Entrepreneurial failure is a common occurrence and almost inevitable before many entrepreneurs succeed (Mueller & Shepherd, 2016). Entrepreneurs who reenter entrepreneurial activities following failure account for almost half of all entrepreneurs (Ucbasaran et al., 2006; Westhead & Wright, 2015). Entrepreneurial reentry intention refers to the motivation and tendency to start a new business after experiencing failure (Baù et al., 2017; Hsu et al., 2017). Previous studies have shown that “reentry entrepreneurs” perform well in opportunity identification (Ucbasaran et al., 2009), strategic agility (Vaillant & Lafuente, 2018), innovativeness (Lahiri & Wadhwa, 2021), and venture performance (Gompers et al., 2010), and they have made significant contributions to job creation and economic and societal development (Sarasvathy et al., 2013; Westhead et al., 2005). However, to date, research on the reentry of failed entrepreneurs remains limited (Hsu et al., 2017). Scholars suggest that to develop relevant theoretical insights and their associated practical implications, we need to better understand the antecedents and mechanisms that influence the reentry intention of failed entrepreneurs (Amaral et al., 2011; Baù et al., 2017; Hsu et al., 2017; Lafuente et al., 2019; Simmons et al., 2014).

An abundance of research has focused on the sources of motivation, such as personality traits (Hayward et al., 2010) and psychological factors (Hsu et al., 2017; Singh et al., 2015). Motivation, though a significant driver in entrepreneurial reentry, cannot alone explain the intention to reenter because it’s interwoven with other critical factors such as the entrepreneurs' capability, cognitive resilience, and social dynamics (Costa et al., 2023; Hsu et al., 2017; Liu et al., 2019). It is important to explore whether failed entrepreneurs believe they have the necessary capabilities for reentry (Hsu et al., 2017). For example, to understand how failed entrepreneurs search for new information and leverage existing knowledge and experience to revise their ideas and embark on a fresh entrepreneurial journey (Liu et al., 2019; Ucbasaran et al., 2013). The knowledge, experiences, and skills possessed by individuals that have potential economic value are referred to as human capital (Becker, 1964; Schultz, 1960). Thus, adopting a human capital perspective is well suited to this research direction. To date, most of the research employing such a perspective has focused on understanding how human capital affects initial entrepreneurial intention (Corbett, 2007; Davidsson & Honig, 2003; Ucbasaran et al., 2008); research on its impact on entrepreneurial reentry intention remains scarce and the results mixed. For example, some studies have shown that entrepreneurs with higher levels of education and entrepreneurial skills are more likely to start up a new business despite previous failures (Carrasco, 1999), while others have shown that entrepreneurs with such an educational background and market-transferable skills are likely to switch to paid employment after they experience failure (De Wit & Van Winden, 1989; Gimeno et al., 1997). Scholars argue that when exploring the relationship between human capital and entrepreneurial intention, it is crucial to clarify what kind of human capital is under study (Amaral et al., 2011; Marvel et al., 2016), and it is worthwhile exploring how an emerging human capital, such as the knowledge, experience, and skills generated from the use of digital technology, affects entrepreneurial intention (Youssef et al., 2021) (Chaudhuri et al., 2023; Youssef et al., 2021).

Our study contributes to the research on entrepreneurial reentry by linking it with the literature on digital technology. Emerging literature has studied how the diffusion of digital technology provides an opportunity for individuals to develop digital technology-related capabilities (Zhao et al., 2023). Based on existing research, we define digital technology capability (DTC) as the ability to acquire diverse information and knowledge through digital technology, and identify its potential for improving existing or developing new products, services, and processes (Annarelli et al., 2021; Arkhipova & Bozzoli, 2018; Lenka et al., 2017; Proksch et al., 2021). We argue that a high level of DTC enhances an entrepreneur’s ability to access diverse information, reduce entry barriers, and materialize business ideas effectively. This increases the likelihood of entrepreneurs engaging in new business.

Further, we shed light on the mechanism underlying DTC’s impact on reentry intention by examining the mediating role of entrepreneurial alertness (EA), which is the ability to perceive and identify opportunities ignored by others (Kirzner, 1979). EA was further conceptualized by Tang et al. (2012) as a cognitive capability that involves three components: scanning and search, association and connection, and evaluation and judgment. In our study, we submit that DTC stimulates the development of EA by extending and sensitizing search, triggering the formation of unique associations among various sources of information and business connections, and enhancing the proficiency of opportunity evaluation. We propose that enhanced EA could contribute positively to failed entrepreneurs’ reentry intention, and we test our arguments with a sample of 263 Chinese entrepreneurs who have experienced at least one entrepreneurial failure.

This study also provides a nuanced understanding of the associated contingencies by introducing and testing social costs and entrepreneurial resilience as critical moderators for the mediated relationship. The social costs incurred by failure can lead to negative impacts on entrepreneurs’ personal and professional relationships (Simmons et al., 2014; Singh et al., 2007; Ucbasaran et al., 2013). Given that the shame and stigma of entrepreneurial failure are particularly strong in the East Asian context (Begley & Tan, 2001; Damaraju et al., 2021), we examine how high social costs of failure might limit entrepreneurs’ potential to convert DTC into EA and thereby affect reentry intention. The ability of individuals to overcome adversity and persist in the entrepreneurial process is referred to as entrepreneurial resilience (Korber & McNaughton, 2017). Our study shows that this adaptative and transformative ability enables entrepreneurs to better convert DTC into EA and thus influences reentry intention.

Theoretical background

Failed entrepreneurs, human capital, and reentry intention

Entrepreneurial failure is an unavoidable experience for most entrepreneurs (Headd, 2003; Mueller & Shepherd, 2016). It refers to the cessation of entrepreneurship as a result of not reaching the expected minimum threshold for economic viability (Ucbasaran et al., 2013). In the face of such failure, individuals display differing attitudes. Some regard their failure as “evidence” that they are not suited to entrepreneurship and thus cease such efforts. Their failure leads to self-doubt, negative emotions, and risk avoidance, impairing their motivation for reentry (Lafuente et al., 2019; Shepherd et al., 2009). By contrast, others regard failure as a learning opportunity and a step toward future success. They overcome negative emotions and use the information about why their business failed as feedback to revise their existing knowledge and embark on fresh business (Shepherd, 2003; Yamakawa et al., 2015).

Prior research has provided insights into the antecedents of entrepreneurial intention, with research drawing from human capital theory making an important contribution (Marvel et al., 2016). However, research on the impact of human capital on entrepreneurial reentry intention provides mixed evidence. Some studies suggest that high levels of general and specific human capital promote entrepreneurial reentry intention. For example, individuals with high levels of general human capital, such as better education, are endowed with better skills and capabilities and thus have a higher probability of entrepreneurial entry than less well-educated ones (Amaral et al., 2011; Carrasco, 1999; Davidsson & Honig, 2003). Individuals with specific human capital, such as market knowledge and technology capability, improve the quantity and quality of their opportunity identification, thereby stimulating additional entrepreneurial activity (Corbett, 2007; Ucbasaran et al., 2008).

By comparison, other studies suggest that both general and specific human capital may have negative effects on entrepreneurial reentry. For example, after experiencing entrepreneurial failure, compared to individuals with lower education levels, those with higher levels are more likely to switch to paid employment with attractive salaries and better career development options than enter a new entrepreneurial project (De Wit & Van Winden, 1989; Gimeno et al., 1997). Amaral et al. (2011) demonstrated that negative founding experiences decrease the likelihood of reentry. These mixed findings regarding the impact of human capital on reentry intention suggest the need for a closer specification of the form(s) of human capital involved (Amaral et al., 2011). In the following section, we argue that DTC, as a new form of specific human capital in the digital age, deserves consideration (Setia et al., 2013; Zhou & Wu, 2010).

The notion and characteristics of DTC

Technology capability, encompassing the utilization of various technologies to establish organizational infrastructure, create products and services, and facilitate decision-making and business operations (Levallet & Chan, 2018; Sambamurthy et al., 2003), has been recognized as a specific form of human capital crucial for enterprise development (Proksch et al., 2021; Sambamurthy et al., 2003). When integrated into daily work, this capability becomes indispensable. Consequently, with the advent of digital technology, the significance of digital technology-related capabilities for venture development has grown, attracting increasing scholarly attention (Annarelli et al., 2021; Arkhipova & Bozzoli, 2018; Lenka et al., 2017; Proksch et al., 2021).

To advance this area of study, our definition of DTC incorporates research on digital capability and digitalization capability. Digital capability pertains to the use of digital technology, emphasizing foundational skills like digital literacy, data management, and adaptability to technological changes (Arkhipova & Bozzoli, 2018; Proksch et al., 2021). Digitalization capability, on the other hand, focuses on integrating digital assets with business resources, stressing the efficient implementation of digital technologies to optimize business processes (Annarelli et al., 2021; Lenka et al., 2017). Building upon the works of Annarelli et al. (2021), Arkhipova and Bozzoli (2018), Lenka et al. (2017), and Proksch et al. (2021), we define DTC as the ability to acquire diverse information and knowledge through digital technology, and identify its potential for improving existing or developing new products, services, and processes.

While DTC can be developed at both individual and firm levels, our study focuses on DTC as individual human capital in the digital era. Individual-level DTC encompasses skills and knowledge in digital technologies, such as computer programming and data analysis (Arkhipova & Bozzoli, 2018; Proksch et al., 2021). This capability is essential for job efficiency, market competitiveness, and adapting to digital shifts. DTC, as a novel and specialized form of human capital, offers distinct advantages compared to traditional forms such as general education and management skills. This is particularly evident in the digital economy era, where proficiency in digital technology is not just advantageous, but essential. Traditional education and managerial experience, although valuable, may not provide the specific digital skills required in today’s digital economy. Broad education often lacks in-depth knowledge of digital technology, while managerial experience may lack technological proficiency unless in a tech-oriented environment. In the rapidly evolving digital landscape, these traditional forms of human capital may fall short in navigating digital landscapes and seizing tech-driven opportunities. On the other hand, DTC is specifically tailored for this era. A high level of DTC empowers entrepreneurs to effectively leverage digital technologies for innovative business models, optimized operations, and compelling value propositions. More specifically, the communicability, associability, and generativity of digital technology generate rich and diverse information and knowledge, enabling entrepreneurs to develop the capability to “abstract and generalize across contexts, to recognize patterns and build relationships between different situations and events” (Cope, 2005). As examples, the communicability of digital technology advances interaction between diverse products/services (Günther et al., 2017; Yoo, 2010), its associability enables continual interactions with business partners and customers (Günther et al., 2017; Yoo, 2010) from whom entrepreneurs can obtain relevant knowledge and information, and its generativity redounds to a new combination and organization of digital technology elements and industry knowledge (Cennamo & Santaló, 2019), which can conduce to grasping fresh market demand (Amit & Han, 2017; Nambisan, 2017). The DTC accumulated on the basis of these features enhances entrepreneurs’ competence when it comes to leveraging valuable, idiosyncratic, and inimitable sources for value creation and decision-making.

Entrepreneurial alertness

According to Kirzner (1973, 1979), EA is the ability to perceive new opportunities ignored by others. Building on Kirzner’s (1973, 1979) work, scholars describe EA as a unique cognitive ability that involves searching for information, connecting prior knowledge and new information, and deploying resources to evaluate new ideas (Gaglio & Katz, 2001; Tang et al., 2012). Thus, it is a crucial component of opportunity identification (Gaglio & Katz, 2001; Tang, 2008) but does not involve the actual launching of a venture, only whether an opportunity exists (Tang et al., 2012).

On the basis of Kirzner’s (1973, 1979) work and that of McMullen and Shepherd (2006), Tang et al. (2012) proposed that EA includes three different but complementary components: the first is “scanning and search”, referring to continuous scanning and searching for information and changes that might suggest new opportunities; the second is “association and connection”, which refers to creating valuable solutions and identifying new opportunities by unifying disparate information; the third is “evaluation and judgment”, referring to assessment of whether the identified opportunities have the potential to create value. The first component provides the necessary information input; the second and third make creative links and identify promising business opportunities.

A good amount of research has explored the antecedents and consequences of EA. In terms of antecedents, scholars have generally adopted two approaches: an environment-centered one and a person-centered one (Lanivich et al., 2022; Sharma, 2018). In the first of these, scholars examine how environmental factors affect the level of EA (Tang, 2008), while in the second they study how personality traits (Obschonka et al., 2017), psychological capital (Levasseur et al., 2022), human capital (Unger et al., 2011), and social capital (Ardichvili et al., 2003) affect individuals’ responses to information and opportunities. Research shows that general human capital such as education (Westhead & Solesvik, 2016), specific human capital such as technology and market knowledge (Ma & Huang, 2016), and domain-specific experience (Valliere, 2013) all make positive contributions to EA.

In terms of the consequences of EA, researchers have found that it contributes positively to people’s innovativeness (Jiao et al., 2014) as well as organizational performance (Adomako et al., 2018) and success (Amato et al., 2017), but the most important consequences identified to date are EA’s contributions to initial entrepreneurial intention (Arenius & Minniti, 2005; Neneh, 2019) and the number of ideas and opportunities identified (Ardichvili et al., 2003; Gaimon & Bailey, 2013).

Overall, extant research has mainly studied how general human capital and some specific human capital affect EA and how EA affects opportunity identification and initial entrepreneurial intention. Despite the existing insights, the impact of specific human capital like DTC on EA, as well as the relationship between EA and reentry intention, remains unknown. Recognizing this significant gap and the growing importance of DTC for information gathering and analysis, we adopt a person-centered approach to examine how EA mediates the link between DTC and reentry intention, while also investigating the contingencies that influence this relationship. Our full theoretical model is shown in Fig. 1, and we develop our accompanying hypotheses below.

Fig. 1
figure 1

Theoretical model

Full size image

Hypotheses development

DTC and entrepreneurial reentry intention

Based on the unique characteristics of DTC discussed above, we propose a positive relationship with entrepreneurial reentry intention. First, a high level of DTC enables entrepreneurs to acquire diverse and comprehensive information while also adapting to dynamic technological and market conditions, thus enhancing their likelihood of starting a new business despite previous failures. Entrepreneurs may reflect on past failures and identify novel opportunities by connecting previous entrepreneurial knowledge/experience with digital technology information/knowledge (Zahra et al., 2023). Moreover, entrepreneurs equipped with an extensive knowledge base can effectively learn from past failures, developing an improved understanding of technological requirements and market demands. By fostering technology intelligence and market agility (Mathiassen & Pries-Heje, 2006; Sambamurthy et al., 2003; Zaheer & Zaheer, 1997), they can actively seek out opportunities to test new business ideas, thus bolstering their reentry intention.

Second, a high level of DTC can reduce reentry barriers for entrepreneurs, enabling them to pursue new entrepreneurial opportunities with renewed confidence. Market research prior to initiating entrepreneurial activities is typically a costly and time-consuming but essential process (Ibeh et al., 2019). Entrepreneurs with a high level of DTC can utilize digital tools and platforms for swift and inexpensive collection of business data (Henfridsson et al., 2014). This helps them efficiently conduct market research, predict industry trends, and analyze competitive landscapes, thereby generating more targeted business ideas (Rizvanović et al., 2023). Furthermore, entrepreneurs with a high level of DTC can leverage cost-effective digital marketing technologies such as social media, search engines, and e-commerce platforms to reach a larger customer base (Chen & Guo, 2022). This enables them to identify untapped customer needs with minimal financial costs, thus providing crucial support for financially constrained failed entrepreneurs to reenter entrepreneurial activities.

Third, a high level of DTC enhances the potential of entrepreneurs to materialize business ideas and improve their market feasibility, making them more prone to start their own businesses again. Research has demonstrated that many aspiring entrepreneurs have a wealth of ideas but struggle to convert them into fully-fledged and profitable products/services due to inadequate collection and analysis of customer feedback (Blank, 2013). Entrepreneurs with high levels of DTC can gather and evaluate customer feedback, particularly from previous unsuccessful attempts, which in turn enhances their learning and reflective experiences. Consequently, this empowers entrepreneurs to improve existing products/services and create novel ones with increased confidence (Birch-Jensen et al., 2020; Guinan et al., 2019).

Although entrepreneurs with a high level of DTC may be more likely to start digital technology ventures, we argue that DTC also enables other types of venture. The increasing omnipresence of digital technology facilitates the capture, modeling, and prediction of data in general. For example, DTC enables individuals to access diverse channels and information and apply digital technologies to better understand core components of business models, such as value propositions, market segments, customer feedback, and revenue streams (Vial, 2021; Warner & Wäger, 2019). Every occasion of access and interaction with digital information provides an opportunity for the individual to connect and update existing ideas and knowledge so that they may learn better from failure, monitor market and technology dynamics, and thus improve the desirability and feasibility of new business ideas. Taken together, despite previous failures, entrepreneurs with a high level of DTC have the potential advantages over other entrepreneurs in gathering rich and timely information, and better materializing products and services. Therefore, we propose:

  • H1: DTC is positively related to entrepreneurial reentry intention.

The mediating role of EA

We propose that EA plays a mediating role in the positive relationship between DTC and entrepreneurial reentry intention. While both DTC and EA are shaped by an individual’s ability to connect information, knowledge, and expertise, they differ in key aspects. DTC focuses on digital-related capabilities acquired through information accumulation, technological trend identification, and the application of technology to products and services (Fellnhofer, 2022; Hund et al., 2021). On the other hand, EA emphasizes the interpretation and evaluation of potential opportunities based on market feasibility and economic prospects, considering uncertainty factors (Korsgaard et al., 2016; Lanivich et al., 2022; McMullen & Shepherd, 2006; Tumasjan et al., 2013). In essence, individuals with DTC possess the necessary information and technological advantages, but may not transcend this capability and develop reentry intention without the presence of EA (Fellnhofer, 2022; Si et al., 2023). By shedding light on the mediating role of EA, we can gain a deeper understanding of how DTC influences entrepreneurial reentry intention.

First, DTC stimulates the development of cognitive capability by extending and sensitizing scanning and search. Entrepreneurs who possess a high level of DTC can build a sensory store with a vast array of new information and knowledge provided by digital technology and its integration with prior knowledge, experience, and learning (Busenitz, 1996; Fellnhofer, 2022; Hund et al., 2021; Reed, 2004). This sensory store helps build “unique preparedness” (Kaish & Gilad, 1991) in terms of consistent scanning of the dynamic technological environment and the sensitization of entrepreneurs to its (technological) connotations in relation to potential business ideas (Lanivich et al., 2022; Scuotto et al., 2022). For example, entrepreneurs could scan social media for information on customer demands (Chuang, 2020) and utilize various digital channels to mitigate information asymmetry, enabling them to uncover information overlooked by others (Chatterjee et al., 2022; Kahle et al., 2020).

Second, DTC triggers the formation of unique connections among various sources of information and associations with other market entities to explore business opportunities. A high level of DTC enables individuals to access new and diverse information, which allows them to diverge from cognitive routines and perceive associations between seemingly unrelated events (Tang et al., 2012). The recursive relationship between search and association inspires individuals to see a bigger picture, build concepts, and recognize meaningful patterns in relation to business opportunities (Baron, 2006; Grégoire et al., 2011). For instance, by connecting diversified information and knowledge encompassing government policies, technological innovations, and market demands, entrepreneurs with a high level of DTC can identify opportunities that have been ignored by others (Gielnik et al., 2012; Ozgen & Baron, 2007). Moreover, DTC empowers entrepreneurs to interact more effectively with suppliers, customers, and fellow entrepreneurs (Yli-Renko et al., 2001). During this process, information and feedback are exchanged and new ideas are generated to satisfy the demands of these market entities (Audia & Goncalo, 2007; Gaimon & Bailey, 2013).

Third, DTC enhances the proficiency of opportunity evaluation and judgment. As a form of dynamic capability (Chakravarty et al., 2013), DTC helps entrepreneurs master the latest information and knowledge and discern multiple opportunities in a dynamic environment (Gaimon & Bailey, 2013). With this acumen, entrepreneurs filter out non-essential information, envision the future, and make cost–benefit analyses, thereby selecting the opportunity with the greatest potential to create value (Kirzner, 1985; Tang et al., 2012, 2021). For example, research shows that entrepreneurs with a high level of DTC benchmark similar products and services to evaluate opportunities on the basis of economic value, novelty, and property rights protection (Gaimon & Bailey, 2013; Shane, 2001).

Considering the amplified influence of DTC on EA, we postulate that failed entrepreneurs with higher levels of EA are more inclined to embark on new business ventures. Extensive research has consistently demonstrated the positive relationship between EA and the intention of individuals to initiate their first business endeavor (Ardichvili et al., 2003; Baron, 2007; Tang et al., 2012). We contend that enhanced EA could positively contribute to the reentry intention of failed entrepreneurs through the identification of “first-person opportunity”. These are potential opportunities that a prospective entrepreneur decides are for him or her and for which they are willing to bear the associated uncertainty (McMullen & Shepherd, 2006). We suggest that failed entrepreneurs with a high level of EA are keen to discover this type of opportunity. They learn from their failures, transfer existing knowledge and experience, and leverage search, association, and evaluation, thereby enhancing their confidence as to the feasibility and profitability of starting a business again. Based on the arguments presented above, we propose:

  • H2: The positive relationship between DTC and entrepreneurial reentry intention is mediated by EA.

The moderating effect of social costs

The social costs caused by entrepreneurial failure can result in negative impacts on entrepreneurs’ personal relationships with close ones and their professional relationships with suppliers, investors, employees, and business partners (Simmons et al., 2014; Singh et al., 2007; Ucbasaran et al., 2013). Social costs also lead to reputational damage and social status decline (Cardon et al., 2011; Shepherd, 2003). Begley and Tan (2001) show that the shame of entrepreneurial failure is stronger in East Asian countries than in Anglo-Saxon ones. The accompanying loss of face and reputational damage causes failed Chinese entrepreneurs to seek to avoid further shame from letting down their family and/or business partners (Begley & Tan, 2001). Other scholars find that the stigmatization of failure discourages entrepreneurial risk-taking in Western countries as well (Damaraju et al., 2021; Simmons et al., 2014; Singh et al., 2015). For example, Simmons et al. (2014) used data from the Global Entrepreneurship Monitor and found that in Western countries with higher levels of failure stigmatization (such as Spain, Hungary, and Iceland), failed entrepreneurs are less likely to reenter entrepreneurial activities.

Thus, we propose that the perception of higher social costs of failure will limit entrepreneurs’ potential to convert DTC into EA. When entrepreneurs are very concerned about the negative impact of failure on their social and professional relationships and the damage to their reputations, they prolong their recovery time (Shepherd, 2003) and are less likely to proactively search for and sensitize information (Cope, 2011; Shepherd & Haynie, 2011) for potential business ideas. Furthermore, fearful of losing status and of negative reactions from others, they are less likely to exchange information and ideas with other market entities, such as investors, suppliers, and business partners, in pursuit of potential business ideas. Last but not least, worrying about others’ opinions can result in a loss of confidence in their own entrepreneurial judgment and thus missed opportunities to develop nascent business ideas. Given the mediation mechanism proposed in H2, we hypothesize that:

  • H3: Social costs negatively moderate the indirect relationship between DTC and entrepreneurial reentry intention via EA, such that the mediation relationship is weakened when the level of social costs is high.

The moderating effect of entrepreneurial resilience

Entrepreneurial resilience refers to the psychological ability of individuals to overcome strong entrepreneurially-related challenges and persist in the entrepreneurial process in the face of adversity and unexpected results (Korber & McNaughton, 2017). It is an adaptive and transformative capability that enables entrepreneurs to learn from failure, reduce its negative impact, and find flexible solutions to move forward (Haynie et al., 2012; Korber & McNaughton, 2017; Lengnick-Hall & Beck, 2005).

We propose that a high level of entrepreneurial resilience will enable failed entrepreneurs to convert DTC to EA. First, these entrepreneurs are equipped to maintain a positive attitude amidst adversity, driving their persistence to overcome challenges (Luthans et al., 2006), and enabling proactive use of the new knowledge and information brought by digital technology to recognize entrepreneurial opportunities. Second, resilient entrepreneurs can recalibrate and reframe their perception of failure, influencing their initiative to seek new entrepreneurial opportunities. They navigate through challenges, drawing connections from disparate information and leveraging business networks to identify potential business opportunities. Lastly, these entrepreneurs are likely to build confidence in making informed judgments and identifying opportunities despite uncertainties. Their resilience manifests in resisting negative emotions and fostering self-recovery (Masten, 2001). On the basis of the mediation mechanism proposed in H2, we hypothesize that:

  • H4: Entrepreneurial resilience positively moderates the indirect relationship between DTC and entrepreneurial reentry intention via EA, such that the mediation relationship is stronger when the level of entrepreneurial resilience is high.

Methodology

Sample and data collection

To select respondents for our survey, we employed the following criteria: first, respondents had to be individuals who had experienced at least one entrepreneurial failure in the preceding six months (e.g., Jenkins et al., 2014; Schermuly et al., 2021), and entrepreneurial failure defined as the termination of entrepreneurial processes and activities, including bankruptcy, closure, and ownership termination of a new venture (Jenkins & McKelvie, 2016; Khelil, 2016); second, respondents had to be individuals who were not running a new venture at the time of the survey; third, respondents were selected from different regions and industries in China to ensure representativeness. We designed a questionnaire based on measurements established in prior studies and ensured that the scale as translated from English to Chinese was accurate and applicable to our research context (Craig & Douglas, 2006).

The survey was conducted between August and October 2022 with the assistance of a professional survey agency. We adopted a random selection method and identified 650 entrepreneurs from a list provided by the survey agency as our initial sample frame. The sample from China was across different regions and multiple industries. A total of 305 questionnaires were returned, with a response rate of 46.92%. After excluding unqualified and incomplete surveys, 263 responses remained with a valid response rate of 40.46%. Table 1 presents detailed sample characteristics.

Table 1 Sample characteristics (N = 263)
Full size table

Measures

We adopted a seven‑point Likert scale (1 = strongly disagree; 7 = strongly agree) to measure the main variables. Table 2 lists the items used to measure these variables, which we elaborate on below.

Table 2 The reliability and validity of measures
Full size table

Dependent variable

To measure entrepreneurial reentry intention, we adopted and modified Liñán and Chen’s (2009) six-item measure of entrepreneurial intention and Hsu et al.’s (2017) measurement of subsequent entrepreneurial intention following business exit. We added a clause (i.e., “Although I experienced this entrepreneurial failure”) before each of the original items to fit our research context. These items comprehensively capture whether respondents have the motivation to start a new business venture and become an entrepreneur again after experiencing failure.

Independent variable

We measured DTC using a five-item scale, adapted from Zhou and Wu (2010). The original scale required respondents to assess their technological capability relative to their competitors. To retain the items’ original meaning while fitting our context, we asked respondents to evaluate their own DTC. The scale includes five items: obtaining digital technology information; identifying new digital technology opportunities; coping with digital transformation; mastering the most advanced digital technology; and developing new products/services/processes using digital technology.

Mediator variable

EA was assessed with a 13-item scale developed by Tang et al. (2012), covering three components: scanning and search, association and connection, and evaluation and judgment.

Moderator variables

A four-item scale was used to measure social costs (Ucbasaran et al., 2013). This evaluates the impact of entrepreneurial failure on respondents’ relationships with family, investors, and other stakeholders and their concerns about the stigma attached to their future development.

A four-item scale was also used to measure entrepreneurial resilience (Yao et al., 2021). This evaluates respondents’ abilities to find solutions, control reactions, and maintain a positive attitude in the face of setbacks or difficulties.

Control variables

We controlled variables at three different levels. First, individual-level variables were controlled, including demographic variables (i.e., gender, age, education level, and marital status) (Baù et al., 2017), personality traits (i.e., risk-taking, inventiveness, self-confidence, proactivity, desire for competition, and optimism) (Walter & Block, 2016), and prior experiences (i.e., work experience, venture experience, the number of failures, the nature of failure, internal attribution of failure) (Lafuente et al., 2019; Morgan & Anokhin, 2020; Yamakawa et al., 2015). Next, family-level variables were controlled, including family entrepreneurial experience, family entrepreneurial attitude, and annual household income (Athayde, 2009; Baù et al., 2017; Kautonen et al., 2011). Finally, industrial- and environmental-level variables were controlled, including previous industry type, previous location, and COVID-19 impact (Bergenholtz et al., 2023; Wu et al., 2022).

Reliability and validity

Table 2 also presents the reliability and validity indicators for all of the main variables. First, confirmatory factor analysis (CFA) was used to test the fit between the model and the data: χ2 = 657.524, degrees of freedom (df) = 437, p = .000, standardized root mean square residual (SRMR) = .047, root mean square error of approximation (RMSEA) = .044, comparative fit index (CFI) = .950, incremental fit index (IFI) = .951, Tucker–Lewis index (TLI) = .944. Next, the standardized factor loading of all items was shown to be higher than .60, which indicates that the relationship between each variable and its corresponding items was acceptable. Similarly, the average variance extracted (AVE) value for each variable was found to exceed .50, and the square roots of these AVE values were found to be greater than the Pearson correlation coefficients between them and other variables, showing that the discriminant validity of the scales was acceptable (Fornell & Larcker, 1981). Further, the composite reliability (CR) value of all variables was higher than .70, indicating that the scales used offer good convergent validity (Hair et al., 2010). Finally, Cronbach’s alpha coefficient for each variable was found to exceed .70, indicating that the internal consistency of the scales was in line with requirements.

Common method bias

Because self-reported data from a single source may result in common method bias (CMB; Podsakoff et al., 2003), several precautions were taken to reduce such bias: we used a pilot survey and expert feedback to improve the comprehensibility and clarity of the questionnaire, and then we randomly arranged the position of the items therein to reduce respondent speculation about the study’s purpose. Finally, to minimize any social-desirability bias, we informed respondents that data collection was anonymous.

In addition to the above precautions, Harman’s single-factor test was used to assess CMB. The results showed that there were seven factors with eigenvalues greater than 1.0, accounting for 67.55% of the total variance. Moreover, the leading factor accounted for 29.71% of the variance, significantly lower than the recommended threshold of 50% (Podsakoff & Organ, 1986), showing that no single factor can explain a majority of the variance. Thus, CMB is not of significant concern in our study.

Results

Hypothesis testing

We checked multicollinearity via variance inflation factors (VIFs). The results showed that the maximum value of any VIF was 2.13, well below the threshold of 10, indicating that multicollinearity is also not a concern in this study. Table 3 presents the means, standard deviations, minimum and maximum values, and Pearson correlation matrix for all variables.

Table 3 Descriptive statistics and correlations
Full size table

We examined the hypotheses via ordinary least-squares (OLS) regression analysis, the results of which are presented in Table 4. H1 predicts that failed entrepreneurs’ DTC is positively correlated with reentry intention. As shown in Model 1 of Table 4, the coefficient was significant and its direction was consistent with that expected (β = .187, p < .001). Thus, H1 was supported.

Table 4 Results of OLS regression analysis
Full size table

Mediating effect testing

H2 posits that EA mediates the relationship between DTC and reentry intention. EA was added to Model 1 to create Model 2. The results showed that the direct effect of DTC on reentry intention was weakened after EA was incorporated into the model. The mediating effect of EA was formally examined using 5,000 bootstrap samples in the SPSS PROCESS macro, and bias-corrected 95% confidence intervals (CIs) were derived. The results showed that the indirect effect of DTC on reentry intention through EA was significant and in the expected direction (βindirect effect = .110, 95% CIs = .045 to .185), thus verifying H2.

Moderated mediation effect testing

H3 and H4 respectively propose that social costs and entrepreneurial resilience moderate the indirect effect of DTC on reentry intention via EA. Before examining moderated mediation effects, we first performed hierarchical regression analysis (Models 4 and 5 in Table 4) to test the moderating effects of social costs and entrepreneurial resilience. The results showed that social costs weakened the positive effect of DTC on EA (β = -.062, p < .01), while entrepreneurial resilience significantly strengthened this relationship (β = .186, p < .001). In addition, we plotted the interaction effect diagram. As illustrated in Fig. 2a, a simple slope analysis showed that the positive relationship between DTC and EA was significant at low levels of social costs (β = .395, p < .001) and decreased at high levels (β = .234, p < .001). A similar simple slope analysis, illustrated in Fig. 2b, showed that the positive effect of DTC on EA was significant at low levels of entrepreneurial resilience (β = .211, p < .001) and increased at high levels (β = .412, p < .001).

Fig. 2
figure 2

Interaction of digital technology capability and entrepreneurial alertness according to: a) social costs; b) entrepreneurial resilience

Full size image

To further test moderated mediation effects, we calculated the index of moderated mediation using the bootstrap approach with 5000 bootstrap samples and bias-corrected 95% CIs (Hayes, 2015). As shown in Table 5, under the moderation of social costs, the mediating effect of EA was significant (index = -.019, 95% CIs = -.041 to -.007), and with the moderation of entrepreneurial resilience, the mediating role of EA was also significant (index = .057, 95% CIs = .028 to .100). Moreover, we explored these relationships by estimating the indirect effects of social costs and entrepreneurial resilience at three different levels: mean, one standard deviation below the mean, and one standard deviation above the mean (Preacher et al., 2007). The results in Table 5 showed that as social costs increased, the indirect effect of DTC on reentry intention via EA gradually decreased. In contrast, as entrepreneurial resilience increased, the indirect effect of DTC on reentry intention via EA gradually increased. Thus, hypotheses H3 and H4 have been verified.

Table 5 Results of moderated mediation effect
Full size table

Robustness checks

We performed additional analyses to test the robustness of our results. A potential concern in this study is endogeneity; unobservable factors may affect reentry intention despite us controlling for multiple variables. In addition, there may be reciprocal causation between DTC and reentry intention, which may also lead to endogeneity. We adopted digital technology training as an instrumental variable to mitigate the endogeneity concern. Because digital technology training is beneficial in improving an individual’s capability in learning and using digital technology (Chen et al., 2021), we can infer that such training is closely related to DTC.

We used a four-item scale adapted from Walter and Block (2016) to assess individuals’ digital technology training. Cronbach’s alpha coefficient for this variable was .828. Table 6 presents the estimation results of the two-stage least-squares (2SLS) method for the instrumental variable. These showed that there was a significant positive correlation between digital technology training and DTC (β = .421, p < .001). Moreover, to date, no research has shown digital technology training to have a direct effect on reentry intention, which ensures, at least to some extent, the exogeneity of the instrumental variable. According to the test, the corresponding p-value of the Durbin–Wu–Hausman test is greater than .05, indicating that the instrumental variable is exogenous. In addition, the results of the weak instrumental variable test showed that the minimum eigenvalue statistic in the first stage is 86.421, which is higher than 16.38 (i.e., a critical value corresponding to 10%) in the 2SLS size of the nominal 5% Wald test. This shows that the weak instrumental variable problem is not a threat and it is reasonable, therefore, to use digital technology training as an instrumental variable. Finally, after the treatment of endogeneity with the instrumental variable method, the second-stage estimates showed a significant positive correlation between DTC and reentry intention (β = .313, p < .001). These additional analyses reassured us that endogeneity was of minimal concern for our results.

Table 6 The estimation results for the instrumental variable by 2SLS
Full size table

Discussion

Theoretical implications

For many entrepreneurs, failure is a common occurrence and a stepping stone toward entrepreneurial reentry and development. However, reentry is not guaranteed if the entrepreneur lacks motivation and capability. To better understand why some entrepreneurs bounce back from failure while others do not, our study explores the role of DTC in reentry intention. Through a survey study of failed Chinese entrepreneurs, we offer the initial evidence that entrepreneurs with a high level of DTC are more likely to start up new businesses despite previous failures. Further, we find evidence for the mediating role of EA in bridging this positive relationship. In addition, our research shows that entrepreneurs’ perceptions of high social costs negatively impact their reentry intention while entrepreneurial resilience positively impacts them.

Our study makes the following contributions to the research on reentry intention, EA, and digital technology. First, by exploring the role of DTC as a specific human capital, our study investigates how digital-related capability affects entrepreneurial reentry intention. Prior research has yielded inconsistent findings regarding whether failed entrepreneurs with high levels of general human capital (e.g., education) and specific human capital (e.g., market and industrial knowledge, management experience) are inclined to embark on a new entrepreneurial journey or seek paid employment instead (Amaral et al., 2011; Carrasco, 1999; De Wit & Van Winden, 1989; Gimeno et al., 1997). By synthesizing the literature on digital technology and entrepreneurial reentry, our study views DTC as a form of specific human capital that enables entrepreneurs to develop capabilities in acquiring information and knowledge, as well as spotting the potential for innovation through the utilization of digital technology (Annarelli et al., 2021; Arkhipova & Bozzoli, 2018; Lenka et al., 2017; Proksch et al., 2021). Our study demonstrates that entrepreneurs with a high level of DTC are keen to reengage in entrepreneurial activities due to their ability to access diverse information, reduce reentry barriers, and enhance their potential to materialize business ideas. We contend that DTC, as an emerging and distinctive form of human capital, will play an increasingly significant role in reentry intention.

Our second contribution delves into the underlying mechanism that explains the relationship between DTC and reentry intention by elucidating the mediating role of EA, contributing to understanding the mechanisms through which DTC influences entrepreneurial decision-making processes. Previous studies have identified various psychological, personality, and environmental factors as important determinants of EA (e.g., Levasseur et al., 2022; Obschonka et al., 2017; Tang, 2008). By situating EA in the context of the digital age, we have investigated why DTC is also conducive to EA. Furthermore, moving beyond extant research on the impact of EA on initial entrepreneurial intention (Arenius & Minniti, 2005; Neneh, 2019), our findings demonstrate that enhanced EA also increases the likelihood of reentry intention. We contend that entrepreneurs armed with DTC can leverage past failures, apply their existing knowledge and experience, and leverage search, association, and evaluation processes to bolster their confidence in reentering entrepreneurship. This revelation extends our comprehension of DTC’s role beyond its direct effects and furnishes valuable insights into the mechanism through which DTC influences reentry intention. In essence, by emphasizing the mediating role of EA, we illuminate the cognitive processes and information-processing mechanisms underlying the relationship between DTC and entrepreneurial reentry intention.

Third and finally, we provide a nuanced understanding of the relationship between DTC and reentry intention by focusing on social costs and entrepreneurial resilience as important moderators. Previous research suggests that entrepreneurial failure incurs loss of face, reputational damage, and relational disruption (Begley & Tan, 2001; Damaraju et al., 2021). Our study shows that perceived high social costs constrain entrepreneurs in converting their DTC to EA, leaving them less motivated for reentry, probably because they are less likely to proactively search for information and exchange ideas with other market entities, losing confidence in their own entrepreneurial judgment. In addition, entrepreneurs with high levels of resilience are better equipped to overcome challenges, setbacks, and failures, and are more likely to persist in their entrepreneurial pursuits. This resilience enables them to leverage their DTC to enhance their alertness, recognize opportunities, and maintain their intention to reenter entrepreneurship. Overall, our research adds nuance to the understanding of the interplay between DTC and entrepreneurs’ social context and psychological ability in shaping their reentry intention.

Practical implications

Our study yields several practical implications. First, entrepreneurs can benefit from embracing digitalization and consciously developing their capabilities to search, interpret, and apply digitally related information and knowledge. This is particularly important for entrepreneurs who have encountered failure and adversity because it could help them leverage new information to update knowledge, build connections among disparate information, develop a better understanding of market demands and customer preferences, and thus enhance their alertness to potential new business opportunities. Second, rather than being an inherent personality trait, entrepreneurial resilience is an ability that can be developed. Failed entrepreneurs could do so by developing self-compassion (Engel et al., 2021), learning from failure, developing market and technological capabilities, and receiving entrepreneurial training (Amankwah-Amoah et al., 2022). Lastly, considering that the social costs associated with failure often rely on the degree of societal tolerance for failure and the institutions supporting failure recovery (Cardon et al., 2011; Ucbasaran et al., 2013), it would be beneficial for policymakers to cultivate an environment that is more accepting of failure. To promote a failure-accepting environment, policymakers could offer financial incentives like tax breaks or grants for startups. Streamlined bankruptcy laws can act as safety nets for businesses, facilitating recovery. Additionally, government-backed innovation hubs can be established as spaces for experimentation, acknowledging failure as a facet of innovation. These approaches can aid entrepreneurs in rebounding from failure and overcoming adversity.

Limitations and suggestions for future studies

This preliminary effort to study the relationship between DTC, EA, and entrepreneurial reentry intention does, nevertheless, have several limitations, some of which offer directions for future research. First, it should be noted that the generalizability of our findings might be limited since our research model was tested using a sample of Chinese entrepreneurs. The experience and tolerance of failure, a common occurrence in entrepreneurship, can greatly vary across diverse social and cultural contexts (Cardon et al., 2011), potentially influencing entrepreneurs’ reentry intentions. Furthermore, factors such as variations in national culture, specifically long-term orientation (Lumpkin et al., 2010), entrepreneurship policies (Liu et al., 2019), and bankruptcy systems (Schulz et al., 2021) across different countries can impact an entrepreneur's motivation to reenter in different countries. As such, we suggest future studies to increase their sample size and scope and consider transnational comparative research to validate our findings.

Second, our study employed a cross-sectional design based on a questionnaire survey, which imposes limitations on our ability to explore the dynamic mechanisms between variables and increases the risk of CMB. Entrepreneurship is inherently a dynamic process that evolves over time and in response to various circumstances (Boudreaux et al., 2019). The cross-sectional design can only provide a snapshot of the entrepreneur’s capabilities and intentions, preventing us from deeply understanding the dynamic developmental mechanisms among entrepreneurial capabilities, psychological factors, entrepreneurial alertness, and reentry intentions. Although we have implemented statistical remedies to mitigate CMB, it is crucial to interpret the causal relationships in our study with caution. Thus, future research should aim to collect more objective data from multiple sources and incorporate time dimensions to support longitudinal analyses. Additionally, it is important to note that survey-based measurements of psychological variables are inevitably subject to social desirability bias to some extent (Podsakoff et al., 2003). We encourage future studies to consider the presence of social desirability bias when conducting surveys on psychological variables.

Conclusion

Recognizing that we are now living in a digital economy and appreciating that human capital is crucial to entrepreneurial behavior, our study explores how DTC affects reentry intention after entrepreneurs have experienced failure. Our findings show that there is significant heterogeneity in reentry intention: entrepreneurs equipped with a high level of DTC are more likely to start up new ventures again. We explicate the underlying mechanism by demonstrating the mediating role of EA and expose its contingencies by examining the moderating roles of social costs and entrepreneurial resilience. We hope our study will stimulate further study of this topic.