Introduction

Science and technological innovation are important not only for economic and social development, but also for the long-term competitive advantage of entrepreneurial enterprises. Since 2012, China has adhered to the innovation-driven development strategy and strived to build an innovative economy. According to China’s 14th Five-Year Plan (2021–2025) and 2035 Vision the country will improve the market-oriented mechanism of technological innovation, strengthen the status of enterprises as the principal engine of invention and promote the concentration of various innovation resources among enterprises. In the so-called VUCA (volatility, uncertainty, complexity, and ambiguity) era, the exploration, development and application of technology entrepreneurship are contributing to business innovation and social progress (Troise et al., 2022). The current socioeconomic environment is invariably influenced by the emergence and rapid changes of new technologies, such as social media, big data, cloud computing, artificial intelligence and blockchain, which provide small and medium enterprises (SMEs) with a wide range of entrepreneurial opportunities and cheap access to information (Elia et al., 2020; Rodriguez-Lluesma, García-Ruiz and Pinto-Garay, 2021). It is apparent that in this process, firms are involved in the transformation of critical business operations, which in turn affects their innovation performance (Rippa & Secundo, 2019; Oppong, Singh and Kujur, 2020; Zhang, van Gorp and Kievit, 2022).

Research on technology entrepreneurship can be broadly divided into two categories. The first category focuses on how, why, and when technology entrepreneurship affects the socio-economic development of a region. This, of course, includes social performance at the macro level and firm innovation from the micro perspective. Scholars generally believe that technology entrepreneurship is one of the significant sources of economic growth and contributes significantly to wealth and job creation (Mosey et al., 2017; Gu & Wang, 2022; Mackiewicz & Błoch, 2022). It is argued that innovative entrepreneurship and the development of new technologies contribute to technological change, which facilitates productivity growth and thus the upward shift of national production frontiers (Lafuente et al., 2020).

The second category explores the formation of technology entrepreneurship and interdependence between technology path and small technology firm formation and growth. Previous research has established that external organizations, national entrepreneurial culture, resource environment and regional knowledge production all influence the creation and development of new technology firms (Colino et al., 2014; Seguí-Mas et al., 2018; Hülsbeck & Pickavé, 2014). Furthermore, from an internal organizational perspective, the entrepreneur is a decisive factor influencing technology entrepreneurship, which involves entrepreneurial traits such as education background (Fayolle et al., 2021; Bolzani et al., 2021; Blankesteijn et al., 2021), human capital (Ganotakis et al., 2012; Zane, 2022), and entrepreneurial capability (Reese et al., 2021). As communication technologies and digitalization have evolved, the external environment driving technology entrepreneurship has also changed. Scholars are not only interested in how supportive regulatory and normative environmental conditions affect the configuration of technology entrepreneurship initiatives but also began to explore the link between digital transformation and technology entrepreneurship innovation (Hanoteau & Rosa, 2019). Several studies show that the organizational will of start-ups is primarily influenced by the individual traits of technology entrepreneurs. However, we know very little about the mechanism of how technical founders affect innovation performance of enterprises from the founders’ perspective. In particular, there is still a gap in research on the strategic choices adopted by founders of technology start-ups to reap the rewards of innovation in the context of digitalization.

To fill the above-mentioned research gap, this paper adopts a text analysis approach to measure digital transformation of enterprises, and empirically analyzes the relationship between technical founders and innovation performance by adopting a moderated mediating model. Our empirical results indicate that digital transformation mediates the positive relationship between technical founders and innovation performance. Therefore, this study makes a major contribution to research on upper echelons theory and imprinting theory. It shows that technical founders tend to favor digital transformation and hence confirms the impact of founder identity on the digital transformation of companies. Meanwhile, the paper provides new insight into technology entrepreneurship. It expands the literature on the interdependence between technology path and innovation and explains the dynamics underlying the differential performance of firm innovation performance from the perspective of founder heterogeneity. In addition, this study presents a comparison of the differences in innovation performance between technical founders, academic founders and business founders, and examines the microscopic mechanism of the role of technology-based founders in corporate technological innovation.

The remainder of this paper is organized as follows. In Sect. 2, we describe the literature review and hypothesis development. Section 3 provides a description of the models, data collection and variable measurement. In Sect. 4, we present the empirical results and interpretation. The robustness of the baseline modelling results is checked in Sect. 5. Finally, we present the conclusions and discuss the contributions, implications and limitations of this paper in Sect. 6.

Literature review and hypotheses development

Technical founder and enterprise innovation performance

According to the upper echelons theory, CEOs’ values, experiences and personalities may influence their access to information in decision-making, which in turn affects the organizational actions and performance (Wei et al., 2018; Wang & Chen, 2020). Specifically, the behavior of a CEO is based on his or her personal, individualized interpretation of the strategic environment. The individualized cognitive construction of the external environment stems from the combined influence of the CEO’s experience, values, and personality traits. Technical founders, as the protagonists of corporate innovation investment and decision-making, are one of the most innovative groups (Sun et al., 2022), often serving as both chairman and CEO. Studies have shown that the greatest strength of technology entrepreneurs, whether working alone or in teams, is that they usually have a deep technical understanding of their business foundation and therefore have a clear understanding of how the new technologies they create can be applied to the business environment (Oakey, 2012). In the actual R&D process, the technical founder spontaneously participates in innovation activities and shows better innovation performance due to their dual roles of R&D and management (Jiang et al., 2021).

Furthermore, companies with technology-based founders tend to have coordinated IT teams and establish a communication environment with customer interaction, and once changes in user value attributes are identified, they may launch technology upgrades, thus showing better innovation performance (Qian et al., 2021). Therefore, when founders with technical backgrounds are involved in entrepreneurial activities, their knowledge capabilities and technical skills can be transformed into intangible assets that are constantly used and updated during the entrepreneurial process, opening up more innovative fields and providing high-quality problem solutions for corporate innovation (Marvel et al., 2020). Similarly, a study by Yeganegi et al. (2021) validates that potential technology entrepreneurs are at an advantage when information is more readily available, leading to higher technology entrepreneurship. Nielsen’s (2015) study also shows that technical entrepreneurs are found to perform better in both profitable and uncertain industries, while non-technical entrepreneurs perform better only in profitable industries. Therefore, we may conclude that technical founders are able to leverage their technical expertise, innovative thinking and entrepreneurial experience to effectively organize and manage teams, identify market opportunities and continuously drive technological innovation. These also explain why technical founders are the primary drivers of superior technological innovation performance. Based on the above arguments, we propose the following hypothesis.

Hypothesis 1

Firms led by technical founders show better innovation performance than those led by non-technical founders.

Technical founder and enterprise digital transformation

Imprinting theory suggests that individuals and organizations may experience many sensitive periods during their development, generating imprints that match environmental characteristics and persistently influence individual and organizational decision-making behavior (Pieper et al., 2015). The entrepreneurial imprint of technical founders also has an impact on the enterprise’s digital transformation. Numerous studies highlight that digital transformation should be a top management priority and a defining trait of corporate business strategy. Companies raising digital consciousness may get one step ahead on their transformation journey (Saarikko et al., 2020; Porfírio et al., 2021; Gurbaxani & Dunkle, 2019).

Since technology-based founders are relatively proactive and adaptive in the face of digital changes and dynamic environments, they are more likely to leverage their resources to spur business growth (Yeganegi et al., 2021). The study by Weber, Büttgen and Bartsch (2022) indicates that leadership is crucial to the success of digital transformation in companies. The skills and competencies of individual founders have a significant impact on the acceptance of digital transformation (Chatterjee et al., 2022). At the innovation capabilities level, technical founders are specialized in integrating and leveraging digital technologies for innovation. They transform organizations into flexible enterprises, thus reducing the possibility of rigidity caused by digital technologies (Tsou and Chen, 2022). At the same time, as organizations go digital, technical founders understand the enormous challenges that technological change poses to their employees. They strengthen employees’ belief in technological change and adopt a set of HR practices that facilitate innovative employee behavior, thus achieving a match between employee competencies and digital technologies (Nicolás-Agustín et al., 2021; Solberg et al., 2020; Kozanoglu and Abedin, 2021). Therefore, this people-oriented leadership behavior can buffer the potential negative effects of digital transformation-oriented leadership behavior, which is mandatory to stay competitive in the digital era (Weber et al., 2022).

In addition, at the level of corporate R&D and cooperation, technical founders have rich knowledge assets and are more familiar with R&D and production processes, so they can better grasp the technological development and cross-organizational collaboration. At the policy perception level, the government encourages technological innovation by giving SMEs guidance, support and incentives. Accordingly, technical founders can obtain relatively more financial support and R&D subsidies from various policies, alleviate corporate financing constraints, increase investment in digital technology exploration, and tend to make digital transformation decisions. Therefore, the educational background and experience of technical founders in the technology field imprint a deeper understanding of digital and information technology and enhance their forward-looking vision and sensitivity towards technology innovation and application. Moreover, the inventor identity of technical founders in technology exploration and exploitation also makes them to pay more attention to the application and practice of technology and display greater tolerance for failure and a tendency to integrate technological innovation with business innovation. All these are helpful for the identification and pursuit of opportunities for digital transformation (Bostan & Mian, 2019). Based on these arguments, we propose the following hypothesis.

Hypothesis 2

Firms led by technical founders are more likely to choose digital transformation than those led by non-technical founders.

Digital transformation and enterprise innovation performance

Digital transformation is a revolution and restructure that is driven and built on digital technology. Within the enterprise, digital transformation is defined as an organizational shift towards big data, business analytics, cloud computing, mobility, and social media platforms. The utilization of digital technologies has been widely studied, and scholars have explored the relationship between digital transformation and corporate innovation performance from different perspectives. Drawing on the resource dependence theory, Li, Rao, and Wan (2022) examined the impact of digital technology on corporate innovation using data from Chinese A-share listed manufacturing firms from 2011 to 2019 and found that corporate digital transformation has a significant positive effect on corporate innovation. Zhao et al. (2022), using a fixed effects model based on a panel of 584 listed manufacturing firms in China from 2016 to 2020, also found that digital transformation significantly improves firms’ innovation capabilities. Furthermore, Tsou et al.‘s (2022) survey of 227 Taiwanese financial industry regulators found that digital technology adoption facilitated corporate digital transformation and organizational innovation, which in turn influenced corporate innovation performance. Similarly, Peng and Tao’s (2022) study revealed that digital transformation not only significantly improved firm performance but also further motivated firms to innovate. In contrast, Lin and Yi (2022) explored the combination of antecedent conditions for digitally enabled enterprise innovation through the lens of configuration and holism. Their results show that digital infrastructure is the underlying support for digitally enabled enterprise innovation, while cross-boundary collaboration plays a crucial role in innovation.

Despite the above scholars affirming the positive effects of digital transformation on corporate innovation, there are also scholars who find that digital transformation cannot play a significant role as a single tool and must be synergized with technology, information, institutions, and policies to be sustainable. Nasiri et al. (2020) show that the digital transformation of companies alone cannot improve innovation performance and needs to be combined with intelligent technologies to achieve this goal. A similar view is that firms can apply digital technologies to enhance novel internal and external processes and integrate them into new business models, thereby advancing a new wave of innovation in the firm (Ricarda et al., 2021). Wu et al. (2022) also specifically highlight resource integration as a bridge between digital capabilities and open innovation. Digital transformation enables companies to access and transfer knowledge both within and outside the company, which facilitates resource matching and utilization. In summary, the above-mentioned scholars have clarified from the perspective of resource integration that the application of digital technology can directly help enterprises to obtain valuable information resources, reduce their innovation risks, further optimize and improve the allocation efficiency of enterprise resources, and enhance innovation performance. Based on the above arguments, we propose the following hypothesis.

Hypothesis 3

Digital transformation is positively related to firm innovation performance.

The above hypotheses together demonstrate the mediating roles of digital transformation. Specifically, organizations and entrepreneurs with a consciousness of digital transformation are more likely to build digital technology infrastructure. In addition to the technical expertise and human capital that technical founders have, the rapid changes in technology are forcing them to think outside the box. Meanwhile, the external and internal relationships of digital technology support in companies and the strategic alignment of digital technology with business enhance the ability of entrepreneurs to cope with the turbulence of the market environment and are important for companies to achieve innovative performance (Li et al., 2020). Therefore, when external technology changes, technology founders have a positive impact on innovation performance by virtue of digital transformation technology application and information integration and adjusting innovation strategies accordingly to accelerate resource sharing within the organization, shorten innovation cycles, and reduce corporate innovation costs (Chierici et al., 2021). Based on the above arguments, the following hypothesis is proposed.

Hypothesis 4

Digital transformation mediates the relationship between technical founders and firm innovation performance.

It has been shown that venture capital (VC) and/or private equity (PE) plays a role in financing, selection, information gathering, embedding and signaling in the growth of start-ups (Ferrary & Granovetter, 2009). Among them, business and technical embeddedness of venture capital is an effective means to enhance the competitive development and innovation performance of start-ups, which explains why VC/PE support empowers corporate digital transformation and technological innovation (Andersson et al., 2002). Most studies have confirmed that network embeddedness is positively related to firms’ innovation capabilities and innovation performance. Dyer and Singh (1998) believe that network embeddedness enriches firms’ knowledge sharing routes and enhances corporate effective governance and complementary resources through joint learning and knowledge interaction. Furthermore, VC/PE-backed companies typically exhibit certain characteristics in their board structure, such as a lower number of internal and instrumental directors, and a higher number of independent board members (Baker & Gompers, 2003). This board structure can promote innovation performance. Research has shown that independent board members usually possess broader experience and expertise which enable them to provide valuable strategic advice and market insights to companies, and thereby facilitate the identification and development of new opportunities (Li et al., 2020). Therefore, on the one hand, Sci-Tech enterprises leverage the relational capital and structural optimization of the network embedded by VC/PE and hence draw on novel external resources to promote digital transformation and technological innovation. On the other hand, VC/PE participation in corporate governance may affect the board structure, provide more support and guidance to the management and operations of a company, help the company optimise its innovation processes and resource allocation, and thus maximise the promotion of digital transformation in innovation. Based on the above arguments, the following hypothesis is proposed.

Hypothesis 5

VC/PE support positively moderates the mediating role of digital transformation in the relationship between technical founders and firm innovation.

Finally, Fig. 1 is presented to summarize the above-discussed conceptual framework and show the hypotheses to be tested in the empirical analysis.

Fig. 1
figure 1

Conceptual framework and hypotheses

Full size image

The modelling and data issues

The empirical models

The following empirical models are adopted to analyze the relationships among technical founders, digital transformation, and innovation performance:

$$Innovation={\alpha _0}+{\alpha _1}Technical+{\alpha _2}Controls+\varepsilon$$
(I)
$$Digitalization={\beta _0}+{\beta _1}Technical+{\beta _2}Controls+\delta$$
(II)
$$Innovation={\gamma _0}+{\gamma _1}Technical+{\gamma _2}Digitalization+{\gamma _3}Controls+\varphi$$
(III)

where Innovation is the dependent variable representing firm innovation performance, Technical is the independent variable capturing the role of technical founders, Digitalization is the mediating variable which quantifies digital transformation and Controls represent a set of control variables. ε and δ are the error terms.

To test whether the mediating effects of digital transformation are moderated by VC support, the following moderated mediation model is considered (Muller et al., 2005):

$$Innovation={\lambda _0}+{\lambda _1}Technical+{\lambda _2}Digitalization+ \\{\lambda _3}VC+{\lambda _4}Digitalization \times VC+{\lambda _5}Controls+\pi$$
(IV)

where VC is a moderating variable and π is the error term.

Data sources

In this study, the initial research sample covers enterprises registered and declared on the Science and Technology Innovation Board (STAR Market) from March 22, 2019 to May 1, 2021. The sample excludes state-owned enterprises controlled by either central or local governments and enterprises with missing relevant variables. The final sample has 197 observations from 124 companies listed on the STAR Market. The data of corporate innovation performance is obtained from the patent library of the State Intellectual Property Office of China and Qichacha. The founder characteristics and corporate financial data are mainly obtained from founder CVs and company prospectuses. The measurement of digital transformation is based on textual analysis of corporate annual financial reports; and the data of VC/PE support is obtained from the PEdata of Zero2IPO. The complete list of data sources is detailed in Table 1.

Table 1 Variable definition and measurement
Full size table

Variable measurement

Innovation is the main dependent variable which measures corporate innovation performance and is used to evaluate the efficiency and effectiveness of firm innovation activities. It belongs to a specific form of organizational performance (Wang & Hu, 2020). In this paper, we use the total number of invention and utility patents granted in the current year to measure the innovation performance. Like Hu, Pan and Huang (2020), we also use the invention patents and all patents (the invention, utility model and design patents) in the current year as proxy variables for the quality and quantity of innovation of enterprises.

Technical represents technical founders who are business founders with technical professional background and creative characteristics. They often have a deep technical understanding of their business basis and of how the new technology they create is applied to the business environment. This study classifies founders into academic founders, business founders, and technical founders. Specifically, we first structured and stored the textual information of the collected CVs of 2775 founders. We then invoke the LDA topic model to train the corpus and use the Coherence Model to evaluate the model for consistency under a number of topics (for details, refer to Mimno et al., 2011). After several training sessions, the number of topics with the highest consistency score is determined to be 5, and then the initial topic type and the corresponding top 20 representative words with the highest frequency of occurrence are selected to obtain the initial seed word set. Then, we combine the linkage of keyword lexical meanings among topics and classify them into three identity attributes, namely, technical, business, and academic. Next, we supplemented the candidate word sets using core vocabulary semantics and literature related to founder identity. Finally, the construction of the final lexicon of the founder identity was completed. Based on the built-in split-word dictionary in LDA (see Appendix 1), we adopt the jieba library to split-word process all founder resumes and count the number of keyword disclosures (see Appendix A) from academic, business, and technical categories, respectively, for each founder. The highest number determines the founder’s identity. For founders with the same word frequencies (no clear identification of the highest number), we manually check their resumes and integrate relevant information from the official company websites, Wind’s person database, and online news to ultimately determine their founder identity.

Digitalization is a proxy for digital transformation. According to Wu et al. (2021), digital transformation is measured in two dimensions, namely digital technologies and digital application (see Appendix B). Digital technologies include artificial intelligence, blockchain, cloud computing, and big data technologies and the application of digital technologies is based on the commercial operation of these technologies. Specifically, we employed the digital transformation lexicon developed by Wu et al. (2021) as the initial corpus and utilized the Chinese pre-trained BERT model to construct the word vector for each keyword (for details about BERT, refer to Devlin et al. 2018). Subsequently, we incorporated these digital transformation keywords into the jieba segmentation library and applied accuracy mode and lcut algorithm to tokenize 197 annual reports of companies. Stopword removal was conducted to construct the vocabulary of the annual reports. Next, we extracted words from the annual report vocabulary that had 80% similarity to the digital transformation word vectors, and manually screened and combined them with expert opinions, resulting in 24 extra keywords representing digital transformation in companies (see Appendix B), which were added to the built-in dictionary with 76 keywords (thus a total of 100 keywords are used). Finally, we counted the frequency of disclosure for these keywords from the company business summary, discussion and analysis of operating conditions, and corporate governance to determine the level of digital transformation in companies. This gives us the definition of Digital Transformation I (Digitalization_1). At the same time, we also constructed Digital Transformation II (Digitalization_2). In addition, we extracted the before and after texts related to digital transformation from manual reading and evaluation and eliminated the number of word frequencies that were not associated with digital transformation strategies, operations and business models. We use this new data to construct Digital Transformation II (Digitalization_2).

VC reflects support from venture capital or private equity which is the primary source of equity financing for technology start-ups. Companies with technical founders are uniquely positioned for equity financing, as they not only understand the market and customer needs, but also track cutting-edge technology to keep up with the times. Therefore, VC/PE support alleviates the problem of financing availability on the one hand and provides heterogeneous information resources on the other hand. VC has a value of 1 if a firm gets VC/PE support and 0 otherwise.

Like Islam et al. (2022) and Jiang, Wang, and Zeng (2021), variables such as the founder characteristics and basic financial indicators of the firm which are likely to affect the firm’s innovation performance, are used as control variables in this study. At the individual level, we control the age, gender, education, overseas background, and duality of the founder. At the firm level, we control firm scale, ROA, LEA and ONCF. The definitions of the variables are shown in Table 1.

Descriptive statistics and correlation tests

Table 2 reports the descriptive statistics of the variables. The percentage of technical founders is 41.2%, and the percentage of academic and business founders is 22.3% and 36.5%, respectively. The average keyword frequency of digital transformation is 70.44 and 50.09, which reflects the active digital transformation for the listed enterprises in China’s STAR Market. However, the standard deviations are 84.65 and 67.79 respectively, indicating that there are large differences in digital transformation among enterprises. In addition, the sample founders are generally middle-aged (mean = 52.43) and predominantly male (mean = 0.959) with a master’s degree (mean = 3.888). The founders with overseas background accounted for 28.4%, and most of them are also chairmen and CEOs (mean = 0.619). In addition, this paper also tested the variance inflation factor, whose maximum value is 1.99, which is much lower than the critical value of 10.00, indicating that no significant covariance exists among the independent variables.

Table 2 Descriptive statistics and VIF test
Full size table

The results of the correlation analysis are presented in Table 3. It is shown that technical is significantly and positively correlated with firm innovation performance at a 0.1% level. Technical is significantly and positively correlated with digital transformation at a 0.1% level. Digital transformation is also positively correlated with firm innovation performance at 0.1% level. The above results tentatively verify hypotheses H1, H2 and H3.

Table 3 Summary of the correlation test
Full size table

Empirical results

Benchmark regression analysis

Table 4 reports the results of the benchmark regression of the relationship between the technical founder and firm innovation performance. Model (1) is to verify the relationship between technical founders and firm innovation performance without control variables. Model (2) reflects that technical founders positively affect firm technological innovation at the 5% level (regression coefficient of 0.484), which supports hypothesis H1. Meanwhile, compared with Model (1), the R2 in Model (2) increases, which also verifies the rationality of the model construction. The above findings validate that firms led by technical founders show better innovation performance.

Table 4 Results of mediation analysis
Full size table

Analysis of mediating effect of digital transformation

Model (3) in Table 4 is to test the relationship between the technical founder and digital transformation. The results show that the effect of technical founders on digital transformation is positive and significant at the 5% level, which reveals that firms with technical founders tend to favor digital transformation. Furthermore, we compare the differentiated performance of different types of founders on digital transformation. Model (4) further reveals the relationship between founders and corporate digital transformation. The results show that technical founders’ companies have the most prominent digital transformation, followed in turn by companies with business and academic founders. The above results confirm hypothesis H2.

To examine the mediating effect of digital transformation, we add both technical founders and digital transformation to model (5) to test their effects on firm innovation. The results show that the regression coefficient of digital transformation is 0.188 and significant at the 5% level. Model (5) also illustrates the regression coefficient of technical founders on firm innovation performance is 0.396 at the 5% significance level, which is smaller than the benchmark regression of 0.484. Thus, we can conclude that digital transformation partially mediates the relationship between technical founders and firm innovation performance and the mediating effect of digital transformation is 15.38%. Therefore, hypotheses H3 and H4 are verified.

Finally, following the Fang et al. (2021) approach, a moderated mediating model is used to further test whether the mediating effects of digital transformation are moderated by VC/PE support. Model (1) in Table 5 confirms a significant positive effect of technical founders on firm innovation performance. Models (2) and (3) reveal that digital transformation plays a mediating role in the above relationships. Next, we test whether VC/PE moderates the relationship between technical founders and innovation performance or the relationship between technical founders and digital transformation. The regression results of model (4) and model (5) show that the interaction term between VC/PE and technical founders are both insignificant.

Table 5 Results of moderated mediation analysis
Full size table

Meanwhile, The interaction term of VC/PE support and digital transformation in Model (6) and Model (7) has a positive contribution to firm innovation performance (coef.=0.433, p < 0.1; coef.=0.389, p < 0.1), and VC/PE support has a moderating effect on the relationship between digital transformation and innovation performance. Specifically, VC/PE support moderates the second half of Technical Founders→Digital Transformation→ Innovation Performance. Therefore, VC/PE support positively moderates the mediating role of digital transformation in the relationship between technical founders and innovation performance. The mediating effect of digital transformation has increased from 15.38 to 29.27%. So, hypothesis H5 is verified.

Robustness checks

To further ensure the validity of the findings, this study performs robustness tests against the above regression results by using alternative dependent variables and mediating variables.

Alternative measures of innovation performance

Considering the potential bias caused by the type of patents on the innovation performance measurement, this paper also uses the number of invention patents of the firm to measure the quality of technological innovation. In addition, the total number of invention, utility model and design patents is employed as a proxy for the quantity of innovation of a firm. The results are shown in models (1) and (3) in Table 6. It can be concluded that the relationship between technical founders and firm innovation remains significantly positive when innovation quality and quantity are considered separately. In addition, model (2) also shows that digital transformation positively and significantly affects the quality of firm innovation. Similarly, model (4) explains a positive and significant relationship between digital transformation and the quantity of firm innovation. Consequently, we can assume that digital transformation plays a partially mediating role in the relationship between technical founders and the quality and quantity of firm innovation. Thus, technical founders not only contribute to the quantity of innovation in firms, but also promote high-quality innovation activities. In summary, hypotheses H1, H3, and H4 are still validated and supported after replacing the dependent variable.

Table 6 Regression analyses of the robustness test
Full size table

Alternative mediating variable

As mentioned in the discussion of variable measurement, we obtained Digitalization_2 by simultaneously collecting the frequencies of words excluding those unrelated to digital transformation strategies, operations and business models. The regression results are illustrated in model (5) in Table 6, and still show a significant positive relationship between technical founders and firm digital transformation (coef.=0.455, p < 0.05). Model (6) further reveals that digital transformation plays a partially mediating role between technical founders and innovation performance. In summary, hypotheses H2, H3, and H4 are still validated and supported after replacing the mediating variable. Therefore, we may conclude that all the regression results in the above exercises to be stable and reliable.

Endogeneity test

To reduce possible regression bias due to sample self-selection, we further performed an endogeneity test using propensity score matching (PSM). According to Colombo et al. (2022), firm size, founder age, education, and equity concentration are selected as confounding variables. In this paper, the samples are matched 1:4 under the nearest neighbour with replacement. Clearly, after matching, the differences between the control and treatment groups are no longer statistically significant, suggesting that the balancing test is validated. Additionally, the pseudo-R2 is fairly low and the joint significance test is not rejected before but after matching, which verifies the common support assumption of PSM. Finally, we drop the samples that are off support and the comparative regression analyses of the PSM results in Table 7 show that technical founders still show a robust and significant positive relationship with the innovation performance of Sci-Tech enterprises.

Table 7 The PSM results
Full size table

Conclusion and discussion

Conclusion

In this paper, we use 197 annual reports of 124 companies listed on China’s STAR Market for 2019–2020 to measure the digital transformation of firms using textual analysis and further investigate its mediating effect on the relationship between technical founders and innovation performance. There are several interesting findings.

First, it is found that technical founder-led firms have shown better innovation performance, not only in terms of innovation quantity, but also in terms of innovation quality. This may be due to the fact that the experience of technical founders promotes the organizational identity of start-ups in innovation activities, enhances the firm’s tolerance for failure, and increases its participation in exploratory search strategies that exploit new technology trajectories (Bostan & Mian, 2019). The study by Jiang, Wang and Zeng (2021) confirms the positive impact of the founders’ dual role of R&D and management on firm innovation performance. The more the founders of a firm are involved in R&D activities and the higher their centrality in the R&D network, the better the firm’s technological innovation performance. This is because high network centrality gives founders a stronger ability to innovate and identify opportunities and attracts their attention to innovation.

Second, founders with different identities have different impacts on the digital transformation of enterprises, specifically in this order technical founder > business founder > academic founder. In SMEs, the influence of certain key individuals may have a greater impact on the direction chosen. This view is a recognition of the centrality of the entrepreneurs and is also consistent with the higher echelon theory on how the managerial characteristics and beliefs of individuals influence the strategic choices of the firm (Barrett et al., 2021). Therefore, the origins of the founders are systematically linked to the technological direction of their newly formed organizations. Technical founder affirms entrepreneurs to engage in behaviors that are consistent with their identity, which in turn further reinforces their technical identity (Zuzul & Tripsas, 2020). That is why, founders contribute to the inertia and flexibility of the firm through the mechanism of identity affirmation, exhibiting different digital transformations.

Finally, digital transformation plays a partially mediating role in the relationship between technical founders and corporate innovation performance. VC/PE support positively moderates the mediating role of digital transformation in the relationship between technical founders and innovation performance. Digital transformation has undoubtedly become a key driver of innovation and has also fundamentally changed profit models and business models. Considering these shifts and the high ambiguities and uncertainties, most academic founders and business founders initially found the adoption of digital ideas challenging, if not completely unattractive. The organizational role identity and organizational domain identity of technical founders interactively explain how companies should respond to digital transformation (Kammerlander et al., 2018). Thus, the image of entrepreneurs and inventors as well as the adoption of new digital processes by leaders contribute to the dynamic capabilities of the company (Ferreira et al., 2019). At the same time, digital transformation requires specific managerial actions of the founders that can help companies to foster and develop new technological innovations (Urbinati et al., 2020).

Theoretical contributions

First, we provide a novel perspective to study the impact of founder identity on firm technological innovation. Importantly, our study reveals the mechanism underlying the influence of technical founders on firm innovation, where firms led by technical founders show optimal innovation performance through digital transformation. This finding extends the literature on technology entrepreneurship and firm innovation, which differs from the previous literature on the impact of internal executive characteristics (Fayolle et al., 2021; Bolzani et al., 2021; Blankesteijn et al., 2021; Ganotakis et al., 2012; Zane, 2022; Reese, Rieger and Engelen, 2021) and external institutional environment (Colino et al., 2014; Seguí-Mas et al., 2018; Hülsbeck & Pickavé, 2014) on firm innovation performance.

Also, our findings suggest that VC/PE support positively moderates the positive effect of digital transformation on firm innovation, thereby enhancing the mediating effect of digital transformation between technical founders and firm innovation performance. Finally, our study enriches the imprinting theory literature by confirming the impact of founder imprinting on identity formation and is also consistent with the higher echelon theory on how individual founder’s traits and beliefs influence the firm’s strategic decisions (Wei et al., 2018; Wang & Chen, 2020; Pieper et al., 2015; Barrett, Dooley and Bogue, 2021). More broadly, our study highlights the specificity of founder identity and suggests that the recognition of the centrality of entrepreneurs in strategic choices may be crucial to explain innovation performance heterogeneity among Sci-Tech enterprises.

Practical implications

The results of this study have implications for the aim of supporting technology entrepreneurship for technological innovation. The empirical evidence suggests that the strategic choices of Sci-Tech start-ups, such as digital transformation, directly affect firm innovation performance under founder identity heterogeneity. Therefore, national policies should first focus on developing technology entrepreneurship, supporting investment in engineering entrepreneurship education and technology incubation in technology transfer offices, which is especially significant to facilitate the construction of innovative national strategies.

Secondly, strengthening the knowledge diversity of startup teams and bringing in partners with technical backgrounds to participate in technology development and application of nascent companies is an important part of their adaptation to sustainable growth in the digital context. Companies cannot achieve digital transformation and continuous innovation without a co-founder team with complementary capabilities. For technical founders, they have to balance short-term and long-term concerns, keep an eye on changing market dynamics, and make strategic transitions at critical times. In addition to continuously tracking technology frontiers and updating technology reserves, technical founders should also enhance knowledge flow and technology sharing within the organization. For non-technical founders, attracting external high-tech talent to join, not blindly pursuing digital transformation and technology-oriented changes, and leveraging their intellectual strengths and business talents to synergize with organizational changes and business models are the prerequisites for achieving robust growth.

Limitations and future research

This paper measured founder identity only from a single dimension, ignoring the joint effect of multiple founder identities on firms’ strategic choices and innovation performance. In the future, we can further expand the classification criteria of founder identity to reveal the innovation evolution process and the corresponding dominant image of Sci-Tech enterprises in different stages from the perspective of multiple identities. At the same time, it is possible to further track the innovation behavior of firms listed on the STAR Market, especially whether the performance of corporate innovation heterogeneity follows the pre-IPO sequencing after firms with technical founders and academic founders are freed from financing constraints.