, these findings suggest that both phases and tools in innovation process management, along with the innovation ecosystem, significantly predict sustainable and responsible innovation.

Regression Coefficients

Model 1: Innovation phases

In the first regression model (see Table 5), which examined the role of IPM phases and Innovation ecosystem characteristics on Sustainable and responsible innovation, the results showed that Innovation project management phases were a significant and positive predictor of Sustainable and responsible innovation across both blocks. Specifically, in Block 1, the standardized coefficient for IPM phases was β = 0.556, t = 5.63, p < 0.001, and in Block 2, β = 0.516, t = 5.06, p < 0.001. This indicates that organizations that more rigorously implement structured innovation project phases tend to achieve higher levels of sustainability and responsibility in their innovation outcomes. In contrast, Innovation ecosystem characteristics were not a statistically significant predictor in either block of Model 1, and the interaction term Innovation phases × Innovation ecosystem characteristics in Block 2 also failed to reach significance (p = 0.141), suggesting no moderation effect of the innovation ecosystem on the relationship between innovation phases and Sustainable and responsible innovation.

Table 5. Standardized and unstandardized regression coefficients for moderated regression with innovation phases (Model 1).

Predictor	B	SE B	β	t	p
Block 1
(Constant)	4.159	0.054	—	77.62	<0.001
Innovation Project Management Phases	0.697	0.124	0.556	5.63	<0.001
Innovation Ecosystem Characteristics	0.157	0.113	0.138	1.40	0.166
Block 2
(Constant)	4.189	0.057	—	73.55	<0.001
Innovation Project Management Phases	0.646	0.128	0.516	5.06	<0.001
Innovation Ecosystem Characteristics	0.089	0.121	0.078	0.74	0.463
Innovation Phases × Innovation Ecosystem Characteristics	−0.142	0.096	−0.145	−1.49	0.141

Model 2: Innovation tools

In the second model (see Table 6), which focused on Innovation tools, the results demonstrated that both Innovation tools and Innovation ecosystem characteristics were statistically significant predictors of Sustainable and responsible innovation. In Block 1, Innovation tools had a standardized coefficient of β = 0.434, t = 4.79, p < 0.001, while Innovation ecosystem characteristics also showed significance with β = 0.280, t = 3.10, p = 0.003. When the interaction term Innovation tools × Innovation ecosystem characteristics was added in Block 2, Innovation tools remained significant (β = 0.332, t = 3.25, p = 0.002), as did Innovation ecosystem characteristics (β = 0.276, t = 3.09, p = 0.003). Importantly, the interaction term Innovation tools × Innovation ecosystem characteristics also became a significant negative predictor of sustainable and responsible innovation (β = −0.190, t = −2.02, p = 0.046), indicating a significant moderation effect. This implies that while the use of innovation tools positively contributes to sustainable and responsible innovation, this relationship is slightly weakened in stronger innovation ecosystems, suggesting a potential ceiling effect where tools alone may be less impactful when external ecosystem support is already high.

Table 6. Standardized and unstandardized regression coefficients for moderated regression with Innovation phases (Model 2).

Predictor	B	SE B	β	t	p
Block 1
(Constant)	4.159	0.056	—	74.92	<0.001
Innovation Tools	0.491	0.103	0.434	4.79	<0.001
Innovation Ecosystem Characteristics	0.320	0.103	0.280	3.10	0.003
Block 2
(Constant)	4.185	0.056	—	74.55	<0.001
Innovation Tools	0.376	0.116	0.332	3.25	0.002
Innovation Ecosystem Characteristics	0.315	0.102	0.276	3.09	0.003
Innovation Tools × Innovation Ecosystem Characteristics	−0.147	0.073	−0.190	−2.02	0.046

5. Hypotheses Testing, Interpretation and Discussion

The first research hypothesis predicted that the extent to which companies go through IPM phases would have a significant positive effect on SRI outcomes. This hypothesis was supported. The regression analysis confirmed a strong and statistically significant positive effect, indicating that companies that systematically implement structured phases, such as idea generation, feasibility analysis, development, implementation, and evaluation, achieve better performance in SRI. These results highlight the importance of a process-oriented approach to innovation management, where the structured progression through stages contributes to more thoughtful, inclusive, and impactful innovation outcomes. These findings align with prior research demonstrating that structured innovation processes foster sustainability outcomes. For example, Hallstedt et al. [74] identified several key phase-related elements, such as internal process alignment and formal evaluation steps, that significantly enhance the strategic integration of sustainability perspectives in product innovation [74]. Likewise, Klein et al. [1] found that operationalizing sustainability through strategic orientations and structured innovation workflows positively mediates sustainable business model innovation. Klein et al. [1] model suggests that a commitment to sustainability can lead to innovation which enforces the managers to develop sustainability policies in order to differentiate their firms from their competitors with possible business model innovation. Knowledge about these effects can be valuable to guide firms in dedicating resources to deepen firms’ technological knowledge and in promote a proactive search for technological solutions throughout the firm, not limiting this to R&D. These results are also consistent with meta-analytic evidence confirming a moderate but positive link between eco-innovation practices (including structured project phases) and SMEs’ sustainable performance [117]. That comprehensive review of 99 studies covering 134,841 SMEs demonstrated that eco-innovation positively influences economic, social, and environmental performance, especially when internal processes are supportive of sustainability [117]. Oduro’s [117] findings imply that investing in eco-innovation is worthwhile for SMEs. For SMEs, leadership matters. Executives should set a clear sustainability vision, invest in employee training on environmental practices, build systematic stakeholder engagement, and encourage staff participation in sustainability initiatives. At the same time, the evidence is not uniformly one-sided. Jiang and Tol [118], using an extended Crépon–Duguet–Mairesse (CDM) framework that tackles endogeneity by predicting dependent variables at each stage, find that green innovation does not consistently outperform non-green innovation in raising productivity. In non-heavy-pollution sectors, green patents show limited economic effects; only in high-pollution industries do green innovations deliver productivity gains comparable to conventional innovations [118]. In manufacturing and other high-pollution settings, both forms of innovation appear to enhance productivity to a similar degree. Such findings suggest that while phase-based management is beneficial, its effect on SRI performance may depend on contextual factors such as industry type, regulatory environment, and firm maturity.

The second research hypothesis hypothesized that the extent to which companies use IPM tools would positively affect SRI. This hypothesis was supported. The use of tools such as Design thinking methodology, Lean Startup principles, Business model canvas, SWOT and PEST analyses, as well as Key Performance Indicators (KPIs) showed a significant positive relationship with SRI outcomes. Taken together, the evidence reinforces that internal methodological capabilities, especially tools for planning, monitoring, and strategic alignment, are pivotal for delivering innovations that are economically viable and socially and environmentally responsible. In other words, creativity and ideation matter, but disciplined execution enabled by effective tools is what turns sustainability goals into results. This conclusion echoes a growing literature on the value of systematic, tool-based approaches to sustainable innovation. Oduro [117], for example, shows that sustainability-oriented KPIs, balanced scorecards, and structured monitoring frameworks can measurably improve SMEs’ triple-bottom-line performance, highlighting the practical payoff of these methods [117]. Likewise, research on sustainability-specific SWOT analyses finds that firms use such tools to surface environmental and social opportunities and translate them into concrete sustainable innovations [119]. Gerlach [119] further argues that, when combined with the Sustainable Business Model Canvas and the Impact Canvas, a Sustainability SWOT forms a core toolkit for sustainable innovation and management. Mangelkramer [120], through a systematic literature review, argues that without embedding responsibility (e.g., via tools, dashboards, KPI systems) into process structures such as anticipation, inclusion, and responsiveness, innovations, even sustainable ones, risk becoming “partially-sustainable, ‘irresponsible’ socio-technical system change” [120], (p. 1). This reinforces the idea that tools not only serve technical needs but also operationalize normative commitments across project stages. However, there is a more critical perspective suggesting that governance tools can sometimes lead to procedural lock-in. Lubberink et al. [101] found in their review of sustainable and responsible innovation practices that while tools like stakeholder mapping and responsibility checklists can improve accountability, they can also become tick-box exercises if not reflexively applied. This suggests that the mere presence of tools is insufficient, how they are used, and whether they foster reflective and adaptive behavior, is equally important. Additionally, Adubofour et al. [121] conducted a large-scale panel study which found that, in some high-maturity industries, intensive use of innovation monitoring tools (e.g., dashboards, KPI systems) did not necessarily lead to better sustainability outcomes and, in certain cases, higher levels of sustainable innovation had no significant link to environmental performance [121]. These findings suggest that while tools are valuable, their effectiveness depends heavily on contextual integration, long-term flexibility, and alignment with broader ethical and social values.

The third research hypothesis proposed that Innovation ecosystem characteristics would moderate the relationship between IPM phases and SRI, weakening the relationship in highly supportive ecosystems. However, this hypothesis was not supported, as the interaction term was not statistically significant. This suggests that the influence of structured innovation phases remains robust across different levels of ecosystem support. This finding coheres with the view that well-defined internal innovation processes retain their importance irrespective of external ecosystem strength. In their qualitative case studies, Boyer [122] and Gu et al. [123] emphasized that even in mature, sustainability-oriented innovation ecosystems, firms that apply structured internal phases, such as anticipation, design, prototyping, and evaluation, consistently outperform peers, highlighting the intrinsic value of disciplined project management in sustainable innovation [122,123]. Some scholars argue that ecosystem support can amplify, or even substitute for, internal innovation processes. Zheng and Cai [88], for example, show that well-designed policy frameworks and external collaborations can markedly improve innovative outcomes, pointing to interaction effects between ecosystem maturity and firm practices [88]. In our results, however, the lack of a significant moderation effect suggests that structured internal phases function as a foundational capability: a necessary condition for SRI that holds regardless of the level of external support. As such, our findings contribute to the literature by reinforcing the primacy of internal IPM structures even when ecosystem backing is strong, such structured processes remain essential and cannot be fully outsourced or replaced by external support.

The fourth research hypothesis proposed that innovation-ecosystem characteristics would moderate the link between innovation tools and SRI, specifically, that strong ecosystem support would weaken the tools’ positive effect. The results support this claim. Our regression shows a significant negative interaction; innovation tools are helpful on average, but their incremental benefit shrinks in ecosystems already rich in support. The likely reason is overlap and redundancy. Mature ecosystems supply funding, policy incentives, talent pipelines, collaboration networks, and institutional infrastructure, lowering firms’ reliance on formal internal tools. As a result, the marginal contribution of methods such as design thinking, Lean Startup, the Business Model Canvas, SWOT analyses, and KPI systems to SRI becomes less pronounced in highly developed contexts. This pattern aligns with the resource-substitution perspective; when one resource is abundant, dependence on another tends to fall [20]. Consistent with this view, Misra and Wilson [124] find that actors in mature ecosystems lean less on digital decision tools and more on shared knowledge flows and human networks; similarly, research on green-innovation alignment shows that well-matched external partners can stand in for internal capabilities, dampening the added value of internal tools when external support is high [22]. Practically, these findings argue for context-sensitive strategies. In supportive ecosystems, firms may achieve greater sustainable-innovation impact by prioritizing external collaboration, absorptive capacity, and network-based learning rather than further formalizing internal toolkits. In short, ecosystem strength should guide the balance between investing in internal tools and leveraging external resources.

These findings add to a growing body of evidence showing that innovation performance results from the interplay between firm-specific capabilities and broader ecosystem enablers [21,33,84]. Our results refine this understanding by revealing that, in transitional economies, ecosystem maturity does not always enhance innovation outcomes. Instead, it can sometimes produce a substitution effect, where strong external support reduces the added value of internal tools. This dynamic underscores the need for context-sensitive innovation strategies; firms must evaluate not only the strength of their internal capacities but also how these capabilities interact with the maturity of the surrounding ecosystem. Ultimately, achieving sustainable and responsible innovation requires a careful balance between internal structure and external collaboration, as both dimensions must develop hand in hand as ecosystems evolve.

6. Conclusions and Implications

This research paper sought to explore and empirically assess how structured IPM practices and ecosystem-level support mechanisms jointly contribute to the advancement of sustainable and responsible innovation (SRI) among firms in EU widening countries. Grounded in the theoretical frameworks of the resource-based view (RBV) and innovation ecosystem theory, the study illuminates both internal organizational dynamics and the external environmental conditions that facilitate or constrain sustainable innovation. The findings offer critical insights into how micro, small, and medium-sized enterprises (MSMEs) in transitional economies can harness innovation capabilities to address pressing environmental, social, and economic challenges, despite limited R&D capacity and systemic ecosystem weaknesses. To explore these dynamics, we have developed a conceptual model supported by a structured survey instrument tested through two-stage Delphi study targeting business representatives across various sectors from 100 companies from four Southeast European developing transitional countries. The preliminary conceptual framework is designed to be empirically tested through multiple regression analysis, examining both main and moderating effects. The results of our research offer several implications.

6.1. Theoretical Contributions

This study advances theory by combining the Resource-Based View (RBV) with innovation-ecosystem perspectives to explain Sustainable and Responsible Innovation (SRI) in the under-examined context of EU widening countries. First, it advances the integration of the Resource-Based View (RBV) and Innovation Ecosystem (IE) perspectives by empirically demonstrating how internal resources (project management phases and tools) interact with external ecosystem maturity to shape Sustainable and Responsible Innovation (SRI) outcomes. While prior RBV-based studies have mainly focused on internal dynamic capabilities [29,41], our findings demonstrate that these capabilities do not operate in isolation but are contingent on the quality and maturity of the surrounding ecosystem. Second, by uncovering a negative moderation effect between ecosystem support and the impact of innovation tools, the study extends resource-interaction theory, revealing a resource-substitution mechanism rarely tested empirically in the context of emerging and transitional economies. Third, the results contribute to the contextualization of innovation theory [125] by situating the RBV–IE interplay within EU-widening countries, where institutional voids and resource scarcity alter the conventional dynamics observed in mature economies. Finally, the paper conceptualizes structured innovation phases as foundational dynamic capabilities that retain significance across ecosystem conditions, thereby bridging micro-level projects.

Collectively, these contributions enrich the theoretical understanding of how internal and external resources co-evolve, offering a more integrative lens for studying sustainable innovation in varying institutional contexts.

6.2. Policy and Managerial Implications

From a policy standpoint, the results point to the centrality of ecosystem-building in transitional economies. Governments and regional authorities in EU widening countries should foster supportive environments for innovation by combining targeted finance, smart regulatory reforms, infrastructure investments, and stronger academia-to-business (A2B) links. At the same time, there is a need for programs that build firms’ internal innovation capabilities, especially where external support is thin or still emerging. Policymakers should treat structured innovation processes as a cornerstone of sustainable entrepreneurship. Capacity-building initiatives that train entrepreneurs and managers in design thinking, project planning, and agile methods can materially improve the sustainability performance of local businesses. Expanding access to practical innovation tools, particularly for SMEs, can also help close the gap between high-level sustainability ambitions and day-to-day execution. Institutional intermediaries, innovation hubs, accelerators, and university-based incubators, are critical to this agenda. They can both strengthen internal capabilities through training and mentoring and weave the systemic connections that make the broader ecosystem more effective. Prioritizing the development and resourcing of these intermediaries should be a core element of national and regional innovation strategies.

For business leaders and innovation managers, several practical lessons stand out. Treat innovation as a managed system rather than a one-off or linear effort. Structure work into clear phases with defined decision points, meaningful stakeholder input, and room for iteration. This builds organizational discipline and improves both predictability and impact. Use tools with sensitivity to context. Design thinking, agile methods, canvases, and KPI systems can be highly effective, but their payoff depends on the surrounding ecosystem. In mature settings, channel more effort into external partnerships and shared knowledge flows instead of endlessly expanding internal toolkits. In weaker ecosystems, including parts of the Western Balkans, robust internal tools can serve as substitutes for missing external support. Managers should assess the maturity of their local ecosystem and tune their approach accordingly, choosing methods for strategic fit, not just intrinsic appeal, with local institutions, networks, and infrastructure. Finally, treat sustainability and responsibility as core performance dimensions, not peripheral compliance. Firms that weave social and environmental considerations into their innovation strategies are better positioned to create durable value and sustain legitimacy with stakeholders at home and abroad.

6.3. Limitations and Suggestions for Future Research

This study offers useful early insights, but several limitations should be noted. Firstly, its cross-sectional design limits causal claims; longitudinal designs would better capture how innovation practices and SRI performance evolve over time. Secondly, the measures rely on self-reports from business representatives, which may introduce social desirability and perception biases. Future work should triangulate these views with objective indicators (e.g., audited performance metrics). Third, because participation was voluntary and recruitment relied on professional networks, coverage and self-selection bias are possible [99]. To assess selection on observables, we added firm size, sector, and country as covariates in GLM specifications for both the Innovation phases and Innovation tools models. These controls were not jointly associated with the outcome and effects were small, while the focal coefficients remained stable (including a significant Innovation tools × Innovation ecosystem characteristics interaction). This indicates that our findings are not driven by observable composition differences; nonetheless, bias from unmeasured factors may remain. Fourthly, the moderation by ecosystem characteristics was only partly supported, suggesting value in a more fine-grained look at specific ecosystem elements, such as particular institutions, incentive types, and collaboration mechanisms. Finally, the analysis focuses on EU widening countries in the Western Balkans, which constrains generalizability. To build on these findings, comparative studies with developed EU regions and other emerging economies would help validate and contextualize results. Future research might also examine the roles of digital transformation, social-innovation practices, and cross-border collaboration as additional drivers, or contingencies, of sustainable innovation outcomes.