How Technological, Organizational, and Environmental Factors Drive Enterprise Digital Innovation: Analysis Based on the Dynamic FsQCA Approach

Abstract

Improving the level of digital industry innovation is of great significance to enhance the competitiveness of China’s digital industry and improve the sustainable development advantages of the digital economy. Based on the technology–organization–environment framework (TOE framework), this paper establishes a multi-stage configuration analysis model of factors affecting enterprises’ digital innovation, selects six antecedent variables from technology, organization, and environment, including R&D investment, high-level talents, organizational size, top management team heterogeneity, industrial development speed, and regional digitalization level, and explores the driving path for improving enterprise digital innovation intention and digital innovation performance under the influence of multi-factor combinations. The results are presented that single factors or single dimensions struggle to stimulate high digital innovation, and the combination configuration of multiple factors has a significant impact. The configuration paths that drive high digital innovation intention include the technology–environment type (TE type) driven by technological and environmental factors, organization–environment type (OE type) driven by organizational and environmental factors and technology–organization–environment type (TOE type) driven by technological, organizational, and environmental factors; the configuration paths that drive high digital innovation performance include the technology–organization type (TO type) driven by technological and organizational factors, organization–environment type (OE type) driven by organizational and environmental factors. With the evolution of time, there are configuration solutions with relatively high stability, such as the TE type and OE type that drive high digital innovation intention, and the TO type that drives high digital innovation performance. The results provide a reference for revealing the key influencing factors and driving paths of enterprise digital innovation, and improving the vitality and output level of enterprise digital innovation.

1. Introduction

A new round of scientific and technological revolution centered on digital technologies such as big data and cloud computing is accelerating its widespread application, triggering the reshaping of the global innovation economic system [1]. As a green, innovative, and sustainable high-quality economic paradigm [2], the digital economy has become a key theme for China’s economic growth. According to the “Digital China Development Report (2022)” released by the State Internet Information Office of China, the scale of China’s digital economy reached CNY 50.2 trillion in 2022, ranked second in the world in terms of total volume, and its proportion in GDP has increased to 41.5%, China’s digital industrialization and industrial digital development have achieved remarkable achievements. At the same time, according to the “Global Digital Economy Development Index Report” released by the Institute of Finance and Economics of the Chinese Academy of Social Sciences, although China’s digital economy has significant advantages in the digital market and digital infrastructure, ranking 2nd and 3rd in the world respectively, there is still a certain gap in digital technology and digital governance, ranking 15th and 41st in the world respectively, lacking development advantages. Compared with other advanced digital economy countries, the core competitiveness and the independent innovation ability of key core technologies of China’s digital industry still have a significant gap. Therefore, to accelerate the construction of digital China, it is more important to promote continuous innovation in the digital technology industry. As important micro-subjects of economic development, enterprises are explorers, organizers, and leaders of digital innovation, and play an important role in breaking through technological bottlenecks and tackling scientific and technological problems. In the wave of digital development, those companies that have seized the opportunities of digital technology have achieved curve overtaking, and companies that struggle to adapt to the development of digital transformation have been eliminated by the times. Therefore, how to promote enterprises to actively participate in digital innovation and improve digital innovation performance is of great significance for improving the international competitiveness and sustainable development capabilities of China’s digital industry and promoting the high-quality development of China’s digital economy.

The concept of digital innovation was first introduced by Yoo in 2010 [3], who thought digital innovation is an innovative process of combining digital and physical components to produce new products, services, and business models. Subsequently, it aroused widespread concern in academic circles. In terms of the connotation and measurement of digital innovation, some studies based on output theory proposed that digital innovation refers to the innovative results produced by applying digital technology in the innovation process, including the use of digital technology to bring new products, improving production processes, changing organizational models, creating and changing business models, etc. [4]. Based on process theory, some scholars proposed that digital innovation is the combination of digital technologies such as information, computing, communication, and connection used in the innovation process [1]. There are also studies based on the theory of synthesis, pointing out that digital innovation should combine the application of digital technology and the results, including both the efficiency in the innovation process and the generation of innovation results [5].

In terms of research on the driving factors of digital innovation, there are studies on resource, organizational, and environmental factors. Some studies have showed that digital innovation requires advanced human capital, human capital can be integrated and iterated with technological and organizational resources to accelerate digital innovation [6]; knowledge management can also promote digital process innovation [7,8]. Regarding organizational factors, some scholars have found that digital innovation is affected by organizational strategy and organizational change [9], digital innovation is driven by many organizational factors such as opportunity search, business intelligence, organizational change, and organizational adaptation [10]. For example, Beatrice et al. [11] used the cases of four multinational companies operating in different industries to explore how existing companies can adjust their business models to cope with digital innovation. Other scholars have further proved that the capabilities of young entrepreneurs during economic turmoil have different impacts on digital innovation in micro-, small-, and medium-sized enterprises [12]. Regarding environmental factors, some studies have found that government financial support and training can enhance enterprises’ understanding of digital innovation [13], while human capital such as economic level and population size have little effect on the development of digital industries, and regions with poor economic foundations can use information infrastructure construction to promote industrial development [14].

In terms of the role of digital innovation, due to the reprogrammability of digital technologies, which can be continuously embedded into the innovation process to increase the fault tolerance and unpredictability of products, firms can make continuous adjustments to the innovation strategies in a dynamic innovation environment [15,16]. Based on this, some scholars pointed out that digital innovation can enhance the ability of enterprises to perceive the threat of potential competitors and the direction of industrial change, strengthen the continuous monitoring and reconnaissance of changes in the market environment [17], and create prerequisites for enterprises to grasp the window of opportunity of digitization and realize pioneering development [18]. Some scholars proposed that enterprises with strong digital innovation capabilities generally have strong development resilience [19]. Enterprises can use intelligent data to build resource protection walls to help enterprises adapt to changes in the face of external uncertainties and corporate crises, and even build new competitive advantages, and use this as a lever to achieve disruptive development [20,21]. In addition, some studies have found that companies that actively participate in digital innovation can more actively undertake social responsibilities, improve social reputation [22], and build organizational culture, risk quality management, and other organizational capabilities necessary for sustainable development [23].

From the above research, it can be found that digital innovation is a core and effective way to achieve digital empowerment and promote enterprises to enhance their competitiveness and achieve sustainable development [24]. Scholars have conducted various studies on digital innovation, but there are still the following deficiencies: first, the current measurement of enterprise digital innovation is mostly based on result theory, considering that digital innovation is a new product or new income produced by the enterprise, ignoring the consideration of the initiative and enthusiasm of enterprises participating in digital innovation from the perspective of process theory; second, although the existing literature has initially explored the impact of resources, organizational, and environmental factors on digital innovation in enterprises, few studies have integrated the internal and external factors into a holistic and systematic framework; third, most of the existing research focused on the influence and effect of a single factor or single dimension on enterprise digital innovation and used regression analysis to confirm the net effect of a single variable, neglecting the comprehensive consideration of the combination and superposition effect of multiple factors, and failing to reveal the group effect of multi-factor interaction on enterprise digital innovation.

Therefore, this paper starts from a holistic and systematic theoretical framework—technology–organization–environment(TOE) framework, constructs a configuration analysis model of the impact of technological, organizational, and environmental factors on enterprise digital innovation intention and digital innovation performance, takes the listed companies in China’s Shanghai and Shenzhen A-share digital economy core industries as the research object, conducts multi-stage dynamic fuzzy set qualitative comparative analysis (fsQCA), and analyzes the configuration effects of the antecedent variables on enterprise digital innovation intention and digital innovation performance from 2017 to 2020, discovering what combination of factors drives companies to be more actively participate in digital innovation activities and achieve high-quality innovation performance. This paper attempts to answer the following questions: (1) What are the key factors that enhance enterprise digital innovation intention and digital innovation performance? (2) How do different antecedent variables interact to form their configuration effects on enterprise digital innovation? (3) What are the development paths for high digital innovation intention and high digital innovation performance of enterprises, and how do they evolve? The answers to these questions have important reference significance for analyzing the internal operation mechanism of enterprise digital innovation, enhancing the competitiveness of enterprise digital innovation, and building an internationally competitive digital industry cluster.

This paper makes the following theoretical contributions. First, based on the perspective of configuration, the TOE framework is introduced into the research of enterprise digital innovation, and the interaction among variables from the technology, organization, and environment is analyzed, which provides a holistic and systematic research framework. Second, this paper integrates output and process theories and divides enterprise digital innovation into digital innovation intention and digital innovation performance, which is more conducive to enterprises accurately positioning their digital innovation strategies and optimizing the allocation of innovation resources. Third, this paper introduces the time dimension into the study of configuration effects, adopts the multi-stage dynamic fsQCA method, and divides multiple time windows to study and compare the realization paths of digital innovation in different time windows, which enriches the research tools in the field of innovation management, and provides a reference for subsequent configuration research.

2. Literature Review and Theoretical Framework

The TOE framework was initially applied to the research on factors affecting enterprise technology adoption, and the factors were summarized into three dimensions: technology, organization, and environment [25]. Among them, technological factors refer to characteristics related to technological innovation, such as technological advantages, technological compatibility, technological cost, and technological complexity, etc. [25,26]; organizational factors include organizational scale, organizational type, and organizational support [27]; and environmental factors refer to characteristics of the organization’s external environment, including political environment, economic environment, social environment, etc. [28]. As the framework is applied to different research scenarios, the framework connotations are continuously enriched. For example, Wang analyzed the impact of enterprise technology preparation, organizational preparation, and environmental preparation on the performance of green innovation based on the TOE framework [26]; Lei combined innovation environment to explore the impact of different types of digital technology and organizational characteristics on enterprise service diversification [27]; Lexutt used the TOE framework to investigate how technology transfer and geographical location affect regional economic growth [29]. Under the penetration of digital technology, enterprise innovation not only requires organizational resources as the basis for digital technology application but also requires strong support from the external environment, such as regional and industry influences, so as to release innovation vitality in the synergy of internal and external factors [30]. Therefore, this paper takes the TOE framework as the logical basis of this study, incorporates the antecedents from technology, organization, and environment into the same systematic and holistic research framework, constructs a theoretical model of the impact mechanism of enterprise digital innovation, and explores the impact of antecedents at different levels and their combinations on the digital innovation of enterprises.

2.1. Technological Dimension

The TOE framework first focuses on the impact of the technological characteristics of the enterprise itself on the development and application of digital technology [31]. The dynamic and self-growth features of digital technology have increased the uncertainty of digital innovation. New technologies will not only bring new products and solutions but also may bring all-new changes or disruptions to industry, placing higher demands on the speed and quality of digital innovation [32]. As one form of internal resource allocation in enterprises, R&D investment represents the level of intention to digitally innovate and is an important condition for enterprises to start digital innovation [33]. The Chinese Internet company Alibaba’s total R&D investment in 2022 has exceeded CNY 120 billion. In the past three years, 60% of its patents have been concentrated in digital innovation fields such as artificial intelligence and cloud computing. Its Alibaba Cloud, which relies on its self-developed Feitian operating system, has the world’s leading market share and has achieved market share growth for six consecutive years. It can be seen that R&D investment can accelerate digital innovation output by helping enterprises to build digital resource systems.

In addition, the rapid development of emerging technologies such as artificial intelligence, cloud computing, big data, and blockchain, which are characterized by digitization and intelligence, requires more high-level technological talents. As the core of the enterprise’s digital innovation capability, high level talents have strong digital technology learning ability can continuously focus on the connection between corporate resources, capabilities, and product markets, actively respond to dynamic changes in the company and the market, and help accelerate technology development and market expansion [34]. At the same time, in the process of digital innovation, high-level talents can better understand the user needs, actively conduct critical evaluation and improvement of product design and functions, and finally promote product landing. Therefore, this paper investigates the impact of the technological dimension on enterprise digital innovation from two factors: R&D investment and high-level talents.

2.2. Organizational Dimension

The TOE framework also pays attention to the influence of organizational structural characteristics and organizational strategy on the development of digital innovation [35]. Studies have shown that the size of the organization affects the digital innovation of enterprises [36]. Compared with medium-sized businesses (SMEs), large enterprises have more innovative resource advantages, stronger risk resistance, and easier to achieve cross-border operations. Dell’s IdeaStorm, Haier’s crowdsourcing platform, and Xiaomi Community all take advantage of the strong organizational scale to gather heterogeneous innovation groups, thus achieving benefit sharing and risk sharing of digital innovation. However, some scholars have also found that SMEs are more flexible in their organization and could seize technological opportunities promptly to achieve breakthroughs in the rapid iteration of digital technologies [37].

In addition, in the process of digital innovation, as the lead of the organization, the composition of the top management team also has a significant impact on the deployment of the company’s overall digital strategy and the development of innovation actions [38]. First of all, the greater the difference in the professional background of the executive team, the more channels and ways the organization can obtain information, the more quickly the company can perceive the changes in the external environment, and it is easier to identify innovation opportunities and potential risks to adjust the digital innovation strategy on time [39]; second, a more heterogeneous executive team has different knowledge backgrounds, decision-making styles and professional perspectives, which make the whole team more creative and inclusive, meaning it is easier to form an active digital innovation atmosphere and open up innovation boundaries through collective brainstorming [40]. Therefore, this paper studies the impact of the organizational dimension on enterprise digital innovation from two factors: organizational size and top management team heterogeneity.

2.3. Environmental Dimension

The TOE framework also focuses on the impact of the dynamically evolving external environment on digital innovation [41]. As digital technology triggers a “digital butterfly” in various economic sectors, more and more industries are recognizing the importance of digital innovation. Generally speaking, industries with faster development have a more advanced understanding of the R&D and application of digital technology. Based on the isomorphic effect, the improvement of the digitalization level in the industry promotes the formation of new competition rules, the survival of the fittest, and the elimination of the inferior [42]. To obtain quality suppliers and customers and increase the innovation efficiency of enterprises, enterprises need to have digital resources and capabilities that match the development of enterprises in the same industry, thus stimulating the motivation and innovative energy of digital innovation.

In addition, the advancement of digital technology systems and digital industries is inextricably linked to local digital infrastructure and policy support [14]. In recent years, all parts of China have continued to make efforts in the field of the digital economy, extensively building digital infrastructure and promoting the local digital economy development. The government attaches great importance to regional digital innovation activities and carries out the regional layout of digital innovation through the master plan of urban services and urban governance [43]. Usually, the better the construction of regional digital infrastructure and the more improved the local digital policy governance system, the better it is to promote the digital innovation activities of enterprises. Therefore, this paper investigates the impact of the environmental dimension on enterprise digital innovation from two factors: industrial development speed and regional digitization level.

2.4. Model Construction

The above analysis reveals that each of the six antecedent conditions from technology, organization, and environment has an important impact on the digital innovation activities of enterprises. However, from the perspective of configuration, the influence of various conditions on enterprise digital innovation are not independent of each other, but synergistically exert their effects through linkage and matching. In this process, the application of digital technology becomes the basic medium and transformer of digital innovation [44,45]. However, digital technology must be closely integrated with traditional innovation resources such as capital, equipment, and talents to fully exert its innovation-enabling role [46]. R&D investment and high-level talents constitute the resource base of digital innovation, which to a large extent determines the feasibility and adaptability of an organization’s application of digital technology and can further affect the organization’s specific technological requirements. Organizational size and top management team heterogeneity affect the allocation of corporate organizational resources and the interaction effect, they also affect the realization of innovation results under certain technological characteristics. At the same time, under the interaction of internal and external factors, enterprises need to continuously absorb digital innovation resources from the environment, identify, organize, and apply effective information, dynamically adjust innovation behavior, facilitate resource integration, and then promote digital innovation activities.

In sum, this research adopts the TOE framework as the theoretical basis to establish an analytical framework for the influencing factors of enterprise digital innovation and chooses six influencing factors from the three levels of technology, organization, and environment, including R&D investment, high-level talents, organizational size, top management team heterogeneity, industrial development speed, and regional digitalization level, discussing how multiple factors of technology, organization, and environment affect the digital innovation activities of enterprises through mutual linkage and concurrent synergistic effects. The theoretical framework is shown in Figure 1.

3. Research Methods and Data Collection

3.1. Research Methods

Qualitative comparative analysis (QCA) considers a case as a whole composed of antecedent conditions [47]. This method explores multiple concurrent relationships among multiple conditions and mines the equivalent effects of different combinations of antecedent variables on outcome variables through multiple case studies [48]. QCA can be classified into crisp set QCA (csQCA), multi-value QCA (mvQCA), and fuzzy-set QCA (fsQCA). According to the data type, compared with csQCA and mvQCA, fsQCA is not only able to deal with dichotomous and multi-valued variables but also suitable to deal with continuous-type data. Considering that the variables selected in this study are all continuous, we chose fsQCA as the research method.

Although the traditional static fsQCA method can analyze the concurrent relationship among multiple factors, it does not discuss the dynamic evolution of configuration results. Different from traditional innovation, digital technology is developing rapidly. Intelligent technologies such as artificial intelligence and cloud computing are emerging one after another. The collaboration between innovation subjects is accelerating, and digital innovation is also iterating rapidly. Many latecomer enterprises have seized the opportunity window of technology and achieved “curve overtaking” from catching up to surpassing. Therefore, it is necessary to incorporate the time factor into the QCA to explore the dynamic evolution trend of the configuration. Recent studies have introduced the time factor into the QCA: Hino used the mean value of multi-stage sample data to measure variables [49], and Garcia summarized and calibrated multi-stage sample data [50]; some scholars also used multi-stage data to analyze separately, such as Witt discussing the impact of institutional background and organizational conditions in different countries on organizational governance from 2014 to 2017 [51], and Sun using multi-stage panel data to research the development and changes in the configuration path of the high-quality development of Chinese state-owned enterprises [52]. In contrast, the multi-stage qualitative comparative analysis method can further uncover the evolution trend of conditions or configurations over time, and reveal the changing track of configuration in multiple stages. If a configuration appears stably over multiple periods, it indicates that the paths played a dominant role during that time. Therefore, based on Witt’s research, this paper used multi-stage dynamic fsQCA to analyze the collective relationship and joint effect among technological, organizational, and environmental factors, and found the paths to improve enterprise digital innovation intention and digital innovation performance, as well as the evolution of these paths over time.

To better reveal the digital innovation path of Chinese enterprises, we took 2017–2020 as the research time window and divided it into four-time windows: 2017, 2018, 2019, and 2020. This period was chosen because: ➀ In 2017, the digital economy first appeared in China’s “Government Work Report”, marking that the development of the digital economy has risen to the height of China’s national strategy, and the digital industry has begun to flourish under the guidance of the government, so, we chose 2017 as the starting point for this research; ➁ The period from 2017 to 2020 belongs to China’s 13th Five-Year Plan period (2016–2020). During the “13th Five-Year Plan” period, the development of China’s digital economy has achieved leapfrog development, and much experience in enterprise innovation and development has been formed, which provides a sufficient research basis for this paper. The research conclusions of this paper will also provide a reference for the development direction of digital industry enterprises in the 14th Five-Year Plan period.

3.2. Sample and Data

As the key support for promoting the development of China’s digital economy, the core industries of the digital economy continue to play a leading role in driving breakthroughs in “stranglehold” technologies, improving the core competitiveness of China’s digital innovation and the potential of the value chain, have become the most active industries in digital innovation. “The Digital Economy and Core Industries Statistical Classification” released by the National Bureau of Statistics of China in 2021 divided the core industries of the digital economy into computer communication and other electronic equipment manufacturing, telecommunications and satellite transmission services, Internet and related services, and software and information services. Therefore, we used listed companies in these four industries as the research objects excluding Special Treatment (ST), *Special Treatment (*ST), and data-missing companies, and determined 388 listed companies in the core industries of the digital economy as the case samples.

The data sources were collected from multiple sources: digital innovation intention data came from the annual reports of listed companies, digital innovation performance data came from the Patsnap, industry development speed data came from the “White Paper on the Development of China’s Digital Economy”, regional digitalization level data came from the “White Paper on China’s Urban Digital Economy Index”, and the rest of data came from the China Stock Market & Accounting Research Database (CSMAR).

3.3. Variable Measurement

3.3.1. Outcome Variable

Digital innovation refers to the innovation activities around digital technology [3]. Regarding how to measure digital innovation, some studies take the innovation process as the basis and use the keyword frequency of digital innovation activities in the annual report materials of enterprises to measure the digital innovation activity [53]; some studies take the innovation results as the basis and measure the digital innovation output of enterprises from the perspective of digital innovation patents [54]; and some studies use the business income to measure the innovation outcome [55]. At present, digital innovation has become the focus of attention in all industries, and whether enterprises are willing to actively participate in digital innovation under the dual drive of government and industry is a prerequisite for enhancing digital innovation competitiveness [56]. Some enterprises respond to the innovation-driven development strategy, which may increase R&D expenditure investment in the short term, but excessively carry out low-end innovation with low investment and a short cycle, which fails to significantly improve the quality of innovation and is not conducive to the medium- and long-term development of enterprises [57]. Therefore, this paper explored the impact of different combinations of antecedent variables on firms’ digital innovation in terms of both digital innovation intention and digital innovation performance. Digital innovation intention refers to the degree of acceptance of digital thinking, digital technology, and digital model in the innovation process, reflecting the initiative and enthusiasm of enterprises to carry out digital innovation activities; digital innovation performance is the quantity and quality of new digital technologies created by enterprises, reflecting the level of output of digital innovation activities.

Digital innovation intention (DII). The annual report of a listed company can reflect the strategic layout and overall arrangement of the company. The higher the frequency of a certain type of keyword appearing in the annual report, the more attention and investment the company has paid to this aspect. Therefore, we used the frequency of keywords related to digital innovation in the annual reports of listed companies to measure digital innovation intention. Drawing on the research of Wu [58], first, we took the “Digital China Development Report 2022”, “Digital Transformation Report 2022”, and relevant important policy documents and research reports such as the “Government Work Report” in recent years as the blueprint, and extracted keywords related to digital innovation. On this basis, seven experts engaged in digital technology innovation and digital industry innovation management from government, universities, and enterprises were invited to form an expert group, we provided the experts with the background information of this research, and the experts carried out three rounds of verification, supplementation and deletion of the keywords to finalize the dictionary of keywords, as shown in

Table 1. Keyword dictionary for digital innovation.

Keyword Category	Keyword Dictionary
Artificial Intelligence Technology	Artificial Intelligence, Business Intelligence, Image Understanding, Investment Decision Support System, Intelligent Data Analysis, Intelligent Robot, Machine Learning, Deep Learning, Semantic Exploration, Biometric Technology, Face Recognition, Speech Recognition, Identity Verification, Automatic Driving, Natural Language Processing
Big Data Technology	Big data, Data Mining, Text Mining, Data Visualization, Heterogeneous Data, Credit Reporting, Augmented Reality, Mixed Reality, Virtual Reality
Cloud Computing Technology	Cloud Computing, Stream Computing, Graph Computing, In-memory Computing, Multi-Party Secure Computing, Brain Like Computing, Green Computing, Cognitive Computing, Converged Architecture, Billon-level Concurrency, EB-level Storage, Internet of Things, Cyber-Physical Systems
Blockchain technology	Blockchain, Digital Currency, Distributed Computing, Differential Privacy Technology, Intelligent financial contracts

. Finally, referring to the research method of Li [59], we used Python 3.7 software to obtain the annual reports of listed companies from 2017 to 2020 and searched according to the keyword dictionary, after excluding the negative words such as “no” and “none” before the keywords and the keywords of non-sample companies, the keyword frequency was summarized and calculated.

Digital innovation performance (DIP). As a direct manifestation of digital innovation results, digital innovation patents can objectively measure the digital innovation output of enterprises [60]. Therefore, we used digital innovation patents to measure digital innovation performance. Firstly, we used the “Digital Economy and Core Industries Unified Classification and International Patent Classification Reference Table (2023)” to obtain the patent classification numbers corresponding to the industries covered by the digital economy subcategories; then, we extracted the patents corresponding to each digital economy subcategory from the enterprise patent data to obtain the number of digital innovation patents of each enterprise. At the same time, considering the reverse causality between variables and the hysteresis of the input and output of innovation activities, this paper lagged the performance of digital innovation by one year. That is, the time of the antecedent variable data is from 2017–2020, and the time of digital innovation performance data is from 2018–2021.

3.3.2. Condition Variables

1.

Technological Dimension

R&D investment (R&D). Generally, the greater the proportion of an enterprise’s R&D investment in operating income, the higher its R&D intensity [61]. Therefore, we measured the enterprise’s R&D investment by the proportion of enterprise R&D investment in the current year’s operating income.

High-level talents (H-I). Employees with higher education backgrounds tend to have a stronger ability to perceive, utilize, integrate, and reconstruct digital technology. Therefore, the proportion of employees with a master’s degree or above is selected to represent the high-level talents [62].

2.

Organizational Dimension

Organizational size (OS). Drawing on existing research, we measured the size of the enterprise by the operating income of the enterprise [63]. The higher the operating income, the larger the size of the enterprise.

Top management team heterogeneity (H-TMT). We used the Blau coefficient to calculate the H-TMT, as shown in Equation (1). Where n represents the category of the professional background of executives, including 10 categories such as production, R&D, design, human resources, management, and marketing, and Pi is the percentage of members with the i-th background in the executive team. The Blau values range from 0 to 1, with closer to 1 indicating higher heterogeneity in the executive team [64].

$$\mathrm{Blau}=1-{\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{P}}_{\mathrm{i}}^2$$

(1)

3.

Environmental Dimension

Industry development speed (ID). We measured industry development speed by using the growth rate of industry business income compared with last year. This data came from the “China Digital Economy Development White Paper” published by the China Academy of Information and Communications Technology.

Regional digitalization level (RD). We used the urban digital economy index to measure regional digitalization level. The index covers many urban digital indicators such as urban information infrastructure, digital policy planning, digital construction operation, and digital operation effectiveness. The data came from the “China Urban Digital Economy Index White Paper” published by H3C Digital Economy Research Institute.

3.4. Calibration

The fsQCA method is based on the fuzzy set theory to explore the set relationship between the antecedent conditions and the outcome variables. Therefore, the original data need to be converted into the membership score of the set, that is, calibration. Data calibration needs to set a target set and calculate the degree of membership of each variable to the target set. After calibration, the set membership score ranges from 0 to 1. A score equal to 1 means complete membership, and a score equal to 0 means no membership at all [65]. Due to the large number of samples in this paper and the complexity of the situation, it is difficult for researchers to fully grasp the sample situation, and it is impossible to summarize the membership calibration value through case feature extraction. To solve the above problems and reduce the influence of subjective bias on the research conclusions, this paper chose the method of objective quantile value to determine the calibration value. Referring to relevant research, we used the direct calibration method, the 90% quantile is used as the full membership point (F), the 10% quantile is selected as the full non-membership point (N), and the mean value between the 90% quantile and the 10% quantile is used as the intersection point (I) [66]. The calibration points are shown in

Table 2. Calibration anchors.

Years	2017			2018			2019			2020
Variable	F	I	N	F	I	N	F	I	N	F	I	N
DII	95.2	27.7	2.0	99.9	32.0	2.0	107.3	33.1	3.0	96.6	33.6	4.0
DIP	195.7	43.2	1.0	205.9	51.8	2.0	235.3	54.1	2.0	115.9	27.1	1.0
R&D	19.0	8.1	3.3	19.4	8.0	3.5	19.8	9.2	3.6	19.7	9.4	3.6
H-I	19.9	6.1	1.0	20.0	6.2	1.0	20.7	6.5	1.2	20.4	6.6	1.2
OS	73.2	17.8	3.1	77.4	20.1	3.4	87.0	21.9	3.8	92.4	23.0	3.8
H-TMT	70.0	59.1	43.0	71.5	59.5	44.4	71.6	59.2	44.4	71.9	58.9	44.4
ID	14.2	13.6	13.2	12.4	10.4	9.0	16.4	9.9	4.5	13.3	10.6	8.3
RD	88.3	80.9	59.5	89.4	81.6	61.3	90.2	82.9	64.9	91.2	84.4	67.3

Note: F refers to full membership, I refers to intersection, and N refers to full non-membership.

.

4. Results

4.1. Necessity Conditions Analysis

Before configuration analysis, we conducted a necessity test on all variables to check whether a single conditional variable is a necessary condition for the outcome variable [67]. The consistency results of each antecedent variable on enterprise digital innovation intention and digital innovation performance are shown in

Table 3. Analysis results of necessary conditions.

Condition Variable	Digital Innovation Intention				Digital Innovation Performance
	2017	2018	2019	2020	2017	2018	2019	2020
R&D	0.633	0.660	0.643	0.662	0.529	0.563	0.559	0.574
H-I	0.615	0.611	0.628	0.616	0.585	0.583	0.605	0.618
OS	0.511	0.508	0.517	0.511	0.703	0.682	0.694	0.703
H-TMT	0.654	0.634	0.655	0.661	0.666	0.654	0.685	0.693
ID	0.659	0.668	0.664	0.661	0.334	0.340	0.349	0.384
RD	0.761	0.744	0.746	0.750	0.667	0.638	0.656	0.650

.

It can be observed that the consistency of each condition variable is not higher than 0.9, which does not constitute a necessary condition for the outcome variable [68]. This indicates that digital innovation intention and digital innovation performance in enterprises are jointly affected by multiple conditions, not the inevitable result of a single condition. It is necessary to further explore the linkage and synergy of various conditional variables of technology, organization, and environment.

4.2. Configuration Analysis

If the consistency between the configuration and the result is greater than or equal to 0.75, the configuration can be regarded as a sufficient condition for the outcome variable [69]. According to the distribution of consistency scores in the truth table, the original consistency threshold was set to 0.8, and the raw consistency benchmark (PRI) consistency threshold was set to 0.75 [70]. According to the number of samples, the frequency threshold was set to three, and the proportion of cases included in the QCA exceeded 80%.

4.2.1. Configuration Analysis of Digital Innovation Intention

The configurations of high digital innovation intention are shown in

Table 4. Configuration of high digital innovation intention.

Antecedent Conditions	2017			2018					2019					2020
	TE-a	TE-a	OE-a	TE-a	TE-a	OE-a	OE-a	TOE	TE-b	TE-c	OE-a	OE-a	OE-b	TE-c	OE-a
R&D	$•$			$•$	⊗	$\otimes$	$•$	●		●	$•$		$\otimes$	●
H-I	●	●		●	●	$\otimes$	$•$	⊗	●	●		$•$	⊗	●
OS			●	⊗		●	●	●			●	●	●		●
H-TMT	⊗	⊗		$\otimes$	⊗			$\otimes$	⊗		$\otimes$		$•$
ID	●	●	●	●	●	●	●	●	●	●	●	●	●	●	●
RD		$•$	●	⊗	$•$	●	●		●	●	●	●		●	●
Raw coverage	0.255	0.268	0.285	0.131	0.176	0.170	0.196	0.164	0.261	0.291	0.187	0.244	0.165	0.286	0.283
Unique coverage	0.032	0.011	0.083	0.021	0.028	0.022	0.034	0.015	0.010	0.038	0.011	0.014	0.034	0.092	0.090
Consistency	0.882	0.878	0.885	0.954	0.926	0.943	0.950	0.945	0.897	0.893	0.946	0.940	0.925	0.888	0.895
Solution coverage	0.383			0.321					0.395					0.375
Solution consistency	0.849			0.898					0.868					0.864

Notes: ● indicates presence of a condition, ⊗ indicates its absence. Large circles indicate core conditions, small circles indicate peripheral conditions. Blanks indicate “does not matter”.

. If a condition appears in both the parsimonious solution and the intermediate solution, it is a core condition that can play a significant role to the outcome variable; if it only appears in the intermediate solution, it can be seen as an edge condition, indicating an auxiliary contribution to the outcome variable [71]. To further explore the combination of the main antecedent conditions that affect the intention of enterprise digital innovation from 2017 to 2020, we classified the configuration results according to the TOE dimension to which the configuration core conditions belong [72]. There are three types of configurations in the four years: the technology–environment type (TE type) driven by technological and environmental factors, organization–environment type (OE type) driven by organizational and environmental factors and technology–organization–environment type (TOE type) driven by technological, organizational, and environmental factors.

1.

TE type driven by technological and environmental factors

This type of configuration takes technological and environmental factors as the core conditions, including TE-a type (High-level talents–Industry development speed), TE-b type (High-level talents–Industrial development speed–Regional digitization level), and TE-c type (R&D investment–High-level talents–Industrial development speed-Regional digitization level). Returning to the sample cases, we found that among the enterprises that achieved high innovation intention in the four years, 13 enterprises chose the TE type as the development path, and 11 of them have evolved from the TE-a type to TE-b or TE-c type.

In 2017 and 2018, the TE type represented itself as the TE-a type, indicating that enterprises with more high-level talents have a better understanding and application of digital technology, and enterprises can grasp the rich opportunities brought by the rapid development of the industry, improve the application of digital technology, and actively participate in digital innovation activities. A typical case under this configuration is Qingdao Eastsoft Communication Technology Co., Ltd., located in Qingdao, China. The company’s high-level talents account for nearly 20%. In 2017, Eastsoft launched a new round of equity incentive plans, and 89% of the restricted stocks were for the company’s core management employees and technological employees, maintaining the company’s attractiveness to high-level talents. In recent years, the power grid industry has generated an amount of chip demand. Eastsoft’s carrier technology team has deployed chip products in advance, and independently developed a new generation of carrier communication chips with talent advantages, promoting enterprises to actively participate in digital innovation.

In 2019 and 2020, the TE type was shown as the TE-b type (covering 26.1% of cases in 2019) and TE-c type (covering 29.1% of cases in 2019 and 28.6% of cases in 2020). This shows that in recent years, with the continuous in-depth development of digital innovation, the emerging new business formats have triggered new technological demands, and the regional digitalization level and R&D investment have significantly promoted the intention of enterprises in digital innovation. A typical case in this configuration is Wisesoft Co., Ltd., located in Chengdu, China. It used to be the Image Research Institute of Sichuan University and has abundant high-level talent resources and scientific research funds. In 2020, the scale of China’s artificial intelligence industry increased by 16.7% year-on-year. In the same year, Chengdu was approved as a national experimental zone for the innovation and development of new-generation artificial intelligence. Driven by the industry and the region, the company focused on promoting the transformation of traditional civil aviation products and services business, combined with market demand and technological competitive advantages, independently developing five innovative products, including high-precision 3D face recognition and controller training flight simulators that have achieved mass production application.

2.

OE type driven by organizational and environmental factors

This type of configuration takes organizational and environmental factors as the core conditions, including OE-a type (Organizational size–Industrial development speed–Regional digitization level) and OE-b type (Organizational size–Industry development speed). Returning to the sample cases, it can be found that among the enterprise groups that achieved high innovation intention in the four years, 12 enterprises chose the OE type as the development path.

The OE-a type appeared in each of the four years (Covering 28.5%, 36.6%, 43.1%, and 28.3% of the cases from 2017–2020, respectively), with high robustness. It shows that the dual drive of industry and region can effectively promote the active intention of large-scale enterprises in digital innovation. A typical case is Taiji Computer Corporation Limited, which provides integrated IT services for government, public security, national defense, and other enterprises. With the rapid development of digital government in recent years and the Beijing municipal government having also actively carried out digital transformation, Taiji, a leading enterprise in e-government and key industry informatization, has won bids for key projects such as the second phase of the central government portal, Beijing’s big data platform, and the Beijing Health Treasure support platform. In 2020, more than half of Taiji’s operating revenue came from government and institutional units, and about 50% came from Beijing.

The OE-b type appeared in 2019 (covering 16.5% of the cases), indicating that the rapid development of the industry can effectively promote large-scale enterprises to actively participate in digital innovation. A typical case is KAISA JiaYun Technology Inc., located in Dongguan, which is an enterprise in the field of Internet marketing. In 2019, the Internet industry still maintained a relatively high growth rate, among which the growth rate of information services and video services took the lead in the industry, with a year-on-year increase of 22.7%. The company has actively promoted the reform of the business structure and achieved the full coverage of mobile media, search engines, and information streaming media, strengthened cooperation with OPPO, Vivo, Xiaomi, Samsung, etc., and has won the affirmation and appreciation from several advertisers, such as “Xiaomi 2019 Best Partner Award” and Baidu five-star advertising agency qualification.

3.

TOE type driven by technological, organizational and environmental factors

The TOE type (R&D investment–Organizational size–Industrial development speed) appeared in 2018 (covering 16.4% of cases), indicating that high R&D investment, large organization scale, and fast industry development can promote the intention of enterprises in digital innovation. A typical case is Hand Enterprise Solutions Co., Ltd., located in Shanghai, China, which is a digital transformation service provider. In 2018, cloud products and cloud computing market opportunities grew rapidly, and intelligent manufacturing became an important development strategy for the manufacturing industry. To continuously improve the intelligent manufacturing software and cloud product services, HAND has invested CNY 403 million in research and development that year, an increase of 45.49% year-on-year, promoting the company’s transformation from information-based management to intelligent reform, expanding the company’s competitive advantage in the industry, and becoming a key player in digital innovation.

4.2.2. Configuration Analysis of Digital Innovation Performance

The configuration results of high digital innovation performance are shown in

Table 5. Configuration of high digital innovation performance.

Antecedent Condition	2017				2018				2019			2020
	TO-a	TO-a	TO-a	OE	TO-a	TO-b	TO-b	OE	TO-a	TO-a	TO-b	TO-b	TO-c
R&D		$\otimes$	$•$	$\otimes$		●	●	$\otimes$		$•$	●	●
H-I	●	●	●		●	$•$			●	●	$•$	$•$	●
OS	●	●	●	●	●	●	●	●	●	●	●	●	●
H-TMT		$\otimes$	$•$	⊗			$\otimes$	⊗	$•$		$•$		●
ID	⊗	⊗	⊗	⊗	⊗	⊗	⊗	⊗	⊗	⊗	⊗	⊗	⊗
RD	$•$			●	⊗		$\otimes$	●	⊗	⊗			⊗
Raw coverage	0.248	0.198	0.172	0.237	0.197	0.225	0.171	0.218	0.168	0.177	0.197	0.243	0.186
Unique coverage	0.044	0.018	0.011	0.066	0.025	0.026	0.022	0.065	0.020	0.029	0.049	0.084	0.027
Consistency	0.932	0.940	0.969	0.927	0.959	0.972	0.981	0.909	0.968	0.984	0.968	0.948	0.965
Solution coverage	0.351				0.349				0.246			0.270
Solution consistency	0.913				0.923				0.962			0.939

Notes: ● indicates presence of a condition, ⊗ indicates its absence. Large circles indicate core conditions, small circles indicate peripheral conditions. Blanks indicate “does not matter”.

. As above, according to the TOE dimension to which the configuration core conditions belong, we classified the configuration results of enterprise digital innovation performance from 2017 to 2020. It can be seen that there are two types of configurations in the four years: the configuration paths that drive high digital innovation performance include the technology–organization type (TO type) driven by technological and organizational factors, organization–environment type (OE type) driven by organizational and environmental factors.

1.

TO type driven by technological and organizational factors

This type takes technological and organizational variables as the core conditions, including the TO-a type (High-level talents–Organizational size), TO-b type (R&D investment–Organizational size), and TO-c type (High-level talent–Organizational size–Top management team heterogeneity). Returning to the sample cases, we found that among the enterprises that achieved high innovation performance in the four years, 11 enterprises chose the TO type as the main development path, and 5 of them evolved from the TO-a type to the TO-c type.

From 2017 to 2019, the main TO type was shown as the TO-a type, and this type of configuration sample accounted for the most in four years (61.8%, 16.8%, and 34.5% of the cases covered in 2017, 2018, and 2019, respectively). A typical case is Tsinghua Tongfang Co., Ltd., located in Beijing, China, which ranked 20th in China’s electronic information list in 2017. As the first university-run listed company of Tsinghua University, relying on the support of talents from universities, Tongfang has built several integrated platforms of industry, academia, and research institution, including two academician workstations and eight joint laboratories. A large number of high-level talents has ensured that the company is in the leading innovation level at home and abroad in the field of the Internet of Things, security system, and big data.

In 2020, the TO-a type evolved into the TO-c type covering 18.6% of the cases. Both TO-a type and TO-c type indicate that high-level talents could promote digital innovation performance. The difference is that in the TO-c type, in addition to the high-level talents team, the heterogeneous top management team also plays a key role. It shows that as the digital economy driven by scientific and technological innovation enters a period of accelerated development, the role of the top management team is becoming increasingly prominent in the decision-making regarding corporate innovation direction and innovation strategy. A typical case still is Tsinghua Tongfang Co., Ltd. On 31 December 2019, Tsinghua University and the China National Nuclear Corporation jointly promoted the university–enterprise reform of Tongfang. The controlling shareholder of Tongfang was changed from Tsinghua Holdings to China National Nuclear Corporation, driving the development strategy of high-quality technological innovation with the dual-core of “intelligent manufacturing +resources”. In 2020, the company made outstanding innovations in digital information, civil nuclear technology, energy conservation, and environmental protection, and won many awards such as the China Patent Award and the first prize of the Chinese Institute of Electronics Science and Technology Progress Award.

The TO-b type appeared in 2018–2020 (39.6%, 19.7%, and 24.3% of the cases were covered in 2018, 2019, and 2020, respectively), indicating high R&D investment and large organizational scale firms are more likely to produce high digital innovation performance. A typical case is Goodix Technology, Inc., which has continued to increase R&D investment over the years. In 2020, the company’s R&D expenditure was CNY 1.754 billion, an increase of 62.55% over 2019, providing stable financial support for corporate innovation activities. The company’s low-power Bluetooth technology, multi-functional interactive sensors, and other new businesses are in the stage of mass production and commercial promotion. The total number of patent applications and authorizations at home and abroad has exceeded 6000. It has a multi-business layout in the electronic equipment manufacturing industry and has continuously made breakthroughs.

2.

OE type driven by organizational and environmental factors

The OE type (Organizational size–Regional digitization level) appeared in 2017 (covering 23.7% of the cases) and 2018 (covering 21.8% of the cases), indicating that a large organization and a high level of regional digitalization can effectively improve the performance of corporate digital innovation. A typical case is Konka Group Co., Ltd., located in Shenzhen, China, which is a leading enterprise in China’s color TV industry. Shenzhen’s digital economy development ranks first in China. In recent years, it has been vigorously building a science and technology system reform pilot zone and promoting innovative mechanism system reform. From 2017 to 2018, the company fully utilized the rich local digital innovation resources and innovation environment, implemented a new development strategy oriented by transformation and upgrading, and determined the positioning of a platform company driven by technological innovation. The 8K decoding chip, face recognition, far-field voice, and other technologies launched by the company all have independent intellectual property rights. The “Digital TV Broadcasting System and Localization of Core Chips” project won the “Science and Technology Progress Award” in 2018, which is the highest national science and technology award.

4.3. Robustness Test

We tested the robustness of the configuration results by increasing the PRI consistency threshold and case frequency threshold [73]. The results show that there is a clear subset relationship between the original configuration and the new configuration, the coverage and consistency of the configurations in each year have relatively small changes, and the interpretation of the digital innovation by the original configuration has not changed, so the results are relatively reliable.

Taking the 2020 data as an example, we initially raised the PRI consistency threshold to 0.8, and high digital innovation intention produced three configurations. The first configuration is a proper subset of the original configuration TE-c, while the other two configurations are proper subsets of the original configuration OE-a. As for the high digital innovation performance, two configuration patterns align with the original TO-b and TO-c configurations. Furthermore, we increased the frequency threshold to 4, and three high digital innovation intention configurations emerged. The first configuration is completely consistent with the original configuration TE-c, and the other two configurations are proper subsets of the original configuration OE-a. High digital innovation performance has two configurations: the first configuration is a proper subset of the original configuration TO-b, and the second configuration is largely consistent with the original configuration TO-c, as shown in

Table 6. Robustness test.

Antecedent Condition	Digital Innovation Intention						Digital Innovation Performance
	Increasing the PRI Consistency Threshold			Increasing the Frequency Threshold			Increasing the PRI Consistency Threshold		Increasing the Frequency Threshold
R&D	●			●	⊗	⊗	●		●
H-I	●	●		●	⊗		$•$	●	$•$	●
OS		●	●		●	●	●	●	●	●
H-TMT	⊗		⊗			⊗		●		$•$
ID	●	●	●	●	●	●	⊗	⊗	⊗	⊗
RD	●	●	$•$	●	●	●		⊗	$•$	⊗
Raw coverage	0.210	0.235	0.224	0.286	0.171	0.178	0.243	0.186	0.194	0.186
Unique coverage	0.050	0.047	0.036	0.149	0.013	0.012	0.084	0.027	0.071	0.063
Consistency	0.927	0.931	0.926	0.888	0.934	0.948	0.948	0.965	0.942	0.965
Solution coverage	0.320			0.351			0.270		0.257
Solution coverage	0.913			0.877			0.939		0.939

Notes: ● indicates presence of a condition, ⊗ indicates its absence. Large circles indicate core conditions, small circles indicate peripheral conditions. Blanks indicate “does not matter”.

.

5. Conclusions and Contributions

5.1. Conclusions and Discussion

How to effectively utilize digital technology, allocate internal and external resources of the organization, formulate a digital innovation strategy, and build core competitiveness in a digital environment are the keys to promoting the high-quality development of the digital economy. This paper takes the TOE framework as the logical basis of the research, takes China’s Shanghai and Shenzhen A-share digital economy core industry listed companies as the research object, and constructs a configuration analysis model of enterprise digital innovation from three aspects: technology, organization, and environment. We incorporate the time factor into fsQCA, conducting multi-stage dynamic fsQCA, analyzing the influence of synergistic effects of multiple factors on enterprise digital innovation intention and digital innovation performance from 2017 to 2020, and finding out what combination of driving factors companies are more willing to actively participate in digital innovation activities and achieve high-quality innovation output. The conclusions are as follows:

First, this paper finds that technological, organizational, and environmental conditions constitute the three types of configuration to achieve high digital innovation intention, which are the TE type driven by technological and environmental factors, the OE type driven by organizational and environmental factors, and the TOE type driven by technological, organizational and environmental factors. The TE type and OE type cover more years and samples, and they appear in all four time windows. Among them, the TE type evolves with high-level talents and industrial development speed as the core conditions, including TE-a type (High-level talents–Industry development speed), TE-b type (High-level talents–Industrial development speed–Regional digitization level), and TE-c type (R&D investment–High-level talents–Industrial development speed–Regional digitization level), which indicate that the dual roles of technological and environmental factors are the effective ways to increase the enterprises’ digital innovation intention, in which high-level talents and industrial development speed are the keys to drive the enterprises to actively participate in digital innovation under the TE path. In the OE type, the organizational size, regional digitalization level, and industrial development speed are taken as the core conditions, and are stable during the four-year sample period, including OE-a type (Organizational size–Industrial development speed–Regional digitization level) and OE-b type (Organizational size–Industry development speed), indicating that driven by the environmental factors of both industries and regions, it can effectively enhance the enthusiasm of large-scale enterprises to participate in digital innovation. This resonates with the results of several studies, for example, Nadkarni et al. [74] and Verhoef et al. [75] found that when enterprises are in an environment with high digitalization, large-scale enterprises have a stronger willingness to undertake digital challenges. In the TOE type, R&D investment, organizational size, and industrial development speed are the core conditions, indicating that driven by the industry environment, large-scale enterprises with higher R&D investment are more actively involved in digital innovation [10].

Second, this paper finds that technological, organizational, and environmental conditions constitute two types of configuration to drive high digital innovation performance of enterprises, which are the TO type driven by technological and organizational factors and the OE type driven by organizational and environmental factors. Among them, the TO type covers more years and samples and appears in all four time windows, including the TO-a type (High-level talents–Organizational size), TO-b type (R&D investment–Organizational size), and TO-c type (High-level talent–Organizational size–Top management team heterogeneity). This shows that the dual role of technological and organizational factors is an effective way to drive the high digital innovation performance of enterprises, large-scale enterprises can achieve high digital innovation performance by attracting high-level talents or increasing R&D investment, and in the current digital industry, large-scale enterprises can easier achieve high digital innovation performance. The OE type takes organizational size and regional digitization level as the core conditions, and appeared in 2017–2018, indicating that the dual role of organizational and environmental factors is an effective way to drive the high digital innovation performance of enterprises, and regional digital policies and digital infrastructure construction have a significant effect on improving the digital innovation performance of large-scale enterprises. This is consistent with the studies of Ciriello and Nambisan [56,76], whose research show that a good external environment provides complementary resources for digital innovation entities, speeds up the transfer of digital innovation value to end users, and achieves the sustainable development of digital innovation.

5.2. Theoretical Implications

First, this paper introduces the TOE framework into the study of the digital innovation mechanism, which enriches the toolbox of digital innovation mechanism research theories. Based on the group state perspective, the theoretical model of digital innovation affecting enterprises in the core industries of China’s digital economy is constructed from six antecedent conditions at three levels of technology, organization, and environment, which is a supplement and improvement to the traditional regression model of digital innovation mechanism research, and also enriches the research application of multi-level factors under the TOE framework.

Second, most of the existing studies on digital innovation driving mechanisms focus on the outcomes of digital innovation and use business revenue measures without considering the new technological output outcomes of digital technologies, and ignore the fact that an enterprise’s intention for digital innovation is an important representation for predicting enterprise digital innovation behavior. This paper integrates behavioral and outcome theories, divides digital innovation into digital innovation intention and digital innovation performance, and uses fsQCA to explore the key antecedent conditions that affect the digital innovation intention and digital innovation performance of enterprises as well as the linkage between the conditions, deepening the research on the driving mechanism of enterprise digital innovation.

Third, this paper applies the dynamic QCA research method to the study of digital innovation mechanisms to explore the configuration changes under the time dimension. Previous studies have used regression analysis more often, ignoring the dependence between cause and effect; on the other hand, the QCA method has used cross-sectional data more often. This paper uses the dynamic QCA method to explore and compare the digital innovation paths of enterprises under multiple time windows, analyze the changes in the combination of antecedent conditions, and reveal the dynamic realization paths of digital innovation intention and digital innovation performance of Chinese digital economy industry enterprises under the time dimension.

5.3. Practical Implications

Through the research in this paper, it can be found that there is a phenomenon of “different paths to the same destination” in the digital innovation intention and digital innovation performance. It is necessary to strengthen the three-dimensional element construction and synergy of “technology-organization-environment”, promote enterprises to actively participate in digital innovation activities and improve their digital innovation performance, in order to maintain the competitiveness and sustainable development advantages of enterprises in a complex market environment, which are as follows:

(1) There are linkages among technological, organizational, and environmental dimensions in stimulating enterprises’ intention to digital innovation. Firstly, there is a linkage between technological and environmental dimensions, and the digital technology is changing rapidly, so enterprises should pay attention to the introduction and training of high-level digital talents, strengthen digital technology, digital processing, and digital business training, and increase the strength and level of joint training of talents by industry, universities, and research institutes, which can effectively improve the insight of enterprises into industry development, promote digital innovation to a deeper level and effectively enhance the enterprise’s ability to resist risk. Secondly, there is a linkage between organizational and environmental dimensions, and the government and other relevant departments should accelerate the construction of digital infrastructure, smooth digital resources, build institutional mechanisms for data management, unleash the value potential of commercial data, develop standards for cross-fertilization development and other applications in various industries, and promote large-scale enterprises to unleash the vitality of digital innovation. Finally, there is also a linkage between technology, organization, and environment, facing the booming development of the digital economy, enterprises should continuously strengthen innovation investment, enhance innovation support and incentive, and create a relaxed innovation atmosphere, which can effectively improve the belief and intention of enterprises to innovate digitally and actively explore new business models and expand new fields.

(2) There are linkages among the technological, organizational, and environmental dimensions in enhancing the digital innovation performance of enterprises. Firstly, there is a linkage between technological and organizational dimensions: enterprises should increase market share and achieve economies of scale through corporate financing, brand extension, and chain operations, strengthen the gathering of talent, capital, and other technological innovation elements, and play the role of talent advantages and financial security, such as promoting new breakthroughs in digital innovation in large-scale enterprises by implementing equity incentive plans and setting up special funds for research and development; enterprises should also create a diversified top management team structure, connect various business links, strengthen top-level design, guarantee a highly flexible organizational structure with innovative iterative upgrades. Secondly, there is also a linkage between organizational and environmental dimensions, indicating that enterprises should focus on cooperating with industry leaders or leading enterprises and on actively participating in the construction of key government digital projects while actively expanding their business areas and playing to their organizational strengths; at the same time, the government should also study and formulate measures to promote the high-quality development of the digital industry, create digital industry clusters with international competitiveness, improve the digital economy governance system, enhance the level of government, industry, and market supervision, strengthen intellectual property protection, establish an international exchange and cooperation system in the digital field with multi-level synergy, multi-platform support, and multi-subject participation, and promote the application and promotion of enterprises’ digital innovation achievements.

5.4. Limitations and Prospects

This research also has some limitations. First, patents are only one form of digital innovation output for enterprises. In the future, other innovation outputs such as papers, products, and processes can be considered to comprehensively measure the performance of enterprise digital innovation. Second, since this study considers the effect of time on configurations, there are many configurations generated, and in order to mine the typical configurations of enterprise digital innovation intention and digital innovation performance from 2017 to 2020, the configurations are classified according to the TOE dimension of core condition membership, meaning there is also less argumentation for edge conditions and non-conditions, so future research can combine edge conditions and non-conditions to classify configurations under multiple periods.