Decision-Making on Selection of Talent Management Methods in the Era of Digitalization

Abstract

The application of digital technologies has a significant impact on organizational performance by way of different talent management methods, thereby enabling the maintenance of the organization’s continuous competitive advantage. Therefore, this paper studied the four key factors that influence organizational performance: digital technology application (DTA), inclusive talent management (ITM), exclusive talent management (ETM), and non-equilibrium investment (NET), aiming to investigate how digital technology application and talent management methods positively affect organizational performance, explore how this relationship is regulated by NET, and provides suggestions for selecting appropriate talent management methods. To conduct quantitative analysis, questionnaires were used with a sample size of 534 middle and senior managers as well as human resources practitioners from different enterprises. The structural equation model (SEM) was employed along with 5000 iterations to test the research hypotheses. The results indicate a positive correlation between digital technology application and organizational performance. Furthermore, ITM and ETM act as intermediaries between digital technology application and organizational performance, whereas NET plays a regulatory role in relation to ITM and organizational performance. This paper offers comprehensive insights into the factors influencing organizational performance and sheds light on how organizations make decisions regarding data-driven talent management methods at different stages of development.

1. Introduction

As times have evolved, the use of data in human resource management has shifted from basic indicator analysis to big data analysis. Forward-thinking managers now prefer using big data and artificial intelligence for talent management because they recognize that talents are crucial strategic resources for organizations and key drivers of sustainable competitiveness [1,2,3]. For instance, Google utilizes its own forecasting algorithm to adjust employee salaries in a timely and flexible manner to prevent talent attrition [4]. Amazon employs artificial intelligence (AI) to monitor the productivity of warehouse workers. In one warehouse alone, over 300 employees were terminated due to failing to meet productivity targets. Public information suggests that AI automation decision-making leads to a more than 10% annual dismissal rate in this factory [5]. Inclusive talent management (ITM) emphasizes equal investment in all employees, whereas exclusive talent management (ETM) differentiates employees based on their relative potential for enhancing competitive advantage. Therefore, addressing the issue regarding the decision on talent management selection arising from the digital transformation of human resources is both theoretically and practically significant for organizations seeking to maintain and enhance their competitive edge. The key innovation of this study lies in its approach to talent management by integrating both ITM and ETM methodologies. This innovative approach addresses the gap in empirical research on talent management methods and sheds light on the underlying relationship between them. By doing so, it offers valuable insights to inform decision-making in talent management practices.

In recent years, there has been growing interest among scholars and managers in the digital transformation of human resources. Organizations are increasingly integrating talent management with sustainable competitive advantage [6,7,8]. ITM emerges as a comprehensive approach that focuses on promoting equal access to company resources, fulfilling employees’ material and spiritual needs, and ultimately enhancing their sense of responsibility toward contributing to organizational growth [9]. Consequently, many organizations have adopted inclusive talent management as a strategy for future development [10,11]. However, ITM also faces criticism due to potentially unnecessary and costly investments in human resources. Previous studies have shown that employee abilities and contributions are not evenly distributed within an organization; a few highly talented individuals can significantly impact organizational performance [12]. This has led to the practice of ETM—a distinct approach to management that is centered around harnessing the distinctive contributions of talented individuals to an organization. It places a strong emphasis on the attraction and retention of top-tier employees, as well as enhancing the overall value of an organization’s competitive advantage. This methodology has gained significant traction in the field of talent management and has been adopted extensively in practice [13]. The United States Office of Personnel Management recognizes this approach as focusing on managing incumbents and potential personnel crucial for achieving the organization’s mission rather than all employees. ETM is widely acknowledged in theory and extensively implemented in practice [14].

In previous studies, it has been suggested that talent management in organizations should focus on equal resource allocation, talent absorption, and the creation of an inclusive and equitable work environment. However, some researchers argue that a more differentiated approach is necessary when it comes to investing resources in high-performing employees who have the potential to contribute significantly to the organization’s success in the present and future. The debate over which talent management mode to adopt has largely been discussed at a theoretical level, with limited empirical evidence to support either approach. Furthermore, there have been few studies that directly compare and examine the two talent management modes simultaneously. Therefore, this paper aims to explore the internal logical relationship between these different talent management modes through empirical research, with a specific focus on their impact on organizational performance. Drawing on the fundamental principles of talent management, this paper argues that organizations must adapt to the rapidly evolving digital landscape and adopt talent management strategies that leverage digital technologies. Building upon this premise, the paper incorporates the theory that the environment determines organizational structure [15] and the decision-making research on the application of digital technology in human resources. It also explores the choice of talent management mode under the theory of unbalanced growth. Specifically, this study seeks to answer the following questions:

• How does the implementation of digital human resources technology affect organizational performance?
• How do ITM and ETM influence this relationship?
• What is the interplay between ITM and ETM?
• What moderating effect do different talent management methods have on organizational performance?

The research framework of this paper is based on the literature on talent management styles driven by the digitalization of human resources, which provides guidance for studying the antecedents and consequences of talent management style decisions and how to use these elements to achieve high organizational performance. Therefore, this paper focuses on enterprises that utilize digital technology in their human resources practices. The aim is to provide insights into talent management strategies that can enhance organizational performance. The structure of the paper is as follows: Section 1 introduces the research variables and provides a comprehensive explanation of the research theme. Section 2 presents the theoretical hypothesis and research model. Section 3 and Section 4 describe the research methods and measurements employed. Finally, in the last section, we discuss the findings, analysis, and draw conclusions.

2. Theoretical Basis and Research Hypotheses

2.1. Digital Technology Application in Talent Management

Talent management is a typical practical application that precedes theoretical research, and not all employees possess skills that are of equal strategic importance to the organization [16]. It has received groundbreaking empirical research in academia [1]. Although the application of digital technology in human resource management still requires significant data support for improved efficiency, it has become increasingly prevalent in various aspects of organizational management, including talent selection [17], training and development, assessment and supervision, rewards and incentives, talent evaluation, employee relations, and other areas of human resource management [18,19]. Research has shown that the introduction of artificial intelligence technology into human resource management functions can effectively change and manage organizations through their workforce, which is beneficial for improving internal and external performance as well as the work attitude of employees. Consequently, it positively impacts overall organizational performance [20,21]. IBM has launched Watson Recruitment, an innovative application powered by artificial intelligence (AI), aimed at revolutionizing the candidate selection process for organizations. By employing digital technology, organizations can enhance their employer brand, expand the pool of potential candidates, and ultimately improve the overall selection process. However, despite its numerous benefits, the adoption of AI technology in recruitment is hindered by concerns regarding security, privacy, and the so-called “black box problem [21].” Although there are ethical considerations surrounding its use at an individual level or with regard to employee emotions; at an organizational level digital technology is becoming increasingly prevalent.

In line with the theory that posits the influence of the environment on organizational structure, three key factors shape the design of organizational structures: overall ecology, system dynamics, and resource dependence. Specifically, the competitive environment, technical and task requirements and the size of the organization are key drivers that shape the organizational structure. In the context of human resource management, the introduction of digital technology necessitates adjustments to the organizational structure. This adaptation prompts changes in the management system, impacts the utilization of external resources, and ultimately affects talent effectiveness and organizational performance. It becomes evident that digital transformation is not merely a targeted adjustment but rather a Darwinian evolution process for organizations. When an organization fails to adapt its overall capabilities to meet competitive environmental demands adequately, it risks being replaced by more resilient entities. China’s “Digital China Development Report” from 2022 reveals that as global digital technology continues to develop rapidly, China’s digital economy scale amounts to 45.5 trillion yuan (39.8% of GDP). In light of this trend, numerous human resource management practices have emerged due to advancements in digital technology usage within organizations’ operations. Therefore, undertaking a digital transformation in talent management becomes an inevitable choice for evolutionary development based on Darwinian principles. Hence, this study presents several assumptions:

H1a: 

Digital technology application in human resource management has a positive influence on organizational performance.

Digital technology allows for the seamless integration of real-time data into talent management, connecting an organization’s human resource management system with external market dynamics, internal production information, and other management systems. This resolves issues related to delayed, inaccurate, or incomplete data analysis in traditional talent management approaches. Additionally, it addresses the limitations of talent management in terms of forecasting and its impact on team management [22,23]. The Uber platform utilizes real-time analysis of a driver’s braking and acceleration data to assess their driving behavior and provide timely reminders for necessary rest breaks. Previous studies suggest that ITM places significant emphasis on the strategic role of human resources in organizational development. It encompasses various aspects of talent management, including attraction, identification, development, retention, and deployment through systematic projects. ITM advocates for the interactive integration of people and technology to ensure equal development rights for every employee, and this approach promotes a people-oriented mindset that fosters harmonious development between organizations and employees [24]. Additionally, it highlights using digital technology to enhance individual work efficiency and team performance [25]. Research has shown that traditional talent management relies on static information stored within human resources information systems. In contrast, digital age talent analysis can span across entire business processes by accurately capturing dynamic data on talent characteristics. This allows for the efficient identification of proprietary talents within an organization—high-potential individuals who contribute directly or have a significant future impact on organizational performance [26]. By investing more resources into these talents’ growth and commitment to the organization, overall performance can be improved.

H1b: 

Digital technology application in human resource management has a positive influence on ITM.

H1c: 

Digital technology application in human resource management has a positive influence on ETM.

2.2. ITM and ETM

This paper examines the impact of digital technology on talent management, contributing to the existing empirical research in this field. Talent management encompasses two main areas: ITM and ETM. ITM focuses on analyzing human capital at the organizational level, aiming to improve individual efficiency and overall organizational performance. On the other hand, ETM concentrates on managing high-performance and high-potential employees within an organization [27]. These individuals are considered exclusive talents due to their relative value to the organization, leading to focused investment in their development. Digital talent management enables the establishment of a dynamic team operation model [22]. By leveraging comprehensive talent data prediction and analysis through digital technology, organizations can effectively identify and attract top talents as well as provide targeted training for new employees [28,29]. This approach improves decision-making by analyzing risk and sensitivity data. It efficiently meets business needs while seamlessly integrating human resources with front-end operations, transforming HR into effective business partners [30,31]. This study argues that both ITM and ETM have the potential to serve as valuable business partners. Therefore, this study puts forward the following assumptions:

H2a: 

ITM has a positive influence on organizational performance.

H2b: 

ETM has a positive influence on organizational performance.

After implementing digital technology and artificial intelligence equipment, organizations can transform the work environment for employees. This transformation provides effective technical support to enhance work efficiency and encourages employees to actively explore and adapt to changes in their roles. Inclusive talent management in organizations, where team members are treated fairly and their strengths are fully utilized, can significantly impact team performance [32]. The Annual Observation Report on Human Resource Management in China in 2022 issued by Beisen highlights that enterprises currently face a significant challenge of talent shortage, leading to intensified competition among them. Instead of relying solely on high salaries to attract external talents from the market, it is more practical for organizations to identify, develop, and retain exclusive talents within their existing teams using digital technology and talent inventory [33]. These efforts should primarily focus on key positions and proprietary talents within the organization. By leveraging big data and artificial intelligence technology, organizations can effectively label and externalize the work behavior of proprietary talents based on network node characteristics, connections, and content. This enables the efficient identification of talented individuals [34] who exhibit positive emotional responses toward their work while demonstrating higher dedication to the organization. They often find ways to complete tasks faster with lower costs while making contributions beyond their assigned roles. Both ETM—which enhances organizational competitive advantage through employee potential—and ITM should be implemented according to the organization’s development stage [35]. Hence, this study puts forward the following assumptions:

H3a: 

ITM plays an intermediary role between digital technology application and organizational performance.

H3b: 

ETM plays an intermediary role between digital technology application and organizational performance.

2.3. The Regulatory Role of NET

According to the resource-based theory, differences in organizational performance are attributed to variations in resource investment and efficiency levels. The uniqueness and scarcity of exclusive talents play a crucial role in achieving high performance and competitive advantage. Technology application across different organizational levels, as well as the varying contributions of talents, justifies uneven investment [14]. Albert Hirschman’s unbalanced development theory suggests that investing more resources in proprietary talents promotes inclusive talent development through the diffusion effect. Williamson’s inverted “U” theory further explains how different talent management approaches impact organizational performance. During the early stages of organizational development, growth disparities among talents are necessary for performance improvement. As organizations mature, talent level differences gradually diminish, allowing for inclusive talent management practices. In practice, having employees with high creative potential is the key to improving an organization’s innovation performance, and talent retention is proven to provide the best value to customers [36], Google’s talent analysis team maximizes retention of proprietary talents through salary policies but does not extend similar material and spiritual incentives to ordinary employees [37]. The following assumptions are put forward:

H4a: 

NET has a positive influence on organizational performance.

H4b: 

NET has a positive influence on the relationship between ITM and organizational performance.

In summary, the research model of this study is shown in Figure 1.

3. Research Design

3.1. Research Samples and Data Collection

From the perspective of urban development, this study focuses on analyzing and comparing first-tier developed cities and second-tier developing cities. From the perspective of geographical selection, this study selected cities in eastern, central, and western China, including Shanghai, Xi’an, and Zhengzhou, for data analysis. The data were gathered through various channels including the Shanghai University MBA Exchange Group, Xi’an Jiaotong University MBA Exchange Group, Zhengzhou Human Resources Exchange Group, and Shanghai Human Resources Exchange Group. To ensure the reliability and validity of the data, middle and senior managers along with human resources practitioners were selected to complete the questionnaire due to their clear understanding of organizational talent management. Before distributing the questionnaire formally, a Delphi research method was employed to review it with human resources experts. Additionally, small-scale interviews were conducted to fill out the questionnaire and make necessary revisions for clarity. A total of 534 questionnaires were collected; however, only 359 were considered valid after excluding responses from non-middle-/high-level executives and non-human resources practitioners.

Table 1. Basic characteristics of samples.

Variable	Option	Frequency	Percentage
Gender	Male	174	48.50%
	Female	185	51.50%
Age	Under 26 years old	13	3.60%
	26–30 years old	68	18.90%
	31–35 years old	106	29.50%
	36–40 years old	108	30.10%
	Over 40 years old	64	17.80%
Education Level	Technical secondary school and college	15	4.20%
	Bachelor’s degree or equivalent	199	55.40%
	Master’s degree or above	139	38.70%
	Other	6	1.70%
Years of Work	1–5 years	57	15.90%
	6–10 years	95	26.50%
	10–20 years	172	47.90%
	Over 20 years	35	9.70%
Position	Middle to senior managers	208	57.90%
	Human resources practitioners	151	42.10%
Company Nature	State-owned enterprise	88	24.50%
	Private enterprise	215	59.90%
	Foreign enterprises	42	11.70%
	Freelancer	4	1.10%
	Other	10	2.80%
Company Size	More than 1000 people	131	36.50%
	500–1000 people	81	22.60%
	200–500 people	33	9.20%
	Less than 200	114	31.80%

provides specific details about the sample characteristics.

3.2. Variable Measurement

To ensure the reliability and validity of the questionnaire, we used established scales from the existing literature for all variables in this study and followed standard procedures of translation, back-translation, and revision for these scales. Additionally, we made necessary adjustments based on the specific requirements of our study. The scale was designed as a 5-point Likert scale with options ranging from 1 = strongly disagree, 2 = disagree, 3 = uncertain, 4 = agree, and 5 = highly agree.

• Digital Technology Application. This paper draws on the work of Carvalho [38] and Carvalho and Dubey [33] on the Big Data-Driven AI Adoption Scale, which consists of the following key items: “The adoption of digital technology by a company improves decision making”, “The adoption of digital technology by a company enables easy integration of information from different data sources”, “The adoption of digital technology by a company enables better utilization and optimization of resources”, and “The goal of the adoption of digital technology by a company is based on a dynamic business environment”, etc.
• ITM. The ITM framework aims to harness and leverage the skills and abilities of all individuals within a given organizational setting. Based on the inclusive talent development model by Fang Yangchun [39], this scale consists of four main items. These include “building a diverse talent team within the company”, “addressing and learning from the company’s shortcomings and failures in inclusivity towards employees”, and “focusing on showcasing employee strengths, fairness, and mutual benefit”.
• ETM. The ETM framework aims to select a specific employee group that plays a decisive role in the current and future achievements of the organization. Drawing from Mousa and Ayoubi’s [40] ETM scale, it consists of five key items. These include statements such as “the company’s management training is limited to a few employees”, “the company categorizes employees into talented and non-talented groups”, “the company makes efforts to retain identified talented employees”, and “the company does not treat all employees equally”.
• NET. This paper examines the staff salary scale of values proposed by Shao Jianping [41] in 2017 and draws important insights. Investing in employee resources is a way to recognize their abilities, which can be achieved through five key aspects: aligning resource investment with employees’ skill levels, position levels, job performance, and development potential, etc.
• Organizational performance. This paper examines the organizational performance scale, known as the balanced scorecard proposed by Chen Guoquan [42] in 2009. The scale consists of four dimensions: finance, operations, customers, and learning development. It encompasses 12 items that assess various aspects such as profitability advantage, operational efficiency advantage, customer satisfaction advantage, and employee satisfaction advantage.

3.3. Statistical Analysis

In this study, we used SPSS 26.0 for descriptive statistical analysis and AMOS 26.0 for confirmatory factor analysis. We also included demographic factors such as gender, age, education level, and years of work in the questionnaire as control variables to account for their potential impact on talent management style and organizational performance.

4. Data Analysis and Results

4.1. Reliability and Validity Tests

The overall scale in this study has a KMO value of 0.902, indicating good sampling adequacy. The spherical test also yields a highly significant Sig value, allowing for factor analysis to be conducted. All items have standardized factor loads ranging from 0.58 to 0.82, which meets or exceeds the acceptance standard of 0.6. The variables demonstrate good internal consistency with Cronbach’s α coefficient greater than or close to 0.8. Additionally, each variable has an AVE value greater than or close to 0.5, indicating high convergence validity (see

Table 2. Reliability and validity analysis.

Variable	Combined Reliability		Convergent Validity
	CR	Cronbach’α	AVE
Digital technology application	0.815	0.810	0.470
ITM	0.818	0.816	0.530
ETM	0.793	0.788	0.494
NET	0.849	0.847	0.529
Organizational performance	0.918	0.917	0.481

).

The standardized correlation coefficients between pairwise dimensions in this discriminant validity test are all less than the square root of the corresponding AVE value. This indicates that all dimensions exhibit good discriminant validity, as shown in

Table 3. Results of differentiated validity test for each dimension.

Variable	Digital Technology Application	ITM	ETM	NET	Organizational Performance
Digital Technology Application	0.470
ITM	0.578	0.530
ETM	0.168	0.008	0.494
NET	0.308	0.428	0.258	0.529
Organizational Performance	0.377	0.498	0.218	0.592	0.481
Square Root of AVE Value	0.686	0.728	0.703	0.727	0.694

.

The results of the confirmatory factor analysis are presented in

Table 4. Results of validation factor analysis.

Model	χ2	df	χ2/df	RMSEA	IFI	TLI	CFI
Five-factor model	850.586	395	2.153	0.057	0.906	0.895	0.905
Four-factor model	1294.505	399	3.244	0.079	0.814	0.796	0.813
Three-factor model	1520.403	402	3.782	0.088	0.768	0.747	0.766
Two-factor model	2009.761	404	4.975	0.105	0.667	0.639	0.665
Single-factor mode	2429.249	405	5.998	0.118	0.580	0.546	0.577

Note: There are different models that can be used to analyze the relationship between digital technology applications, ITM, ETM, NET, and organizational performance. These models include the five-factor model (digital technology application, ITM, ETM, NET, organizational performance), four-factor model (digital technology application, ITM + ETM, NET, organizational performance), three-factor model (digital technology application + ITM + ETM, NET, organizational performance), two-factor model (digital technology application + ITM + ETM + NET, organizational performance), and single-factor model (digital technology application + ITM + ETM + NET+ organizational performance). Each of these models includes varying combinations of digital technology applications, ITM, ETM, NET, and organizational performance.

. Among the four factor models, the five-factor model demonstrates the best-fit index, suggesting that there is no significant multicollinearity effect among variables and that the discrimination validity of the five constructs is satisfactory. The fitting indexes for the five-factor model are as follows: χ² = 850.586, df = 395, RMSEA = 0.057, IFI = 0.906, TLI = 0.895, and CFI = 0.905. All these values meet the requirements of the fitting standard and indicate a good overall fit for the measurement model.

4.2. Descriptive Statistics and Correlation Analysis

The descriptive statistical results of the main variables are presented in

Table 5. Descriptive statistics and correlation analysis.

Variable	M	SD	1	2	3	4	5
1. Digital Technology Application	4.057	0.562	1
2. ITM	3.861	0.670	0.483 **	1
3. ETM	3.262	0.834	0.138 **	−0.008	1
4. NET	3.716	0.612	0.255 **	0.348 **	0.216 **	1
5. Organizational Performance	3.711	0.580	0.323 **	0.420 **	0.189 **	0.535 **	1

Note: ** At the 0.01 level (double-tailed), the correlation is significant.

. Pearson correlation analysis is conducted to examine the relationships between these variables. The analysis reveals a significant positive correlation between digital technology application and ITM, proprietary talent management, NET, and organizational performance. Additionally, ITM shows a significant positive correlation with NET and organizational performance, whereas ETM exhibits a similar relationship with these factors. Furthermore, there is also a significant positive correlation observed between NET and organizational performance.

4.3. Hypothetical Test Results Analysis

4.3.1. Imaginary Relation Test

Based on the regression results presented in

Table 6. Model regression analysis 1.

Variable	ITM				Organizational Performance
	M1	M2	M3	M4	M5	M6	M7	M8
Gender	0.023	0.038	0.017	0.028	0.007	0.056	0.014	0.038
Age	−0.093	−0.024	0.088	0.135	0.128	0.122	0.144	0.147
Education Level	−0.072	−0.064	−0.067	−0.062	−0.036	−0.066	−0.039	−0.046
Years of Work	0.242 *	0.183 *	−0.065	−0.106	−0.17	−0.002	−0.170	−0.088
Digital Technology Application		0.481 ***		0.328 ***			0.159 **	0.131 **
ITM					0.431 ***		0.352 ***	0.257 ***
NET						0.552 ***		0.441 ***
ITM × NET								0.170 ***
R2	0.031	0.261	0.007	0.114	0.187	0.303	0.206	0.394
ΔR2	0.020	0.251	−0.004	0.102	0.175	0.294	0.192	0.380
F	2.818	24.933	0.629	9.093	16.209	30.753	15.202	28.410

Note: N = 500, * represents p < 0.05, ** represents p < 0.01, and *** represents p < 0.001.

, several conclusions can be drawn. Firstly, we tested the relationship between digital technology application and ITM. Model M1 includes all control variables, whereas model M2 focuses solely on digital technology applications. After controlling for other factors, it was found that digital technology application has a significantly positive influence on ITM (β = 0.481, p < 0.001). Additionally, ΔR² is 0.251 and significant, further supporting the notion that digital technology application positively influences ITM (H1b).

Secondly, we examined the relationship between variables and organizational performance. Building upon the control variables in model M1, we introduced a digital technology application to create model M4. The results revealed a significant positive influence of digital technology application on organizational performance (β = 0.328, p < 0.001), thereby confirming H1a.

Thirdly, by incorporating ITM into model M5 alongside the control variables, we assessed its influence on organizational performance. The findings indicated a significant positive effect of ITM on organizational performance (β = 0.431, p < 0.001), providing support for H2a.

Lastly, when investigating the link between NET and organizational performance, NET was added to model M1 to form model M6. The analysis demonstrated that NET has a substantial positive influence on organizational performance (β = 0.552, p< 0.001), thus corroborating H4a.

Based on the regression analysis in

Table 7. Model regression analysis 2.

Variable	ETM			Organizational Performance
	M11	M12	M13	M14	M15	M16	M17
Gender	0.005	0.009	0.017	0.016	0.026	0.059	0.043
Age	−0.129	−0.11	0.088	0.114	0.152	0.156	0.158
Education Level	0.048	0.05	−0.067	−0.077	−0.07	−0.067	−0.05
Years of Work	0.009	−0.008	−0.065	−0.067	−0.104	−0.033	−0.089
Digital Technology Application		0.135 *			0.307 ***	0.195 ***	0.115 *
ITM							0.277 ***
ETM				0.198 ***	0.156 **	0.076	0.090 *
NET						0.475 ***	0.384 ***
ITM × NET							0.171 ***
ETM × NET						−0.023	−0.073
R2	0.016	0.034	0.007	0.045	0.138	0.346	0.403
ΔR2	0.005	0.021	−0.004	0.032	0.123	0.331	0.386
F	1.446	2.503	0.629	3.364	9.353	23.177	23.496

Note: N = 500, * represents p < 0.05, ** represents p < 0.01, and *** represents p < 0.001.

, we can conclude the following findings regarding the relationship between digital technology application and ETM. Firstly, when controlling for gender, age, education, and working years using model M11, adding digital technology application to form model M12 shows a significantly positive influence coefficient on ETM (β = 0.135, p < 0.05). Additionally, ΔR² is found to be 0.021 with statistical significance. Secondly, when examining the relationship between variables and organizational performance, we incorporate ETM data into model M11 to create model M14. The results indicate that ETM has a significantly positive influence coefficient on organizational performance (β = 0.198, p < 0.001), thus supporting H2b.

4.3.2. ITM and ETM Mediation Effect Test

According to Wen Zhaolin’s [43] mediating effect test method, the mediating effect of ITM is examined. Firstly, after controlling for variables, it is found that digital technology application has a significant positive influence on organizational performance (β = 0.328, p < 0.001). Secondly, in model M2, there is a significant positive correlation between digital technology application and ITM (β = 0.481, p < 0.001). Additionally, ITM has a significantly positive influence on organizational performance (β = 0.431, p < 0.001). Finally, in model M7, where independent and intermediate variables are included in the regression equation simultaneously, it is observed that ITM positively affects organizational performance (β = 0.352, p < 0.001), whereas the regression coefficient for digital technology application decreases from 0.328 (p < 0.001). This indicates that ITM serves as an intermediary function, supporting H3a.

In testing the mediating effect of ETM using the same method, we observed that ETM has a positive influence on organizational performance (β = 0.156, p < 0.01) when both independent and intermediary variables are included in the regression equation in model M15. Additionally, the regression coefficient for digital technology application on organizational performance increased from 0.135 (p < 0.05) to 0.307 (p < 0.001), as shown in Table 7. This indicates that ETM serves as an intermediary function, supporting H3b.

The statistical effect of the mediation effect is improved using the bias-corrected percentile bootstrap method to test the mediation effect of two talent management methods (see

Table 8. Mediating effect analysis.

Path	Mediating Effect	Standard Error	95% Confidence Interval		Effect Proportion
			Lower Limit	Upper Limit
TOTAL	0.207	0.037	0.136	0.279	61.04%
Ind1: Digital Technology Application—ITM—Organizational Performance	0.182	0.036	0.113	0.254	53.71%
Ind2: Digital Technology Application—ETM—Organizational Performance	0.032	0.014	0.008	0.063	9.37%
Ind3: Digital Technology Application—ITM-ETM—Organizational Performance	−0.007	0.006	−0.019	0.005	−2.07%
Ind1–Ind2	0.150	0.040	0.072	0.230	——
Ind1–Ind3	0.189	0.036	0.121	0.261	——
Ind2–Ind3	0.039	0.018	0.008	0.079	——

). The data results indicate that:

-

The total mediation effect, with a 95% confidence interval [0.136, 0.279], is found to be significant as it does not include 0.

-

The indirect effect Ind1 (0.182), which consists of “digital technology application → ITM → organizational performance”, also shows a significant mediating effect with a 95% confidence interval [0.113, 0.254] that does not include 0. It indicates that ITM has a significant mediating effect, confirming that the mediating effect is established and H3a is supported.

-

Similarly, the indirect effect Ind2 (0.032), composed of “digital technology application → ETM → organizational performance”, has a significant mediating effect with a 95% confidence interval [0.008,0.063] that does not include 0. It indicates that ETM has a significant mediating effect, confirming that the mediating effect is established and H3b is supported.

-

The mediating effect value of Ind1-Ind2 is calculated as 0.150 with a bootstrap-based 95% confidence interval of [0.072, 0.230] that does not include 0. This indicates that the mediating effect of Ind1is significantly greater than the one for Ind2, and the mediating effect of ITM on organizational performance is significantly greater than that of ETM.

-

The mediating effect value of Ind1-Ind3 is calculated as 0.189 with a bootstrap-based 95% confidence interval of [0.121, 0.261] that does not include 0. This indicates that the mediating effect of Ind1 is significantly greater than the one for Ind3, thus further supporting H1a and H3a.

-

The mediating effect value of Ind2-Ind3 is calculated as 0.039 with a bootstrap-based 95% confidence interval of [0.008, 0.079] that does not include 0. This indicates that the mediating effect of Ind2 is significantly greater than the one for Ind3, thus further supporting H1a and H3b.

4.3.3. The Moderating Effect of NET

Based on M7, the variable of NET is included in the model to create M8. The interaction between ITM and NET is found to be significant (β = 0.170, p < 0.001), indicating that NET has a positive moderating effect on the relationship between ITM and organizational performance. To further analyze this relationship, we conducted a simple estimation using the average value of NET as well as one standard deviation above and below it as adjustment variables.

Table 9. Moderating effect value and significance.

Dependent Variable	Independent Variable	Moderator Variable	Effect Value	Standard Error	t-Value	95% Confidence Interval
						Lower Limit	Lower Limit
Organizational Performance	ITM	Low NET	0.133	0.047	2.837	0.041	0.226
		Medium NET	0.239	0.046	5.193	0.149	0.330
		High NET	0.345	0.060	5.750	0.227	0.464

shows that under high levels of NET, the impact coefficient is 0.345 (p < 0.001), whereas under low levels of NET, it is 0.133 (p < 0.001). These results demonstrate that the adjustment effect of NET is significant.

Figure 2 shows that when NET is low (M-1SD), ITM has a negative impact on organizational performance (t = 2.837, p < 0.001). Conversely, when NET is high (M + 1SD), ITM has a positive influence on organizational performance (t = 5.750, p < 0.001). This suggests that as NET increases, the influence of ITM transitions to ETM and becomes stronger in impacting organizational performance. Figure 2 visually represents this transition with a larger positive slope for high NET.

5. Research Conclusions and Discussion

5.1. Research Conclusions and Theoretical Contributions

This paper examines the influence mechanism of NET on talent management and organizational performance in the context of digital technology application. The findings reveal that increasing NET in talent management weakens the negative impact of ITM on organizational performance. This shift leads to a change in talent management mode from ITM to ETM, resulting in improved organizational performance through the intermediary role of talent management. Additionally, NET regulates the relationship between ITM and organizational performance, further enhancing the positive influence of talent management on organizational performance. The theoretical contributions of this study are as follows:

Firstly, this study expands the research perspective on organizational performance. Previous studies have mainly focused on examining the mechanisms of organizational performance from the perspectives of human capital theory, social exchange theory, and resource-based theory. However, this paper introduces the development theory of economics into talent management to explain that ITM and ETM are different talent management methods adopted by organizations at various stages of development. This finding aligns with previous research conducted by Maniam Kaliannan [4] and Fang Yangchun [16], which concluded that both ITM and ETM significantly impact organizational performance. Therefore, this paper contributes to enriching our understanding of organizational performance.

Secondly, this study explores the boundary conditions for how talent management modes affect organizational performance from a NET perspective. In today’s digital transformation era, organizations respond to ITM through balanced investment in talent management while adopting ETM mode due to NET practices [8]. By considering NET as a moderating variable, this study explains the influencing factors between talent management modes and organizational performance while revealing their contingency relationship. Our findings support the contingency view and contribute to further developing research on NET within organizations. Additionally, based on the perspective of NET, we argue that organizations should selectively invest in talent management according to their developmental stage—a key focus for company reform efforts. This study demonstrates that a significant level of NET in talent management enhances its impact on organizational performance, thus expanding our understanding of its boundary conditions.

5.2. Implications for Practice

The promotion of organizational performance is crucial for organizational development. This paper examines the impact of digital technology on organizational performance through talent management and provides the following insights for management practice:

Firstly, digital transformation is an inevitable direction for future organizations, as it reveals the relationship between digital technology application and talent management, which ultimately impacts organizational performance. To avoid being surpassed by more innovative organizations and to maximize limited resources, managers must carefully consider how to allocate resources. In the early stages of an organization’s development, investing in high-potential and high-performance talents who can significantly impact performance is crucial. As the organization matures, attention should shift towards improving overall efficiency on a per capita basis while also focusing on ITM. It is important to tailor talent management methods based on the specific needs of each organization.

Secondly, this study reveals that different talent management methods can have both positive and negative effects on organizational performance. Talents serve as the foundation for organizations to achieve sustainable competitiveness, and the impact of various talent management methods varies at different stages of organizational development. To mitigate these effects, managers should prioritize employees’ individual characteristics, unleash their potential, motivate them through a combination of material and spiritual incentives, enhance the incentive system, and ultimately improve organizational performance by enhancing employee performance. Furthermore, when an organization heavily invests in a select few high-level talents, it may yield positive results for organizational performance. However, this approach also carries the risk of losing key talents which can negatively impact the organization. As managers, it is crucial to gradually establish a robust talent pipeline to prevent intermittent disruptions in organizational performance caused by the loss of key talents.

5.3. Research Deficiencies and Prospects

This study has made progress in researching the influence of digital technology application on organizational performance under talent management. However, there are still some shortcomings and limitations that need to be addressed. Firstly, this paper is a cross-sectional study, which limits its ability to fully understand the process mechanism of how talent management affects organizational performance. Future research should consider collecting data from multiple time points and adopting the qualitative method to overcome this limitation. Secondly, the data obtained carries uncertainty, and in future research, robust optimization methods can be employed to mitigate the risks associated with this uncertainty. Thirdly, it is important to note that digital technology application not only impacts organizational performance but also influences the development and performance of individual talents. Therefore, further investigation is needed to understand how digital technology affects personal talent performance. Fourth, whereas ETM has been successfully validated in practical organizational management settings, this study found a relatively low intermediary effect of ETM on organizational performance due to the limited use of widely accepted scales. Subsequent research should explore alternative evaluation scales for ETM. Lastly, under the context of digital technology applications, there is currently no accurate definition or measurement for talent management and NET. It is crucial for future research to address these gaps in order to keep up with the rapid development of emerging technologies such as digital technology applications.