The aim of this paper is to examine how digitalization capabilities related to technological, individual, and managerial dimensions influence organizational ambidexterity, specifically the balance between exploration and exploitation activities.
An empirical survey was conducted among 370 manufacturing and service firms operating in Greece. Initially, Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were applied. Finally, the structural relationships among the latent factors were determined through Structural Equation Modelling (SEM).
The findings indicate that digitalization capabilities are positively associated with both exploitation- and exploration-oriented activities, with a stronger effect on exploitation, thereby underscoring their critical role in fostering sustainable organizational performance.
Unlike prior studies that primarily focus on broader industrial contexts, this research provides a regionally specific yet widely applicable perspective by examining the Greek manufacturing and service sectors. By doing so, we offer novel insights into the mechanisms through which I4.0 capabilities drive organizational ambidexterity, thereby bridging theoretical gaps and offering practical implications for firms navigating digital transformation.
This study is limited by its focus on Greek firms, which may restrict the generalizability of the findings to other economic, cultural, or industrial contexts. The use of self-reported survey data introduces potential response bias and limits causal inference. Future research should employ longitudinal or cross-country designs and include objective performance indicators to validate and extend the model across diverse organizational settings.
1. Introduction
The global industrial landscape is experiencing profound shifts fueled by the ongoing wave of digital transformation and technological progress (Martínez-Román, 2025). Digital technologies have the potential to improve coordination and efficiency across organizational functions, particularly in operational contexts (Hoque et al., 2025). The technological framework of Industry 4.0 (I4.0) includes technologies that collectively drive the transition toward a highly interconnected and intelligent industrial ecosystem, enabling organizations to achieve unprecedented levels of efficiency, customization, and agility (Tortorella et al., 2020; Kurt, 2019). However, the realization of these benefits depends on the development of complementary organizational capabilities, including managerial practices and workforce competencies.
The contemporary business environment requires firms to balance two competing yet complementary goals: exploration (pursuing innovation and new opportunities) and exploitation (optimizing existing operations) (Annamalah et al., 2025). This dual focus, known as ambidexterity, is increasingly recognized as a critical determinant of organizational success in volatile and technology-driven markets. Exploration and exploitation differ fundamentally in their objectives and resource demands, often competing for the same limited organizational resources and managerial attention. This inherent tension makes balancing these activities a significant organizational challenge (Lindskog and Magnusson, 2021).
Although prior research acknowledges that digital technologies can support both efficiency and innovation, empirical evidence remains limited regarding how specific I4.0 capabilities enable firms to simultaneously pursue exploration and exploitation. More specifically, the literature reveals several unresolved issues. First, much of the literature adopts a technology-centric perspective, implicitly assuming that digital technologies alone drive ambidextrous outcomes, while underestimating the role of individual and managerial capabilities (Belhadi et al., 2021; Rialti et al., 2019). Second, limited research has examined the mechanisms through which these capabilities translate into exploration and exploitation at the organizational level, particularly in terms of whether firms can pursue these activities simultaneously or sequentially during digital transformation. Third, empirical evidence remains concentrated on large firms and advanced economies, leaving a gap in understanding how I4.0-driven ambidexterity develops in Small- and Medium-sized Enterprises (SMEs) and under-researched national contexts such as Southern European economies (Jing et al., 2023). Finally, research has primarily focused on short-term operational outcomes, leaving longer-term effects such as resilience, sustainability, and innovation quality underexplored (Aslam et al., 2024; Gomez et al., 2022).
Based on these gaps, the present study challenges the implicit assumption that digital transformation follows a linear pathway from efficiency to innovation, particularly in resource-constrained contexts. It examines whether exploration and exploitation can be simultaneously supported by digital capabilities even in SMEs operating under economic uncertainty. Although the term “I4.0” is commonly used to describe advanced cyber–physical production systems, the firms in our sample primarily represent SMEs engaged in incremental and resource-constrained digitalization efforts. Moreover, the study questions the assumption that technological capabilities alone drive ambidexterity by explicitly incorporating individual and managerial capabilities as co-determinants of exploratory and exploitative outcomes. To isolate these relationships, the analysis controls for key contextual and organizational factors, including firm size, sector (manufacturing versus services), and national context. By holding these factors constant within a single-country setting, the study reduces contextual variability and enables a more focused examination of how distinct I4.0 capability dimensions relate to exploration and exploitation. This approach strengthens internal consistency while acknowledging that external validity remains bounded by the study context.
This study contributes to the literature on I4.0 and organizational ambidexterity in several ways. First, it provides empirical evidence on how I4.0 capabilities influence the development of organizational ambidexterity in the context of SMEs operating in an emerging digital transformation environment such as Greece. While prior research has widely acknowledged the role of digital technologies in enhancing organizational performance, empirical evidence on how specific I4.0 capabilities translate into ambidextrous organizational behavior remains limited, particularly in SME contexts.
Second, the study advances ambidexterity theory by proposing a sequential interpretation of the relationships between I4.0 capabilities, exploitation, and exploration. The findings indicate that I4.0 capabilities are more strongly associated with exploitation-oriented activities, such as efficiency improvements and process optimization, which are, in turn, also positively related to exploration. While the empirical model examines only direct relationships, these results are consistent with a potential sequential pathway, whereby exploitation may provide a foundation for subsequent exploration. However, this sequence is not explicitly tested and should be understood as a theoretical proposition. Third, the study provides practical insights for managers by demonstrating that digitalization investments do not only improve operational efficiency but may also gradually support innovation and adaptability, thereby enabling organizations to build ambidextrous capabilities that are essential for long-term competitiveness and addressing broader challenges associated with Industry 5.0.
The structure of this article is as follows: Section 2 provides an overview of the theoretical foundation and outlines the hypotheses. Section 3 explains the methodology used in the study. Section 4 presents the research model, which is developed based on the theoretical framework and validated through a large-scale empirical analysis. Section 5 discusses the findings in detail. Finally, Section 6 concludes the study by summarizing the key insights, highlighting its limitations, and proposing directions for future research.
2. Theoretical background and hypotheses formulation
2.1 Industry 4.0
I4.0 is frequently used in the literature to describe advanced digital technologies in industrial and operational settings, in which digital technologies enhance efficiency and competitiveness by seamlessly connecting all resources – data, people, and machinery – across the value chain (Sharma et al., 2024). Over time, I4.0 has captured the attention of policymakers, business leaders, and researchers, primarily due to its numerous advantages (Dalenogare et al., 2018). The integration of human involvement, production systems, and physical assets with smart machines and processes plays a vital role in creating a highly efficient industrial value chain (Schumacher et al., 2016).
While much of the existing literature frames digital transformation through the lens of I4.0, recent discourse has increasingly emphasized the principles of Industry 5.0, including resilience, sustainability, and human-centricity (Hammad et al., 2026). Rather than replacing I4.0, Industry 5.0 builds upon its efficiency-oriented foundations by emphasizing the strategic use of digital technologies to enhance organizational adaptability and robustness in the face of uncertainty.
From this perspective, efficiency should not be interpreted as an end in itself, but as a foundational capability that enables firms to reallocate resources, absorb disruptions, and support resilient organizational responses. In digitally-enabled SMEs, improvements in process efficiency and operational reliability create the necessary slack and informational transparency required to support resilience-oriented outcomes, such as rapid reconfiguration of operations, decentralized decision-making, and enhanced employee involvement. Accordingly, this study positions efficiency-enhancing digitalization not as contradictory to industry 5.0 principles, but as a prerequisite for their realization in SMEs.
As digital transformation progresses, the industry is poised to witness further disruptions and innovations in the coming years. I4.0 signifies the fusion of physical and digital systems within manufacturing and production, offering a novel approach to managing processes through real-time synchronization and product customization. This transformation is driven by digital technologies that enable seamless integration and coordination across disciplines and functions, thereby enhancing efficiency and reducing costs (Singh et al., 2026). As I4.0 progresses, it is anticipated to transform industries and economies, driving greater efficiency, productivity, and innovation not only in manufacturing but across various sectors. I4.0 technologies promise to revolutionize production methods, paving the way for a new era of scientific and technological advancements. They enable companies to analyze real-time data, allowing them to swiftly adjust to production needs and align production schedules with order demands. Prior research commonly conceptualizes digitalization initiatives across technological, individual, and managerial dimensions. These capabilities empower organizations to innovate their processes and effectively navigate the fast-changing industrial environment (Belhadi et al., 2021).
Technological Capabilities: According to Attaran and Attaran (2020) I4.0 leverages advanced technologies – including the Internet of Things (IoT), artificial intelligence (AI), big data analytics, and digital twins – to integrate and connect the physical and digital realms. I4.0 harnesses state-of-the-art technologies such as the IoT, AI, big data analytics, and digital twins to bridge the gap between the physical and digital worlds. These innovations enable breakthroughs like autonomous robots and intelligent systems. IoT devices gather real-time data, optimizing processes such as supply chain management and predictive maintenance to boost efficiency and minimize downtime (Mudunuru et al., 2024).
Individual Capabilities: A pivotal aspect of I4.0 is empowering the workforce to adapt and excel in this digital era. Upskilling and reskilling employees to navigate emerging technologies are essential (Li, 2024). I4.0 requires competencies in data analysis, decision-making driven by analytics, and operating sophisticated machinery. Training initiatives, that leverage augmented reality (AR), equip workers with real-time, step-by-step guidance, enhancing safety and operational efficiency (Méndez and Velázquez, 2024). Furthermore, Kamble et al. (2020) highlighted that the integration of Information and Communication Technology into I4.0 workplaces offers considerable potential to enhance sustainability, particularly in the social dimension (Nwaobia and Akintoye, 2024). Key benefits include improved collaboration and engagement among employees, opportunities for personal and professional growth, more effective learning processes, and enhanced well-being and work-life balance.
Managerial Capabilities: Leadership in I4.0 requires aligning technological innovations with organizational objectives through strategic planning (Hidayat and Basuil, 2024). This includes reshaping business models to capitalize on digital technologies and driving a cultural transition toward agility and innovation. Managers must focus on allocating resources effectively for digital transformation, addressing challenges like resistance to change, integration complexities, and standardization hurdles. Some researchers, such as Luthra et al. (2020), note that the managerial capabilities of I4.0 can help leaders adopt modern technological processes to drive sustainable improvements across value chains from both a managerial and organizational perspective. For example, Lean Supply Chain Management (LSCM) and digital lean combine both lean principles with I4.0 technologies (e.g. IoT-driven real-time monitoring) which enhances efficiency and reduces waste (Núñez-Merino et al., 2020). Additionally, Belhadi et al. (2021) used the ISM (Interpretive Structural Modeling) technique to emphasize that the human capabilities within I4.0 serve as the foundation for developing sustainable strategies in organizations.
2.2 Organizational exploitation and exploration
Talaja et al. (2023) described ambidexterity as an organization's capability to maintain alignment and operational efficiency in addressing current business needs, while also being flexible and responsive to environmental changes. This concept highlights a company's ability to develop both exploration and exploitation capabilities (Knight and Harvey, 2015), in other words, organizations must stay aligned with ongoing activities and operate efficiently to meet current demands while simultaneously adapting to and preparing for future changes (Mathias et al., 2018). Previous research defines organizational ambidexterity as a firm's capacity to engage in and harmonize both exploratory and exploitative innovation at the same time (Luger et al., 2018). This capability enables organizations to overcome the stagnation caused by an overemphasis on exploitation while also capitalizing on the benefits of accelerating exploration efforts (Miller and Le Breton-Miller, 2006). Organizations need to excel in current markets by leveraging their skills to achieve short-term gains while identifying opportunities that enable adaptation to emerging markets for long-term success. Exploitation is essential for performing effectively in established markets and technologies, whereas exploration is key to navigating and succeeding in new markets and technological landscapes. Ambidextrous organizations possess the ability to manage both incremental (exploitative) and radical (exploratory) changes simultaneously (Martini et al., 2013). Exploration focuses on driving organizational growth by pursuing novel and alternative approaches and involves a wide search for, and assimilation of, new knowledge and technologies to enhance innovation and radical development of new solutions. Conversely, exploitation focuses on optimizing and extending existing skills and capabilities. It allows organizations to apply their current knowledge effectively through processes such as selection, implementation, production, and execution. Exploitation often involves localized searches for familiar knowledge and technologies, aiming to deepen the current knowledge base and drive continuous improvement and incremental advancements in existing solutions (Lee et al., 2019). Ambidexterity is conceptualized as the balance between exploration and exploitation (Luger et al., 2018). Pursuing both simultaneously is not only possible but also advantageous for enhancing organizational performance (Annamalah et al., 2025).
2.3 Research hypotheses
In the current competitive landscape, organizations operate within a complex environment shaped by continuous technological advancements (Yuksel, 2022). The shift towards I4.0 is primarily driven by the need to enhance competitiveness by leveraging productivity improvements and achieving more efficient management of supply chain functions and processes. While these factors are critical for success, achieving an environmentally sustainable competitive edge requires companies to effectively utilize their existing resources while simultaneously exploring innovative approaches to create value (Tedaldi et al., 2021).
Digitalization capabilities are expected to influence both exploitation- and exploration-oriented activities, although the strength and nature of these relationships may vary across contexts (Janssen et al., 2017; McAfee et al., 2012). Researchers (Talaja et al., 2023) have identified various exploratory and exploitative activities driven by the value creation potential of I4.0 capabilities. Halse and Ullern (2017) emphasize the importance of openness to external partner networks alongside organizational ambidexterity for a successful I4.0 transformation, while Gastaldi et al. (2018) examine how digital technologies can support organizations in managing the exploration–exploitation paradox over time. Research has also explored the roles of human ambidexterity and ambidextrous business process management, as well as the interplay between ambidexterity and organizational agility (Rialti et al., 2019).
However, despite this growing body of research, the extent to which I4.0 capabilities consistently support both exploration and exploitation remains debated. Szalavetz (2019) highlights the dual effects of I4.0 capabilities on organizational processes, noting their ability to simultaneously boost productive capacity and enhance research and development capabilities, yet also cautions that these outcomes may depend on organizational readiness and contextual factors. While I4.0 capabilities can support exploratory activities by uncovering new opportunities, attracting new customers, and analyzing demand patterns (Xu et al., 2018), they may also reinforce existing routines and efficiency-oriented behaviors that favor exploitation (Al-Khatib, 2023). This raises questions about whether I4.0 technologies inherently promote ambidexterity or whether trade-offs emerge in practice.
Moreover, it is widely recognized that smart technologies help companies improve efficiency within production sites and operations, thereby enhancing the exploitation of assets (Xu and Duan, 2019). For example, IoT platforms enable real-time monitoring and predictive maintenance, reducing downtime and extending equipment lifespan (Suthar et al., 2024). Drawing on Dynamic Capabilities Theory, I4.0 technologies such as AI, big data analytics, and cyber-physical systems can enhance firms' ability to sense and seize new opportunities, fostering exploration through innovation and adaptive decision-making (Lichtenthaler, 2020). At the same time, technologies such as automation, IoT, and digital twins may strengthen exploitation by optimizing existing resources and processes (Frank et al., 2019).
Nevertheless, empirical findings remain mixed, particularly regarding whether these technologies enable firms to pursue exploration and exploitation simultaneously or whether organizational constraints limit their ambidextrous potential. While some studies report higher levels of ambidexterity among firms leveraging I4.0 technologies (Büchi et al., 2020), others emphasize the mediating role of organizational factors such as leadership, digital culture, and absorptive capacity (Mathias et al., 2018). This suggests that the impact of I4.0 capabilities on exploration and exploitation cannot be assumed a priori and warrants further empirical investigation.
Based on these competing perspectives, the following hypotheses are developed.
I4.0 capabilities are expected to positively influence exploitation
I4.0 capabilities are expected to positively influence exploration
To effectively embrace ambidexterity, companies must balance exploration and exploitation (Luger et al., 2018). As March (1991) argues, organizations must engage in sufficient exploitation to maintain current viability while dedicating adequate effort to exploration to ensure future viability. However, scholars disagree on how these two activities interact. Hughes (2018) suggests that excessive emphasis on exploitation may enhance short-term efficiency but undermine long-term survival by discouraging innovation. Over time, this may lead to a “capacity trap,” characterized by overreliance on existing knowledge and routines.
Exploitation often consumes substantial organizational resources, such as time, capital, and human effort (Clauss et al., 2019). Given resource constraints, a strong focus on exploitation may crowd out exploratory initiatives, reinforcing short-term performance at the expense of long-term adaptability (March, 1991). This dynamic may reduce risk-taking and increase organizational inertia, limiting the willingness to explore new opportunities (Pietsch et al., 2023).
Conversely, other studies argue that exploitation can enable exploration. Piao and Zajac (2016) note that successful exploitation generates financial and knowledge resources that can be reinvested in exploratory activities. Similarly, He and Wong (2004) suggest that learning from exploitative activities may guide and improve exploratory efforts, supporting a complementary relationship between the two. At the same time, an excessive focus on exploration without adequate exploitation may lead to inefficiencies and weakened competitiveness (Mollenkopf et al., 2011).
Taken together, the literature presents conflicting views on whether exploitation constrains or supports exploration. This unresolved tension highlights the need for empirical examination of their relationship, particularly in I4.0-driven contexts. Therefore, the following hypothesis is proposed.
Exploitation is expected to positively influence exploration
2.4 The conceptual framework
Based on the above, a research framework is illustrated in Figure 1, presenting the hypotheses derived from the literature review. This framework explores how the three distinct dimensions of I4.0 - technological, individual and managerial capabilities – impact the two dimensions of organizational ambidexterity, exploitation and exploration. The overall structure of the model can also be conceptually interpreted through the lens of sequential ambidexterity. Specifically, the proposed relationships suggest a potential ordering in which I4.0 capabilities enhance exploitation, which in turn is associated with exploration. Nevertheless, it is important to emphasize that the present study does not explicitly model or test such sequential relationships. Therefore, sequential ambidexterity is introduced here as a theoretical proposition that is consistent with the hypothesized links, rather than as an empirically validated mechanism. The structural model is intentionally parsimonious, focusing specifically on the relationships between I4.0 capabilities and the two dimensions of organizational ambidexterity – exploration and exploitation. This approach allows the study to isolate the role of digital capabilities in shaping ambidextrous behavior within SMEs.
3. Research methodology
3.1 Sample and questionnaire development
Given the study's aim to examine theoretically grounded relationships between I4.0 capabilities and organizational ambidexterity across a broad sample of firms, a quantitative survey-based approach was considered most appropriate. The use of EFA and CFA allowed for rigorous validation of latent constructs, while SEM enables simultaneous examination of multiple interrelated relationships and indirect effects.
The proposed model and hypotheses were tested using a questionnaire survey developed through a comprehensive literature review and interviews with industry experts. Prior to distribution, the survey underwent a pretest with 15 participants from Greek firms, followed by personal interviews with managers. The experts involved in the questionnaire development were purposively selected to ensure sectoral and geographic diversity within the Greek context. They represented both manufacturing and service sectors and were drawn from firms operating in different regions of the country, capturing variation in firm size, digital maturity, and I4.0 adoption. This diversity helped ensure that the measurement items were not context-specific to a single sector or locality, thereby strengthening the robustness of the instrument and supporting cautious generalization to comparable economic and institutional settings (Hair et al., 2022).
Based on their feedback, minor revisions were made to the questionnaire. The final survey instrument consisted of two pages and included a total of 22 questions. The data for the empirical analysis were sourced from the ICAP database, the largest business information and consulting firm in Greece. The database provided a list of all companies operating within each prefecture. A web-based questionnaire was created as the primary data collection tool and distributed to 1,100 randomly selected manufacturing and service firms in Greece. The questionnaires were sent via email, each accompanied by a cover letter outlining the survey's purpose and assuring confidentiality and privacy. Respondents were asked to answer the questions based on their experiences and the most recent project they completed that involved quality issues. This method allowed the collection of firsthand, up-to-date information about the specific challenges and quality concerns encountered in the respondents' projects.
A seven-point Likert scale was employed to assess all questionnaire items, with respondents indicating their level of agreement from “1” (strongly disagree) to “7” (strongly agree). The survey was administered over a six-month period, from March to August 2024. Data were collected in two waves: the first yielded 207 responses, and the second 163, resulting in a total of 370 usable responses. The sample characteristics are presented in Table 1.
Sample characteristics
| Demographic characteristics of sample | Number | Percent |
|---|---|---|
| Firm size (number of employees) | ||
| 11–49 | 246 | 66 |
| 50–250 | 81 | 22 |
| 251–500 | 43 | 12 |
| Agricultural sector | ||
| Manufacturing | 260 | 70 |
| Services | 110 | 30 |
| Demographic characteristics of respondents | ||
| Male | 231 | 62 |
| Female | 139 | 38 |
| Education | ||
| High school | 32 | 9 |
| University | 178 | 48 |
| Msc/PhD | 160 | 43 |
| Job position | ||
| Senior executive | 147 | 40 |
| Manager | 223 | 60 |
| Experience (years) | ||
| 5> | 110 | 30 |
| 5–10 | 86 | 23 |
| 10< | 174 | 47 |
The questionnaire was designed to be simple, relevant, and well-structured, with a clear scale and logically ordered questions. It was divided into three sections: the first section collected demographic information about the respondents, the second section included a set of thirteen variables measuring I4.0 capabilities (Belhadi et al., 2021; Kamble et al., 2020) and the third section included nine variables assessing organizational ambidexterity, split into exploration and exploitation.
Because the data were collected using a single survey instrument, the potential for common method bias (CMB) was evaluated. Harman's single-factor test was conducted through exploratory factor analysis. The results indicated that the first factor accounted for less than 40% of the total variance, suggesting that CMB is unlikely to significantly affect the findings. Additionally, procedural remedies were implemented during the survey design stage, including ensuring respondent anonymity and separating measurement items, in order to reduce the risk of response bias.
3.2 Analyses
Data analysis was performed using SPSS version 29.0 and AMOS 2.0 software. A correlation matrix among technological, individual and managerial capabilities, as well as exploration and exploitation, was generated in order to examine the relationships among factors. Table 2 presents descriptive statistics and correlations among the study variables.
Descriptive statistics
| Variables | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| 1. Technological capabilities | – | ||||
| 2. Individual capabilities | 0.545 | – | |||
| 3. Managerial capabilities | 0.706 | 0.506 | – | ||
| 4. Exploration | 0.647 | 0.625 | 0.605 | – | |
| 5. Exploitation | 0.457 | 0.487 | 0.419 | 0.719 | – |
| Mean | 4.37 | 3.91 | 4.95 | 5.18 | 5.76 |
| S.D | 1.36 | 1.47 | 1.13 | 1.14 | 0.86 |
| Cronbach's alpha | 0.868 | 0.912 | 0.901 | 0.885 | 0.869 |
Using the measured variables representing I4.0 dimensions, exploration and exploitation, the latent constructs were aggregated into unified dimensions. The analysis began with Exploratory Factor Analysis (EFA), employing the principal component extraction method paired with varimax orthogonal rotation to uncover the underlying patterns among variables.
Next, CFA was utilized to refine the scales identified during the EFA and confirm that the latent constructs and the measured variables (indicators) align with theoretical expectations. Additional evaluations for multicollinearity, unidimensionality, reliability, and construct validity are conducted in line with the methodologies outlined by Hair et al. (2022). The model and hypotheses are subsequently tested using SEM with path analysis, a multivariate approach that examines relationships between variables (Fynes and Voss, 2001).
4. Results
4.1 Measurement unidimensionality, reliability and validity
Although all measures had been previously validated in the literature, an EFA was performed to uncover the underlying constructs. The EFA process identified five latent constructs. Sampling adequacy was assessed using the KMO test, which yielded a value of 0.923, and Bartlett's test of sphericity, with a result of 6079.807, confirmed the suitability of the data for factor analysis. All factor loadings exceeded 0.60, indicating strong item reliability that at least 50% of the variance in each item was explained by the associated latent construct. To determine unidimensionality and whether all the latent factors show acceptable fit to the empirical data, CFA was applied. The results of CFA provided a good fit to the data (see Table 3).
The fit indices of the measurement and structural model
| Fit indices | Measurement Model (CFA) | Structural Model | Levels of acceptance* |
|---|---|---|---|
| Absolute fit indices | |||
| Chi-square (χ2) | 361.011 | 379.462 | 0 ≤ χ2 ≤ 2df |
| Degrees of freedom (df) | 176 | 180 | >0 |
| Root mean square residual (RMR) | 0.068 | 0.064 | <0.08 |
| Root mean square of approx. (RMSEA) | 0.053 | 0.051 | <0.08 |
| Incremental fit indices | |||
| Incremental fit index (IFI) | 0.969 | 0.967 | >0.90 |
| Tucker-Lewis coefficient (TLI) | 0.963 | 0.961 | >0.90 |
| Comparative fit index (CFI) | 0.969 | 0.967 | >0.90 |
| Parsimonious fit indices | |||
| Chi-square/degrees of freedom (χ2/df) | 2.051 | 2.108 | <3.0 |
| Parsimonious normed fit index (PNFI) | 0.789 | 0.805 | >0.50 |
| Goodness of fit index (GFI) | 0.916 | 0.913 | >0.50 |
To evaluate reliability, both Cronbach's alpha and composite reliability (CR) were calculated. CR, being a more precise measure that accounts for unequal item weighting, showed values above the recommended threshold of 0.7 for all latent constructs. Both indices met the acceptable standard, confirming the reliability of the scales (see Table 4).
Evaluation of measurement model
| Constructs | Items | Factor loading | AVE | CR | Corr2 |
|---|---|---|---|---|---|
| Technological capabilities Managerial capabilities Individual capabilities | Cloud computing | 0.809 | 0.711 0.756 0.689 | 0.950 0.954 0.940 | 0.498 0.366 0.390 |
| Big data analytics | 0.874 | ||||
| Internet of Things | 0.871 | ||||
| Additive manufacturing | 0.819 | ||||
| Enforce adequate plans for the utilization of I4.0 technologies | 0.629 | ||||
| Monitor the performance of the I4.0 function | 0.922 | ||||
| Adjust I4.0 plans to better adapt to changing conditions | 0.948 | ||||
| Create appropriate conditions for the implementation or application of robotic systems (RS) | 0.938 | ||||
| Programming and data management | 0.856 | ||||
| Understanding of technological trends | 0.910 | ||||
| Ability to learn new technologies | 0.819 | ||||
| Decision support systems | 0.725 | ||||
| Exploration Exploitation | Thinking “outside the box” | 0.882 | 0.648 0.586 | 0.928 0.924 | 0.516 0.516 |
| Explore new technologies | 0.880 | ||||
| Products or services that are innovative to the firm | 0.718 | ||||
| Ventures into new market segments | 0.724 | ||||
| Improve quality and lower cost | 0.712 | ||||
| Improve the reliability of its products and services | 0.844 | ||||
| Increase the levels of efficiency in its operations | 0.742 | ||||
| Constantly survey existing customers' satisfaction | 0.708 | ||||
| Fine-tunes what it offers to keep its current customers satisfied | 0.812 |
The validity of the scales was examined through content, convergent, and discriminant validity. Content validity was ensured through a comprehensive literature review and the results of a pilot study, confirming the instrument's appropriateness. Convergent validity was evaluated by analyzing the factor loadings of items on their respective latent constructs. All factor loadings exceeded the recommended threshold of 0.5, demonstrating that the items effectively measured their intended constructs. Additionally, each indicator's correlation with its latent construct (the loading) was higher than its correlation with other constructs (cross-loadings).
The Average Variance Extracted (AVE) values, presented in Table 4, all exceeded 0.50 (Fornell and Larcker, 1981) and were statistically significant at p < 0.001, supporting the claim that the model's constructs possess acceptable convergent validity. Discriminant validity was assessed by examining the contribution of items to their theoretical constructs. To confirm discriminant validity, CR, AVE, and the square root of AVE were calculated. For a measurement model to demonstrate strong discriminant validity, the maximum squared value of the correlation coefficient should be less than the minimum AVE value. In this study, all AVE values exceeded the squared correlations, confirming that the constructs are distinct from one another.
4.2 Hypotheses testing
After evaluating the goodness of fit of the measurement model, the research hypotheses (H1-H3) were tested using SEM to estimate the causal relationships among the latent constructs. This approach was used to test the study's proposed hypotheses. The overall results indicated that the structural model demonstrated an acceptable fit (Table 3). Bootstrapped confidence intervals (95% CI) also confirmed the robustness of the results. Figure 2 illustrates the assessed structural model, showing the calculated path coefficients.
Structural basic model for the total sample. Source(s): Figure created by authors
Structural basic model for the total sample. Source(s): Figure created by authors
The findings reveal that I4.0 has a significant and positive effect on exploitation, with strong support for hypothesis H1 (b = 0.567, 95% CI [0.45, 0.68], p < 0.001). Additionally, the standardized regression weights indicate robust relationships between I 4.0 and exploration (b = 0.564, 95% CI [0.44, 0.67], p < 0.001), also confirming hypothesis H2. H3 is further supported by evidence demonstrating that exploitation has a positive and significant impact on exploration (b = 0.400, 95% CI [0.28, 0.52], p < 0.001).
5. Discussion and conclusion
5.1 Findings
While organizational ambidexterity is often examined as a mediating capability linking organizational resources to performance outcomes, the present study intentionally focuses on its antecedents in the context of I4.0 transformation. Specifically, the objective of the research was to examine whether I4.0 capabilities can foster the simultaneous development of exploitation and exploration within SMEs. In this context, ambidexterity is treated as a proximal organizational outcome of digital transformation rather than as an intermediate variable in a broader performance model. This focus allows the study to isolate the mechanisms through which digital capabilities influence firms' ability to balance efficiency and innovation, which constitutes a fundamental organizational capability that underlies multiple downstream outcomes such as competitiveness, resilience, and long-term performance.
The results suggest that digitalization capabilities are associated with both exploitation and exploration, with a stronger relationship observed for exploitation. This finding highlights the dual role of I4.0, reinforcing the view that digital transformation is not merely an operational upgrade but a strategic enabler of organizational renewal (Hammad et al., 2026).
The positive relationship between I4.0 capabilities and exploitation suggests that technologies such as IoT, automation, and data analytics enhance firms' ability to optimize existing processes, reduce downtime, and improve asset utilization (Al-Talib et al., 2025). This aligns with prior research emphasizing the efficiency-enhancing role of smart manufacturing technologies (Frank et al., 2019; Xu and Duan, 2019). By enabling real-time monitoring and predictive maintenance, I4.0 technologies allow firms to extract greater value from existing resources, thereby strengthening firms' exploitative capabilities.
At the same time, the results indicate that I4.0 capabilities also support exploration by enabling firms to identify new opportunities, develop innovative products, and respond more effectively to changing market conditions. This finding is consistent with prior studies suggesting a link between digital tools and exploratory activities (Lichtenthaler, 2020; Xu et al., 2018). The ability to process large volumes of data and simulate alternative scenarios appears to reduce the uncertainty associated with exploratory activities, making innovation more manageable and strategically viable.
Importantly, the findings also show that exploitation positively affects exploration. This result suggests a complementary rather than conflicting relationship between the two dimensions of ambidexterity. Consistent with Piao and Zajac (2016) and He and Wong (2004), the study indicates that successful exploitation generates financial resources, operational stability, and organizational knowledge that can be reinvested in exploratory initiatives. In this sense, exploitation can be interpreted as a foundation that enables sustained exploration, particularly in digitally enabled environments.
However, this finding contrasts with studies that emphasize the risk of over-exploitation leading to rigidity and inertia (Hughes, 2018). The difference may be explained by the context of I4.0, where digital technologies reduce the traditional trade-offs between efficiency and innovation by increasing flexibility and responsiveness. Thus, the results suggest that under conditions of advanced digitalization, exploitation may no longer constrain exploration to the same extent as in more traditional organizational settings.
5.2 Theoretical implications
This study makes several important theoretical contributions to the literature on I4.0 and organizational ambidexterity. While prior research has often examined ambidexterity independently of digital technologies, this study integrates the two streams and demonstrates how I4.0 capabilities reshape the exploration–exploitation relationship. Although the positive association between digitalization and organizational ambidexterity may appear intuitive, this study provides a more nuanced insight into the mechanisms through which I4.0 capabilities shape this relationship. Specifically, the findings indicate that digital capabilities are strongly associated with exploitation-oriented activities, which are also positively related to exploration. This pattern of relationships can be interpreted as consistent with a sequential ambidexterity perspective, whereby improvements in efficiency and process optimization may create conditions that support exploratory initiatives. However, as the present study does not explicitly model temporal or sequential effects, this interpretation should be treated as a theoretical explanation rather than a directly tested mechanism. This interpretation extends existing research by demonstrating that digital technologies do not merely support exploration and exploitation independently, but may also reshape their relationship by suggesting that exploitation may function as a platform that enables exploration.
Such dynamics appear particularly relevant in resource-constrained SME environments, where firms rely on incremental digitalization to gradually build ambidextrous capabilities. The positive relationship between exploitation and exploration contributes to the ongoing debate on whether ambidexterity is characterized by trade-offs or complementarities. The results support a complementary perspective, suggesting that in digitally mature environments, exploitation and exploration may reinforce each other rather than compete for resources. This finding challenges traditional ambidexterity assumptions and calls for a contextualized understanding of exploration–exploitation dynamics in I4.0 settings. This insight is particularly important in SME contexts, where limited financial and technological resources often prevent simultaneous large-scale investments in both efficiency and innovation.
5.3 Practical implications
From a practical perspective, the findings offer several actionable insights for managers and policymakers. The results indicate that investments in I4.0 technologies can yield dual benefits by enhancing both operational efficiency and innovation capacity. Managers should therefore view digital transformation not as a choice between efficiency and innovation, but as an opportunity to pursue both simultaneously.
To strengthen exploitation, firms can begin with modular and cost-effective I4.0 applications, such as IoT-based equipment monitoring or basic analytics for process control. For example, sensor-enabled condition monitoring on critical machinery can reduce downtime and improve maintenance efficiency without requiring extensive system integration. Similarly, cloud-based platforms can support the automation of routine processes in service firms, even where Information and Communication Technology (ICT) infrastructure remains limited.
In support of exploration, I4.0 capabilities enable firms to experiment with new products, services, and business models in a controlled and data-driven manner. Managers may operationalize exploration through small-scale digital pilots, analytics-driven market analysis, or customer feedback platforms that allow firms to test new ideas before committing significant resources. Such approaches are particularly relevant in developing economies, where uncertainty and financial constraints make large innovation investments risky.
A key challenge in these contexts is the heterogeneity of ICT infrastructure and resource availability. To address this, managers should prioritize scalable and interoperable digital solutions, such as cloud-based and platform-oriented technologies, which are adaptable to evolving infrastructure conditions. Strategic partnerships with technology providers, industry networks, or public institutions can further help firms overcome financial and technical limitations.
Finally, human and managerial capabilities are critical for translating I4.0 investments into performance outcomes. Incremental workforce upskilling in digital literacy and analytical skills, combined with strong managerial commitment and strategic alignment, enables firms to balance efficiency and innovation objectives. By sequencing digital investments and leveraging exploitation gains to support exploratory initiatives, organizations can progressively build ambidextrous capabilities despite infrastructural and resource constraints.
6. Conclusions
This study provides insights into how digital transformation supports both efficiency-oriented and innovation-oriented organizational activities. The findings confirm that I4.0 capabilities positively influence organizational exploitation and exploration, and that exploitation, in turn, supports exploration, thereby highlighting a complementary relationship between the two. By demonstrating that I4.0 capabilities support both exploitation and exploration, the study extends existing research that has often examined these outcomes in isolation. Furthermore, this study contributes to ambidexterity theory by providing evidence in favor of a complementary relationship between exploitation and exploration in an I4.0-enabled context. Whereas earlier studies have emphasized the trade-offs between these two activities, the findings suggest that digital technologies may reduce such tensions by enhancing flexibility, information transparency, and resource reconfiguration. While the findings are consistent with a complementary and potentially sequential relationship between exploitation and exploration, the study does not explicitly test sequential dynamics, which remain an important direction for future research.
Importantly, while this study empirically focuses on efficiency-oriented outcomes, the findings can also be interpreted as foundational for emerging Industry 5.0 requirements. In particular, the results highlight the role of organizational ambidexterity, reflected in the complementary relationship between exploitation and exploration, as a critical capability enabling firms to balance operational efficiency with innovation and adaptability. In SMEs, efficiency-enhancing digitalization improves process transparency, decision speed, and resource utilization, which strengthen exploitative capabilities while simultaneously supporting exploratory initiatives. In this sense, ambidexterity emerges as the organizational mechanism through which I4.0 capabilities can support the broader objectives associated with industry 5.0, such as resilience, human-centricity, and sustainable transformation. By stabilizing core operations while fostering experimentation and innovation, ambidextrous organizations are better positioned to respond to disruptions, engage employees in problem-solving, and adapt to volatile environments.
From a practical perspective, the findings offer valuable implications for managers, practitioners, and policymakers. Investments in I4.0 technologies can yield dual benefits by improving operational efficiency and supporting innovation. Managers should therefore approach digital transformation as a strategic initiative aimed at achieving ambidexterity, rather than as a purely technological upgrade. The findings highlight the importance of supporting integrated digital transformation strategies that combine technological adoption with organizational and managerial development, particularly in resource-constrained and transitional economic environments.
7. Limitations and future research
This study was conducted within the economic and industrial context of Greece, which presents both opportunities and constraints. Cross-national studies comparing industrial organizations in economies with varying levels of technological development, financial stability, and institutional support would help assess the broader applicability of the proposed model. While this study provides a foundational framework, further refinement is required to assess its applicability across different organizational and economic landscapes. Moreover, future research may extend the framework of this study by incorporating additional outcome variables – such as firm performance, growth, or resilience – to further examine the downstream implications of ambidextrous capabilities in digitally transforming organizations. Furthermore, although the findings are consistent with a sequential ambidexterity interpretation, the present study does not explicitly model or test temporal or process-based relationships among variables. Future research should employ longitudinal designs or mediation-based approaches to directly examine the existence and direction of such sequential mechanisms. Addressing these limitations in future research will enhance the theoretical depth of ambidexterity studies but also offer actionable insights for practitioners navigating digital transformation.
