Introduction

Digital healthcare innovations have stimulated considerable advancements in the healthcare industry, providing entrepreneurs with ample market opportunities1,2. As a highly interdisciplinary area, the development of the healthcare industry is heavily dependent on key technologies and cutting-edge scientific knowledge1,2,3. However, the high cost and long period of product design and development make enterprises confront great risks in the process of healthcare innovation, and the technology and knowledge possessed by enterprises alone can no longer meet the demand for innovation4,5,6. Furthermore, the situation in healthcare entrepreneurship is marked by uncertainty and unpredictability, together with the intrinsic challenges of novelty and smallness in startups7. The disadvantages encompass insufficient resources (financial, market reputation, legitimacy, etc.), incomplete organizational and operational structures, and obstacles such as intense competition and product homogenization in the digital healthcare industry7,8. The social network in which an enterprise is embedded can help it access new resources, reach new heterogeneous network partners, and better cope with these difficulties and challenges9. As a result, digital healthcare firms tend to establish relationships with academic and research institutions, medical institutions, biotechnology enterprises, and other subjects to integrate the high-quality innovation resources through network embeddedness10,11,12,13. It can be seen that in the current context, utilizing embedded relational networks to access information and resources is more important for the growth of digital healthcare firms rather than controllable and limited internal resources.

Attaining a competitive advantage in the evolving digital environment is crucial for the sustainability and long-term success of digital healthcare14. Embeddedness theory holds that relational embeddedness is an important factor influencing corporate growth and gaining competitive advantage15. Relational embeddedness provides companies with access to a diverse set of such knowledge and innovation resources, which can be used to build partnerships and identify new development opportunities for rapid growth16,17. Organizations use knowledge as a strategic asset to achieve a sustainable competitive advantage. To increase innovation, organizations must promote effective knowledge management processes18. Organizations are motivated to continuously create and apply knowledge to sustain competitive advantage through innovation19. Even though numerous studies have been done on how relational embeddedness affects the performance20,21,22, some studies have also shown that relational embeddedness has some drawbacks, such as homogenization and cognitive lock-in, which reduces the sensitivity of firms’ access to information and therefore can constrain firm growth23,24. In summary, it is recognized that relational embeddedness has an important impact on firm growth, but there are divergent views on its specific effects and insufficient explanations of the key mechanisms. Moreover, few studies have examined how relational embeddedness affects the competitive advantage of digital healthcare ventures.

This study aims to elucidate how organizational learning ambidexterity influences the indirect impact of relational embeddedness on competitive advantage in digital healthcare. Organizational learning ambidexterity denotes the process by which enterprises acquire, understand, and apply knowledge, information, and technology through both exploitative and explorative learning25. In organizations, learning depends on how employees persevere to acquire and apply novel expertise and skills26. Ambidextrous learning is important for getting information, putting it together, sharing it, and making new information. It also helps cross-border innovations by giving people the different kinds of information they need27,28. Ambidexterity in organizational learning enables firms to promptly identify new opportunities and adjust quickly in competitive markets29,30. As a result, learning organizations tend to be more adaptive and proactive, leading to improved organizational performance31,32. The rapid progression of information technology has profoundly altered the healthcare sector through digital innovation1,2,33. The healthcare sector is unequivocally demanding in terms of expertise, resources, technology, and talent. Building organizational learning, which is essential for improving healthcare systems and patient outcomes and has been linked to a range of positive outcomes, including improved financial and clinical outcomes34,35. Organizational learning is a core element for digital healthcare ventures to drive growth and enhance competitiveness36. In the complex healthcare environment, exploratory learning prompts enterprises to continuously update their knowledge base, gain precise insights into the application potential of cutting-edge technologies such as artificial intelligence and big data, and develop innovative solutions in healthcare34,35. They are also able to refine existing products and services by mastering existing medical technologies, clinical experiences, and market information through exploitative learning37. The complex landscape exerts implicit pressure on digital healthcare ventures to force them to learn from a range of partners and competitors across many industries. Digital technologies must combine all their resources and knowledge. They also need to be on the lookout for new technological and market opportunities and understand how big data can help organizations come up with new ideas38.

Organizational learning theory posits that enterprises can enhance their long-term viability and growth by collecting, assimilating, and converting pertinent knowledge into actionable competencies25,26,27. Given the continuous external transformations, digital innovation subjects acquire digital value through knowledge transfer and sharing in the digital innovation ecosystem, thus augmenting competitive advantage7,39,40. The ongoing expansion of organizational boundaries and the rising trend of enterprise networking render the connectivity a potential source of organizational learning, facilitating the continuous transfer of both routine and novel knowledge between organizations22,26. Organizations strive to develop higher organizational learning capabilities to equip employees with advanced skills and knowledge to meet challenges at work, such as disruptive technologies and increasingly competitive markets26. Relational embeddedness provides a gateway for the exchange of information needed for firms to innovate and facilitates the assessment and acquisition of innovation opportunities41. In the network, digital healthcare ventures are capable of amalgamating diverse knowledge, skills, and experiences, while effective learning can foster innovation and assist firms in identifying new growth opportunities, thereby enhancing their competitive advantage in a highly competitive market environment15,23. However, there have been few studies that explore in more detail how relational embeddedness affects competitive advantage by acting on ambidextrous learning in digital healthcare.

The external environment profoundly impacts firms’ strategic objectives and their execution38,42. Market volatility and technological progress pose challenges and impacts on the present operational and inventive activities of digital healthcare ventures33,43. Environmental dynamism refers to a continuously dynamic and unexpected external environment, indicating the speed and volatility of changes within the industry setting in which a company operates44,45. The emergence of advanced technologies, such as cloud computing, the Internet of Things, 5G technology, and artificial intelligence, has revolutionized production and collaboration practices in the healthcare sector. The amalgamation of digital technology and healthcare will continue to advance significantly1,33,46. Digital healthcare ventures face challenges unique to their sector, such as inadequate technology and expertise, difficulties in obtaining funding and investment, rapid shifts in market demand, and vulnerability to external environmental changes2,8,10. A thorough examination of environmental dynamics is crucial for digital healthcare. The research questions are as follows: (1) Does relational embeddedness increase competitive advantage in digital healthcare? (2) Do explorative and exploitative learning play mediating roles in the influence of relational embeddedness on competitive advantage in digital healthcare? (3) Under environmental dynamism, how does relational embeddedness impact explorative and exploitative learning in digital healthcare? To answer the questions, this paper carries out rigorous empirical research based on the cross-sectional survey data collected from digital healthcare ventures.

This study has the potential to make three significant contributions. First, this study adds to the body of research on relational embeddedness in the digital economy by looking at the connection between relational embeddedness and competitive advantage in digital healthcare. This paper will help the embeddedness theory grow by showing how relational embeddedness can be used to gain a competitive advantage in digital businesses. Second, this study, grounded in organizational learning theory, broadens the theoretical examination of organizational learning ambidexterity, unveils the mediating roles of explorative and exploitative learning, sheds light on the complex association between relational embeddedness and competitive advantage in digital healthcare, and facilitates the integration of embeddedness and organizational learning theories. Third, this study adds to our knowledge of organizational learning ambidexterity by looking at the role of environmental dynamism as a moderator. This helps us understand how relational embeddedness affects competitive advantage, expands our knowledge of the conditions under which relational embeddedness works in digital settings, and makes it clear how environmental dynamism affects the effect of relational embeddedness on competitive advantage through explorative and exploitative learning.

Theoretical background and conceptual model

Embeddedness theory

The embeddedness theory originated with Polanyi’s research, was further expanded by sociologist Granovetter, and has since become a major theory in the study of modern economic sociology47,48. Firms’ social capital has two dimensions: relational embeddedness and structural embeddedness49. The study of Moran49 found that both elements of social capital have an impact on managerial performance, albeit in different ways: structural embeddedness is more important in explaining routine, execution-oriented tasks (managerial sales performance), whereas relational embeddedness is more important in explaining new, innovation-oriented tasks. The digital healthcare business is rapidly evolving and transforming, and innovation is critical for firms seeking a competitive advantage1,2,3. In the digital age, digital healthcare businesses must be innovative and constantly seek new markets. As a result, relational embeddedness is a more intriguing topic to investigate than structural embeddedness. Granovetter defined relational embeddedness as the extent to which individuals in a network value the needs or goals of others, trust each other, and share information48. The theory contends that economic action is integrated in and compatible with a specific network of social relations48. Relational embeddedness emphasizes the direct linkage between subjects in a social network and is an important characterization of the strength and quality of inter-subjective relationships, which have a major impact on subject behavior and performance50. Companies operate in a network of social ties, and their relational embeddedness impacts their ability to access, integrate, and use network resources51. Innovation entails bringing unique ideas, products, processes, and services that can be identified by efficiency, speed, new technology, and marketing strategies18. Relational embeddedness helps the assessment and acquisition of innovation prospects, hence improving business performance52.

Traditional business models emphasize the value of competition in corporate activities53,54. However, changes in the business environment, rapid development of digital technologies, and the frequency of public health incidents make it difficult to sustain a competitive edge1,4. Digital healthcare ventures are rapidly realizing that a strategy based purely on competition is insufficient to meet the demands of practice development55,56. In the digital economy, a number of firms have interdependent ties and collaborate with one another to create value and achieve synergistic development57. Since the embeddedness theory has been an important premise in strategic tools for accessing resources, mature enterprises have been the primary focus of relational embeddedness research15,23,58, as opposed to digital start-ups, which have received insufficient attention. Entrepreneurial organizations suffer greater environmental unpredictability than traditional businesses and have strong competitive motivations stemming from a desire to survive2,10. Zhang et al. discovered that small and medium-sized firms (SMEs) may develop and adjust their resource bases through relational embedding59. Li and Fei found that relational embeddedness has a considerable favorable impact on enterprise performance60.

Although these studies initially highlight relational embeddedness, they do not provide a thorough study of the competitive benefits of digital healthcare. Indeed, digital healthcare ventures have collaborated with a variety of Internet companies, pharmaceutical companies, health insurers, and others10, but the business models for creating value are unclear, with unprecedented consequences and obstacles. Konopik stated that under Healthcare 4.0, ecosystem participants collaborate to create value through resource development, collaborative competition, and the use of digital platforms, radically altering the delivery of healthcare services and products33. Combining the essential features of the digital healthcare scenario, this paper sees relational embeddedness in digital healthcare ventures as an approach that allows enterprises to use digital technology to provide complete, personalized solutions based on customer needs in data-driven collaboration with multiple stakeholders, including healthcare providers, pharmaceutical companies, and regulatory agencies.

Organizational learning theory

Simon61 first introduced the concept of learning in relation to organizational theory, and Argyris and Schon formalized organizational learning as the process of recognizing and correcting errors. In academia, research on organizational learning is conducted from several perspectives62. Nevis et al. presented three metrics for assessing organizational learning, including the sharing, diffusion, and use of knowledge31. Hult and Ferrell examined organizational learning capacity using four dimensions: team-oriented, system-oriented, learning-oriented, and memory-oriented63. Sinkula et al. proposed the effect of dedication to learn, shared vision, and open-mindedness in the context of market information64. Tippins and Sohi identified five elements of organizational learning: information acquisition, information interpretation, information dissemination, organizational memory, and procedural memory65.

March66 established the “exploration-exploitation” theoretical model in the development of organizational learning theory, which is based on distinctions in learning styles and positioning. It explains the fundamental mechanism and process of organizational learning, which has huge theoretical implications. This classification is widely accepted in academia and has been utilized as a research paradigm in a variety of disciplines, including strategic management, organizational theory, and managerial economics25,27,28. Explorative learning focuses on the organization’s pioneering activities, such as searching, discovering, experimenting, risk-taking, and creating, which fuel the process of discovering and capturing new knowledge and opportunities. Acquiring and expressing new information can help organizations in a variety of ways, including strengthening inventive capacities and management skills, transferring tacit knowledge, and adapting new knowledge to introduce fresh perspectives67. On the other hand, exploitative learning focuses on the organization’s behaviors, such as optimizing, selecting, manufacturing, executing, and implementing, which aim to improve organizational efficiency, maximize benefits, and dominate the process of exploiting knowledge and opportunities68,69. This categorization is straightforward to measure and has been utilized by many relevant studies, and this work on dimensional classification of organizational learning uses March’s perspective66. March emphasized that in the business process of enterprises, focusing primarily on exploration while ignoring exploitation might result in a lack of effective industrialization and commercialization, preventing the enterprise from realizing the benefits of its exploratory activities66. In contrast, if an enterprise focuses solely on exploitation and ignores exploration, it may achieve satisfactory short-term performance, but it will lack the motivation to continue its growth and risk falling into the “capability trap.” Liang et al.69 discovered that proportionally balanced, rather than perfectly balanced, combinations of IT exploitation and exploration maximize organizational agility. Shao et al.28 expanded on ambidextrous learning in the digital arena by introducing the terms “exploitative” and “explorative” use. They developed and tested an integrative theoretical framework to better understand how female and male employees advance from ambidextrous learning in digital technology to digital creativity. Specifically, women want to refine and exploit digital technology, whereas men are eager to experiment and innovate with it to achieve digital creativity.

Competitive advantage refers to a company’s superior position over its competitors in terms of cost, efficiency, market presence, and product innovation70. This is critical for businesses because it can put the organization in a better position in a highly volatile entrepreneurial landscape and in the face of fierce market competition42, allowing the venture to survive and expand in the long run. In contrast to established, mature businesses, young initiatives have inherent difficulties such as insufficient resources, legitimacy, and user acceptance71, which severely limits their capacity to gain a competitive edge. By combining the ideas of relational embeddedness and organizational learning in a dynamic environment, digital healthcare ventures can reduce R&D costs and accelerate technical developments through partnership with other healthcare entities or technology companies. These organizations continue to research digital healthcare technologies and improve their existing products and services, resulting in an effective learning framework that promotes adaptable and innovative competencies10,11,72. This ambidextrous learning allows businesses to quickly respond to changes in market demand and revise their plans, increasing their competitive advantage73. This study proposes a mechanistic model to demonstrate the impact of relational embeddedness on competitive advantage in digital healthcare. It also examines the relationship between relational embeddedness, organizational learning ambidexterity, and competitive advantage, as well as the effect of environmental dynamics (as depicted in Fig. 1).

Fig. 1
Fig. 1
Full size image

Research model.

Relational embeddedness and competitive advantage in digital healthcare

Competitive advantage is vital for an organization’s survival and growth in a dynamic environment, and in the era of the knowledge economy, it vanishes at an ever-increasing rate. Establishing and maintaining a competitive advantage is at the heart of every firm74. Therefore, a large number of strategic activities are needed to gain a competitive advantage75. The means by which firms achieve a competitive advantage for sustainable growth has been a vital inquiry in strategic management75. The embeddedness theory claims that relational embeddedness plays an important role in the formation of competitive advantage15. There is a certain amount of existing research on relational embeddedness and performance. In21 Swierczek indicated that relational embeddedness of logistics service providers has a positive effect on the supply chain performance in transitive triads. Albis, N., et al. in22 revealed that external relational embeddedness in isolation has a positive impact on performance. Guo et al. in41 showed that relation embeddedness is significantly positively correlated with shared mental model and innovation performance.

Relational embeddedness can also be defined based on “weak ties” versus “strong ties,” or “intra-industry networks” versus “inter-industry networks,” or “supplier ties” versus “customer ties.” Different types of embeddedness may impact competitive advantage differently. With regard to the tie strength, Zhong and Werner in76 believed that having strong ties and a highly cohesive network with those who help them establish their businesses and regular suppliers is critical for traders, enabling them to respond to crises, reduce risk and uncertainty, and improve performance, which does not mean that weak ties are unnecessary; rather, vendors need a combination of strong and weak relationships with a diverse range of customers, including small businesses and consumers. Wilson and Tonner77 also suggested that the competitive advantage of family firms has been shown to benefit from wider networks and strong and weak ties between family members. As for the relational configurations, intra-industry networks focus primarily on linking companies within the same industry, bringing the advantage of specialized resource integration, while the greatest advantage of inter-industry networks is the promotion of diversified innovation and expansion of business boundaries78,79. Considering the position in the chain, linking with suppliers can bring significant cost advantages, reducing procurement costs and optimizing supply chain management through centralized purchasing and joint R&D80, while connecting with customers helps companies create differentiated advantages and explore the market potential81. However, research on relational embeddedness and its impact on competitive advantage in the digital health sector is currently scarce.

The healthcare sector in the digital era exhibits intricate and varied trends, while the ongoing incorporation of emerging technologies like big data, artificial intelligence, and digital twins offers numerous prospects for advancing the digital healthcare domain14. In the digital age, relationships serve as a source of innovative ideas, resource accumulation, and novel resource combinations33,36. The conventional business model of enterprises generating value independently has become obsolete as digital healthcare ventures transition toward a paradigm of collaboration and mutual benefit, shifting from a focus on productivity to one on relationships32,60. The primary objective of digital healthcare ventures is survival, exhibiting traits typical of general businesses; however, their nascent stage renders them perpetually vulnerable to failure33,55. The abrupt emergence of the pandemic has adversely affected conventional offline healthcare facilities, resulting in an increased need for online healthcare. Many conventional healthcare organizations are progressively venturing into digital healthcare via digital transformation, as the digital healthcare services industry expands intermittently and the market becomes more competitive1. Consequently, digital healthcare ventures must engage with various stakeholders in social networks to generate synergistic effects, enhance efficiency, and secure a competitive advantage.

Digital technology has transformed the operations and interactions of healthcare ventures, with the swift advancement of emerging digital technologies eliminating constraints of time and space. Connectivity has become a crucial component in the execution of digital healthcare ventures’ strategies3. Enterprises are always in a social network that influences and interacts with stakeholders. By establishing close ties with multiple stakeholders, such as competitors, complementors, and customers, enterprises can obtain the resources they need from the network to promote their own development60. Firstly, collaboration among competing enterprises within the same industry is becoming increasingly frequent in the digital economy56. Digital healthcare ventures offer digital healthcare products and services similar to their counterparts, and through collaboration in peer-to-peer networks, they can share industry insights, co-develop new products, and innovate technologies while also distributing R&D expenses, facilitating swift market entry for their products, and establishing competitive advantages2,36. Peers working together can help a market grow, take advantage of industrial clustering and scale effects, set technical standards, and make the industry environment better overall57. Secondly, due to the intricacy of products in the digital era, digital healthcare ventures with constrained technological resources and expertise struggle to independently deliver comprehensive solutions to customers and must proactively collaborate with complementors to enhance the functional scope of their products or services55,82. The expansion of the function space enhances customer operations and utilization, offers a more comprehensive solution, fulfills customer needs, improves customer satisfaction, increases customer retention, and provides a competitive advantage in the market8,83. Finally, customers have transitioned from mere service recipients to active co-creators of value. In the digital age, business activities are primarily governed by a customer-centric approach, leading to increased interactions between enterprises and their customers46. Digitalization enables customers to access specific healthcare resources and services, engage in dialogue, and complete transactions at any time and from any location, thereby enhancing the convenience and speed of interactions. Additionally, digital healthcare ventures can gather market insights from customers, refine market perception and sensitivity, and offer more personalized and tailored healthcare services, which aids organizations in stabilizing markets and bolstering customer loyalty, ultimately improving competitive advantage3,11,43,52.

Therefore, this paper argues that digital healthcare ventures utilizing relational embeddedness can not only reduce costs and improve the sharing of risks but also better understand user needs, come up with new ideas and solutions, and develop novel products and services that better meet users’ expectations. We believe that digital healthcare ventures can improve their competitive advantage through relational embeddedness. We thus propose:

H1: Relational embeddedness is positively associated with competitive advantage in digital healthcare.

The mediation of organizational learning ambidexterity

Different types of embeddedness can have different impacts on learning. In terms of the strength of the relationship, Jiang and Zhao84 suggested that strong ties have a direct positive effect on exploitation learning, and the effect of weak ties on exploration learning is much greater than that on exploitation. Bettis-Outland et al.85 argued that micro-networks with strong ties will lead to generative learning, but interactions between micro-networks with weak ties will result in transformative learning. In terms of the types of relational configurations, companies can explore new opportunities in the inter-industry environment with diverse knowledge while, at the same time, exploiting homogeneous knowledge sources available via the intra-industry. Almahendra86 highlighted the importance of balancing these two conditions to sustain organizational learning. Given that the position in the chain, connecting with suppliers, opens a window of learning for organizations to optimize operations and deepen expertise, while establishing ties with customers builds a learning platform for organizations to gain insight into market needs and drive product innovation87,88. Liao et al.75 have proposed organizational learning as a fundamental strategic process and the only sustainable competitive advantage of the future. For new ventures, the race for survival and growth is like a race for learning because they must quickly overcome their “liabilities of newness and adolescence” and establish for themselves the legitimacy and reduce the uncertainty enjoyed by more established firms84.

According to the organizational learning theory, ambidextrous learning can serve as a bridge connecting relational embeddedness and competitive advantage30,32,89. Based on the logic of the above assumptions, it can be seen that the higher the degree of enterprise relational embeddedness, the higher the degree of competitive advantage. Gaining a competitive advantage is possible for digital healthcare ventures because they can share, learn, and integrate knowledge through learning mechanisms that create and deliver better customer value36,75. However, ineffective knowledge sharing or inadequate knowledge utilization can hinder innovation and degrade an organization into an obsolete one18. In social networks, a higher degree of relational embeddedness implies mutual trust, increases willingness to share knowledge and information, and facilitates knowledge transfer60. Relational embeddedness, characterized by a high degree of knowledge sharing and internalization, can help firms gain access to heterogeneous knowledge bases, where relevant knowledge can be identified, assimilated, transformed, and ultimately exploited by the firm15,90. High involvement in the acquisition, sharing, and application of knowledge results in higher learning in organizations26. In the R&D process, organizations require communication and collaboration to effectively utilize knowledge due to the vast volumes and complexity of information91. Only well-informed firms can effectively recognize and adapt to changes in digital technology and client requirements. Understanding consumer preferences enhances product development capabilities, hence augmenting customer satisfaction30. Numerous research studies highlight the necessity for firms to perpetually enhance their internal abilities and knowledge to adopt their competitive posture32. Consequently, we contend that relational embeddedness needs to go through the mediating function of organizational learning ambidexterity to create competitive advantage. Research indicates that firms employing both explorative and exploitative tactics can attain superior outcomes in establishing a competitive advantage29. Learning involves the acquisition and sharing of routine and novel information and knowledge that improves the existing knowledge base of the receiver and increases the understanding of developments in technology, goods, processes, and markets by enhancing learning through exploitation and exploration26,30. We hypothesize that organizational learning ambidexterity mediates the connection between relational embeddedness and competitive advantage in digital healthcare.

Relational embeddedness promotes explorative learning and, thus, the attainment of competitive advantage. The utilization of relational embeddedness can provide access to extensive new information and promote the exchange of best practices and knowledge transfer9,36. Upon acquiring pertinent knowledge, enterprises evaluate, process, and integrate this information into the creation of new products and services to establish core competencies and sustain a competitive advantage75. Firstly, explorative learning is predicated on acquiring new knowledge or diverging from established knowledge to create innovative healthcare products or services, novel designs, or alternative distribution channels25,29. To improve their abilities, skills, and competencies, firms strive to create recombined knowledge lakes by gaining new knowledge. These knowledge bases serve as the foundation for new competitive advantages, which are necessary for better performance and financial gain74. The diverse knowledge base can enable ventures to attain radical innovations by collaborating to integrate various perspectives, identifying and selecting valuable information sources from extensive multidimensional data, and effectively disseminating, sharing, and institutionalizing this knowledge within the organization to integrate new insights, thereby enhancing the organization’s capacity to innovate and achieve its innovation objectives18,26,28,30,51. Secondly, explorative learning seeks to address the needs of emerging users or healthcare markets, with ventures utilizing relational embeddedness to discover new developmental pathways through the exchange of inter-organizational knowledge regarding new users or markets, enhancing the possibility of enterprises obtaining valuable digital healthcare technology and other resources, and the establishment of competitive advantages in delivering customer value72. Thirdly, explorative learning enables enterprises to generate innovative strategies, concepts, and insights for the attainment of sustained advantages27,30,41. Digital healthcare ventures cultivate productive, cooperative relationships with stakeholders and, through dynamic interactions, leverage digital technology to extensively seek external knowledge and extract value from this acquired knowledge via data analysis. This enables them to forecast future market trends and customer requirements, identify new opportunities, devise innovative solutions, proactively adapt to fluctuations in technological and market landscapes, and achieve long-term objectives17,29.

Furthermore, relational embeddedness facilitates exploitative learning and ultimately the acquisition of competitive advantage. Exploitation as an organizational capability is based on the routines that allow firms to refine, extend, and leverage existing competencies or to create new ones by incorporating acquired and transformed knowledge into their operations67. A strong ability to transform acquired knowledge is important when developing innovations, i.e., when functions start to make connections among apparently disparate pieces of market information6. Transforming acquired knowledge is also key to gaining a more comprehensive understanding and establishing broader perspectives72. Firstly, relational embeddedness enhances the breadth and depth of a firm’s knowledge search, updating, recombining, and reconfiguring a firm’s existing knowledge, and enabling firms to arrive at economical solutions for product development or extension67. Digital healthcare ventures achieve incremental innovation by refining and incorporating existing knowledge into healthcare products or services28. Second, exploitative learning tries to meet the needs of existing customers or markets by encouraging cooperation between businesses, which makes it easier for them to share information about current customers or markets. Digital healthcare ventures increasingly integrate, internalize, and apply existing information to convert information superiority into a competitive advantage in the marketplace27,29. Thirdly, relational embeddedness is conducive to long-term cooperation between enterprises and lower transaction costs to attain short-term profitability. Close connections between firms increase the costs of normative cooperation and violation of network agreements, reducing the speculative behavior in the transaction process60. Enterprises utilizing relational embeddedness can more effectively contribute to knowledge sharing, swiftly acquire information and technological trends, consistently achieve self-improvement, and secure a competitive advantage in the marketplace in the short term2,28.

In conclusion, we expect that digital healthcare ventures undertake explorative and exploitative organizational learning and achieve organizational ambidexterity through relational embeddedness, which eventually increases competitive advantage. We thus propose:

H2a: Explorative learning mediates the positive relationship between relational embeddedness and competitive advantage in digital healthcare.

H2b: Exploitative learning mediates the positive relationship between relational embeddedness and competitive advantage in digital healthcare.

The moderation of environmental dynamism

Researchers have examined the contingency factors influencing the relationship between embeddedness and organizational performance12,20,23,92. For example, Xie et al.92 indicated that resource orchestration capability strengthens the positive effects of both market and technology network embeddedness on green innovation performance. However, there remains a paucity of research regarding the boundary role of the relationship between relational embeddedness and organizational learning ambidexterity, particularly within the digital healthcare context. The influence of external environmental factors on the relationship between relational embeddedness and both explorative and exploitative learning has not been thoroughly explained.

Environmental dynamics refers to the rate of change and degree of instability in the environment45. Prior research has demonstrated that environmental dynamics are reflected in both the amount and the unpredictable nature of change44. Dynamic environments may be characterized by changes in technology, changes in customer preferences, and a proliferation of new products or services, with a stream of business model innovations42,93. In rapidly evolving environments characterized by swift technological advancements and fluctuating consumer preferences, sustaining product competitiveness through existing knowledge is challenging. The dynamic nature of business landscapes can render current products and services obsolete, necessitating ventures to acquire new knowledge and innovate products and services to adapt to these changes94. To mitigate the risk of obsolescence in the market, enterprises are supposed to improve external communication, broaden information acquisition channels regarding customers, suppliers, and competitors, actively pursue new knowledge in social networks, and integrate this knowledge through learning. This approach will enhance the knowledge base, facilitate new product development and technological innovation, and identify innovative strategies to adapt to changes in the business environment5,72. Organizations engaged in exploratory learning can leverage dynamic environments to develop innovative products and services or address the demands of emerging markets, thereby generating opportunities for above-average returns by penetrating new high-end markets and establishing unexplored niches83. Consequently, in a highly dynamic external environment, digital healthcare enterprises use an exploratory learning model to leverage relational embeddedness and acquire new knowledge, thereby maintaining a competitive edge and staying ahead of competitors.

On the contrary, organizations leveraging the benefits of exploitative learning emphasize utilizing existing knowledge at the expense of knowledge acquisition; the dynamic environment is likely to be detrimental to their development as they are unable to remain competitive in a rapidly changing environment73. Such organizations tend to continuously extract (mine) hidden and useful knowledge and seek ways to cope with changes in the business environment from existing practices and solutions27,94. Under high levels of environmental dynamism, digital healthcare ventures are likely to fall behind if they remain with the status quo; existing product, service, and solution offerings become too quickly outdated, and focusing too much on executing routines could make them lag behind in the digital age95. Because of this, a changing environment will lessen the effects of relational embeddedness on exploitative learning in digital healthcare.

In general, a dynamic environment may strengthen relational embeddedness’s impact on explorative learning in digital healthcare while weakening relational embeddedness’s influence. Accordingly, we suggest the following hypotheses:

H3a: Environmental dynamism increases the strength of the positive relationship between relational embeddedness and explorative learning.

H3b: Environmental dynamism decreases the strength of the positive relationship between relational embeddedness and exploitative learning.

Methodology

Data collection and the sample

In this study, questionnaires were used to collect data. Members of this research team worked from July 2024 to November 2024 to distribute the questionnaires. This study was approved by the Qiqihar Medical University Ethics Committee with the approval number of (Qi) Ethics Review [2024] No. 43. Informed consent was obtained from all subjects and their legal guardians. All the procedures were performed in accordance with the Declaration of Helsinki and relevant policies in China. Participants were assured of their privacy and informed that the data would be used solely for research purposes, with no commercial intent. Prior to data collection, the researchers were trained to understand and master the questionnaire content and interview techniques. The researcher contacted potential respondents by phone or WeChat to explain the study’s purpose and methodology, arrange their participation, increase their willingness to participate, and maximize the response rate. Respondents were told their data would be used for research only, with no commercial intent. Respondents’ personal information will be kept confidential, and the results of the study will be shared upon request. The sources of the survey samples are distributed in regions with different degrees of innovation and entrepreneurship activity. According to the Index of Regional Innovation and Entrepreneurship in China (IRIEC) compiled by Peking University’s Center for Enterprise Research, we selected regions with high levels of such activities, like Guangdong Province, Jiangsu Province, and Zhejiang Province, as well as regions with low levels of such activities, like Hainan Province and Heilongjiang Province, as the research regions. A combination of purposive sampling and random sampling was adopted in the study. In collaboration with local government institutions, industry associations, and other management organizations, we first used purposive sampling to select multiple regional and technology business incubators and parks, and then randomly selected sample ventures from a list of digital healthcare ventures provided by these incubators. We initially used purposive sampling to identify suitable industrial parks and incubators for the following reasons. First, when the research target is a population with specific characteristics, random sampling may be difficult to achieve due to the low proportion of that population in the total population96. The target of this research is digital healthcare ventures, which possess distinct industry characteristics. Purposeful sampling can directly identify ventures using specific criteria, ensuring that the questionnaire data concentrates on the core research target and prevents the inclusion of invalid samples. Second, random sampling is costly and complicated to implement. Purposeful sampling allows researchers to directly reach the target population through their experience or channels, greatly reducing sampling difficulty and time costs96. In this paper, based on positive working relationships with local government bodies, industry associations, and other groups in the early stages, we focused on digital healthcare ventures in enterprise zones, technology business incubators, and hackerspaces in these regions as our research subjects, effectively saving time, manpower, and money. At least two questionnaires were distributed to each firm. On one hand, the research team distributed questionnaires on-site to enterprise concentration areas like technology parks and office buildings. On the other hand, the electronic version was distributed to the enterprises by e-mail and WeChat through online channels. In the data collection process, the researcher explained the research process and addressed questions. If respondents were unable to complete the questionnaire on-site, we provided an electronic version and requested that it be returned via WeChat or email within a specified timeframe.

The interviewees are general managers of these enterprises or middle and senior managers who have worked in them for longer periods of time. These employees are more aware of the operation of the enterprise and the entrepreneurial process and can help us obtain information accurately. To ensure the reliability and validity of the measurements, the questionnaire was appropriately modified based on previous research to make it applicable to the digital healthcare environment of this study. Since the items on the scale are mostly borrowed from foreign languages, to ensure the accuracy and consistency of the questionnaire, we adopted a back-translation methodology. It was first written in English and then translated into Chinese with the help of a professional bilingual translator. Then the scales were refined until the Chinese and English versions were closely aligned. In addition, we conducted a pilot study with corporate executives and revised the questionnaire based on their opinions until no new feedback could be obtained from the study, and finally a total of 20 corporate executives participated in the pilot study. We distributed 656 questionnaires to 300 ventures, of which 523 returned, yielding a response rate of 79.73%. After eliminating all invalid cases, 356 effective questionnaires remained, resulting in an effective response rate of 54.27%. The majority of responding companies were founded by males, with an average company age of 5.6 years, and 60.7% of the enterprises were established for less than 6 years. Small-sized enterprises with fewer than 50 employees, predominately private, made up the majority of responding companies, followed by foreign-owned, joint, and state-owned enterprises. Geographically, 21.3% of the enterprises were located in Guangdong Province, 22.5% in Jiangsu Province, 20.2% in Zhejiang Province, 20.2% in Hainan Province, and 15.7% in Heilongjiang Province. Also, an independent sample t-test that compared the means of early and late responses on the variable “firm size” did not demonstrate any statistically significant differences. This suggests that non-response bias is not a problem in this study97. The descriptive statistics for the respondents are given in Table 1.

Table 1 Company profiles.
Full size table

Constructs and measurement

All constructs were assessed using a five-point Likert scale, ranging from “Strongly Disagree” to “Strongly Agree,” with corresponding values of 1 to 5 points.

Independent variable-relational embeddedness: Relational embeddedness has been portrayed as a construct comprising elements including closeness, trust, and communication frequency48,49,54. These elements can serve as a critical pillar of the mutualistic and cooperative orientation that emerges in highly dependent relationships98. We employed the scale by Li and Fei60, which demonstrated satisfactory reliability with a Cronbach’s α of 0.787.

Dependent variable—competitive advantage: In the digital age, digital healthcare ventures face a large number of risks from different sources, such as globalization, environmental changes, complex financial models, and corporate governance changes. In a dynamic context, increasing competitive advantage becomes one of the main challenges for digital healthcare ventures. The scale by Saeidi et al.70 was used, yielding a Cronbach’s α of 0.898, indicating good reliability.

Mediator variable—Organizational learning ambidexterity, as an effective way for enterprises to expand their knowledge base, can provide sufficient support for the application of digital technologies40, embodying two different underlying logics of exploration and exploitation in digital innovation30. The study adopted the scale by Jiang et al.27, with Cronbach’s α coefficients of 0.846 and 0.884, respectively, reflecting the scale’s reliability.

Moderator variable—environmental dynamism: The dynamic environment is accompanied by the rapid iteration of digital technologies, customer demands, and product/service volume fluctuations in the digital era95. Environmental uncertainty was measured using scales from Meulenaere et al.44, resulting in a Cronbach’s α of 0.816, confirming the scale’s reliability.

Controls: Given their documented influence on the entrepreneurial venture10,57, we included founder gender, firm age, firm size, and firm ownership as control variables. Firm age was determined by years since establishment, and firm size by employee count. Firm ownership was defined by the dominant shareholder type, including private, joint, state-owned, and foreign-owned enterprises, with a dummy variable for firm ownership.

Result

Reliability and validity of measures

We used SPSS 26.0 and AMOS 26.0 to assess the reliability and validity of the constructs in this study. Consistent with Anderson and Gerbing99, we first conducted a measurement model test before evaluating the conceptual model. Exploratory factor analysis (EFA) using SPSS 26.0 identified the latent constructs, employing principal axis factoring and varimax rotation with Kaiser Normalization. Factors with eigenvalues greater than one were retained, while those with eigenvalues less than one were considered insignificant. The EFA extracted five factors with eigenvalues exceeding 1.0, explaining 68.375% of the variance in the data: relational embeddedness (18.262%), explorative learning (16.165%), exploitative learning (14.235%), competitive advantage (10.605%), and environmental dynamism (9.107%). The Kaiser-Mayer-Olkin (KMO) test yielded a value of 0.904, above the threshold of 0.8, and Bartlett’s test of sphericity achieved a chi-square of 4798.456 with 253 degrees of freedom, p < 0.01, indicating suitability for factor analysis100. All items used in the constructs are presented in Table 2. All standard factor loadings were 0.5 or higher, exceeding the threshold required for confirmatory factor analysis (p < 0.001)99. We then used a confirmatory factor analysis (CFA) with AMOS 26.0 software involving these five constructs. A more precise metric (i.e., χ2/df) was assessed, which showed satisfactory model fit, which is < 5101. The Standardized Root Mean Square Residual (SRMR) and Root Mean Square Error of Approximation (RMSEA) are key measures, with SRMR ≤ 0.08 and RMSEA ≤ 0.08 indicating good fit102,103. Incremental fit indices, including the Incremental Fit Index (IFI), Tucker-Lewis Index (TLI), and Comparative Fit Index (CFI), need thresholds of ≥ 0.90 for acceptable fit102,104,105. Table 3 shows the measurement model’s fit for the five constructs. The measurement model demonstrates a satisfactory fit to the data:χ2/df = 2.607 < 5, IFI = 0.925 > 0.9, TLI = 0.913 > 0.9, CFI = 0.924 > 0.9, RMSEA = 0.067 < 0.08, SRMR = 0.061 < 0.08. Comparative analyzes with alternative models (four-factor, three-factor, two-factor, and one-factor) supported the superiority of the five-factor model. All fit indicators meet the required standards. All Cronbach’s alpha values exceeded the 0.7 threshold for internal consistency reliability, confirming the acceptability of the measures106. Composite reliability (CR) values surpassed the 0.70 benchmark, and average variance extracted (AVE) values exceeded 0.50, indicating that convergent validity was achieved106. As shown in Table 4, the AVE square root values were higher than the inter-construct correlation coefficients, which showed that it was a good discriminant107.

Table 2 Measurement model: constructs, items, SFL, AVE and CR.
Full size table
Table 3 Confirmatory factor analysis.
Full size table

Common method bias

Common method bias (CMB) arises when variables are measured contemporaneously using a single instrument108. We implemented both procedural and statistical controls to mitigate CMB. Procedurally, we assured respondents of anonymity, clearly communicated the purpose of the questionnaire, and collected at least two questionnaires per firm57. We conducted Harman’s single factor test statistically to assess potential CMB. The first factor extracted from exploratory factor analysis accounted for less than the critical threshold of 50%109. Additionally, as indicated in Table 3, Harman’s single-factor model was tested and found to fit the data poorly: χ2/df = 10.285, IFI = 0.544, TLI = 0.496, CFI = 0.542, RMSEA = 0.162, SRMR = 0.136. The single-factor model fit was inferior to our measurement model, suggesting that CMB was not a significant concern in this study110.

Descriptive statistical analysis and correlations

Table 4 presents the descriptive statistics and correlations, offering support for the variables and hypotheses’ credibility. These correlations indicate significant relationships among relational embeddedness, explorative learning, exploitative learning, environmental dynamism, and competitive advantage. These findings align with the research direction and provide a solid basis for hypothesis testing. Table 4 presents a prerequisite analysis for descriptive statistics. The mean of all observed variables ranged from 3.543 to 4.061, with a standard deviation spanning from 0.705 to 0.860.

Table 4 Correlation matrix.
Full size table

Results of analysis

We used regression analysis and the bootstrap method to verify our research hypotheses. First, we discovered that H1 was supported by a regression analysis on the effect of relational embeddedness on competitive advantage in digital healthcare (β = 0.434, t = 9.053, p < 0.001). Second, we used Hayes’ PROCESS plug-in to do bootstrap analysis with 5,000 iterations for mediation path tests, which you can see in Table 5. We employed Model 4 to assess the mediating effects of explorative learning and exploitative learning while controlling for variables such as founder gender, firm size, firm age, and firm ownership. Explorative learning positively mediates relational embeddedness’s effect on competitive advantage (β = 0.070, p < 0.05), as the 95% confidence interval {0.025, 0.131} did not include 0. Similarly, exploitative learning positively mediates relational embeddedness’s effect on competitive advantage (β = 0.075, p < 0.05), as the 95% confidence interval {0.010, 0.147} did not include 0. The direct effect of relational embeddedness on competitive advantage was 0.326, and the confidence interval {0.204, 0.447} did not include 0, indicating partial mediation. Therefore, we support H2a and H2b.

Table 5 Mediation model analysis.
Full size table

We utilized Model 1 in the PROCESS plug-in for moderation effects, taking into account the same control variables. As shown in Table 6 (Moderation model 1), the interaction of relational embeddedness and environmental dynamism had a significant positive effect on explorative learning (β = 0.086, p < 0.05), as the 95% confidence interval {0.006, 0.167} did not include 0, affirming H3a. As shown in Table 7 (Moderation model 2), the interaction between relational embeddedness and environmental dynamism, on the other hand, had a negative impact on exploitative learning (β=−0.228, p < 0.05), with a confidence interval of {−0.305, − 0.151} that did not include 0, which supports H3b.

Table 6 Moderation model 1 analysis.
Full size table
Table 7 Moderation model 2 analysis.
Full size table

Discussion

Discussion of findings

This study theoretically contributes by highlighting the synergistic effect of the relational embeddedness-organizational learning ambidexterity relationship on competitive advantage in digital healthcare. Firstly, we develop and evaluate hypotheses by examining the digital healthcare ecosystem to yield insights into the scenarios necessary for attaining a competitive advantage in digital healthcare. H1 supports the significant and positive impact of relational embeddedness on competitive advantage. The findings of the study are in line with recent research suggesting that relational embeddedness promotes firm performance and competitive advantage12,13,22,51,90. The crucial issue concerning the development of enterprises is the improvement of their integration with multiple actors to provide them with mutual benefits and ultimately increase their performance20,56,111. These findings corroborate previous studies21,41,60, affirming relational embeddedness helps the enterprises to establish more ways of information acquisition and resources, improve interorganizational collaboration by offering value-adding activities and innovative solutions, and then increase competitive advantage. In particular, firms in emerging market countries such as China, which face complex market competition, have a stronger perception of competition112. By focusing on emerging markets22,113, this study deepens the research on relational embeddedness and contrasts with previous assessment studies that focused on mature economies, thereby filling the gap in research on emerging economies24,51.

Secondly, we contend that in a dynamic digital ecosystem, it is desirable for digital healthcare ventures to engage in exploratory learning regarding external information and exploitative learning concerning existing technologies and paradigms to achieve innovation for sustained competitive advantage36. H2a and H2b support the significant and positive impact of relational embeddedness on competitive advantage through the mediating effect of explorative and exploitative learning. Consistent with the findings of existing literature25,27,114, we found that organizational learning is an important mediator of competitive advantage. For example, Bending et al.32 suggested that organizational learning played a mediating role between network relations and organizational performance. In addition, we found that relational embeddedness promotes organizational learning ambidexterity, which, as existing research has pointed out, is the effect of the external embeddedness on the firm’s ambidexterity51,52,58. Furthermore, our findings indicate that organizational learning ambidexterity is essential for achieving competitive advantage within business ecosystems. This finding agrees with the conclusions of existing literature25,27,32, which assert that organizational learning is essential for improving company performance and competitive advantage. For example, Zhang et al.114 found that enhancing organizational learning, including both exploratory and exploitative learning, contributes to sustainable competitive advantage in Chinese high-tech enterprises.

Thirdly, environmental dynamics exert varying influences on relational embeddedness and organizational learning ambidexterity. This study posits that enterprises inside the digital healthcare ecosystem utilize different learning approaches contingent upon the extent of change in the external environment. For digital healthcare ventures, it is significant to select suitable learning trajectories in varying external contexts, with empirical findings indicating that the beneficial effect of relational embeddedness on organizational learning is shaped by dynamic environments. Our research aligns with earlier studies10,115,116, which demonstrate that in unstable environments, firms must adapt to environmental changes by engaging in exploratory learning. Srikanth and Ungureanu117 claimed that firms should explore more in more dynamic environments. When they explore, they should always choose options further away from their status quo. In contrast, under relatively stable environmental conditions where consumers prioritize product availability and usability over diversity and innovation, firms have diminished incentives to innovate and face reduced risks of competitive activity. As argued by Gong et al.118, the stable environment prefers organizations with a practice strategy of high exploitation and low exploration. However, there are different perspectives in existing studies. Some scholars argue that in a fast-changing environment, some firms may still benefit from exploitative learning77,119,120,121. For example, Schulze and Dada74 suggested that high technology turbulence supports the exploitation of existing competitive advantages. When the technological environment in a given industry is considered to be turbulent or even hostile, firms might foster internal R&D activities or strategic entrepreneurship activities to concentrate on their core competencies74. Posen and Levinthal119 posited that in the condition of environmental dynamics, the appropriate response to environmental change is a renewed focus on exploiting existing knowledge and opportunities.

We suggest that most digital healthcare ventures face the “liability of newness and smallness,” and these firms face great difficulty in pursuing high levels of exploration and exploitation simultaneously due to resource constraints in an environment of high dynamics2. It might be the case that in dynamic environments, digital healthcare ventures with limited resources switch from a dual focus (i.e., a focus on both exploration and exploitation) to a more exploration-related focus. This is because in dynamic environments, it is difficult for firms to respond to competition through exploitation-based learning with their existing knowledge, skills, and related processes10. They are likely to fall behind because they become consistently better at performing routines that are decreasingly valued by the environment45. Firms sticking to and refining existing experiences can lead to company losses as the value of these capabilities gradually diminishes. Although exploitation can ensure system effectiveness and stable cash flow, it can also lead to firms being driven out of the market at this point in time. Our research advances the work of March66 by showing that firms do not necessarily focus on dual application in all cases but rather carefully adjust their strategies and operations to the impact of their environment. In addition, the differentiated impact of factors such as firm size and resource constraints on exploratory and exploitative learning in dynamic environments deserves further discussion. As a frequently examined variable, firm size has been the subject of a great deal of ambidextrous learning122. Large firms are often able to provide important resource support for the successful implementation of innovation activities, such as investing more capital, human resources, and materials. Compared to large firms, small firms usually lack the resources and hierarchical administration systems that could help them manage their contradictory knowledge processes, in turn affecting the attainment of ambidexterity123. Generally speaking, when a firm has abundant resources, creative thinking and improvisation increase; conversely, when resources are scarce, firms are more strict in terms of cost and risk control, including management, financial and technical risks, which is not conducive to innovation activities, especially exploratory innovation124. According to Cao et al.125, firm size is indicative of the resources a firm has at its immediate disposal. A relative abundance of available resources may provide larger firms with a buffer that mitigates the effects of risk and shocks. Therefore, the size of a firm reflects the abundance of its resources to a certain extent. The larger the firm, the fewer resource constraints it faces, and the more conducive it is to achieving ambidextrous capabilities122.

Theoretical implications

This study adds to our understanding of theories by using embeddedness theory, organizational learning theory, and knowledge management theory to look closely at the ways that digital healthcare ventures can gain a competitive advantage in the digital economy. First, this research enhances the embeddedness theory in digital healthcare by analyzing the relationship between relational embeddedness and competitive advantage. The digital economy has fostered cooperative relationships involving numerous healthcare enterprises, healthcare providers, pharmaceutical companies, and regulatory agencies. Previous studies have shown that relational embeddedness can help them gain a better performance20,51. However, existing literature has selected different types of firms, such as energy services providers, logistics services companies, and internationalized small and medium-sized enterprises, to participate in research on how relational embeddedness maintains competitive advantage15,126, but they have not yet been extended to the field of digital healthcare. Furthermore, existing literature tends to focus more on the impact of embeddedness on the innovation performance and enterprise performance of firms41,127,128, with less research on how it affects competitive advantage. Consequently, this study elucidates the mechanisms and rationales of the digital economy that facilitate and enhance relational embeddedness, aiming to induce complex transformations in corporate development. This endeavor seeks to establish a novel theoretical foundation for the implementation of relational embeddedness and to enrich the theoretical exploration of relational embeddedness across diverse contexts.

Second, this study examines the organizational ambidexterity perspective to elucidate the mechanisms by which relational embeddedness influences competitive advantage in digital healthcare. It identifies the mediating roles of explorative and exploitative learning between relational embeddedness and competitive advantage, advances the theoretical framework of organizational learning ambidexterity, and promotes the integration of embeddedness theory, organizational learning theory, and knowledge management theory within digital contexts. This paper integrates these theories to elucidate the mechanisms through which knowledge acquisition and transfer are achieved through relational embeddedness and how these are transformed into competitive advantage through ambidextrous learning. Relational embeddedness has been posited to enhance performance, examining its effects on competitive advantage13,129. While prior research has examined the antecedents of organizational learning ambidexterity to some degree25,27, there are few studies that explore the effects of organizational learning ambidexterity through the lens of relational embeddedness. This study’s incorporation of relational embeddedness addresses the limitations of existing research on the antecedents of organizational learning ambidexterity, thereby integrating embeddedness theory with organizational learning theory and knowledge management theory and advancing the theoretical framework, thus broadening the applicability of organizational learning theory and knowledge management theory. Additionally, existing literature has explored the relationship between relational embeddedness and competitive advantage from different perspectives, such as organizational capabilities and network connections15,126, but has not yet examined this relationship from the perspective of organizational learning. This study explores the relationship between relational embeddedness and competitive advantage from two different types of learning, addressing the limitations of existing research, which tends to focus on static descriptions of existing capabilities from the perspective of organizational capabilities, and refining the explanatory logic of competitive advantage evolution.

Third, this work improves our knowledge about relational embeddedness boundaries in a digital healthcare context, explaining the distinct impacts of relational embeddedness on explorative and exploitative learning amid differing levels of environmental dynamism. Environmental dynamism markedly increases the dependence of explorative learning on relational embeddedness, whereas the dependence of exploitative learning on relational embeddedness is augmented in relatively stable contexts and diminished in dynamic environments. Existing literature has studied the role of relational embeddedness from the perspective of moderating variables such as intermediaries, digitalization, and the complexity and uncertainty in safety risks23,130,131. However, few studies have examined the boundary conditions of the relationship between relational embeddedness and organizational learning ambidexterity within a digital healthcare framework. This study accounts for how digital healthcare ventures foster entrepreneurial activities through the lens of environmental dynamism, contextualizing relational embeddedness and enhancing the comprehension of knowledge-based theory within the digital healthcare ecosystem.

Managerial implications

This paper offers several managerial perspectives for business executives. First, digital healthcare ventures should acknowledge the dual nature of relational embeddedness. Relational embeddedness is a dual-faceted phenomenon, presenting both the advantages of innovation and the disadvantages of homogenization and cognitive lock-in23,24. On the whole, relational embeddedness is advantageous for digital healthcare ventures. The longstanding emphasis on competition has led to a diminished perception of cooperation within enterprises. However, the digital economy has transformed the framework of competitive strategy, making cooperation essential for the survival and growth of businesses. Enterprises that fail to make relationships with stakeholders struggle to adapt to market changes, relying solely on their knowledge and resources. Utilizing relational embeddedness can dissuade enterprises from pursuing excessive economic gains and engaging in “egoistic behavior.” It is vital for digital healthcare ventures to cultivate positive relationships with healthcare organizations, medical professionals, patients, equipment manufacturers, logistics firms, academic institutions, local government bodies, and other relevant stakeholders. By establishing value networks through collaboration via coopetition, alliance, and innovation strategies, and by emphasizing coordination, learning, and communication with partners, the enterprise can cut down on the time and financial expenses associated with resource conversion and acquisition, thereby boosting its competitive edge.

Taking a pioneering force in China’s internet healthcare sector called “WeDoctor Group” as an example, it takes Internet hospitals as the focal point, integrates hospitals, doctors, health insurance agencies, pharmaceutical companies, and other healthcare providers, and creates a “digital health community”132. Integration of “medical treatment, medical insurance, and medicine supply” creates the ecosystem. By establishing a close relationship with all parties and embedding itself in the healthcare industry, the enterprise provides users with online and offline integrated health services, effectively breaking down the barriers of the traditional medical systems and promoting the rapid development of the digital health industry in China133.

Second, it is supposed for digital healthcare ventures to acknowledge the significance of knowledge and make connections to facilitate organizational learning while capitalizing on the beneficial impact of relational embeddedness on learning ambidexterity. Relational embeddedness involves knowledge transfer, and digital healthcare ventures should prioritize learning possibilities in business networks, drawing insights from practical experiences and competitors. They should also acknowledge the distinctions among different learning strategies and justify the use of exploitative and explorative learning to mitigate their respective potential disadvantages. On the one hand, it is necessary for companies to promote employee initiative in learning and experimenting with new technologies, processes, and business models while prioritizing organizational resilience through the establishment of flexible structures and processes, the development of adaptable and innovative employees, and the cultivation of a collaborative and learning-oriented culture. On the other hand, enterprises can leverage existing knowledge resources more effectively, as this form of learning carries a minimal risk of error, thereby enhancing their ability to integrate and utilize the knowledge base shared within the entrepreneurial ecosystem and optimize the use of existing knowledge resources across organizational boundaries. It is necessary to adopt knowledge management strategies to create knowledge-sharing platforms and knowledge bases that provide a rich source of knowledge for organizational learning and also implement resource allocation strategies to rationally guide the direction of organizational learning.

Third, it is supposed for digital healthcare ventures to prioritize the influence of the external entrepreneurial landscape, establish cooperative relationships to react to dynamic environmental shifts in competitive endeavors, and enhance their learning and adaptive competencies. In a relatively volatile market, digital healthcare ventures ought to aggressively pursue exploratory innovation, leveraging insights gained from strategic learning to enter new domains and assume a leadership role. In a relatively stable environment, it is highly desirable if enterprises engage in exploitation-oriented learning, focusing on preserving their existing competitive advantage, reallocating resources acquired through strategic learning, enhancing current capabilities, fostering incremental innovation, and attaining short-term objectives. In highly dynamic environments, enterprises should use adaptive strategies to flexibly adjust their strategies, organizational structures, products, or services to better adapt to changing environments, maintain competitiveness, and achieve sustainable development.

Limitations and future research

This paper examines the impact of relational embeddedness utilized by digital healthcare ventures on their competitive advantage, offering both theoretical and practical insights into current research. This work has certain limitations and offers guidance for future research. Firstly, this study examines the process by which relational embeddedness influences competitive advantage, focusing on digital healthcare ventures. Future research may include data from established organizations for comparative analysis to identify differences. Secondly, this research posits that stakeholders like suppliers, consumers, and competitors remain stable in a firm’s relationships at a specific moment. The interactions between them in digital healthcare ecosystems are dynamic, and different stakeholders’ responsibilities are changing as well. This finding is particularly evident with the rise of cross-border collaboration, where partners may act as competitors in some areas and suppliers in others. Future research may investigate the complex dynamics of partnership interactions and role transitions to draw more significant conclusions. Thirdly, the intensity, amount, and position of relational embeddedness fluctuate in response to strategic changes, intrinsic motivations, and other factors. This study used cross-sectional data to examine the effect of relational embeddedness on competitive advantage, without investigating the influence of different types of embeddedness on learning and competitive advantage. Consequently, future research should involve longitudinal studies or case studies for a more comprehensive analysis. Fourthly, relational embeddedness and structural embeddedness are two dimensions of the social capital of firms. This paper only discusses relational embeddedness, which plays a stronger role in explaining innovation-oriented tasks. Future research can focus on structural embeddedness, which plays a stronger role in explaining execution-oriented tasks.