Integrated AI and Business Analytics for Sustaining Data-Driven and Technological Innovation: The Mediating Role of Integration Capabilities and Digital Platform

Abstract

While integrated Artificial Intelligence and Business Analytics (AI-BA) represents a significant advancement in marketing analytics and greatly influences firms’ innovations, there is a considerable gap in current research regarding its impact on technological innovation. This study addresses this gap by exploring how AI-BA affects data-driven and technological innovation, considering the mediating roles of integration capabilities and digital platforms. A theoretical model has been developed based on the dynamic capability view (DCV) and organizational information processing theory (OIPT). The model has been validated using data from enterprises in Saudi Arabia, and Partial Least Squares Structural Equation Modeling (PLS-SEM) has been employed for analysis. The findings demonstrate that AI-BA directly enhances both technological and data-driven innovation. Additionally, it was discovered that data-driven innovation, integration capabilities, and digital platforms mediate these effects, thereby enhancing technological innovation within the respective industries. These findings provide both theoretical and practical insights into the relationship between AI-BA, data-driven innovation, and technological innovation. They enrich the existing literature and provide actionable guidance for practitioners aiming to align their AI-BA with improved technological innovation outcomes.

1. Introduction

Industry 4.0 refers to the integration of various emerging information and communication technologies, marking a transformative shift in industrial processes [1]. Similarly, Analytics 4.0 represents a significant advancement beyond traditional analytics in the context of business analytics, marking the evolution of traditional analytics alongside the rise of artificial intelligence [2]. While the topic of business analytics (BA) is seen as interesting by practitioners and academics [3], currently, there is a growing perspective that the 5.0 revolution is leading the industry by enabling AI to enhance the development of data analytics [4,5]. The rapid advancement and adoption of AI challenge firms to transition from Big Data and analytics to AI-driven solutions as they consider new transformative technologies [6].

With the beginning of AI, the demand for business analytics (BA) is increasing exceptionally [7,8]. This increased interest in BA and prepared to enhance firms’ capabilities by integrating AI into business analytics (AI-BA) to gather, analyze, and utilize data to make informed decision-making. Yet, data-driven decision-making is now considered an essential competency that ultimately positions firms for sustained growth and innovation and plays a critical role in digital transformation and the development of analytics capabilities [9].

One of the critical aspects of this transformation is linking business analytics (BA) with artificial intelligence, which represents a significant technological progression that empowers firms to analyze data effectively using these emerging technologies [10,11]. For example, although enterprises apply Customer Relationship Management (CRM) systems to inspect customer data for enhanced insights, Chatterjee et al. argue that an abundance of data may pose significant challenges [12]. According to Gnizy, the integration of AI can effectively address these difficulties by providing enterprises with more trustworthy information rapidly, thereby facilitating the success of decision-makers [13].

Zhao et al. examine how the combination of Big Data and artificial intelligence will fundamentally reshape the future of innovation and transformation across various sectors, particularly in businesses and academic institutions [6]. A particularly remarkable aspect of business analytics is its strong connection to artificial intelligence, which enhances data processing, predictive capabilities, data visualization, statistical modeling, and enriching its effectiveness in enhancing decision-making capabilities and gaining deeper insights to make more informed decisions [10,11].

However, current studies primarily emphasize the opportunities and practical applications of artificial intelligence (AI) in business, particularly in data analytics [4,7,14]. Despite this emphasis, a gap remains in addressing and exploring the role of integrated AI and data analytics in enhancing their ability to drive innovation. This gap persists despite technological advancements in data analytics and artificial intelligence, which enable firms to enhance their technological innovation initiatives [15]. Furthermore, Zameer et al. highlighted that although technological innovation is a significant subject, examining the factors that facilitate technological innovation and the associated challenges is essential [16].

In addition, Saleem et al. noted that while firms concentrate on their business operations due to intense competition, they must enhance their capabilities by utilizing big data for technological innovation [17]. This approach enables them to develop innovative products utilizing their existing resources, including organizational culture and IT infrastructure. However, Saleem et al. argue that while current studies focus mainly on linking data analytics with firm performance, there is a lack of exploration into how the essential elements of data analytics are interconnected with technological innovation, particularly within the framework of data-driven decision-making [17]. Also, Zamani et al. emphasize that AI-BA is an important topic that needs to be discovered for firms aiming to ensure business continuity [18]. This suggests the need for a more comprehensive understanding of how new data analytics technologies, such as AI-BA, interact and influence technological innovation while addressing the main enablers within the data-driven decision-making framework.

However, as Business Analytics (BA) and integrated AI-BA are emerging fields that present significant opportunities to explore their benefits and impact to understand their advantages fully [7,10,11,18], this study applies the Dynamic Capability View (DCV) and Organizational Information Processing Theory (OIPT) to demonstrate how AI-BA and data-driven innovation can significantly enhance technological innovation. This exploration is critical, as existing studies have not examined how AI-BA improves data-driven and technological innovation. By addressing this gap, the study aims to show the transformative potential of integrating AI-BA.

Furthermore, this study broadens the discussion on AI-BA initiated by scholars such as Rana et al., Zamani et al., Davenport, Gómez-Caicedo et al., and Conboy et al. [2,7,8,10,18]. It highlights that AI-BA influences data-driven innovation, a key element for technological innovation. Additionally, while 95% of analytics-driven innovation projects fail and technical difficulties are considered as one of the main causes [19], and while the capabilities of IT improve strategic innovation capabilities, which boosts the performance of innovation [20], the study highlights the crucial role of integration capabilities and digital platforms in enhancing the effectiveness of data-driven innovation initiatives by addressing two key research questions: (RQ1) Do integration capabilities mediate the relationship between AI-BA and data-driven innovation? (RQ2) Do digital platforms mediate this relation?.

However, this study represents a significant advancement as it is the first to examine the aforementioned variables within the context of Saudi Arabia. Saudi Arabia is a prominent member of the G20 and is currently undergoing intense digital transformation, driven by the ambitious Saudi Vision 2030 initiative launched in 2016. This initiative has resulted in substantial advancements in digital capability, as reflected in its impressive standing on global indices [21].

Saudi Arabia presents an opportunity to advance integrated AI-BA, particularly within the framework of its national transformation agenda, Vision 2030. As a key initiative, the Kingdom introduced the National Strategy for Data and AI (NSDAI) in 2020, guided by the Saudi Data & AI Authority (SDAIA). This forward-thinking strategy aims to establish Saudi Arabia as a global leader in data-driven economies by recognizing data and artificial intelligence as crucial assets for economic diversification, while supporting 66 of the 96 objectives outlined in Vision 2030 [22]. Also, the National Data Management Office (NDMO) has proactively introduced comprehensive standards that support governance relevant to both the public and private sectors. Furthermore, initiatives like the Saudi Open Data Portal demonstrate a strong commitment to enhancing transparency and fostering data sharing [23]. Together, these efforts create a supportive environment for adopting integrated AI-BA, highlighting the significant role of national strategies, governance frameworks, and digital infrastructures in shaping and advancing organizational capabilities.

The current study is designed to address specific research objectives. The upcoming section presents the theoretical framework, which includes an in-depth literature review of two essential concepts: the dynamic capability view (DCV) and the organizational information processing theory (OIPT). Following this exploration, the proposed model is introduced, and the specific hypotheses derived from it are elaborated on. This model will highlight the critical elements of AI-BA and data-driven innovation, examining how these components can significantly impact technological innovation. Furthermore, the study investigates the roles of integration capabilities and digital platforms, treating them as essential mediating variables that may enhance the primary relationships discussed. The subsequent sections will provide a comprehensive outline of the research methodology, detailing the procedures for data collection and the analytical techniques employed throughout the study. They will then present the results of model testing, accompanied by a thorough discussion of the empirical findings resulting from the statistical analyses conducted. To conclude, the study will summarize its findings by presenting theoretical insights and practical implications drawn from the research. Additionally, the limitations encountered during the study and suggestions for future research are presented in this developing field.

2. Theoretical Background and Hypotheses Development

2.1. Theoretical Background

2.1.1. Dynamic Capabilities View (DCV) and Innovation Performance

This research examines the relationship between AI-BA and firms’ innovation performance through two theoretical frameworks: the Dynamic Capabilities View (DCV) and Organizational Information Processing Theory (OIPT). It aims to investigate how AI-BA can enhance a firm’s ability to innovate and adapt in a rapidly changing business environment. Additionally, the study examines the crucial roles that digital platforms and integration capabilities play in facilitating this process. The research seeks to comprehensively understand how these elements interact to drive innovation and improve organizational outcomes.

A leading and constructive theory related to resource utilization and firms’ ability to adapt to rapidly changing business environments is the dynamic capabilities view (DCV), as outlined by Teece et al. [24]. This framework highlights the importance of developing and leveraging unique capabilities to respond effectively to evolving challenges and opportunities in the marketplace. Rana et al. noted that the dynamic capability view (DCV) is essential for identifying the resources crucial for achieving high levels of firm performance in AI-BA integration [7].

The Dynamic Capability View (DCV) expands on the existing Resource-Based View (RBV) framework by emphasizing the importance of reconfiguring resources and enhancing ordinary capabilities. This approach enables firms to innovate and effectively respond to market changes while facilitating data extraction from various sources [7]. However, Teece articulates the concept of dynamic capability as a firm’s ability to identify, combine, and reconfigure its resources to swiftly adapt to evolving business environments and effectively address new opportunities and emerging challenges [25].

Nowadays, the intersection of new technologies such as business analytics, artificial intelligence, and dynamic capabilities has gathered interest from academics and practitioners. Rana et al. study, for example, employs DCV to address the required skills and training for employees to use an AI-integrated BA solution [7]. Also, it utilizes the DCV perspective to emphasize the necessity of AI-BA adoption to acquire valuable, significant, rare, and important data. In addition, Chaudhuri et al. [26] study explores the DCV framework to illustrate how the firm can effectively harness the capabilities of sensing, seizing, and transforming in leveraging AI-CRM technology, which can significantly enhance an organization’s resilience and sustainability during periods of crisis. This comprehensive approach highlights how businesses can proactively identify challenges and address opportunities for innovation to maintain stability and growth in turbulent times.

Moreover, DCV is increasingly acknowledged as a key enabler of knowledge creation within the scope of IT solutions. It plays a significant role in leveraging big data analytics solutions, which are instrumental in enhancing strategic planning and identifying emerging trends and situations that require attention [27,28]. Given this importance, it is crucial to explore how DCV functions as a facilitator when integrated with advanced AI solutions incorporating data analytics, such as AI-BA solutions. This examination will light on the synergistic relationship between DCV and AI-BA, aiming to optimize decision-making processes and improve innovation performance in a growing business landscape.

In addition, Liu et al. [15] explore how digital platforms can drive technological innovation through the lens of dynamic capabilities. While much of the existing literature emphasizes the importance of a firm’s ability to develop digital platforms, this study highlights the critical role these platforms play in advancing innovation. In addition, while digital platforms can yield innovative outcomes by enhancing network capabilities [29], they may not directly influence firm innovation but aid in developing dynamic capability [30]. This distinction highlights the importance of leveraging digital platforms not just as tools for immediate innovation but to build a more agile and responsive organizational framework [31].

However, Liu et al. [15] argue that business analytics, with their comprehensive approach, effectively respond to changes in market demand by focusing on the integration process involved in managing, processing, and analyzing large amounts of data [30]. The authors emphasize that a firm’s information processing capabilities are fundamental in strategically enhancing its capacity to align innovation initiatives with overarching business goals. This alignment is crucial for creating and sustaining competitive value, as evidenced by Saldanha et al., who note that companies that effectively leverage their data can unlock substantial opportunities for growth and innovation [32]. Additionally, they note a lack of business analytics topics addressing digital platform capabilities from the information management perspective [33]. Therefore, it is essential to incorporate digital platforms and data-driven capabilities into business analytics to leverage the value of information resources fully.

2.1.2. AI-BA and Organizational Information Processing Theory (OIPT)

In the context of the OIPT perspective, Rashid et al. (2024) emphasize the significance of enhancing data processing capabilities in integrated big data analytics (BDA) and artificial intelligence (AI) technologies [27]. They propose that incorporating AI into BDA can optimize the utilization of vast amounts of data necessary for decision-making by integrating various data sources, such as enterprise resource planning (ERP) systems and business process automation, so firms can harness a more comprehensive approach to data analysis, ultimately leading to more strategic and impactful business outcomes.

Organizational information processing theory refers to the systematic approach of collecting raw data, converting this data into meaningful information, and maintaining that information within the firm to ensure that relevant insights are readily available, enabling decision-makers to make well-informed decisions that drive growth and efficiency [34,35]. In addition, Kyagante et al. highlighted the importance of information technology capabilities and information integration, which connect suppliers and customers, and play a crucial role in enhancing a firm’s ability to process information [35]. This capability allows firms to manage uncertainties effectively [36]. As a result, these factors are essential for strengthening firm resilience and reducing risks associated with uncertainties that require information processing for adaptability in a dynamic market environment [35].

Moreover, Monroy-Osorio [37] argues that while OIPT primarily emphasizes the critical information processing necessary for making informed decisions as discussed by [38], a truly effective strategic decision-making system must go beyond individual information processing as it should also encompass the integration of both individual and organizational information processes to improve the overall decision-making capabilities within a firm. Additionally, the study examines the significant role that data analytics tools play in this integrated decision-making framework and how they can significantly enhance decision-making processes within a firm, providing valuable insights [37,39,40]. Furthermore, Wong et al. emphasized that the OIPT serves as a valuable framework for understanding how information integration, the maturity of IT infrastructure, and performance improvements are interconnected in the context of decision-making [41]. This theory helps elucidate the complex relationships that influence how organizations can leverage information and technology to enhance their innovative decision-making.

2.2. Hypotheses Development

This study focuses on the Dynamic Capabilities View (DCV) and Organizational Information Processing Theory as its main framework. It examines the relationships between AI-BA, integration capabilities, digital platforms, and data-driven innovation, as well as their effects on technological innovation.

2.2.1. The Impacts of Artificial Intelligence Integrated Business Analytics (AI-BA)

AI-driven predictive analytics represents an advanced evolution in the domain of analytics, offering firms the ability to enhance their strategic planning and services [37]. By harnessing the power of AI, firms can proactively respond to emerging needs and challenges, ensuring they remain agile and adaptable to address a rapidly changing business environment [37,42].

It can be observed that traditional business analytics has distinct differences from AI-integrated business analytics. While traditional analytics often relies on pre-programmed rules, integrated AI-BA is capable of learning from data autonomously by using advanced techniques, such as machine learning and deep learning [7]. Business Analytics (BA) alone is centered around big data and is defined as a set of techniques used to analyze business data [43]. This analysis helps firms gain deeper insights into their operations, enabling them to make more informed and strategic decisions through descriptive, predictive, and prescriptive approaches to understand their current and past performance and forecast future trends [3,43]. In contrast, AI-integrated business analytics leverages artificial intelligence to enhance learning through sophisticated techniques such as machine learning and deep learning, allowing the recognition of complex patterns and adaptive analyses to derive meaningful insights without constant human intervention and explicit programming [2].

However, while numerous studies have examined the impact of business analytics (BA) in firms, recent research has shifted focus towards the promising contributions of AI-integrated BA. This new direction emphasizes the potential benefits and critically evaluates existing technological solutions to identify any weaknesses [7]. The concept of AI-integrated business analytics (BA) involves a comprehensive approach to data management, which includes the systematic processes of gathering, analyzing, interpreting, and learning from vast amounts of data to accomplish four essential outcomes: diagnostic, descriptive, predictive, and prescriptive insights through the utilization of advanced analytical tools and techniques, such as deep learning algorithms, machine learning models, and neural networks [44]. These sophisticated methods enable firms to understand past trends, forecast future scenarios, optimize decision-making processes, and ultimately drive strategic initiatives.

In addition, Raghupathi and Raghupathi relate AI to business analytics by stating that business analytics consists of three components: AI, statistical modeling, and visualization [11]. Each of these elements contributes to a thorough understanding of business performance and decision-making. However, Gómez-Caicedo et al. highlighted a growing number of studies that examine the role of AI and BA in fostering beneficial development [10]. This expanding field of study emphasizes how these technologies can help firms optimize their operations and improve their overall performance.

Moreover, while analytical capabilities are recognized as a crucial element in a company’s ability to innovate, serving as a foundation for generating new ideas, refining processes, and developing products [45]; Zamani et al. emphasized that the integration of artificial intelligence with data analytics presents an opportunity to improve a higher level of innovation and enhances the efficiency of information processing within firms [18]. They further argue that integrating AI with data analytics, combined with machine learning and simulation techniques, can be viewed as emergent technologies. These technologies enable firms to forecast supply and demand by creating an ecosystem that can respond to disruptions. This adaptability not only influences firm performance but also strengthens overall resilience in the face of challenges [46,47].

In addition, Saleem et al. argue that innovation plays a crucial role in driving superior firm performance, often reflected in higher annual sales and enhanced labor productivity [17], as noted by Wadho and Chaudhry [48]. Despite these advancements, a significant gap remains in the research regarding the impact of data analytics utilization, particularly when combined with effective information integration. This gap underscores the need for studies that examine how these elements can facilitate more informed, data-driven decision-making processes, ultimately leading to enhanced performance outcomes for firms [17]. They further argue that businesses have the potential to drive technological innovations in their operational processes through the effective use of big data analytics. According to Wills and Chen et al., many firms utilize data analytics to enhance their services by analyzing comprehensive client data [49,50]. This data is regarded as a crucial informational asset, providing valuable insights that can significantly reinforce a firm’s performance and competitiveness in the market.

However, to fully use the advantages offered by data analytics technologies for transforming raw data into meaningful business insights that drive decision-making, firms must invest in a strong and reliable IT infrastructure, as Saleem et al. emphasized [17]. Furthermore, Akter et al. highlighted the potential of emerging technologies, such as integrated artificial intelligence (AI), to improve a competitive environment [51]. These advanced technologies can enable firms to create comprehensive platforms that integrate various innovative solutions, allowing businesses to adapt swiftly to market changes and innovate more effectively, as Gill et al. pointed out [52].

In addition, as demonstrated by recent studies, Akter et al. argue that combining AI with other complementary technologies can facilitate digital transformation and integration [51]. To effectively develop both predictive and prescriptive analytics, firms can use the potential of integrating diverse systems, such as customer relationship management (CRM) and supplier relationship management. When paired with innovative business models, this integration positions companies to gain a competitive advantage in the marketplace [51]. For instance, integrating advanced AI platforms—such as Microsoft’s Genee, Salesforce’s Einstein, and Oracle’s Crosswise—results in a more profound and impactful analytics application [53]. This enhanced analytical capability provides deeper insights, empowering firms to make data-driven decisions and ultimately solidify their competitive edge. Based on the argument presented above, the hypotheses can be stated as follows:

H1. 

AI-BA is positively associated with technological innovation.

H2. 

AI-BA is positively associated with data-driven innovation.

H3. 

AI-BA is positively associated with integration capabilities.

H4. 

AI-BA is positively associated with digital platforms.

2.2.2. The Mediating Role of Digital Platforms

Today, digital platforms serve as key drivers of the digital economy’s growth by establishing innovative infrastructures that facilitate interactions among businesses and various stakeholders, promoting continuous innovation and enabling firms to adapt quickly to changing market demands and technological advancements [15,54,55]. The digital platform is an enterprise-level solution designed to facilitate connections among various stakeholders by leveraging information and advanced technology to enhance communication and enable users to share resources effectively [29]. Moreover, digital platforms are crucial as socio-technical entities, utilizing data to generate value and forward innovation in business models [15]. Consequently, a growing body of research across various disciplines has investigated the diverse impacts of digital platforms on markets, industries, and society [15,56]. This multifaceted exploration is essential for understanding how these digital ecosystems shape contemporary business practices and social interactions.

In addition, recent studies emphasize the essential role that digital platforms play in driving innovation for a diverse range of stakeholders [57]. These platforms are increasingly recognized as vital components in enhancing innovation performance within the digital economy, as they serve as hubs for networking, allowing effective collaboration and enabling the absorption of knowledge by providing access to valuable information, ultimately developing a more innovative environment [57,58]. Furthermore, several researchers highlight the significant role that AI and digital platform technologies play in enhancing business operations. They illustrate how these tools enable firms to engage in innovative practices that generate substantial value for their stakeholders [59,60].

In addition, researchers such as Ravichandran and Kroh et al. emphasize the significance of digital platforms for businesses [61,62]. However, they clarify that the impact of these platforms on overall firm performance is not straightforward. Instead, it is mediated by a set of dynamic capabilities, including the ability to innovate, integrate, and conduct thorough environmental scanning [63]. Moreover, Pietronudo et al. emphasize the crucial function of digital platforms as intermediaries that significantly impact digital innovation within organizations [57]. They mentioned that these platforms empower businesses to strengthen their internal coordination processes by seamlessly integrating dynamic capabilities. This alignment facilitates smoother operations, enables firms to coordinate their activities and competencies with their partners, and enhances collaboration in achieving goals [64].

However, Akter et al. highlighted the challenges posed by incompatible technologies in firm platforms designed for data analytics [51]. They pointed out that incompatibility can create significant inconsistencies between internal and external databases. This fragmentation not only hampers the ability to extract valuable and precise information but also increases the risk of making misguided decisions based on incomplete data. However, based on the above argument, the hypotheses are as follows:

H5. 

A firm’s digital platform is positively associated with data-driven innovation.

H6. 

The digital platform of a firm positively mediates the association between AI-BA and data-driven innovation.

2.2.3. The Mediating Role of Integration Capability

Integration capability refers to a company’s ability to enhance overall performance by developing and enhancing product value, which includes effective collaboration and engagement with partners, such as upstream suppliers and downstream customers, to navigate the complexities and fluctuations of a dynamic environment [65]. Wong et al. emphasized that information integration plays a crucial role in enhancing coordination within firms, enabling them to organize activities and make collaborative decisions effectively [41]. As a result, teams can align their objectives towards common goals and improve overall performance and responsiveness. However, several scholars have proposed that integration capability is a crucial precursor to successful teamwork, enabling more productive interactions among team members [41].

Moreover, Chen argues that achieving full integration within an organization is critical to achieving optimal success [4]. However, Zhao et al. caution that a firm must enhance its integration capabilities before it can effectively collaborate [66]. This improvement should encompass integrating systems, data, and processes to create a cohesive working environment. Additionally, Horn et al. emphasize that strong integration capabilities are crucial for fostering effective teamwork and collaboration among employees [67].

Furthermore, Chang et al. stated that an effective integration capability allows a firm to create a remarkable value proposition for innovative products, research and development (R&D), and relationships with suppliers and customers [68]. In addition, Chen emphasized that a firm’s ability to integrate various processes and resources not only helps it distinguish itself from its competitors but also stems from its capacity to make decisive and strategic decisions [4]. He emphasizes that while the insights derived from data analytics empower firms to innovate by generating new ideas, investigating emerging market needs, and developing collaborative relationships with their partners, the effective internal integration of information can significantly improve communication and collaboration across various departments, ensuring that everyone is aligned and working toward common goals [69].

Chatterjee et al. emphasize that firms must integrate their systems, such as Customer Relationship Management (CRM) and Enterprise Resource Planning (ERP), along with all relevant data [70]. They argue that AI technology facilitates these integration processes, ultimately benefiting firms. Conversely, while some studies like those conducted by Saleem et al. focus on the role of information sharing as a mediator that influences the relationship between the utilization of Big Data Analytics (BDA) and innovative outcomes [17], others like Wang et al. investigate into the concept of integration capability, examining it as a mediating factor that affects the performance of innovation within firms [71]. Additionally, Guan and Liu provide insights into how integrating systems and processes directly impacts a firm’s ability to engage in innovative activities [72]. These various perspectives give a more precise understanding of how integration and innovation relationships drive innovation within firms. Based on the argument presented above, the following hypothesis has been formulated:

H7. 

The Integration Capabilities are positively related to data-driven innovation.

H8. 

The integration capabilities of a firm positively mediate the association between AI-BA and data-driven innovation.

2.2.4. Data-Driven Innovation and Technological Innovation Performance

Data-driven innovations (DDI) refer to advancements that prioritize data as a key element in developing innovative solutions and strategies aimed at facilitating informed decision-making and improving operations, thereby helping organizations remain competitive [73]. According to Hossain et al., data-driven innovation is considered one of the most transformative kinds of innovation emerging today [74]. Pietronudo et al. noted that while data-driven culture focuses on a deep understanding of core values and underlying assumptions, it empowers organizations to harness the full potential of product development initiatives [57]. The term “data-driven innovation” refers to the latest advancements in information technology, strong analytical skills, and effective data management systems that are crucial for a firm aiming to thrive in a competitive landscape [12,57,71].

However, in today’s rapidly evolving marketplace, businesses need to prioritize innovation to stay competitive and this requires the development of new products and services that offer unique value and a commitment to continuously updating their knowledge and capabilities. By doing so, companies can better understand and anticipate their customers’ changing needs and preferences, both now and in the future [73]. Eriksson and Heikkilä argue that despite the substantial volumes of data generated by enterprise applications such as Customer Relationship Management (CRM) and other systems, only a few firms successfully exploit this data for meaningful purposes [73]. Many firms struggle to derive tangible benefits or creatively leverage this data asset to drive innovation or enhance their current products and services [75,76]. This highlights a significant gap in data utilization, suggesting that only accumulating data is not enough; effective use is required to translate this data into actionable insights and innovations.

In addition, Pietronudo [57] pointed out that while many researchers actively explore how data influences product development across various industries and its impact on society [77,78], firms encounter challenges related to data that hinder their ability to leverage its full potential. These challenges arise from the complexities associated with data-driven applications, which in turn limit their ability to innovate and improve their offerings. Conversely, to effectively use data for innovation, technological capabilities are crucial in harnessing data for this purpose, as highlighted by scholars like Koski [79]. Also, Eriksson and Heikkilä emphasize the significance of these technological capabilities, noting that assessing them can be challenging because they are intangible, while certain indicators can be utilized to evaluate them [73].

In addition, as digital technologies continue to revolutionize various industries, the concept of innovation is significantly altering the landscape of artificial intelligence (AI) and big data [57]. This transformation highlights the importance of generating new ideas and establishes frameworks for leveraging data effectively within a data-driven industry. As a result, firms are increasingly adopting a data-driven approach to enhance their operations and advance innovation [57].

Although some research, such as that conducted by Sultana et al., highlights the potential value creation associated with DDI capabilities and their influence on firm performance, a significant gap exists in our understanding of DDI’s role at the industrial level [80]. This lack of clarity suggests that more comprehensive investigations are needed to explore this topic further, as Hossain et al. recommended [74]. Additionally, existing literature has not adequately addressed the mediated role of strategic agility in this context, nor has it examined the critical importance of analytics capabilities in enhancing DDI effectiveness [74,80]. Addressing these issues could provide valuable insights into how DDI can be leveraged for improved performance in various industries.

Moreover, scholars have increasingly focused on the performance of technological innovations in their earlier research, seeking to understand the critical influence that various business models employ on this performance [81]. This investigation highlights the essential relationship between innovative practices and the strategic frameworks firms adopt to ensure their success in a competitive environment. Technical innovation refers to the systematic implementation of new ideas and technologies that lead to significant advancements in products, processes, and administrative practices within a firm [82]. Furthermore, it has been argued that the data and knowledge, whether sourced from within the organization or gathered from external environments, play a crucial role in advancing technological innovation, and the effective application of innovative practices has the potential to reshape and redefine the existing dominant paradigms of technological innovation [81].

Moreover, Liu et al. argue that while the global landscape rapidly evolves and expands, we are witnessing a significant surge in technological advancements, such as big data and AI, that impact and strengthen the development of technological innovation [15]. This dynamic relationship between emerging technologies and global progress underscores the transformative potential of these revolutions in shaping future advancements and technological innovation. Some studies emphasize the crucial need for organizations to enhance their capabilities by adopting innovative techniques [83]. These investigations, such as the study by Saleem et al. [17], reveal a notable connection between the use of big data and technological innovation, illustrating how these factors significantly impact the operational effectiveness of firms. Leveraging big data drives technological advancements and transforms various aspects of business operations, leading to improved efficiency and competitiveness [17,83].

However, Saleem et al. highlight an important gap in the existing literature regarding the relationship between data analytics and its impact on technological innovation [17]. While numerous studies have established a connection between big data usage and improved firm performance [84,85], a lack of research remains in specifically examining how the utilization of data can drive innovation in technology, particularly in the context of data-driven decision-making processes. This oversight highlights the need for further investigation into the role of DDI in driving technological innovation within firms. Consequently, based on the aforementioned reasoning, the following hypothesis is proposed:

H9. 

Data-driven innovation is positively related to technological innovation performance.

H10. 

Data-driven innovation positively mediates the association between AI-BA and technological innovation performance.

H11. 

Data-driven innovation positively mediates the association between digital platform capabilities and technological innovation performance.

H12. 

Data-driven innovation positively mediates the association between integration capabilities and technological innovation performance.

2.3. Research Model

Figure 1 presents a conceptual research framework that effectively highlights both direct and indirect relationships, along with their associated hypotheses. To explore these relationships, structural equation modeling (SEM) is utilized, showcasing directional effects through path coefficients. This approach enables a clear visualization of the model, aligning with established SEM practices, rather than relying solely on regression equations.

3. Methods

This study employed a deductive approach, which guided the selection of a quantitative method for gathering data from the identified target population. Glymour et al. noted that a causal research design is instrumental in testing hypotheses and producing numerical outcomes to help validate the research questions [86]. Additionally, this research was explanatory, aiming to provide a comprehensive understanding of theoretical concepts alongside empirical findings. This approach aims to establish necessarily actual knowledge, provided that all the parameters involved are well-defined and accurately measured [27]. By taking this step, the study can enhance outcomes, increase clarity, and make a significant contribution to the existing body of knowledge.

3.1. Sampling and Data Collection

To achieve the objectives set forth in this study, the measurement scales were adapted from existing literature, and a comprehensive theoretical examination and an extensive literature review were conducted. All items were explicitly attached to firm-level practices in the context of integrated AI-BA and innovation capabilities. During the second phase of the project, an English-language survey was developed to be sent to the experts in the field. English was selected due to its widespread use in professional and academic environments in Saudi Arabia, which ensures respondents can effectively engage with the survey instrument, leading to more meaningful and accurate feedback.

To establish content and face validity, a small pre-test was conducted involving three professional respondents outside the main sample to gather valuable insights and ensure the validity of our approach. The experts in the field were tasked with evaluating the survey for clarity and accuracy, ensuring that survey items were clearly understood and interpreted consistently with their intended meaning. This clarity will help ensure accurate responses and valuable insights. Based on their insightful feedback, the necessary modifications were implemented to refine the wording of the instrument, ensuring that it effectively captures the intended data. By following these steps, a solid theoretical foundation has been established, and practical clarity for the instrument, as well as its validity in the context of Saudi business practices, has been ensured.

The data were collected within the Saudi Arabian context. In 2024, the country achieved notable recognition by securing second place in the ICT Development Index, highlighting its commitment to enhancing its technological infrastructure and innovation [87].

The survey comprises several questions that contain categorical variables, including the industry sector, the number of employees in the organization, and the respondent’s position within the company. To evaluate respondents’ opinions effectively, the questionnaire employs a five-point Likert scale ranging from “strongly disagree” to “strongly agree.” This scale is particularly beneficial for expressing levels of agreement or disagreement with the presented statements [88].

The data collection for this study was conducted among firms across various industries in Saudi Arabia, utilizing a purposive sampling approach. This method was chosen to effectively target respondents with specific expertise relevant to the research objective [26]. The focus was on individuals knowledgeable about business analytics (BA), artificial intelligence (AI), digital and integration capabilities, and the potential for data-driven and technological innovation to enhance business value. Also, this sampling approach has proven effective in previous studies, including an AI-CRM study by Chaudhuri et al. [26].

In addition, to ensure a rich and diverse dataset, respondents were recruited through established professional networks and industry associations, and data collection was conducted between October and January 2025. A survey link was shared with approximately 600 professionals, achieving a commendable response rate of 26%. This engagement attracted 230 participants from diverse firms and professions, yielding a robust sample size for hypothesis testing. By adhering to the guidelines set by Gefen et al. [89], we can confidently evaluate the research model. The characteristics of the sample are detailed in

Table 1. Characteristics of the sample.

Sector	No.	%	Employees	No.	%
Telecommunications	37	16.0%	<50	47	20.4%
			50–249	55	23.9%
Wholesale and retail	39	16.9%	250–499	49	21.3%
			>500	79	34.3%
Manufacturing	42	18.2%	Respondent’s position
			CIO	7	3.0%
Healthcare	27	11.7%	Business unit manager	25	10.8%
			Project Manager	59	25.6%
Insurance	17	07.3%	Data Analyst	64	27.8%
			System Analyst	53	23.0%
Banking	59	25.6%	Programmer Analyst	22	9.5%
Other	9	03.9%

below.

In addition, the dataset underwent a thorough quality control process to ensure its accuracy and completeness before analysis began. An analysis of the missing data revealed that all variables had zero missing values, as the survey was conducted online, eliminating the need for imputation. Furthermore, the skewness and kurtosis values for all items fell within the acceptable ranges—±2 for skewness and ±7 for kurtosis, which met the distribution diagnostics. As Hair et al. [90] point out, this supports the assumption of approximate normality, which is essential for valid results. Moreover, all responses fell within the 1–5 Likert scale range, confirming that there were no outliers beyond this interval. However, the alignment of skewness, kurtosis, and observed ranges within these acceptable parameters positively suggests the absence of extreme univariate outliers in the dataset, reinforcing the reliability of our analysis.

To enhance data adequacy assessment, the Kaiser-Meyer-Olkin (KMO) measure has been used and shows that sampling adequacy yielded an impressive overall value exceeding 0.90, with all individual MSAs also scoring above 0.90. This suggests a strong foundation for conducting factor analysis [90]. Moreover, the Principal Component Analysis (PCA) results indicated that the first four components had eigenvalues greater than 1, together explaining 67.5% of the variance, which reinforces the view that the dataset possesses a robust factor structure, providing a solid basis for further analysis [90]. See

Table A1. Components Eigenvalue.

	Eigenvalue	Variance Proportion	Variance Cumulative
Component 1	13.889	0.514	0.514
Component 2	1.967	0.073	0.587
Component 3	1.259	0.047	0.634
Component 4	1.111	0.041	0.675
Component 5	0.788	0.029	0.704
Component 6	0.692	0.026	0.730
Component 7	0.674	0.025	0.755
Component 8	0.654	0.024	0.779
Component 9	0.506	0.019	0.798
Component 10	0.483	0.018	0.816
Component 11	0.459	0.017	0.833
Component 12	0.455	0.017	0.849
Component 13	0.414	0.015	0.865
Component 14	0.376	0.014	0.879
Component 15	0.372	0.014	0.892
Component 16	0.346	0.013	0.905
Component 17	0.325	0.012	0.917
Component 18	0.308	0.011	0.929
Component 19	0.292	0.011	0.940
Component 20	0.269	0.010	0.950
Component 21	0.246	0.009	0.959
Component 22	0.221	0.008	0.967
Component 23	0.206	0.008	0.974
Component 24	0.196	0.007	0.982
Component 25	0.184	0.007	0.989
Component 26	0.163	0.006	0.995
Component 27	0.147	0.005	1.000

and

Table A2. Principal Component Analysis (PCA).

	MSA
DD1	0.951
DD2	0.951
DD3	0.951
DD4	0.960
DP1	0.946
DP2	0.955
DP3	0.958
DP4	0.970
DP5	0.961
DP6	0.964
DP7	0.958
DP8	0.965
IC1	0.932
IC2	0.946
IC3	0.963
IC4	0.954
IC5	0.942
INP1	0.947
INP2	0.937
INP3	0.954
INP4	0.947
INP5	0.950
USEIN1	0.943
USEIN2	0.927
USEIN3	0.917
USEIN4	0.931
USEIN5	0.929

in Appendix C.

3.2. Data Analysis

The Partial Least Squares Structural Equation Modeling (PLS-SEM) analysis technique is valuable for scholars. It enables them to gain detailed insights into variance and facilitates a thorough examination of both the measurement and structural models [91]. This can effectively conduct hypothesis testing, enhancing the robustness and validity of their findings. Additionally, Hair et al. have argued that PLS-SEM is frequently recommended as a superior method to Covariance-Based Structural Equation Modelling (CB-SEM) in various research contexts when the objective is to improve prediction accuracy [91].

However, PLS-SEM was utilized for model testing using Smart Partial Least Squares version 4.1.0.2. This multivariate statistical analysis technique focuses on predictive purposes and incorporates reflective measurement within the model, enabling a comprehensive understanding of the relationships between variables [91]. Additionally, this approach is employed by several relevant studies, including those of Rana et al. and Chatterjee et al. [7,12].

3.3. Data Analysis

This study employed two data analysis approaches, as outlined in the methodology by [91]. The first step involves developing a measurement model to assess validity and reliability. Following this, the structural model aims to explore the relationships and effects between the constructs of the model.

3.4. Measurement Model

To effectively develop the measurement model, the factor loadings and average variance extracted (AVE) values must exceed the threshold of 0.5. This indicates that a significant portion of the variation in the indicators is explained and demonstrates a strong relationship between the constructs [91]. Additionally, the composite reliability must be greater than the recommended threshold of 0.7, reflecting the internal consistency of the measurement [91].

Table 2. Cronbach’s alpha, rho_A, composite reliability, average variance extracted (AVE).

Constructs	Items	Loadings	AVE	Composite Reliability	Cronbach’s Alpha	rho_A
AI-BA	USEIN1	0.834	0.706	0.923	0.896	0.896
	USEIN2	0.856
	USEIN3	0.848
	USEIN4	0.822
	USEIN5	0.839
Data-Driven Innovation	DD1	0.879	0.770	0.930	0.900	0.901
	DD2	0.878
	DD3	0.879
	DD4	0.873
Digital Platform	DP1	0.758	0.667	0.941	0.928	0.929
	DP2	0.810
	DP3	0.817
	DP4	0.844
	DP5	0.823
	DP6	0.844
	DP7	0.813
	DP8	0.820
Integration Capabilities	IC1	0.781	0.665	0.908	0.874	0.875
	IC2	0.852
	IC3	0.835
	IC4	0.830
	IC5	0.779
Technological Innovation	INP1	0.801	0.705	0.923	0.895	0.896
	INP2	0.868
	INP3	0.846
	INP4	0.840
	INP5	0.843

provides a detailed overview of the results obtained for convergent validity. Since all observed loadings, along with the AVE values (ranging from 0.665 to 0.770) and composite reliability values, exceeded these critical thresholds, it can be confidently asserted that convergent validity has been thoroughly established for this study.

Additionally, the discriminant validity test is a technique used to assess the effectiveness and precision of a measurement tool. To achieve this, it is essential to ensure that the underlying variables demonstrate significantly stronger correlations with their constructs than those of competing constructs [92]. The loading factor results presented in Table 2 exceed 0.5, indicating a strong association between the variable and its indicators. Moreover,

Table 3. HTMT Analysis.

	AI-BA	Data-Driven Innovation	Digital Platform	Integration Capabilities	Technological Innovation
AI-BA
Data-Driven Innovation	0.66
Digital Platform	0.69	0.78
Integration Capabilities	0.65	0.80	0.90
Technological Innovation	0.74	0.64	0.72	0.71

demonstrates discriminant validity, as the loading factor for each indicator is higher than its cross-loading values with other variables. This further confirms the distinctiveness of the indicators. Consequently, these indicators can be considered valid and reliable for analysis [92].

Moreover, the discriminant validity test is a technique used to evaluate the effectiveness and precision of a measurement tool. To achieve this, it is essential to ensure that the underlying variables demonstrate significantly stronger correlations with their constructs than those of competing constructs [93]. The loading factor results presented in Table 2 exceed 0.5, indicating a strong association between the variable and its indicators.

Additionally, the HTMT analysis reveals that the values fall within the liberal criterion of 0.90, which is considered an acceptable threshold [94], as shown in Table 3. Additionally, to enhance the credibility of the measurement model, the Fornell–Larcker criterion is incorporated as a dual approach, which strengthens the analysis and reinforces the overall reliability of our findings. As presented in

Table 4. Fornell–Larcker criterion.

	AI-BA	Data-Driven Innovation	Digital Platform	Integration Capabilities	Technological Innovation
AI-BA	0.84
Data-Driven Innovation	0.59	0.87
Digital Platform	0.63	0.72	0.81
Integration Capabilities	0.58	0.71	0.81	0.81
Technological Innovation	0.67	0.58	0.66	0.62	0.84

, the result demonstrates discriminant validity, as the loading factor for each indicator is higher than its cross-loading values with other variables. This further confirms the distinctiveness of the indicators. Consequently, these indicators can be considered valid and reliable for analysis [92].

However, while the correlation between Digital Platforms and Integration Capabilities is relatively strong, it is essential to acknowledge that these concepts are theoretically different, as highlighted by prior research. A Digital Platform serves as the technological foundation that enables interactions among businesses and various stakeholders [15,54,55]. In contrast, Integration Capabilities focus on an organization’s ability to improve overall performance through the enhancement of product value, which involves effective collaboration and engagement [65]. This distinction is supported by HTMT confidence intervals, which remain below 1.00, affirming the validity of differentiating between these two constructs.

To assess the potential impact of common method bias (CMB), both procedural and statistical measures were completed. First, prioritized participant privacy by ensuring anonymity and confidentiality, which fosters a safe environment. Additionally, the survey was structured by organizing items into distinct sections to help alleviate any concerns about evaluation apprehension. On the statistical side, an analysis has been performed of full collinearity variance inflation factors (VIFs), all of which were below 5.0 as shown in

Table 5. Full collinearity VIFs.

Constructs	VIF
AI-BA → DDI	1.704
AI-BA → Digital Platform	1.000
AI-BA → Integration Capabilities	1.000
AI-BA → Technological Innovation	1.550
DDI → Technological Innovation	1.550
Digital Platform → DDI	3.320
Integration Capabilities → DDI	3.020

. This evidence confirms that our data is free from multicollinearity and common method bias, reinforcing the integrity of our findings [95].

Nevertheless, the results provided above offer acceptance values that facilitate the evaluation of the research hypotheses through assessments within the structural model.

4. Result

Structural Model

Structural Equation Modeling (SEM) is a technique used to evaluate the relationships between latent variables and to assess whether the proposed model accurately reflects the underlying data structure and relationships among variables [91]. The Partial Least Squares Structural Equation Modeling (PLS-SEM) method is an analytical technique employed in research to examine complex relationships within data to allow researchers to analyze data and provide a framework for estimating the structural models that represent these relationships [91].

Furthermore, the R² values serve as indicators for evaluating both the strength of the structural paths within the model and the overall effectiveness of joint productivity. The research model exhibits impressive predictive capability, supported by R² values of 0.591, 0.399, 0.339, and 0.502. These values reflect a range of acceptable outcomes, indicating that the model can reliably forecast relationships among the studied variables. According to the guidelines by Hair et al., the value exceeds the minimum threshold and is within the recommended range for these assessments [91].

In addition, the Effect size (f²) was analyzed to understand how each exogenous construct contributes to its associated endogenous constructs. Following Cohen’s [96] guidelines, the categorize f² values of 0.02, 0.15, and 0.35 as indicative of small, medium, and large effects, respectively. The effect size analysis (f²) confirmed that AI-BA exerts large effects on Digital Platform (0.664) and Integration Capabilities (0.514), and a medium-to-large effect on Technological Innovation Performance (0.330), as shown in

Table 6. f² values.

Constructs	f2 Values
AI-BA → DDI	0.054
AI-BA → Digital Platform	0.664
AI-BA → Integration Capabilities	0.514
AI-BA → Technological Innovation	0.330
DDI → Technological Innovation	0.102
Digital Platform → DDI	0.077
Integration Capabilities → DDI	0.093

. The mediating constructs contributed smaller but consistent effects. Together, the robustness checks add significant confidence in the model’s stability and explanatory strength. This aligns well with established recommendations for evaluating both explanatory power and predictive validity in PLS-SEM [91,96].

To strengthen the evaluation of the proposed structural model, an exploration of alternative model specifications, including PLSpredict, has been used (RMSE; MAE). Initially, explored alternative model specifications, such as reversing causal paths (for instance, from Technological Innovation to AI-BA and excluding mediators. While these alternatives provided valuable insights, they resulted in higher prediction errors, as measured by PLSpredict (RMSE = 0.755–0.825; MAE = 0.569–0.639), compared to the original model. This indicates that the proposed model not only maintains lower error rates but also demonstrates stronger predictive relevance, with Q²_predict values ranging from 0.331 to 0.440 [97]. These findings reinforce the robustness of the initial model and its effectiveness in capturing the underlying relationships. See

Table 7. Q² predict, RMSE, MAE values.

	Q2 Predict	RMSE	MAE
DDI	0.347	0.816	0.639
Digital Platform	0.389	0.790	0.594
Integration Capabilities	0.331	0.825	0.636
Technological Innovation	0.440	0.755	0.569

.

In addition, the proposed model’s overall suitability was analyzed using the goodness-of-fit assessment described by Tenenhaus et al. for Partial Least Squares (PLS) path modeling [98]. The results indicated that the model has a strong goodness-of-fit, surpassing the critical threshold of 0.36 recommended by Wetzels et al. [99]. This suggests that the model effectively represents the relationships among the variables, demonstrating a statistically acceptable structural model with a goodness-of-fit rating of 0.565, as shown in Table 4. Also, the SRMR result of 0.046 indicates that the model is a good fit, as an ideal fit is characterized by an SRMR value of 0. A value of 0.08 or less is generally considered acceptable, confirming the model’s effectiveness [100]. See

Table 8. R² and global fit indexes and SRMR.

	R2	Average Variance Extracted (AVE)
AI-BA	--	0.706
Data-Driven innovation	0.591	0.770
Digital Platform	0.399	0.667
Integration Capabilities	0.339	0.665
Technological Innovation	0.502	0.705
Average	0.457	0.702
AVE ∗ R2	0.320
GoF	0.565
SRMR	0.046

.

Twelve hypotheses were developed during the investigation. Based on the statistical validation findings in

Table 9. Summary results of the hypothesis development.

Constructs	Hypothesis	Original Sample	Standard Deviation	T Statistics (|O/STDEV|)	p Values
AI-BA → Technological Innovation	H1	0.505	0.067	7.510	0.000
AI-BA → DDI	H2	0.194	0.074	2.606	0.009
AI-BA → Integration Capabilities	H3	0.583	0.058	10.112	0.000
AI-BA → Digital Platform	H4	0.632	0.059	10.657	0.000
Digital Platform → DDI	H5	0.324	0.092	3.503	0.000
Integration Capabilities → DDI	H7	0.339	0.088	3.869	0.000
DDI → Technological Innovation	H9	0.280	0.066	4.249	0.000

and

Table 10. Summary results of the indirect hypothesis development.

Constructs	Hypothesis	Original Sample	Standard Deviation	T Statistics (|O/STDEV|)	p Values
AI-BA → Digital Platform → DDI	H6	0.204	0.062	3.273	0.001
AI-BA → Integration Capabilities → DDI	H8	0.198	0.058	3.416	0.001
AI-BA → DDI → Technological Innovation	H10	0.054	0.025	2.200	0.028
Digital Platform → DDI → Technological Innovation	H11	0.091	0.034	2.667	0.008
Integration Capabilities → DDI → Technological Innovation	H12	0.095	0.036	2.673	0.008

, all of these hypotheses received strong support, indicating the reliability and significance of the results obtained.

The findings from the study examining the effect of AI-BA on DDI (H1), digital platforms (H2), and integration capabilities (H3) indicate that AI-BA has a more substantial association with digital platforms. This is evidenced by a higher path coefficient of 0.632 and a significance level of *** p < 0.000. Furthermore, the impact of integration capabilities on DDI (H7) is more significant than that of the digital platform (H5), with a path coefficient of 0.339 and a significance level of *** p < 0.000.

In the mediation analysis, the bias-corrected bootstrapping was utilized with 5000 resamples to evaluate significance, implementing two-tailed tests at a 95% confidence level and a fixed random seed [101]. While we acknowledge that some indirect effects are small in absolute magnitude, the bootstrapped analysis (5000 resamples, bias-corrected, two-tailed, 95% CI, N = 230) confirms that they are statistically significant. For example, the indirect paths AI-BA → DDI → Technological Innovation (β = 0.054, t = 2.200, p = 0.028), Digital Platform → DDI → Technological Innovation (β = 0.091, t = 2.667, p = 0.008), and Integration Capabilities → DDI → Technological Innovation (β = 0.095, t = 2.673, p = 0.008) were all significant. In addition, higher-order indirect effects such as AI-BA → Digital Platform → DDI (β = 0.204, t = 3.273, p = 0.001) and AI-BA → Integration Capabilities → DDI (β = 0.198, t = 3.416, p = 0.001) demonstrate meaningful transmission mechanisms within the model. Importantly, these results show that the hypothesized mediating pathways are empirically supported and theoretically consistent, even if the size of some indirect effects is modest.

Additionally, the mediating effects of digital platforms on the relationship between AI-BA and DDI (H6) are stronger than the mediating effects of integration capabilities (H8). This is evidenced by a path coefficient of 0.204, which is statistically significant at *** p < 0.001. Moreover, the analysis shows that the association of AI-BA on technological innovation (H1) is significantly greater than its influence on indirect linkage (H10). This conclusion is supported by the highest path coefficient of 0.505 with a significance level of *** p <0.000.

Furthermore, the mediating effects of DDI play a more significant role in linking integration capabilities to technological innovation (H12) compared to the relationships between digital platforms and technological innovation (H11), as well as AI-BA and technological innovation (H10). This is supported by the highest relevant path coefficient of 0.095 with a significance level of ** p < 0.008.

However, the results below support the hypotheses. For additional details, see Figure A1 in Appendix A.

5. Discussion

This research shows that AI-BA significantly enhances the innovation performance of firms by leveraging both the Dynamic Capabilities View (DCV) and Organizational Information Processing Theory (OIPT). The findings indicate that OIPT plays a crucial role as a mechanism for implementing dynamic capabilities, empowering firms to more effectively sense, seize, and transform opportunities. This integration highlights the positive interplay between DCV and OIPT, providing valuable insights into how AI-BA can drive successful innovation outcomes. However, while this research aims to conduct a detailed empirical analysis of how AI-BA enhances firms’ innovation performance, it additionally explores the significance of data-driven innovation (DDI) in facilitating this enhancement, examining how these elements interact to drive innovative outcomes within firms.

To fulfill the study’s objective, the study thoroughly examined a comprehensive model that incorporates Artificial Intelligence-Business Analytics (AI-BA), Data-Driven Innovation (DDI), and technological innovation performance. This analysis also explored the mediating effects of integration capabilities and the role of digital platforms in enhancing the overall model’s effectiveness. Additionally, this investigation examines the role of DDI and the contributions of digital platforms and integration capabilities in facilitating this process. By leveraging these theories, it is possible to gain a deeper understanding of the complex interplay between technology and organizational innovation performance, enabling firms to thrive in a rapidly changing environment.

While AI-BA is recognized as a vital component for achieving business success, empowering firms to drive growth by advancing innovation and effectively navigating challenging crises [10,37,42], this research has leveraged the concept of AI-BA to develop hypothesis H1, which was subsequently subjected to thorough verification.

The results confirm a correlation between AI-BA and data-driven innovation (H2), as supported by the research of Bahrami and Shokouhyar, Zamani et al., Akter et al., and Kumar et al. [18,45,51,53]. This finding reinforces the idea that integrating AI-BA can enhance innovation levels by facilitating the generation of new ideas, the development of products, and improving the efficiency of information processing within firms to help businesses remain competitive in a rapidly evolving market [18,45]. However, the improved analytical capabilities provide firms with insights into their operations and market dynamics. This empowered understanding allows firms to make informed, data-driven decisions that not only enhance their strategic planning but also significantly reinforce their competitive advantage in the industry [51,53].

The results also confirm a correlation between AI-BA and integration capabilities (H3), as supported by the research of Akter et al., Kumar et al., Saleem et al., Zamani et al., and Wadho and Chaudhry [17,18,48,51,53]. Additionally, while integrating AI provides an opportunity to enhance innovation levels, as suggested by Zamani et al. [18], it could be argued that integrating AI promotes innovation and plays a crucial role in enhancing integration capabilities. By leveraging AI alongside analytics, firms can enhance their integration capabilities and more accurately forecast supply and demand. This enables them to respond more effectively to market dynamics and better analyze comprehensive client data [46,47,49].

Furthermore, the study confirms the effect of AI-BA on digital platforms (H4) by developing a complete platform that creates business value from the benefits of using new data analytics technologies, as discussed by Akter et al. study [51], It could be argued that firms may be motivated to invest in a strong and reliable IT infrastructure, as pointed out by Saleem et al. [17], to enable businesses to harness the full potential of integrating AI with data analytics technologies. This powerful combination allows them to process and analyze vast amounts of raw data, turning it into insightful and actionable decisions. These meaningful insights are crucial in guiding strategic decision-making and helping firms thrive in an increasingly competitive environment.

In addition, while integration capabilities and digital platform elements are regarded as essential resources [4,15,54,55,57,68], the study indicates that integration capabilities (H7) have a stronger positive association with DDI than the digital platform (H5). This suggests that firms with strong integration capabilities may achieve more effective DDI outcomes.

However, it is equally important to note that the digital platform serves as a more significant mediator in the relationship between AI-BA and DDI (H6) than integration capabilities do (H8). This observation underscores the crucial role of digital platforms in facilitating and enhancing data-driven innovation processes, particularly when utilizing AI-BA technologies. Additionally, it can be argued that digital platforms serve as comprehensive infrastructures that facilitate the integration of various data sources and enable seamless interaction among diverse stakeholders involved in the innovation process. They enhance an organization’s capability to leverage data analytics effectively, promoting more agile and informed decision-making.

Overall, it is crucial to acknowledge that integration capabilities and digital platforms are essential mediators and fundamental resources for driving successful data-driven innovation initiatives. This positioning of the mediation role is supported by the findings of various studies in the field, including those by Helfat & Raubitschek, Pietronudo et al., Akter et al., Chatterjee et al., Saleem et al., Wang et al., and Guan and Liu [17,51,57,63,70,71,72]. The study confirms that firms aiming to thrive in an increasingly data-centric environment with AI-BA must harness the power of both digital platforms and integration capabilities to forward innovation and achieve sustainable competitive advantage.

Additionally, this research has empirically validated four hypotheses through statistical analysis. Firstly, it establishes that DDI has a significant influence on technological innovation performance (H9). Furthermore, the findings indicate that DDI positively mediates the relationships among AI-BA, digital platforms, and integration capabilities, enhancing their association with technological innovation performance (H10, H11, H12). This suggests that DDI is crucial in advancing innovation within these technological domains.

The findings suggest that DDI plays a more significant mediating role in the relationship between AI-BA and integration capabilities (H12). This suggests that DDI enhances integration capabilities by enabling firms to improve and leverage their data assets. However, this approach can lead to positive outcomes, allowing the firms to drive innovation, enhance their existing products, and effectively translate this data into actionable insights and novel innovations, as highlighted by Babu et al. and Dubey et al. [75,76].

6. Implications

6.1. Theoretical Implications

This study presents a comprehensive theoretical framework that builds upon the dynamic capabilities view (DCV) and organizational information processing theories to examine the profound implications of AI-BA. The central aim of this framework is to enhance the existing body of literature by exploring in-depth how AI-BA shapes and influences technological innovation within firms.

This investigation has explained the ways in which AI-BA contributes to advancing technological and data-driven innovation, a topic that has not been thoroughly examined in previous studies. By elucidating this relationship, the research seeks to provide valuable insights into the intersection of AI-BA and innovation, thereby filling a critical gap in understanding how firms can leverage AI-BA to drive significant advancements in their technological and data-driven capabilities.

The results of this study underscore the significant importance of AI-BA, digital platforms, integration capabilities, and Data-Driven Innovation (DDI) as essential resources for firms. To thrive in today’s competitive environment, firms must strategically leverage these elements to drive technological advancements and foster innovation effectively. Additionally, the research highlights that digital platforms and integration capabilities are essential for firms seeking to innovate in today’s fast-paced business environment. These aspects are vital for firms seeking to enhance technological innovation and maintain leadership positions in their respective industries.

In addition, this study provides a detailed examination of the crucial role played by intermediaries in bridging the gap between AI-BA and Data-Driven Innovations (DDI). Also, it underscores the importance of digital platforms and integration capabilities as foundational elements that empower firms to drive and sustain technological innovation. By enhancing these capabilities, firms can effectively navigate the complexities of data-driven innovation within the AI-BA framework, ensuring a harmonious balance between technological innovation and strategic goals. Furthermore, thoroughly examining these two critical components in conjunction with existing literature provides unique theoretical insights that enrich our understanding of how digital platforms and integration capabilities can drive innovation within firms. The study shows how those two elements can drive transformative changes within firms, ultimately leading to superior innovation performance and sustained growth.

6.2. Practical Implications

The findings of this study indicate that AI-BA plays a crucial role in enhancing the performance of both data-driven and technological innovation. This suggests that firms adopting AI-BA will likely experience improved effectiveness in their innovative processes and data utilization. In addition, this study presents comprehensive insights for practitioners seeking to advance technological innovation by strategically utilizing their firm resources through AI-BA.

Moreover, the findings underscore the crucial need for practitioners and professionals to prioritize the adoption of AI-BA in their operations. Most firms nowadays, like Amazon, have made substantial investments in AI-BA, aiming to improve their operational efficiency and effectively address customer needs [37]. By adopting AI-BA, a firm can achieve high-performance levels, enhancing the efficiency of its data-driven capabilities and significantly strengthening technological innovation in a dynamic environment.

Furthermore, the study highlights the critical importance of enhancing integration capabilities to improve data and technological innovation within the AI-BA. In this context, managers must recognize the need to develop these integration capabilities. This involves combining various processes and resource aspects, a concept thoroughly discussed by Chen [4]. By developing integration capabilities, firms can gain the full potential of their systems and data. This optimization enables firms to create new data products and improve existing processes.

The digital platform is increasingly recognized as a vital component of AI and innovation [57,58,59,60]. This study highlights the crucial role that digital platforms play in enabling firms to develop and implement technological and data-driven innovations. In line with these findings, managers should focus on leveraging information and adopting advanced technologies, as noted by Cenamor et al. [29]. Accordingly, they can facilitate better collaboration among employees and improve resource sharing, ultimately enhancing data-driven innovation within their firms.

7. Conclusions

Research has highlighted the association of AI-BA on data and technology innovation firms that attempt to enhance both their processes and products. The research has employed the Dynamic Capabilities View (DCV) and Organizational Information Processing Theory (OIPT) to explore how AI-BA helps firms navigate the complexities of innovation in today’s competitive environment and meet their unique needs for technological innovation.

This study fills an important gap by offering valuable insights into the broader implications of AI-BA applications and their association with technological and data-driven innovation. It highlights the significance of integration capabilities in enhancing data-driven innovation and ultimately driving a firm’s success in technological innovation performance. Additionally, the research examines the role of digital platforms in advancing data-driven innovation and enabling firms to achieve exceptional levels of technological innovation performance. Through this investigation, the study demonstrates the interconnectedness of AI-BA, integration capabilities, digital platforms, and DDI, providing a comprehensive perspective on how businesses can leverage these elements to sustain technological innovation.

However, like any other study, it is essential to acknowledge the various limitations of this research that may impact the interpretation of the findings and several opportunities for future research that could expand on these results. Initially, the study primarily used PLS-SEM to make predictions based on the collected data [91], while future studies could employ alternative models, such as covariance-based SEM, for theory testing. This would provide deeper validation of the underlying theories related to the study.

Moreover, given the limited time available for data collection, the framework developed in this study was validated through a targeted survey that covered only firms operating within Saudi Arabia, with a sample size of 230. To enrich the model’s applicability and gain more comprehensive insights, future research should consider expanding and incorporating data from diverse geographic regions to provide a clearer understanding and enhance the overall validity of the findings. Additionally, employing probability-based sampling strategies can improve representativeness [91].

Furthermore, future research could uncover several moderating factors that could impact AI-BA. For instance, it might explore how different environmental pressures, such as market competition and regulatory changes, may affect the implementation of AI-BA and influence overall technological innovation performance. Moreover, explore the algorithmic and governance dimensions and their impact on the implementation of AI-BA, including aspects such as data-sovereignty regulations, state-sponsored AI ethics frameworks, and sector-specific data-sharing mandates across Gulf economies and beyond. Furthermore, future research has the potential to enhance the model by examining additional factors, such as different industry sectors, firm size, or various contextual moderators. Also, future research could employ longitudinal or experimental designs to further enhance the understanding of causal relationships and provide stronger evidence of causal linkages. Finally, future research has a valuable opportunity to enhance measurement precision by incorporating multi-informant designs and multi-method approaches.