Ambidextrous Innovation, Organizational Resilience, and the High-Quality Development of Enterprises: A Dynamic Analysis Based on the Enterprise Life Cycle

Abstract

Ambidextrous innovation, defined as an organization’s capacity to simultaneously pursue exploratory innovation and exploitative innovation, serves as the core driving force for the high-quality development of enterprises. Based on existing studies, this paper explores the mechanisms among ambidextrous innovation, organizational resilience, and the high-quality development of enterprises from the dynamic perspective of the enterprise life cycle. The findings indicate that both exploratory innovation and exploitative innovation positively influence the high-quality development of enterprises. Moreover, at different stages of the enterprise life cycle, the effects of exploratory innovation and exploitative innovation on the high-quality development of enterprises are different. Organizational resilience partially mediates the relationship between the two dimensions of ambidextrous innovation and the high-quality development of enterprises. This study can help enterprises enhance the scientific and effective implementation of ambidextrous innovation activities in the context of coexisting crises, while also providing theoretical guidance and practical implications for their high-quality development.

1. Introduction

Against the backdrop of globalization, digitalization, and rapid technological advancements, enterprises are navigating an increasingly complex and volatile market environment. Achieving high-quality development amidst such uncertainties has emerged as a critical challenge for their survival and growth. High-quality development requires enterprises not only to maintain operational efficiency in the short term [1], but also to cultivate the capacity for continuous innovation and adaptation to environmental changes in the long run [2]. Within this context, ambidextrous innovation—defined as the ability to simultaneously pursue exploratory and exploitative innovation—has been recognized as a pivotal driver of high-quality development [3,4]. Exploratory innovation enables enterprises to venture into new markets and technological domains [5,6], while exploitative innovation enhances operational efficiency through the optimization of existing resources [7,8]. The equilibrium and synergy between these two forms of innovation can significantly bolster an enterprise’s competitiveness and adaptability [9]. Empirical studies suggest that ambidextrous innovation can markedly improve innovation capabilities and market performance [10,11], although some scholars argue for a non-linear relationship between ambidextrous innovation and enterprise value. Despite differing conclusions, a consensus exists that ambidextrous innovation is particularly effective in helping resource-constrained, newly established enterprises strengthen their competitive positions in the market [12].

In practice, enterprises often face the “innovation paradox”, which refers to the inherent conflicts between exploratory innovation and exploitative innovation in terms of resource allocation and organizational structure [4]. Furthermore, the dynamism and uncertainty of the external environment have exacerbated the complexity of enterprise innovation. Against this backdrop, organizational resilience, defined as the ability of enterprises to adapt, recover, and achieve redevelopment in the face of crises [13,14], has emerged as a focal point in both academic research and practical applications. Studies indicate that organizational resilience can support ambidextrous innovation by enhancing enterprises’ resource flexibility and environmental adaptability [15,16]. Additionally, organizational resilience plays a pivotal role in promoting high-quality development, enabling enterprises to mitigate risks, improve operational efficiency, and achieve sustainable growth [17]. Consequently, exploring the relationships among ambidextrous innovation, organizational resilience, and high-quality development holds significant theoretical and practical importance.

Most existing studies have predominantly focused on a single innovation model or adopted a static perspective, leaving a significant gap in systematic research that explores the relationships among ambidextrous innovation, organizational resilience, and high-quality development from the dynamic perspective of the enterprise life cycle. According to the enterprise life cycle theory, enterprises encounter distinct resource constraints, market environments, and development goals at different stages of their development [18]. Consequently, the roles of ambidextrous innovation and organizational resilience in fostering high-quality development may vary significantly depending on the specific life cycle stage of the enterprise. Building on this foundation, this paper aims to investigate the mechanisms through which ambidextrous innovation and organizational resilience influence high-quality development, adopting a dynamic perspective rooted in the enterprise life cycle. By doing so, it seeks to provide both theoretical insights and practical guidance for enterprises striving to achieve high-quality development in complex and uncertain environments.

The contributions of this paper are as follows: Firstly, it comprehensively takes into account the innovation heterogeneity and enterprise life cycle heterogeneity, conducts an in-depth analysis of the differences in the impact of ambidextrous innovation on high-quality enterprise development at different stages, and provides incremental evidence for the empirical research on the relationship between innovation and high-quality enterprise development. Secondly, organizational resilience is incorporated into the research path, where innovation affects high-quality enterprise development, clarifying the mechanism through which ambidextrous innovation influences high-quality enterprise development by enhancing organizational resilience. Thirdly, exploring the relationship between ambidextrous innovation and the high-quality development of enterprises at different life cycle stages can assist enterprises in carrying out exploratory and exploitative innovation activities in a more scientific and targeted manner in the VUCA (Volatile, Uncertain, Complex, and Ambiguous) environment.

2. Theory and Hypotheses

2.1. Ambidextrous Innovation and High-Quality Enterprise Development

Numerous studies have demonstrated that the high-quality development of enterprises is primarily driven by innovation, with green and sustainable development serving as fundamental prerequisites [19]. This approach emphasizes more efficient resource allocation, the production of high-quality products, and the delivery of high-quality services. It underscores the importance of balancing economic value with social value, thereby achieving advanced and sustainable enterprise development [20]. Building upon this conceptual framework, the subsequent discussion will examine the respective impacts of exploratory innovation and exploitative innovation on high-quality enterprise development.

Ambidextrous innovation, comprising exploitative and exploratory innovation, represents a dual approach to organizational development. Exploratory innovation focuses on pursuing novel technologies and knowledge, potentially disrupting existing technological accumulations and knowledge stocks, thereby challenging current market offerings. This form of innovation is instrumental in exploring new markets, developing unique core competitive advantages through the research and development of new technologies, products, and production processes. It delivers more valuable products and services to new markets, significantly enhancing an enterprise’s competitive strength and development capabilities [6]. Exploratory innovation serves dual purposes: it empowers enterprises to strengthen their brand influence and elevate industry entry barriers through the development of unique technologies, innovative products, and cutting-edge services. These distinctive advancements create significant challenges for competitors to replicate or follow, thereby solidifying the enterprise’s competitive edge and securing a dominant position in the market [21]. Additionally, it elevates the enterprise’s overall technological level, aids in identifying new opportunities and markets, enhances organizational adaptability to technological and market changes, and broadens the enterprise’s survival prospects. This study concludes that exploratory innovation not only boosts future earnings and long-term competitiveness but also fortifies the enterprise’s core competitive advantage. It enhances adaptability and responsiveness to market changes, ultimately improving total factor productivity and promoting high-quality development.

Conversely, exploitative innovation leverages existing knowledge and technology to refine and optimize current products, services, and production processes. Characterized by high success rates, short cycles, and low investment, it effectively meets the current demands of the target market and maintains market stability. Exploitative innovation reduces costs and enhances operational efficiency, thereby improving short-term financial performance and stabilizing existing business operations. It also strengthens the competitiveness of existing products, better satisfies mainstream customer needs, increases customer loyalty, and maintains market share, reinforcing the enterprise’s competitive position and core advantages. Thus, exploitative innovation serves as a strategic approach for enterprises to optimize internal operations, reduce costs, and enhance production efficiency. By enabling refined upgrades and aligning with market demands, it significantly improves customer satisfaction. Furthermore, it supports organizations in steadily expanding market shares and solidifying competitive advantages, thereby fostering sustainable growth and high-quality development [22]. Based on the above analysis, this paper puts forward the following research hypotheses:

H1a: 

Exploratory innovation has a significant positive impact on the high-quality development of enterprises.

H1b: 

Exploitative innovation has a significant positive impact on the high-quality development of enterprises.

2.2. The Mediating Role of Organizational Resilience

Exploratory innovation seeks to transcend the enterprise’s existing knowledge base through the research and development of new products and technologies. This approach enhances the enterprise’s adaptability to external market and technological shifts, providing a stronger buffer against the adverse effects of emergencies and bolstering the enterprise’s defensive and adaptive capabilities in crisis situations [23]. By emphasizing the expansion of knowledge search scope, exploratory innovation enables enterprises to accumulate new knowledge and technologies over time. This accumulation helps break through conventional thinking patterns, reduces response times, and offers more diverse and flexible solutions, thereby improving the enterprise’s resilience [24]. Furthermore, exploratory innovation necessitates the collection of extensive information on customer needs, technological advancements, and market dynamics. This process enhances the organization’s predictive capabilities [25] and builds a repository of experience in addressing various challenges. Consequently, the enterprise is better equipped to anticipate risks and respond swiftly to emergencies, mitigating negative impacts and maintaining organizational stability.

Exploitative innovation focuses on addressing customer needs in the existing market by optimizing and upgrading products and services to enhance customer loyalty. Simultaneously, it evaluates opportunities and threats within the current market and predicts near-term industry trends [26], thereby improving the organization’s ability to perceive and defend against market changes. On one hand, exploitative innovation activities heighten enterprises’ sensitivity to shifts in market conditions and customer demand. When confronted with unexpected crises, such enterprises can respond more swiftly than their competitors, making timely adjustments to products and services. This agility mitigates the adverse effects of market fluctuations, stabilizes short-term earnings, and supports enterprise survival during challenging times [27]. Exploitative innovation fosters an organizational culture of consistently monitoring market changes and industry trends. It enables enterprises to continuously refine products and reallocate resources in response to evolving market conditions, thereby enhancing the stability of organizational development [28]. Consequently, both exploratory and exploitative innovation contribute to strengthening organizational resilience. From the perspective of dynamic capabilities, organizational resilience is a core capability for enterprises to navigate dynamic environments, with resource allocation playing a pivotal role in this process. In such environments, where markets, technologies, or external conditions may undergo significant changes, enterprises must strengthen resilience through strategic and efficient resource allocation [29]. The enhancement of organizational resilience underscores the necessity for organizations to internalize learning as a core capability. By establishing robust learning systems, they can ensure the continuity of learning processes, facilitate knowledge acquisition, dissemination, and interpretation, and strengthen organizational memory. These mechanisms collectively form dynamic capabilities that enable continuous renewal and adaptation, allowing organizations to thrive in complex and changing environments [30]. Thus, strengthening organizational resilience signifies a further enhancement of the organization’s comprehensive capabilities, laying a solid foundation for improving the enterprise’s overall strength and development quality. In summary, the implementation of innovation activities during enterprise development facilitates resource integration and restructuring. It also enhances the organization’s ability to anticipate, respond to, and adapt to changes, thereby reinforcing the enterprise’s competitive advantages and resilience. This, in turn, supports the achievement of long-term, stable performance and promotes sustained, high-quality development. Based on the above analysis, this paper puts forward the following research hypotheses:

H2a: 

Organizational resilience mediates the relationship between exploratory innovation and the high-quality development of enterprises.

H2b: 

Organizational resilience mediates the relationship between exploitative innovation and the high-quality development of enterprises.

2.3. The Impact of Ambidextrous Innovation on the High-Quality Development of Enterprises at Different Stages of the Enterprise Life Cycle

Enterprise innovation activities span the entire life cycle of an organization. According to the enterprise life cycle theory, enterprises at different stages of their life cycle exhibit distinct organizational structures, resource endowments, and operate within diverse market environments. They also encounter varying development objectives, business risks, and innovation demands. As a result, the influence of ambidextrous innovation on the high-quality development of enterprises may differ significantly across these stages. Since the sample enterprises are all listed companies and have generally moved beyond the start-up phase, the start-up and growth stages are consolidated into a single growth stage. Accordingly, the enterprise life cycle is categorized into three stages: growth, maturation, and recession. This classification facilitates an analysis of the differential impacts of the two dimensions of ambidextrous innovation on the high-quality development of enterprises across these three stages [31].

Enterprises in the growth stage are typically characterized by a relatively small scale, limited market share, and high production costs. These enterprises exhibit a strong drive for innovation and tend to prioritize the development of new products [32]. At this stage, enhancing market recognition and expanding market reach are critical for the survival and growth of enterprises. On one hand, exploratory innovation serves as a vital strategy to achieve these goals. By allocating resources to research and develop new technologies, create innovative products, or explore untapped markets, enterprises can unlock potential opportunities, strengthen their competitive edge, and achieve differentiated growth. These efforts not only drive business expansion but also foster long-term enterprise development [33]. On the other hand, enterprises can solidify the gains from exploratory innovation by continuously refining existing products and services through exploitative innovation. This approach further reduces production costs, enhances operational efficiency, and sustains current earnings, thereby supporting the enterprise’s high-quality development.

Enterprises in the maturation stage typically operate with a well-established management system. On one hand, they maintain a stable business model and command a relatively high market share. On the other hand, they encounter intense market competition, which often results in a deceleration of development speed and a scarcity of new profit growth points. Compared to enterprises in the growth stage, those in the maturation stage can leverage their resource and experience advantages to not only strengthen innovations within their existing business domains but also strategically pursue cross-field and expansionary innovations. This dual approach enables them to sustain growth momentum, inject vitality into their operations, and effectively avoid stagnation or growth bottlenecks [34]. Exploitative innovation maximizes the potential of existing products and services by fully leveraging original technologies. By capitalizing on technological monopolies and product advantages, enterprises can achieve substantial returns through this form of innovation. Simultaneously, the implementation of exploratory innovation activities enables enterprises to penetrate new markets, uncover additional potential opportunities, and identify new profit growth points, thereby expanding their development horizons. To a certain extent, this can extend the maturity period of the enterprise and infuse continuous vitality into the organization’s sustainable development.

Enterprises in the recession stage face significant challenges, including aging equipment, outdated technology, and declining production efficiency, which lead to increased costs and reduced competitiveness. Simultaneously, their market shares shrink, customer retention weakens, and brand influence diminishes. The inability to innovate and launch market-aligned products further exacerbates their deteriorating market position, pushing them into a cycle of declining performance and relevance [35]. Although exploitative innovation, which involves upgrading and optimizing existing products, can assist enterprises in alleviating the issues of customer loss and capital shortage in the short term, it fails to address the genuine difficulties that enterprises encounter during the recession stage. Therefore, enterprises are increasingly inclined to adopt exploratory innovation strategies. By developing cutting-edge technologies, launching innovative products, or venturing into new markets, they can revitalize their competitive edge, unlock new growth opportunities, and optimize resource allocation. These efforts enable them to adapt to market dynamics, mitigate competitive pressures, and continuously create value by redefining market boundaries [36]. Such transformative initiatives can catalyze a “second startup” for enterprises, propelling them into a new life cycle phase and paving the way for sustainable development. In summary, for enterprises in the recession stage, exploratory innovation proves to be more conducive to their development. Based on the above points, the current study proposed the following hypotheses:

H3a: 

For firms in the growth stage, both exploratory innovation and exploitative innovation contribute to high-quality development. Moreover, the contribution of exploratory innovation to high-quality development is more significant than that of exploitative innovation.

H3b: 

For firms in the maturation stage, both exploratory innovation and exploitative innovation make contributions to high-quality development. Moreover, the contribution of exploitative innovation to high-quality development is more significant than that of exploratory innovation.

H3c: 

For firms in the recession stage, both exploratory innovation and exploitative innovation contribute to the high-quality development of enterprises. Moreover, the contribution of exploratory innovation to high-quality development is more significant than that of exploitative innovation.

3. Methodology

3.1. Sample Selection and Data Sources

This paper selected China’s A-share listed companies in Shanghai and Shenzhen from 2012 to 2022 as the research sample and collected relevant data from the CSMAR database, the WIND database, and the China Research Data Services Platform (CNRDS). To ensure the reliability of the regression results, the sample data collected in this study were processed in the following ways: (1) Enterprises with significant missing data regarding major indicators were excluded. (2) Samples of companies in the financial and insurance industries, as well as ST and *ST companies, were excluded. (3) All continuous variables were winsorized at the 1% level.

We selected the 2012–2022 period in China as it represents a complete economic cycle following the nation’s entry into the “New Normal” phase, characterized by significant structural transformations. This specific decade was chosen to avoid both earlier data limitations and the incomplete economic cycles post-2022. This period’s distinctive economic evolution provides robust and reliable data for analyzing our research questions within China’s unique developmental context.

3.2. Variable Setting

3.2.1. Explained Variables

High-quality enterprise development (TFP): Referring to the research of Bernini et al. [37], this paper employed the single indicator of firm total factor productivity to measure the high-quality development of firms. Enterprise total factor productivity, which represents the comprehensive productivity of various factors within an enterprise, plays a vital role in enhancing the quality of enterprise development. As a result, it is frequently utilized as an important criterion by scholars when evaluating the high-quality development of enterprises. In this paper, the LP method was adopted to calculate the total factor productivity of enterprises for the purpose of measuring the high-quality development of enterprises in the benchmark regression. Additionally, the OP method was utilized to conduct the robustness test. The LP method (Levinsohn–Petrin method) and the OP method (Olley–Pakes method) are both significant econometric methods for estimating the total factor productivity (TFP) of enterprises in economics, aiming to solve the endogeneity problem in the estimation of production functions.

3.2.2. Explanatory Variables

Ambidextrous Innovation: Referring to Guan and Liu [38], this paper employed the number of patent applications filed by the firm in a given year as a variable measurement benchmark. Invention patents signify that enterprises have achieved breakthroughs and innovations in technology, which aligns with the conceptual connotation of exploratory innovation. Meanwhile, design and utility model patents are primarily concerned with improving the functions and quality of existing products, thus reflecting the connotation of exploitative innovation. Consequently, this paper adopted the natural logarithm of the number of invention patents plus one to measure exploratory innovation and utilizes the natural logarithm of the sum of the number of design and utility model patents plus one to measure exploitative innovation.

3.2.3. Mediating Variables

Organizational resilience can be conceptualized as a two-dimensional construct comprising high-performance growth and low financial volatility. Following the methodology proposed by Ortiz [39], these two dimensions were measured through specific indicators.

For measuring long-term performance growth, we employed the three-year cumulative sales revenue growth as the primary indicator. This approach is preferred over year-on-year growth metrics as it better captures the sustained development trajectory of enterprises. The three-year timeframe aligns with established academic practices and has been widely adopted in scholarly research. Financial volatility, the second dimension of organizational resilience, is quantified through stock return volatility. This is calculated as the standard deviation of monthly stock returns over a one-year period, providing a robust measure of financial stability and risk exposure.

3.2.4. Control Variables

Based on the studies conducted by existing scholars, the following indicators were selected as control variables: years of listing (Age), enterprise size (Size), profitability (Roa), solvency (Lev), enterprise growth (Growth), enterprise risk (Risk), and the size of the board of directors (Board). Specific variable descriptions are shown in

Table 1. Variable definitions and descriptions.

Type	Variable Name	Variable Symbol	Variable Definition
Explained Variable	High-Quality Enterprise Development	TFP	Total Factor Productivity of Enterprises by Lp Method
Explanatory Variable	Exploratory Innovation	Exploratory	Ln (Number of Patents for Inventions + 1)
	Exploitative Innovation	Exploitative	Ln (Number of Design and Utility Model Patents + 1)
Mediating Variable	Organizational Resilience	Resilience	Composite Indicator of Growth in Performance and Financial Volatility
Control Variable	Number of Years Listed	Age	Age of Listing
	Enterprise Size	Size	Natural Logarithm of Total Assets
	Profitability	Roa	Return on Assets = Net Profit/Total Assets
	Solvency	Lev	Gearing Ratio = Liabilities/Total Assets
	Corporate Growth	Growth	Revenue Growth Rate
	Business Risk	Risk	Consolidated Leverage = Financial Leverage * Operating Leverage
	Board Size	Board	The Number of Board Members Is Taken as a Natural Logarithm
	Year	Year	Year Dummy Variable
	Industry	Ind	Industry Dummy Variables

.

3.3. Classification of Enterprise Life Cycle Stages

This paper refers to Dickinson’s cash flow approach to classify the life cycle of a firm [40]. In this approach, the determination of the life cycle stage in which a firm is located is based on the signs of the firm’s net cash flow from operating activities, net cash flow from investing activities, and net cash flow from financing activities. Following Wang et al. [31], the business life cycle is divided into three stages—the growth stage, the maturation stage, and the recession stage—as presented in

Table 2. Classification of enterprise life cycle stages.

	Growth Stage		Maturation Stage	Recession Stage
	Inception	Growth Stage	Maturation Stage	Turbulent Stage	Turbulent Stage	Turbulent Stage	Recession Stage	Recession Stage
Cash Flow From Operating Activities Symbol	-	+	+	-	+	+	-	-
Cash Flow From Investing Activities Symbol	-	-	-	-	+	+	+	+
Cash Flow From Financing Activities Symbol	+	+	-	-	+	-	+	-

.

3.4. Modeling

In order to test the relationship between the two dimensions of ambidextrous innovation proposed and the high-quality development of enterprises, this paper constructs models (1) and (2):

$$TFP={\alpha }_0+{\alpha }_{1}Exploratory+{\alpha }_{2}Control+\sum Ind+\sum Year+{\epsilon }_1$$

(1)

$$TFP={\alpha }_0+{\alpha }_{1}Exploitative+{\alpha }_{2}Control+\sum Ind+\sum Year+{\epsilon }_1$$

(2)

To test the relationship between the two dimensions of ambidextrous innovation and organizational resilience, regression models (3) and (4) are constructed:

$$Res={\beta }_0+{\beta }_{1}Exploratory+{\beta }_{2}Control+\sum Ind+\sum Year+{\epsilon }_2$$

(3)

$$Res={\beta }_0+{\beta }_{1}Exploitative+{\beta }_{2}Control+\sum Ind+\sum Year+{\epsilon }_2$$

(4)

For the mediating role of organizational resilience, this paper adopts the step-by-step testing method to verify the relationship among ambidextrous innovation, organizational resilience [41], and the high-quality development of enterprises. This is achieved by adding the mediating variable, organizational resilience, on the basis of models (1) to (4), and constructing the model as follows:

$$TFP={\gamma }_0+{\gamma }_{1}Exploratory+{\gamma }_{2}Res+{\gamma }_{3}Control+\sum Ind+\sum Year+{\epsilon }_3$$

(5)

$$TFP={\gamma }_0+{\gamma }_{1}Exploitative+{{\gamma }_{2}Res+\gamma }_{3}Control+\sum Ind+\sum Year+{\epsilon }_3$$

(6)

where TFP represents the high-quality development of firms, Exploratory represents exploratory innovation, Exploitative represents exploitative innovation, Res indicates organizational resilience, Control represents a series of control variables, α, β, and γ are constants, ε is the residual term, and two dummy variables, industry and year, are controlled within the model.

4. Results

4.1. Descriptive Statistics

This study employed Stata 15 software for statistical analysis, and the results of the descriptive statistics of the main variables are presented in

Table 3. Descriptive statistics for key variables.

Variable Name	Sample Size	Average Value	Standard Deviation	Minimum Value	Median	Maximum Value
TFP	22,418	8.373	1.036	5.204	8.256	13.096
Exploratory	22,418	1.361	1.446	0	1.098	9.028
Exploitative	22,418	1.424	1.589	0	1.098	9.221
Res	22,418	0.891	0.054	0.742	0.899	0.974
Size	22,418	22.290	1.295	20.045	22.085	26.302
Age	22,418	9.660	7.262	0	8	30
Roa	22,418	0.052	0.040	0.010	0.043	0.198
Risk	22,418	2.128	2.139	0.873	1.447	15.48
Growth	22,418	0.189	0.336	−0.412	0.128	1.970
Lev	22,418	0.396	0.195	0.0482	0.387	0.833
Board	22,418	2.127	0.198	1.099	2.197	2.890

. As can be seen from the results in the table, the maximum value of high-quality enterprise development is 13.096, the minimum value is 5.204, and the standard deviation is 1.036, which indicates that there are large differences in the development quality among different enterprises, and there exists the phenomenon of unbalanced development. The maximum values of exploratory innovation and exploitative innovation are 9.028 and 9.221, respectively, and the minimum value is 0. The means and medians of the two are relatively small, which indicates that the overall innovation level of enterprises is low, and there are still large differences in the innovation level among different enterprises, and there are still some enterprises that carry out fewer innovative activities. Moreover, the mean value of exploratory innovation is smaller than the mean value of exploitative innovation, indicating that, overall, enterprises produce more outputs from exploitative innovation activities than from exploratory innovation. The maximum value of organizational resilience is 0.974, and the minimum value is 0.742, indicating that there is also a certain gap in resilience capacity among different enterprises.

From the point of view of the control variables, the sample company asset size and enterprise years of experience of the maximum value and the minimum value have a significant gap, indicating that the selection of sample enterprises is more extensive, the size of the differences between the different enterprises is still more obvious, and the number of years on the market has large differences; for the enterprise revenue growth rate, the maximum value is 1.970 and the minimum value is −0.412, indicating that the gap between the different enterprises is very large and the ability to develop more disparately. The results for the rest of the control variables are also all in line with the results for the other control variables. The results for the rest of the control variables are also within a reasonable range.

4.2. Correlation Analysis

Before conducting the regression analysis of the model in this paper, a Pearson correlation test was first performed on each variable to avoid the influence of the multicollinearity problem.

Table 4. Correlation analysis.

	TFP	Exploratory	Exploitative	Res	Size	Age	Roa	Risk	Growth	Lev	Board
TFP	1
Exploratory	0.141 ***	1
Exploitative	0.113 ***	0.714 ***	1
Res	0.092 ***	0.020 ***	0.011 *	1
Size	0.795 ***	0.149 ***	0.117 ***	0.137 ***	1
Age	0.370 ***	−0.117 ***	−0.141 ***	0.107 ***	0.450 ***	1
Roa	0.024 ***	0.089 ***	0.075 ***	−0.021 ***	−0.097 ***	−0.176 ***	1
Risk	−0.013 *	0.007	0.0110	−0.002	0.095 ***	0.117 ***	−0.436 ***	1
Growth	0.090 ***	0.003	−0.015 **	0.030 ***	0.019 ***	−0.105 ***	0.172 ***	−0.118 ***	1
Lev	0.555 ***	0.007	0.030 ***	0.024 ***	0.567 ***	0.347 ***	−0.398 ***	0.280 ***	0.056 ***	1
Board	0.173 ***	0.024 ***	0.011 *	0.011 *	0.262 ***	0.162 ***	−0.037 ***	0.062 ***	−0.032 ***	0.144 ***	1

Note: *** denotes 1% level of significance, ** denotes 5% level of significance, * denotes 10% level of significance.

shows the results of the correlation analysis of the main variables. From the results, it can be seen that the correlation coefficient between exploratory innovation and high-quality enterprise development is 0.141, and the correlation coefficient between exploitative innovation and high-quality enterprise development is 0.113, and both are significant at the 1% level. This preliminarily finding suggests that there is a positive correlation between the two dimensions of ambidextrous innovation and the high-quality development of the enterprise. The correlation coefficients of exploratory innovation and exploitative innovation with organizational resilience are 0.020 and 0.011, respectively, and both are significant at the 1% level, which indicates that there is a positive impact of ambidextrous innovation on organizational resilience.

4.3. Multicollinearity Test

To ensure the robustness of the research findings, this study employed the Variance Inflation Factor (VIF) method to examine multicollinearity among variables. The VIF value serves as an indicator of the linear correlation between a given variable and other independent variables, with higher values indicating more severe multicollinearity issues. According to established statistical standards, VIF < 5 suggests negligible multicollinearity that can generally be disregarded. When 5 ≤ VIF < 10, it indicates moderate collinearity, necessitating careful attention to the significance of key variables. The analysis results, as presented in

Table 5. Multicollinearity test.

Variables	VIF (Exploratory)	1/VIF (Exploratory)	VIF (Exploitative)	1/VIF (Exploitative)
Exploratory	1.36	0.733086	-	-
Exploitative	-	-	1.42	0.702552
Size	2.12	0.471363	2.06	0.485137
Age	1.48	0.674895	1.49	0.669973
Roa	1.53	0.654056	1.53	0.652210
Risk	1.34	0.745520	1.34	0.745502
Growth	1.12	0.896701	1.12	0.895077
Lev	2.10	0.475794	2.10	0.475158
Board	1.14	0.878931	1.14	0.878914

, demonstrate that all VIF values for the examined variables fall within the range of 1.12 to 2.12, well below the threshold of 5. This finding confirms the absence of significant multicollinearity among the variables. The exclusion of multicollinearity concerns ensures that subsequent empirical regression analyses will not be substantially affected by multicollinearity issues, thereby enhancing the reliability of the research outcomes.

4.4. Benchmark Regression Results

4.4.1. Regression Analysis on Ambidextrous Innovation and High-Quality Development of Enterprises

Column 1 of

Table 6. Benchmark regression results.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	TFP	TFP	Res	Res	TFP	TFP
Exploratory	0.023 ***		0.072 ***		0.023 ***
	(8.07)		(6.12)		(7.93)
Exploitative		0.019 ***		0.088 ***		0.019 ***
		(7.33)		(8.01)		(7.15)
Res					0.532 ***	0.522 ***
					(3.30)	(3.24)
Size	0.558 ***	0.561 ***	0.372 ***	0.372 ***	0.556 ***	0.559 ***
	(140.67)	(143.34)	(22.58)	(22.96)	(138.63)	(141.22)
Age	0.003 ***	0.003 ***	0.025 ***	0.026 ***	0.003 ***	0.003 ***
	(5.27)	(5.40)	(10.07)	(10.48)	(5.04)	(5.16)
Roa	3.772 ***	3.763 ***	−2.987 ***	−3.112 ***	3.788 ***	3.779 ***
	(34.60)	(34.46)	(−6.61)	(−6.88)	(34.72)	(34.58)
Risk	−0.028 ***	−0.029 ***	−0.034 ***	−0.034 ***	−0.028 ***	−0.028 ***
	(−15.28)	(−15.30)	(−4.38)	(−4.42)	(−15.18)	(−15.20)
Growth	0.108 ***	0.109 ***	−0.678 ***	−0.667 ***	0.112 ***	0.113 ***
	(9.77)	(9.86)	(−14.73)	(−14.50)	(10.05)	(10.13)
Lev	1.234 ***	1.229 ***	−1.273 ***	−1.299 ***	1.241 ***	1.236 ***
	(46.53)	(46.30)	(−11.58)	(−11.81)	(46.66)	(46.42)
Board	−0.064 ***	−0.064 ***	0.173 **	0.169 **	−0.065 ***	−0.065 ***
	(−3.39)	(−3.37)	(2.19)	(2.14)	(−3.44)	(−3.42)
Constant	−4.824 ***	−4.879 ***	84.684 ***	84.676 ***	−5.274 ***	−5.321 ***
	(−53.74)	(−55.03)	(227.49)	(230.51)	(−32.29)	(−32.68)
Ind	YES	YES	YES	YES	YES	YES
Year	YES	YES	YES	YES	YES	YES
N	22,418	22,418	22,418	22,418	22,418	22,418
R-squared	0.739	0.740	0.837	0.837	0.729	0.731

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; **, and *** indicate statistical significance at 5%, and 1%, respectively.

presents the regression results between exploratory innovation and the high-quality development of enterprises. The results show that the regression coefficient of exploratory innovation on the high-quality development of enterprises is 0.023, which is significant at the 1% level. This indicates that exploratory innovation has a significant positive impact on the high-quality development of enterprises, thus validating hypothesis H1a. Column 2 of Table 6 presents the regression results of the relationship between exploitative innovation and high-quality enterprise development. The results indicate that the regression coefficient of exploitative innovation on the enterprise’s high-quality development capability is 0.019, which is also significant at the 1% level. This suggests that exploitative innovation has a positive impact on high-quality enterprise development, thereby verifying hypothesis H1b of this paper. This finding indicates that both components of ambidextrous innovation, namely exploratory innovation and exploitative innovation, positively contribute to and facilitate the high-quality development of enterprises. In practical terms, when enterprises implement ambidextrous innovation strategies, they effectively promote their high-quality development.

4.4.2. Analysis of the Regression Results of the Mediating Effect

This paper employs a step-by-step test to verify the mediating role of organizational resilience between the two dimensions of ambidextrous innovation and the high-quality development of enterprises. The first step tests the relationship between the explanatory variables and the dependent variable. According to the regression results of models (1)–(2) in Table 6, it is evident that both exploratory innovation and exploitative innovation have a significant positive correlation with the high-quality development of the enterprise, indicating that both dimensions of ambidextrous innovation contribute to the high-quality development of the enterprise. The second step tests the relationship between the explanatory variables and the mediator variable. The results of models (3)–(4) show a significant positive relationship between both exploitative innovation and exploratory innovation and organizational resilience. The third step involves regression analysis again after adding the mediating variable, organizational resilience. According to the regression results of model (5), the regression coefficients of exploratory innovation for the high-quality development of enterprises and organizational resilience are 0.023 and 0.532, respectively, both significant at the 1% level. According to the regression results of model (6), the regression coefficients of exploitative innovation for the high-quality development of enterprises and organizational resilience are 0.019 and 0.522, also significant at the 1% level. Thus, organizational resilience has a partial mediating role between exploratory innovation, exploitative innovation, and the high-quality development of the enterprise, indicating that ambidextrous innovation can promote the high-quality development of the enterprise through the enhancement of organizational resilience capacity, thereby supporting hypotheses H2a and H2b. To a certain extent, the results demonstrate that the two components of ambidextrous innovation, namely exploratory innovation and exploitative innovation, exert a positive and facilitating effect on the high-quality development of enterprises by strengthening organizational resilience. In other words, when enterprises adopt ambidextrous innovation strategies, they enhance their organizational resilience, thereby further driving their high-quality development.

4.4.3. Grouped Regression Analysis Based on the Enterprise Life Cycle

On the basis of the above research, this paper divides the enterprise into three stages, namely growth, maturation, and recession, for group regression, and the group regression results are shown in

Table 7. Enterprise life cycle grouping regression results.

Variables	Growth Stage		Maturation Stage		Recession Stage
	TFP	TFP	TFP	TFP	TFP	TFP
Exploratory	0.026 ***		0.020 ***		0.018 **
	(6.02)		(5.60)		(2.21)
Exploitative		0.017 ***		0.022 ***		0.011 **
		(4.22)		(5.95)		(2.57)
Size	0.547 ***	0.551 ***	0.564 ***	0.565 ***	0.582 ***	0.584 ***
	(88.82)	(90.64)	(97.77)	(99.76)	(53.39)	(54.16)
Age	0.006 ***	0.006 ***	0.003 ***	0.003 ***	−0.006 ***	−0.006 ***
	(5.78)	(5.80)	(3.75)	(3.90)	(−3.92)	(−3.90)
Roa	3.729 ***	3.769 ***	3.702 ***	3.671 ***	3.424 ***	3.427 ***
	(18.09)	(18.28)	(25.13)	(24.86)	(12.78)	(12.78)
Risk	−0.034 ***	−0.034 ***	−0.026 ***	−0.026 ***	−0.024 ***	−0.024 ***
	(−12.23)	(−12.17)	(−9.01)	(−9.06)	(−4.99)	(−5.00)
Growth	0.135 ***	0.134 ***	0.064 ***	0.068 ***	0.186 ***	0.186 ***
	(8.67)	(8.53)	(3.20)	(3.39)	(6.69)	(6.68)
Lev	1.172 ***	1.169 ***	1.257 ***	1.249 ***	1.514 ***	1.511 ***
	(27.22)	(27.10)	(31.66)	(31.43)	(22.97)	(22.92)
Board	−0.076 **	−0.074 **	−0.070 **	−0.071 **	−0.059	−0.057
	(−2.57)	(−2.50)	(−2.51)	(−2.58)	(−1.17)	(−1.14)
Constant	−4.542 ***	−4.641 ***	−4.969 ***	−4.996 ***	−5.177 ***	−5.231 ***
	(−32.44)	(−33.56)	(−38.03)	(−38.71)	(−21.29)	(−21.72)
Ind	YES	YES	YES	YES	YES	YES
Year	YES	YES	YES	YES	YES	YES
N	9363	9363	8897	8897	4158	4158
R-squared	0.731	0.730	0.784	0.784	0.692	0.692

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; **, and *** indicate statistical significance at 5%, and 1%, respectively.

. As can be seen from the results, the role of exploratory innovation in promoting the high-quality development of enterprises is significant at the 1% level in all three stages, and its promoting effect is the most pronounced in the growth stage. The role of exploitative innovation in the high-quality development of firms is also significant at all three stages (growth, maturation, and recession), with the greatest contribution being made in the maturation stage. For enterprises in the growth stage, the regression coefficients of exploratory innovation and exploitative innovation for the high-quality development of enterprises are 0.026 and 0.017, respectively, both of which are significant at the 1% level, indicating that exploratory innovation promotes the high-quality development of enterprises more effectively than exploitative innovation in the growth stage, thus verifying hypothesis H3a. For firms in the maturation stage, the coefficient of exploratory innovation is 0.020, which is significant at the 1% level, while the coefficient of exploitative innovation is 0.022, which is also significant at the 1% level. This suggests that both exploratory and exploitative innovations can contribute to the high-quality development of the enterprises, and exploitative innovations contribute more than exploratory innovations. In the recession stage, the coefficient of exploratory innovation is 0.018, which is significant at the 5% level, and the coefficient of exploitative innovation is 0.011, which is significant at the 5% level. This also proves that both exploratory and exploitative innovations can promote the high-quality development of enterprises, thereby verifying hypothesis H3c. To a certain degree, this finding indicates that the correlation between ambidextrous innovation and high-quality enterprise development demonstrates distinct intensity characteristics and undergoes stage-specific variations across different phases of the corporate life cycle.

4.5. Robustness Tests

4.5.1. Replacement of Explained Variable Measures

In order to test the robustness of the findings of this paper, the total factor productivity of enterprises calculated using the OP method is used to replace the measure of the explanatory variable for high-quality enterprise development. The model is then re-estimated based on the new indicator after replacement, and the regression results are shown in

Table 8. Regression results with replacement of explanatory variables.

Variables	(1)	(2)	(3)	(4)	(5)	(6)
	TFP	TFP	Res	Res	TFP	TFP
Exploratory	0.005 **		0.072 ***		0.005 ***
	(1.99)		(6.12)		(2.55)
Exploitative		0.007 ***		0.088 ***		0.007 ***
		(2.82)		(8.01)		(2.93)
Res					0.293 ***	0.330 ***
					(3.30)	(3.24)
Size	0.397 ***	0.401 ***	0.372 ***	0.372 ***	0.396 ***	0.400 ***
	(100.85)	(103.36)	(22.58)	(22.96)	(99.45)	(101.86)
Age	0.004 ***	0.003 ***	0.025 ***	0.026 ***	0.004 ***	0.003 ***
	(6.14)	(5.54)	(10.07)	(10.48)	(6.00)	(5.38)
Roa	2.914 ***	2.955 ***	−2.987 ***	−3.112 ***	2.923 ***	2.965 ***
	(26.93)	(27.28)	(−6.61)	(−6.88)	(26.99)	(27.35)
Risk	−0.025 ***	−0.025 ***	−0.034 ***	−0.034 ***	−0.025 ***	−0.025 ***
	(−13.36)	(−13.34)	(−4.38)	(−4.42)	(−13.30)	(−13.28)
Growth	0.148 ***	0.146 ***	−0.678 ***	−0.667 ***	0.150 ***	0.148 ***
	(13.49)	(13.22)	(−14.73)	(−14.50)	(13.60)	(13.36)
Lev	0.990 ***	0.993 ***	−1.273 ***	−1.299 ***	0.994 ***	0.998 ***
	(37.61)	(37.72)	(−11.58)	(−11.81)	(37.64)	(37.77)
Board	−0.108 ***	−0.105 ***	0.173 **	0.169 **	−0.108 ***	−0.106 ***
	(−5.71)	(−5.59)	(2.19)	(2.14)	(−5.73)	(−5.62)
Constant	−2.735 ***	−2.822 ***	84.684 ***	84.676 ***	−2.983 ***	−3.102 ***
	(−30.71)	(−32.09)	(227.49)	(230.51)	(−18.40)	(−19.20)
Ind	YES	YES	YES	YES	YES	YES
Year	YES	YES	YES	YES	YES	YES
N	22,418	22,418	22,418	22,418	22,418	22,418
R-squared	0.637	0.637	0.312	0.311	0.641	0.641

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; **, and *** indicate statistical significance at 5%, and 1%, respectively.

. It can be observed that after changing the measure of the high-quality enterprise development variable, the regression coefficient of exploratory innovation for high-quality enterprise development is 0.005, which is significant at the 5% level. The regression coefficient of exploitative innovation for high-quality enterprise development is 0.007, which is significant at the 1% level. The relationship between exploratory innovation, exploitative innovation, and high-quality enterprise development remains significant and positive, which is in line with the results from the previous section. Consistent with the previous results, after adding the mediating variable of organizational resilience, the coefficients of the explanatory variables of exploratory innovation and exploitative innovation are positive and significant, and the coefficients of the mediating variables are also significant. This proves that the mediating effect of organizational resilience remains valid and is consistent with the findings of the previous study, indicating that the findings of this paper possess a certain degree of robustness.

4.5.2. Bootstrap Method to Test the Mediation Effect

In order to test the robustness of the mediating effect, this paper employs the Bootstrap method to re-test the mediating effect of organizational resilience. The test results are shown in

Table 9. Bootstrap method mediation effect test.

		Coef	Std. Err.	[95% Conf. Interval]
Exploratory	indirect effect	0.0018	0.0008	0.0002	0.0034
	direct effect	0.0657	0.0261	0.0132	0.1174
Exploitative	indirect effect	0.0014	0.0003	0.0009	0.0020
	direct effect	0.0235	0.0033	0.0170	0.0300

. We can observe that the effect value of the direct effect and the indirect effect of the mediating variable within the 95% confidence interval does not include 0, which further verifies that the mediating effect of organizational resilience exists.

4.6. Endogeneity Test

To address the endogeneity problem arising from sample self-selection bias, this study employs the propensity score matching (PSM) method for analysis. The sample is divided into distinct groups based on whether enterprises have engaged in exploratory innovation or exploitative innovation. Covariates such as enterprise scale, years since listing, profitability, asset–liability ratio, growth rate, risk level, and board size are utilized for matching, with the 1:1 nearest neighbor matching method applied. To ensure the reliability of the PSM results, a balance test is conducted on all covariates, all of which pass the test successfully. Subsequently, regression analysis is performed on the matched sample data post-PSM processing. The regression results, as presented in

Table 10. Endogeneity test regression results.

Variables	TFP	TFP
Exploratory	0.030 ***
	(7.25)
Exploitative		0.020 ***
		(5.37)
Size	0.549 ***	0.559 ***
	(94.97)	(101.65)
Age	0.003 ***	0.002 ***
	(2.98)	(2.92)
Roa	3.705 ***	3.647 ***
	(22.84)	(23.73)
Risk	−0.030 ***	−0.030 ***
	(−10.79)	(−11.39)
Growth	0.123 ***	0.126 ***
	(7.50)	(7.92)
Lev	1.232 ***	1.225 ***
	(31.49)	(32.69)
Board	−0.133 ***	−0.055 **
	(−4.73)	(−2.06)
Constant	−4.496 ***	−4.862 ***
	(−34.88)	(−38.93)
Ind	YES	YES
Year	YES	YES
N	10,750	11,668
R-squared	0.733	0.743

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; **, and *** indicate statistical significance at 5%, and 1%, respectively.

, reveal that both exploratory innovation and exploitative innovation exert a significantly positive impact on the high-quality development of enterprises, with coefficients of 0.030 and 0.020, respectively, both significant at the 1% level. This suggests that both types of innovation significantly enhance the high-quality development of enterprises. These findings align with the conclusions drawn from the main regression analysis in the preceding section, thereby confirming that the endogeneity issue has been effectively addressed.

4.7. Further Analysis

4.7.1. The State-Owned Enterprises and the Non-State-Owned Enterprises

Given that state-owned enterprises (SOEs) and non-state-owned enterprises (non-SOEs) differ in terms of their internal and external environments, resource bases, organizational structures, and systems, there are substantial disparities in their implementation of innovation activities and innovation inputs. Thus, when studying the impact of innovation activities, the nature of property rights must be considered.

Consequently, this paper categorizes all sample enterprises into SOEs and non-SOEs based on the property rights nature and conducts group regression. The aim is to examine whether the difference in the property rights nature influences the relationship between exploratory innovation, exploitative innovation, and the high-quality development of enterprises.

Table 11. Test for heterogeneity of SOEs and non-SOEs.

Variables	State-Owned Enterprise Group		Non-State Enterprise Group
	TFP	TFP	TFP	TFP
Exploratory	0.0004		0.043 ***
	(0.07)		(10.12)
Exploitative		−0.00003		0.004 ***
		(−0.03)		(4.50)
Size	0.558 ***	0.563 ***	0.567 ***	0.566 ***
	(88.95)	(90.87)	(106.46)	(107.41)
Age	0.008 ***	0.008 ***	−0.004 ***	−0.004 ***
	(7.60)	(7.46)	(−5.13)	(−4.84)
Roa	3.532 ***	3.529 ***	3.871 ***	3.845 ***
	(15.96)	(15.91)	(31.74)	(31.49)
Risk	−0.028 ***	−0.027 ***	−0.029 ***	−0.029 ***
	(−9.56)	(−9.48)	(−11.93)	(−11.92)
Growth	0.188 ***	0.189 ***	0.078 ***	0.080 ***
	(8.91)	(8.95)	(6.13)	(6.31)
Lev	1.153 ***	1.145 ***	1.213 ***	1.206 ***
	(23.95)	(23.77)	(38.43)	(38.20)
Board	−0.223 ***	−0.226 ***	0.015	0.015
	(−6.55)	(−6.63)	(0.66)	(0.65)
Constant	−13.446 ***	−13.471 ***	−19.997 ***	−19.694 ***
	(−10.68)	(−10.35)	(−18.08)	(−17.10)
Ind	YES	YES	YES	YES
Year	YES	YES	YES	YES
N	7813	7813	14,605	14,605
R-squared	0.297	0.297	0.302	0.298

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; *** indicate statistical significance at 1%.

presents the results of the grouping test for SOEs and non-SOEs. Columns (1) and (2) show the test results for the SOE group. The regression results between exploratory innovation, exploitative innovation, and the high-quality development of enterprises are not significant. Columns (3) and (4) display the regression results for the non-SOE group. In the non-SOE group, the regression coefficients of exploratory innovation and exploitative innovation are 0.043 and 0.004 respectively, and both are significant at the 1% level. This indicates a significant positive correlation between exploratory innovation, exploitative innovation, and the high-quality development of enterprises.

The reason for these divergent results might be that, compared to non-SOEs, SOEs generally enjoy a higher market position and a robust resource base. As a result, they face less survival pressure, tend to pursue stable development, and have a relatively weak innovation willingness. Additionally, their innovation mechanisms may not be well established, and the rates of R&D achievements and result transformation are low. Consequently, the enhancement of the high-quality development ability of SOEs through innovative activities is limited. In contrast, non-SOEs operate in a more competitive market environment. They are under pressure to innovate continuously to maintain their core competitiveness, avoid market elimination, and ensure the survival and long-term development of the enterprise.

4.7.2. High-Tech and Non-High-Tech

Enterprises with different industry natures exhibit significant disparities in terms of market environment, innovation level, and access to innovation resources. Moreover, their reliance on innovation varies as well. To explore whether there are differences in the impact of ambidextrous innovation on the high-quality development of enterprises in high-tech and non-high-tech industries, this paper refers to the method of defining high-tech industries. It categorizes enterprises into high-tech and non-high-tech industry groups and then examines the relationships between exploratory innovation, exploitative innovation, and the high-quality development of enterprises within each industry group separately.

As can be seen from the regression results in

Table 12. Tests for heterogeneity of high-tech industries and non-high-tech industries.

Variables	High-Tech Industry Group		Non-High-Tech Industry Group
	TFP	TFP	TFP	TFP
Exploratory	0.042 ***		0.023 ***
	(8.25)		(4.56)
Exploitative		0.005 ***		0.002 *
		(6.18)		(1.80)
Size	0.518 ***	0.522 ***	0.580 ***	0.583 ***
	(88.67)	(90.84)	(107.81)	(109.88)
Age	0.011 ***	0.011 ***	−0.001	−0.001
	(12.62)	(12.60)	(−1.28)	(−1.30)
Roa	3.911 ***	3.906 ***	3.771 ***	3.763 ***
	(27.49)	(27.42)	(23.49)	(23.38)
Risk	−0.035 ***	−0.035 ***	−0.024 ***	−0.024 ***
	(−12.70)	(−12.74)	(−9.51)	(−9.47)
Growth	0.050 ***	0.051 ***	0.156 ***	0.155 ***
	(3.16)	(3.23)	(10.21)	(10.16)
Lev	1.297 ***	1.285 ***	1.166 ***	1.165 ***
	(36.26)	(35.82)	(30.62)	(30.56)
Board	−0.015	−0.017	−0.075 ***	−0.074 ***
	(−0.60)	(−0.66)	(−2.80)	(−2.73)
Constant	−3.289 ***	−3.362 ***	−5.264 ***	−5.335 ***
	(−6.69)	(−6.84)	(−44.02)	(−45.17)
Ind	YES	YES	YES	YES
Year	YES	YES	YES	YES
N	9657	9657	12,761	12,761
R-squared	0.725	0.725	0.744	0.744

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; *, and *** indicate statistical significance at 10%, and 1%, respectively.

, the regression coefficients of exploratory innovation on the high-quality development of enterprises in the two groups are 0.042 and 0.023, respectively, and both are significant at the 1% level. This indicates that the positive impact of exploratory innovation on the high-quality development of enterprises is more pronounced in enterprises within high-tech industries.

The regression result of exploitative innovation for the high-quality development of high-tech enterprises is 0.005, which is significant at the 1% level, while the regression coefficient for non-high-tech industries is 0.002, which is only significant at the 10% level. From the above results, it is evident that the bimodal innovation in the high-tech industry group generally makes a greater contribution to the high-quality development of firms than that in the non-high-tech industry. This conclusion may stem from the fact that most enterprises in non-high-tech industries focus more on the cost and quality of their products, and innovation is not a top priority. Even if the innovation level is low, it will not have a significant impact on the survival and development of these enterprises. In contrast, the high-tech industry, which is technology-intensive and knowledge-intensive, being based on high technology use, mainly relies on exploratory innovation. Enterprises invest a substantial amount of capital and employ many scientific researchers in the research and development of new products and technologies to provide customers with higher-quality, more diversified, and differentiated products and services in order to capture the market.

Therefore, in the high-tech industry, having a higher level of technology and a faster R&D speed is more crucial, and these aspects are inseparable from innovation activities. Enterprises with a low level of innovation will face the risk of being phased out of the market if they are unable to adapt to technological and product iterations.

4.7.3. Manufacturing Enterprises and Service Enterprises

When investigating innovation indicators across enterprises, notable disparities emerge between the manufacturing and service enterprises in terms of total factor productivity and various innovation metrics. The manufacturing enterprises, which primarily depend on fixed-asset investments for production, direct their innovation efforts towards enhancing production processes and developing new products. This sector leverages technology to boost efficiency, reduce costs, and improve product quality. Conversely, the service enterprises, with their emphasis on intangible services, focus their innovative strategies on refining service models, optimizing service delivery, and expanding service channels. Given these distinct characteristics, an analysis of industry heterogeneity is essential.

This study categorizes the sample into manufacturing and service enterprises based on industry classification and conducts empirical tests using group regression. The regression results, presented in

Table 13. Tests for heterogeneity of manufacturing enterprises and service enterprises.

Variables	Manufacturing Enterprises		Service Enterprises
	TFP	TFP	TFP	TFP
Exploratory	0.022 ***		0.008
	(7.90)		(0.97)
Exploitative		0.019 ***		0.005
		(7.82)		(0.62)
Size	0.560 ***	0.562 ***	0.531 ***	0.532 ***
	(123.44)	(125.85)	(59.88)	(60.65)
Age	0.011 ***	0.011 ***	−0.012 ***	−0.012 ***
	(16.44)	(16.62)	(−8.52)	(−8.51)
Roa	3.840 ***	3.842 ***	4.200 ***	4.201 ***
	(35.39)	(35.40)	(13.49)	(13.49)
Risk	−0.024 ***	−0.024 ***	−0.054 ***	−0.054 ***
	(−12.87)	(−12.78)	(−9.08)	(−9.09)
Growth	0.045 ***	0.047 ***	0.236 ***	0.236 ***
	(3.57)	(3.72)	(9.65)	(9.64)
Lev	1.060 ***	1.052 ***	1.710 ***	1.708 ***
	(37.24)	(36.87)	(26.88)	(26.85)
Board	−0.024	−0.023	−0.154 ***	−0.154 ***
	(−1.15)	(−1.10)	(−3.32)	(−3.33)
Constant	−4.684 ***	−4.735 ***	−3.938 ***	−3.959 ***
	(−49.47)	(−50.77)	(−20.14)	(−20.37)
Ind	YES	YES	YES	YES
Year	YES	YES	YES	YES
N	15,004	15,004	7414	7414
R-squared	0.760	0.760	0.695	0.695

Notes: The numbers in the table are variable regression coefficients; the numbers in the corresponding brackets are t-value; *** indicate statistical significance at 1%.

, reveal significant relationships between both exploratory and exploitative innovation and the high-quality development of enterprises within the manufacturing enterprises. In contrast, the service enterprises group shows no significant regression results between these types of innovation and high-quality enterprise development. This suggests that ambidextrous innovation, namely exploratory innovation and exploitative innovation, effectively promotes high-quality development in manufacturing enterprises, but has a limited impact in the service enterprises.

5. Discussion

Taking organizational resilience as the mediating variable, this paper incorporates ambidextrous innovation, organizational resilience, and high-quality enterprise development into the same research framework. It explores the impact of ambidextrous innovation on high-quality enterprise development at different life cycle stages and reveals the mechanism through which ambidextrous innovation influences high-quality enterprise development capability under the VUCA environment.

Using Shanghai and Shenzhen A-share listed companies as research samples, in line with the research content of this paper and based on the organizational ambidextrous theory, dynamic capability theory, and enterprise life cycle theory, we proposed hypotheses and conducted empirical tests. Eventually, the following research conclusions were drawn:

Firstly, ambidextrous innovation significantly contributes to high-quality development of enterprises. The findings of this study indicate that both dimensions of ambidextrous innovation, namely exploratory innovation and exploitative innovation, can effectively contribute to the high-quality development of enterprises. Exploratory innovation facilitates the formation of sustainable competitive advantages, primarily through its emphasis on novel product development and technological innovation, coupled with proactive market trend anticipation. This strategic orientation enables enterprises to achieve long-term developmental objectives. Conversely, exploitative innovation drives enterprise advancement through the optimization and transformation of existing products, technologies, and service systems. This approach not only enhances immediate financial performance but also strengthens market positioning, thereby establishing a robust foundation for sustained high-quality development.

Secondly, ambidextrous innovation significantly enhances organizational resilience through established behavioral mechanisms. Longitudinal analysis demonstrates that enterprises consistently engaging in both exploratory and exploitative innovation activities develop robust organizational routines in market monitoring and opportunity identification. These routines contribute to three critical capabilities: (1) enhanced adaptability to market environmental changes, (2) improved risk anticipation and crisis identification, and (3) strengthened emergency response and problem-solving capacity. The synergistic effect of these capabilities enables enterprises to maintain operational stability and significantly improve organizational resilience. These findings show that both dimensions of ambidextrous innovation serve as crucial drivers in building organizational resilience.

Thirdly, organizational resilience is a significant partial mediator in the relationship between ambidextrous innovation and high-quality enterprise development. The empirical evidence suggests that through the simultaneous implementation of exploratory and exploitative innovation activities, enterprises establish a robust mechanism for resilience enhancement. This enhanced resilience subsequently contributes to developmental quality improvement through three distinct pathways: (1) facilitating product and service portfolio upgrading, (2) enabling flexible response mechanisms to market risks and challenges, and (3) strengthening core competitive advantages. The continuous execution of dual innovation strategies not only drives direct performance improvements but also cultivates organizational resilience as a critical strategic asset. This mediating mechanism ultimately enables enterprises to navigate dynamic market conditions while sustaining high-quality development trajectories.

Fourthly, this study reveals significant heterogeneity in the impact of ambidextrous innovation on high-quality enterprise development across different life cycle stages. Through comparative analysis of growth, maturity, and decline stages, distinct patterns emerged: exploratory innovation demonstrates significant positive effects across all stages, with its impact being most pronounced during the growth stage. Conversely, while exploitative innovation also shows significant effects throughout all stages, its influence peaks during the maturity stage. The relative strength of these effects varies systematically across stages: during the growth phase, exploratory innovation’s contribution to development quality substantially exceeds that of exploitative innovation, whereas this relationship reverses during the maturity phase. Both innovation types maintain significant yet diminishing effects during the decline phase. These findings demonstrate that the relationship between ambidextrous innovation and enterprise development quality exhibits stage-dependent characteristics, with both the magnitude and relative importance of each innovation type showing systematic variation across the organizational life cycle.

5.1. Implications of the Study

Firstly, exploratory innovation and exploitative innovation are both crucial for enterprise development. Thus, ambidextrous innovation activities should be conducted scientifically to continuously propel enterprise growth. During the enterprise growth process, attention must be given to both stability, efficiency, and long-term competitive advantages. To achieve this, enterprises need to establish an ambidextrous innovation synergy mechanism considering internal and external environments and development characteristics. They should also combine resource advantages and organizational traits to appropriately choose ambidextrous innovation activities and allocate resources rationally between the two types of innovation to attain coordination and balance. In daily operations, enterprises should conduct exploratory innovation to explore new markets and seize opportunities for future development, while also focusing on exploitative innovation to improve existing products, technologies, and capabilities, thereby maintaining current survival and development. Moreover, the relationship between the two should be managed well to realize their balance and complementarity, providing a durable and powerful driving force for enterprise development.

Secondly, enterprises should integrate their development stage with internal and external environments to formulate dynamic and differentiated innovation strategies. In a VUCA environment where the market landscape changes rapidly and technological progress evolves daily, enterprises need to constantly adjust innovation models and development strategies by devising a framework for selecting and adjusting strategy orientation based on diverse environments at each development stage. When determining whether to adopt a radical or conservative innovation strategy, the life cycle stage and development characteristics of the enterprise should be considered. It is essential to comprehensively evaluate the adaptability of exploratory and exploitative innovations at different stages and rationally allocate enterprise resources. Scarce high-quality resources should be appropriately distributed between the two types of innovations to maximize innovation efficiency. This targeted innovation approach enables enterprises to continuously enhance innovation levels and foster core competitive advantages, facilitating better achievement of high-quality development.

Thirdly, when carrying out innovation activities, enterprises should emphasize the cultivation of organizational resilience and balance innovation risks with resilience capabilities. In the current VUCA environment, low-probability crisis events occur frequently. The aim of an enterprise is not only to develop and expand but also to possess the ability to respond promptly to emergencies and progress steadily. During innovation activities, a comprehensive risk control mechanism should be established. Enterprises should also aim to strengthen internal prevention, foster risk awareness, and improve risk perception and decision-making abilities to promote “dualistic” activities. When formulating innovation strategies and decisions, these should be combined with the actual situation, and enterprises should fully analyze the market environment, competitive landscape, and internal and external resources to reasonably formulate plans for exploratory and exploitative innovation for effective risk distribution and management. Simultaneously, enterprises should focus on cultivating resilience during the innovation process. The enhancement of organizational resilience helps enterprises seize development opportunities while handling crises and achieve counter-trend growth. In emergencies, highly resilient enterprises can quickly detect and respond to risks, effectively integrate and coordinate resources, and progress steadily due to organizational resilience. Organizations should build resilience capabilities and adopt a proactive prediction attitude rather than a reactive response approach to face innovation risks, enabling them to respond swiftly to environmental changes and seize market opportunities.

Fourthly, manufacturing enterprises should consistently allocate resources to innovation and R&D. They must not only focus on exploring new technologies and markets but also deepen the optimization and upgrading of existing operations to achieve dual-wheel drive. Additionally, companies can attain economies of scale through mergers and acquisitions, expanding production lines, and other strategies. Simultaneously, they should emphasize knowledge transfer and experience sharing to strengthen their competitive edge. Regarding financial and operational strategies, a strong return on assets, a moderate growth rate, and a reasonable asset–liability ratio all contribute to the high-quality development of enterprises. However, while pursuing growth, companies must prioritize risk management to ensure stable and sustainable operations.

5.2. Limitations and Future Research

As it was constrained by factors such as research methodology, research capability, data processing, and variable measurement, this study has certain limitations.

Firstly, in this paper, ambidextrous innovation is divided into two dimensions, namely exploratory innovation and exploitative innovation, to explore its impact on the high-quality development of enterprises. However, in practice, enterprises usually conduct these two types of innovation activities simultaneously. This paper discusses the dimensions of exploratory innovation and exploitative innovation separately and fails to consider the impact of the balance and complementarity of ambidextrous innovation at a deeper level. Hence, future research could be enriched by including more dimensions and perspectives related to ambidextrous innovation, so as to gain a more comprehensive and in-depth understanding of the impact of ambidextrous innovation on enterprises.

Secondly, the variable measurement of organizational resilience in this paper has some limitations. Organizational resilience is a complex concept encompassing multiple levels and dimensions. In future research, more scientific and rigorous data and indicators could be employed to measure organizational resilience.

Finally, the control variables selected in this paper may have limitations. The development of an enterprise is affected by numerous factors. This paper only chose some of the main factors with a relatively greater impact as control variables, which might lead to the omission of other factors. Moreover, factors such as the external environment and the macroeconomic situation were not taken into account. In future research, it should be considered to incorporate more factors into the model for the study of the high-quality development of the enterprise.