Unpacking the Role of Big Data Analytics Capability in Sustainable Business Performance: Insights from Digital Sustainability Reporting Readiness in Latvia

Abstract

A fundamental debate among sustainable development researchers and practitioners is whether, and through what mechanisms, sustainability contributes to gaining a competitive advantage. To address this consideration, this study draws on the general systems theory (GST), organisational information processing theory (OIPT), resource-based theory and recent studies on big data analytics capability (BDAC), digitalisation, sustainability, digital sustainability reporting (DSR), and business performance. A comprehensive research model was developed to assess the interrelationships and mediating effects of the mentioned constructs; the analyses utilised a sample of 75 large Latvian organisations preparing for the first reporting under Corporate Sustainability Reporting Directive. The results obtained using partial least squares structural equation modelling contribute to the theory by revealing that digital sustainability reporting fully mediates the relationship between sustainability and business performance. Moreover, BDAC emerges as a fundamental enabler, forming the foundation of a “house” where digitainability and DSR serve as its walls and enhanced business performance as its roof. This study also contributes to the sustainability and sustainability accounting literature by presenting a sustainable business development framework that offers practical insights for organisations navigating the integration of sustainability reporting in highly uncertain environments.

1. Introduction

The European Parliament (EP) adopted the Corporate Sustainability Reporting Directive (CSRD) in 2022, in alignment with the sustainable development goals (SDGs) and to promote data transparency and comparability within the European Union (EU) market’s sustainability practices [1]. This directive establishes a standardised framework for sustainability reporting, ensuring informed decision-making among stakeholders across the EU market. The new regulatory framework obliged around 50,000 large EU organisations to embed digitally formatted sustainability reports in their annual reports, following the European Sustainability Reporting Standards (ESRS) and EU Taxonomy delegated act structure [2].

However, not everything related to the EU corporate sustainability agenda progressed as smoothly as initially planned. First, 17 countries failed to transpose CSRD in their local legislation by 6 July 2024 [3], while 9 EU members have not finalised that process as of 2025 [4]. Second, the Sustainable Development Goals Report 2024 revealed that only 17% of the SDG goals are on track to be achieved by 2030, while others are moderately to regressively lagging behind [5]. Finally, while some EU organisations published their first CSRD-compliant sustainability reports in early 2025, the wave of market resistance against the new regulation reached its peak. In response, the European Commission (EC) introduced the Omnibus package, proposing to postpone and simplify the CSRD requirements [6]. As stated by the EC, the main purpose of the Omnibus is to remove 80% of organisations from the CSRD scope and significantly reduce the administrative burden [7].

In the context of potential regulatory changes, organisations that have already published their initial sustainability reports, as well as those that have proactively begun or are planning to begin investing in building sustainability reporting corporate practices, remain in a situation of heightened uncertainty. This study aims to examine the key mechanisms underlying the implementation of DSR, determine their relationship with sustainable business practices, and identify critical precursors for maximising the positive impact on business performance in the changing external environment. To address these gaps, the author grounds the research on the notion of BDAC that is defined as the organisation’s ability to leverage technology, talent, and governance tools to transform and analyse data into strategic insights for effective decision-making [8,9,10]. Consequently, this study seeks to answer the following questions:

• Does DSR provide gains for business performance?
• Does DSR enhance sustainability in the organisation?
• What is the role of BDAC in navigating the effective implementation of DSR?

A literature review was initially conducted to identify prominent articles, research streams, and emerging trends. A further empirical study was conducted based on a survey among large organisations in Latvia. A positivist stance was taken, using a survey method to collect empirical data. The questionnaire was developed following the guidelines in [11]. The author applied the partial least squares structural equation modelling (PLS-SEM) method to evaluate the research model and identify predictors of business performance. This method was chosen for its suitability with non-normal data and mixed construct types and is the most commonly used method in BDAC studies [12]. Content analysis was used to analyse answers to the open-ended question.

This study is the first academic research to bridge OIPT and GST by proposing a framework that captures the alignment of internal organisational changes with external environmental conditions. This alignment is driven by the interconnectivity of subsystems within the organisation, which is conceptualised as an open and dynamic system. It also contributes to the literature by revealing the essential role of DSR that mediates the relationships between both sustainability and business performance, as well as BDAC and performance. While BDAC has a direct positive effect on performance, the impact of digitainability is fully mediated by DSR. Based on these insights, the article’s main practical contribution is a proposed five-step framework designed to guide organisations toward sustainable business development.

To answer these questions, the author presents a theoretical foundation in Section 2, focusing on GST, OIPT, and RBT as a theoretical framework for analysing the complex dynamics of organisational change within the DSR environment. Section 3 describes the methodology, including the research model, measurement constructs, and theoretical underpinnings. Section 4 presents the empirical results derived from the PLS-SEM analysis, open-question content analysis and introduces a sustainable business development driven by BDAC and digitainability framework. Section 5 critically discusses the results and explains their theoretical, practical, and policy implications. Finally, Section 6 and Section 7 provide the conclusions, study limitations, and future research directions.

2. Theoretical Background

Aligned with the studies that have empirically investigated the relationship between BDAC and business performance, the author employed a literature review and content analysis as the core methodology for establishing the theoretical background of this study [9,13,14,15,16]. The Scopus database served as the primary source for the literature review, which was initially conducted in early 2024 and updated through January—March 2025. The following keywords were used for the literature search: “big data”, “big data analytics capability”, “sustainability”, “sustainability accounting”, “digitalization”, and “digitainability”.

The BDAC and big data literature analysis focused on the period from 2014 to 2025, as the research stream only emerged meaningfully around 2014. Notably, interest in the topic has surged over the past four years, warranting detailed analysis of literature from 2020 to early 2025.

Sustainability literature was analysed from its inception in the 1987 Brundtland Report to 2025, following the concept’s ongoing evolution. The analysis of sustainability accounting focused on the period of greatest growth in the literature, between 2020 and 2024.

While digitalisation has attracted scholarly attention for over a century, significant academic growth occurred only in the last decade, with publications rising from 234 in 2014 to 8458 in 2024. Accordingly, the theoretical underpinning for digitalisation focuses on this recent period. Finally, all 25 articles published in the last four years on the emerging concept of digitainability, as indexed in the Scopus database, were reviewed.

2.1. GST

Bertalanffy’s GST provides a comprehensive approach to recognise the relationships and interplays among sustainability, digitalisation, DSR, BDAC, and related external and internal drivers, by viewing the organisation as an open system that includes its subsystems. The connection between GST and a corporate integrated reporting system that combines financial and non-financial reporting was first introduced by [17].

GST aims to describe organisations as a complex system consisting of many mutually interacting elements and their environments [18,19].

Thus, DSR cannot be seen as an isolated phenomenon in the organisation, as it interplays with other elements in the organisation and is closely related to the quality and level of established relations among the existent elements. For the application of GST within organisational research, it is important to define the levels of analyses to state the necessary boundaries that ensure the consistency of the approach [18].

Therefore, the levels reported by [20] were used in this study. They defined nine open systems levels, where each of the subsequent levels included properties of the lower levels—level 1: frameworks, level 2: clockworks, level 3: control systems, level 4: open systems, level 5: blueprinted-growth systems, level 6: internal image systems, level 7: symbol processing systems, level 8: multi-cephalous systems, and level 9: systems of unspecified complexity. Similar to [17], this study applied ‘level 3: control systems’. “Control system models describe regulation of system behaviour according to an externally prescribed target or criterion, as in heat-seeking missiles, thermostats, economic cycles in centrally controlled economies, and the physiological process of homeostasis.” [20]. The core characteristic of BDAC is flexibility, which is determined by operational control criteria over information flow based on big data, ensuring changes in related routines and artefacts [21]. Thus, level 3 is a good fit for studying BDAC in an open organisational system. Novicka and Volkova [22] argue that sustainability and digitalisation converge in characteristics and exert a significant influence on one another. Both elements operate in the same larger open system and can be combined into one subsystem that influences the organisation as an open system. Thus, GST provides a robust theoretical basis for the digitainability concept as a subsystem that interacts with BDAC, thereby creating a path driven by the impact of sustainability and digitalisation interaction on business performance.

To investigate the underlying phenomena of BDAC’s impact on business performance, this study elaborates on GST, asserting that digitainability and BDAC function as two control-level subsystems within the organisation, while market uncertainty, driven by the novel and evolving CSRD framework, represents the external environment interacting with the organisation and the changing relationships between its subsystems.

According to GST, subsystems continuously adapt to environmental forces through feedback mechanisms to achieve a steady state of the system–organisation [18]. Hence, under the high-level GST theoretical framework, when describing an organisation as a system, the efficiency of the entire system is considered to depend on the interconnectedness and synergy of the controlled subsystems. Thus, in this study, the analysis shifts its focus from silo-based elements to the routines and coordination of artefacts to understand the best systemic mechanisms that can introduce DSR in the most efficient way for the organisation.

2.2. OIPT

The SLR by [12] lists OIPT among the top three theories applied in the BDAC literature research stream. OIPT provides the theoretical mechanism for transforming big data into BDAC-driven insights to ensure evidence-based decision-making [10,23,24,25,26]. This theory focuses on an organisation’s ability to consciously adapt to changes by diminishing the uncertainty level [10,27].

Scholars define OIPT as a theory that “characterizes business firms as open social systems that seek to execute business strategy through mitigating uncertainty in decision-making processes”, adapted from [28,29] by [30]. OIPT consists of three theoretical elements: information processing requirement, processing capability, and the fit between those for achieving superior business performance [10,23,29,31].

Grounded on OIPT, the CSRD is viewed as an information processing requirement that, in a coercive manner, triggers an organisation’s information processing capability—BDAC level, and navigates an accumulation of new knowledge to support decision-making through DSR readiness and digitainability as mediation concepts for enhanced business performance. Figure 1 reflects the visualisation of the author’s created OIPT-based conceptual research model.

In line with [10,26], OIPT suggests that an organisation adheres to information processing capacity to fit the needs of the changing environment. External relationships and internal environmental uncertainty, as shown in Figure 1, are framed by the DSR. Hence, the adaptation of BDAC happens through the need for conditional and qualitative information processing and knowledge creation to achieve synergy between the organisational capabilities and external requirements [31]. The author argues that the coercive implementation of DSR has a bidirectional positive effect on digitainability. It drives the demand for mature, modern sustainability accounting systems that are integrated with conventional financial systems, thereby improving the availability and quality of sustainability reporting data and enabling its effective use for enhanced business performance.

2.3. RBT

The RBT is the most commonly applied theoretical framework among scholars investigating BDAC [12]. RBT is frequently used to explain how organisations achieve superior business performance by leveraging resources they own or control [32,33,34,35,36]. BDAC, as an organisational resource, is primarily intangible and closely associated with organisational learning and the development of business intelligence required to execute strategic capabilities [37,38,39,40]. When effectively developed and aligned with strategic objectives, these capabilities play a pivotal role in driving superior business performance, as visualised in Figure 1.

2.4. BDAC

Scholars agree that BDAC plays a pivotal role in enhancing organisational competition advantage [9,13,14,15,16,41,42,43,44,45]. However, the related mechanisms that orchestrate or complement the application of BDAC elements to achieve the desired results have not yet been thoroughly explored [9]. Literature review [46] identified four main BDAC elements and driving forces essential for ensuring robust sustainability reporting and improving business performance. These are technology capability, driven by the unity of IT system; management capability, driven by the dynamic capability of knowledge management; human capability, driven by human analytical competencies; and infrastructure capability, driven by flexibility. Talib et al. [47] found that the presence of infrastructure, management, and human capabilities is significant and necessary to enhance decision-making performance. Based on this, the following hypotheses are proposed:

H1. 

The key success elements for BDAC are the technology, management, infrastructure, and human capabilities.

H2. 

BDAC has a positive effect on business performance.

An SLR by [12] identified a gap in existing literature, namely the limited body of research investigating BDAC’s relationship with digitalisation, sustainability, and sustainability reporting. Recently, Choi and Park, Aziz et al., and Xu et al. [33,48,49] identified BDAC’s positive influence on sustainability performance. Li et al. [44] demonstrated that BDAC has a positive impact on green competitive advantage. Market research by [45] identified the crucial role of digital maturity in advancing BDAC in organisations. However, they did not explore the inverse relationship of BDAC’s impact on digitalisation. Thus, the following hypotheses are proposed:

H3. 

BDAC has a positive effect on digitainability.

H3.1. 

BDAC has a positive effect on sustainability.

H3.2. 

BDAC has a positive effect on digitalisation.

H4. 

BDAC has a positive effect on DSR readiness.

According to [50], sustainable business development in the manufacturing organisation is enabled by Industry 4.0. Yaroson et al. [51], in their SLR of circular economy and SDG linkages from 1991 to 2022, highlight BDA readiness as a key enabler for digitally based, evidence-driven decision-making in the execution of circular economy strategies, which are core to achieving sustainable business development. Halbusi et al. [52] outline the positive role of BDA in achieving enhanced sustainable performance. The bibliometric analysis of sustainable business performance by [53] emphasises the need for further research on the role of Industry 4.0 and BDA as key predictors of sustainable business performance. To the best of the author’s knowledge, the existing academic literature lacks a comprehensive understanding of the pathway through which organisations can achieve enhanced sustainable business development in the context of DSR, thereby highlighting a gap in the literature.

2.5. Digitainability and Sustainability Accounting Framework

To ensure reliable, accountable, and transparent information in sustainability reports, organisations must build proper sustainability accounting systems [54,55]. In light of the growing topicality of sustainability reporting, there is increasing research on sustainability accounting aspects aimed at boosting data transparency [56]. Wagenhofer [57] suggests that adopting generally accepted financial accounting principles could provide a more structured approach to data aggregation to enhance the comparability and usefulness of sustainability accounting data. However, the existing sustainability accounting systems are predominantly built in parallel with conventional accounting and management reporting systems [58,59,60]. They lack proper control systems that are always applicable to conventional accounting systems, leading to inconsistencies in data capture and subsequently, invalid eventual results. Thus, sustainability accounting practices require digital solutions that ensure clear definitions of sustainability data and can be integrated into conventional data validation processes [61].

The CSRD mandates a specific, structured, and open-data format, namely Inline eXtensible Business Reporting Language (iXBRL), to facilitate DSR under the European Single Electronic Format (ESEF) framework [1]. iXBRL is a digital reporting format that integrates machine-readable XBRL tags into the human-readable Xtensible Hypertext Markup Language (XHTML) format, enabling both advanced machine-readable data processing and analyses and human-readable presentation that replicates traditional paper-based reports [62]. This digital format catalyses potential digital transformation initiatives within sustainability accounting and reporting practices [22]. Consequently, this study conceptualises sustainability accounting as the seamless integration of sustainability and digitalisation practices within an organisation under a unified digitainability umbrella [63,64,65].

However, according to [55], SLR on sustainability accounting studies, while researchers mostly agree on adapting the conventional accounting principles for sustainability accounting, they point out its distinct nature. The distinctive characteristic of sustainability accounting lies in its multilinked nature, in which data inputs and generated insights are interconnected across departments and external organisational value chains. This interconnectedness extends far beyond the scope of conventional accounting’s internal organisational limits, enabling a more comprehensive and dynamic approach to decision-making [17]. Therefore, the author contends that integrated sustainability and financial accounting practices increase data reliability, leading to more informed management decisions and improved business performance [22]. Thus, the following hypotheses are proposed:

H5. 

Digitainability has a positive effect on business performance.

H5.1. 

Sustainability has a positive effect on business performance.

H5.2. 

Digitalisation has a positive effect on business performance.

2.6. Sustainability, Sustainability Reporting, and Business Performance Framework

There are divergent views on the impact of sustainability on financial business performance. One strand literature presents evidence that sustainable practices contribute positively to overall business performance [66,67,68]. Based on market analyses of top index organisations between 2011 and 2023, Bhue et al. [69] found that investments in sustainability help reduce uncertainty effects on market value, which supports long-term thinking in corporate sustainability practices. However, based on an empirical analysis of a sample of listed European organisations, Cerciello et al. [70] found that, on average, sustainability practices negatively impact business performance in the short and mid-term. Consistent with earlier findings by [71], these results suggest that market representatives who commit to sustainability practices achieve success primarily because of factors other than sustainability commitment. Thus, “telling firms that being socially responsible will deliver higher growth, profits, and value is false advertising” [71].

Wagenhofer [57] investigated the emerging sustainability reporting frameworks, such as ESRS, ISSB Sustainability Disclosure Standards, and SEC regulations, in comparison with the conceptualisation of financial reporting, and found no direct evidence of the usefulness of extensive sustainability data for information users, but emphasised the significant organisational costs associated with establishing sustainability accounting and producing such reports. According to [57], sustainability reports lack a data aggregation perspective, which makes it difficult to measure sustainability performance, effectiveness, and efficiency. Novicka and Volkova’s [22] literature review and bibliometric analyses imply that a digitainability construct is necessary to ensure qualitative DSR. To better understand the cause-and-effect relationships, this study introduces the bidirectional relationships between DSR and digitainability. Consequently, the following hypotheses are proposed:

H6. 

DSR readiness has a positive effect on digitainability.

H6.1. 

DSR readiness has a positive effect on sustainability.

H6.2. 

DSR readiness has a positive effect on digitalisation.

H7. 

DSR readiness has a positive effect on business performance.

H8. 

Digitainability has a positive effect on DSR readiness.

H8.1. 

Sustainability has a positive effect on DSR readiness.

H8.2. 

Digitalisation has a positive effect on DSR readiness.

3. Research Model

In the context of the evolving regulatory environment, for large organisations, digitainability and DSR mediate the relationship between BDAC and business performance (PERF), while DSR directly influences on digitainability and vice versa. The research model is based on GST, OIPT, and RBT and the emerging literature on BDAC, digitainability, and DSR regulation, as illustrated in Figure 2. This research model illustrates the nine main hypotheses of the study.

Based on a synthesis of various academic studies and following the research model logic, the author seeks to explore the orchestrating mechanisms of the impact of BDAC on sustainability, digitalisation, and business performance within a high regulatory uncertainty environment, which exerts pressure on organisations to implement DSR. Consequently, the following hypotheses related to the mediating effects identified within the research model were proposed:

H2.1. 

Digitainability mediates the positive effect of BDAC on business performance.

H2.2. 

DSR readiness mediates the positive effect of BDAC on business performance.

H9. 

DSR readiness mediates the positive effect of digitainability on business performance.

H9.1. 

DSR readiness mediates the positive effect of sustainability on business performance.

H9.2. 

DSR readiness mediates the positive effect of digitalisation on business performance.

These nine hypotheses serve as the foundational framework for the empirical analysis in this study. To explore the complex and dynamic phenomena of the organisation as an open system, the author conducted a survey targeting large Latvian organisations preparing for DSR in the final months of 2024.

4. Methods

4.1. Survey Methodology and Data Collection

The author adopted a functionalist approach aligned with the GST, OIPT and RBT theoretical framework to test the hypotheses and assess each concept as a subsystem of the complex organisational system [18,19]. An additional reason for applying this approach is the descriptive nature of the hypotheses, which require a systematic foundation for the analysis. Based on this approach, ‘systems appear as objective aspects of reality independent of us as observers’ [72]. Thus, a positivist position was adopted for the empirical observations of an organisation as a system, and subsequently, the survey method was applied to collect data and investigate the relationships between organisational subsystems.

The guidelines of [11] were applied to develop the survey. Furthermore, to improve the response rate within a limited time period, the author followed the recommendations of [73], which included topical salience, consent prescreening, social networks, and anonymity.

Based on [9,13], a 7-point Likert scale (1 = totally disagree; 7 = totally agree) was applied. This is broadly used in large-scale studies where there are no well-established standard measures, such as organisational capabilities or newly introduced constructs of DSR. As in other empirical studies related to BDAC [9,13], to avoid organisational and demographic bias following the suggestions by [74], the author added control variables. Similar to [9], the author controlled for organisations based on the following aspects:

• The sector, based on the 11 sectors stated in the International Sustainability Standards Board (ISSB) of the IFRS Foundation [75];
• Size, based on the EU accounting directive [76];
• Age since the inception of the organisation.

Considering the context of the regulative framework, the author also added organisational control for the year of the first reporting [1] and iXBRL reporting experience based on the [77] regulation.

To avoid demographic bias, in line with [13], the author controlled for respondents’ gender and work area (experience). The survey targeted managers responsible for sustainability reporting, as they are most likely to be familiar with the associated reporting processes, hence, management position was included as a control variable. After a panel review of the questionnaire, a test pilot survey in the Latvian language was sent to 64 large companies in Latvia to examine the content validity of the constructs. Answers were collected within roughly three weeks (7 May 2024 to 24 May 2024). The introductory email clearly stated anonymity and that there are no correct or incorrect answers to the survey. The survey included reverse-worded questions. In the pre-test stage, 22 responses were received, representing 8% of the target audience and 269 firms [78], which complies with the recommended range of 5–10% responses for a pilot survey.

First, the author analysed the feedback on the questionnaire, which was provided by a few respondents via email. Based on the pilot survey results, the validity and reliability of the first-order model were investigated using Smart-PLS4 software (Version 4.1.1.1). Subsequently, based on the pilot survey results, improvements in the model were made to enhance clarity in the final survey. It was then translated into English and reviewed by a panel of two experts.

To test the research model, the survey was administered online using the Webropol 3.0 platform across 206 executives from large organisations who are engaged in preparing sustainability reports in Latvia. The main sample organisations, according to the publicly available information, are subject to the Sustainability Information Disclosure Law (henceforth, Law) for the first reporting period in 2024, 2025, and 2026 [79]. The names and details of the main sample respondents were obtained from the author’s personal contacts, publicly available corporate directories, and professional sustainability forums. Additionally, four professional industry networks in Latvia were used to disseminate the electronic survey. The author emailed the respondents with an initial invitation to fill in the survey, which was later followed up with one or two reminders.

The unique value of this empirical research is that it captures the market in the period when the Law was adopted and preparation for sustainability reports started. However, sustainability reports that were due to be published early in 2025 have not been published yet [80]. It is crucial to examine how the subsystems and their elements interact throughout the implementation of sustainability reporting. These results provide a robust reference point for future longitudinal studies that examine the potential impact of regulatory changes and market progress on the adoption of these requirements. Thus, the survey was conducted over a four-month period between 17 September 2024 and 20 January 2025. The sample in four months comprised 85 responses, 75 of which qualified for further analysis. This complied with the minimum required sample of 73 responses to represent a group of 269 organisations with a confidence level of 95% and a margin error of 10%.

4.2. Measurement Framework

The measurement scales for the constructs were drawn from existing literature and grounded in the GST, OIPT and RBT theoretical framework. Table S1 lists the survey questions with the applied scales and related descriptive statistics (means and standard deviations). Based on the research model (Figure 2), two main variables, BDAC and digitainability, were operationalised based on the hierarchical component model with sub-constructs for the main constructs [81].

BDAC. The first latent variable, BDAC, was constructed as a second-order formative construct with four endogenous sub-constructs: technology (TECH), management (MNG), human (HUM), and infrastructure (INFR) capabilities. According to [46] the key success factors of these elements are unity of the IT system for technology capability, dynamic capability for knowledge management for management capability, human analytical competencies for human capability, and flexibility for infrastructure capability. These characteristics were adopted as the basis for BDAC measurement. Particularly, this research employed BDAC measurement items for the sub-constructs from [9,10,16] which were adapted to align with the four main BDAC typologies identified in the SLR by [12] and tailored to the context of DSR. Each BDAC sub-construct was measured using 10 to 13 formative statements.

Digitainability. Digitainability (DIGITTY) was defined according to [63] as an amalgamation of organisational sustainability and digitalisation with the goal of enhancing business performance and measured as a reflective latent construct. This definition distinguishes sustainability (SUST) and digitalisation (DIGIT) as orchestration elements of digitainability [22]. The measurement items for both sub-constructs were adopted from [26,82], further refined based on the findings from Section 2.5 and Section 2.6, and supplemented with additional items to ensure alignment with the DSR framework.

Sustainability. The measurement of the sustainability endogenous sub-construct is based on triple bottom-line integration in the organisation’s business model [83]. The respondents were asked to assess using 10 formative statements the level to which their organisation has implemented sustainability in their policies and decision-making—for instance, ‘We have a sustainability policy’ or ‘Sustainability is integrated into management’s decision-making’.

Digitalisation. The measurement of digitalisation was based on assessing the level of digital technology integration within the organisation’s business model, which was evaluated through 11 formative statements. Respondents were asked to indicate their organisational level for such statements as ‘Management regularly evaluates new possibilities of the digital technologies’, or ‘The role of digitalisation is clearly reflected in the firm’s strategy’. To ensure a link to the new sustainability reporting regulatory framework, the author integrated statements such as ‘Management is ready to invest in digitalisation to ensure qualitative ESRS reporting’, or ‘We implement new digital tools for sustainability accounting’.

Business performance. The literature demonstrates the positive impact of BDAC on business performance (PERF) [9,13,15,16,84]. In this study, business performance was measured based on traditional economic metrics – sales growth and productivity, and financial metrics—return on investment (ROI), return on assets (ROA), and return on equity (ROE) [9,16,85,86]. The author also added sustainability-related aspects in the scales of the dependent construct—for instance, ‘Using data analytics ensures sustainability projects are profitable’.

DSR. The DSR measure was grounded on the regulative CSRD requirements [1], with a particular focus on assessing progress in double materiality analysis and value chain identification and alignment with financial reports before the first report was published. Additionally, DSR measurements incorporating the iXBRL aspect as a digital reporting regulative mandate could influence organisational reporting behaviour, thereby affecting timeliness, digital report quality, and reporting choices [62]. The DSR construct was reflectively measured. Therefore, respondents were asked to assess their current readiness for core DSR-forming elements to comply with the Law, such as double materiality methodology, related integrated governance controls, or iXBRL reporting. “Readiness is a state that is attained prior to the commencement of a specific activity in relation to psychological, behavioural, and structural preparedness of organizations” [87]. Thus, this study defines readiness for DSR as an organisation’s assessment of its preparedness to understand, adopt, and leverage the elements and methodologies of DSR.

4.3. Survey

Table 1. Descriptive statistics of the sample and respondents.

Factors	Coded in the Model	Sample (N = 75)	Proportion (%)
Organisational factors
Sector of operation	Sector
Consumer goods		6	8%
Extractives and minerals processing		6	8%
Financials		12	16%
Food and beverage		4	5%
Health care		2	3%
Infrastructure		6	8%
Renewable resources and alternative energy		7	9%
Resource transformation		2	3%
Services		9	12%
Technology and communications		9	12%
Transportation		12	16%
Size of the organisation	Size
Medium enterprise (50–249 employees)		20	27%
Large enterprise (more than 250 employees)		55	73%
Age of the organisation	Age
10 to 20 years		23	31%
More than 20 years		52	69%
iXBRL reporting experience in the organisation	iXBRL
Yes		8	11%
No		67	89%
Year of the first sustainability reporting	CSRD
For the year 2024		15	20%
For the year 2025		49	65%
For the year 2026 and later		11	15%
Respondents’ demographic factors
Gender	Gender
Female		57	76%
Male		18	24%
Job position	Position
Top management		24	32%
Middle management		31	41%
Lower management		11	15%
Non-management		9	12%
Work area	Work area
Combining two or more work areas		31	41%
Sustainability		23	31%
Finance		14	19%
Other		7	9%

shows that the survey responses were received from organisations in various sectors. The leading sectors were finance and transportation (16% each), followed by technology and communications and services (12% each). Most respondents indicated that their organisation was large (73%). All organisations were established more than 10 years ago, with 69% of them being established more than 20 years ago. In line with the data provided in the official annotation of the Law [78], majority of the respondents intend to prepare sustainability reports for the year 2025 (65%), followed by 20% of the organisations who have already reported for 2024, while only 11% intend to report for 2026. Only 11% of the organisations had previous experience with iXBRL reporting (Table 1).

The survey was predominantly aimed at managers responsible for sustainability reporting, as they were likely to be more knowledgeable about the related reporting process. The demographic data in Table 1 show that 88% of the respondents had lower, middle, or top management positions in their organisations, which is in line with the target of the study. In terms of gender, 76% of the respondents were female, and 24% were male. The breakdown of respondents’ work areas shows that 41% claimed they were combining two or more work areas, such as finance and sustainability, quality and sustainability, and risk management, compliance, and sustainability. Almost one-third (31%) of the respondents identified sustainability and 19% identified finance as their only work area. Moreover, 9% of the respondents indicated other main work areas such as project management, sales, or executive management (Table 1).

Table 2. Structure of the survey sent to respondents.

Nr.	Section Name	Coded in the Model	Total Number of Statements	Reversed Statements
1.	General information	-	8	-
2.	Business performance	PERF	9	1
3.	Digital sustainability reporting	DSR	10	1
4.	Big data analytics technology capability	TECH	10	3
5.	Please name the main challenge faced when preparing for the digital ESRS reporting.	-	Open-ended question	-
6.	Big data analytics management capability	MNG	13	2
7.	Sustainability	SUST	10	2
8.	Big data analytics human capability	HUM	12	1
9	What is the main department responsible for the digital ESRS reporting in your firm?	-	Open-ended question	-
10.	Digitalisation	DIGIT	11	0
11.	Big data analytics infrastructure capability	INFR	10	0
	TOTAL		93	10

Source: Created by author.

summarises the survey structure, which includes 93 statements measuring 8 constructs. Following the recommendations outlined by Podsakoff et al. (2003), the author introduced procedural remedies to address potential common method bias, which may have been caused by measuring the structural elements of BDAC and other constructs [88]. First, as shown in Table 2, 11% of the statements were reverse coded to ensure a high level of respondents’ attention. Second, the order of the statements was intermixed (the original order of the survey is preserved in Table 2) to avoid a consistency motif among the respondents. Third, two open-ended questions were embedded in the middle of the questionnaire (Table 2) to ensure a minimum level of respondents’ acquiescence, and their answers to these questions were in line with the responses provided to the Likert scale statements. However, the target respondents of the survey were managers. To control for common method bias, 9% of the responses from the non-management level were also accepted (see Table 1).

Finally, the statements and definitions of the constructs were clearly worded to avoid ambiguity. The introductory paragraph ensured the anonymity of the responses and that there were ‘no correct or wrong answers’ in the survey.

The hypotheses were tested using Jamovi (version 2.6.23) and Smart-PLS4 tools. First, data were tested for normality in the Jamovi software using the Shapiro–Wilk test, which showed that the data were not normally distributed. Second, to test the research model (Figure 2) and identify the most essential predictive constructs for business performance and related mediation effects, the PLS-SEM method was applied following the recommendations of [89] on how to conduct and report PLS analyses in information systems studies. This method was chosen considering its capability for multiple regressions in the case of both formative and reflective constructs that do not assume normally distributed data. It is commonly used by researchers because it allows the analysis of complex multiple relationships between exogenous and latent variables when the research has an exploratory and predictive nature aimed at theory building [9,35]. Moreover, according to a systematic literature review by [12], PLS-SEM is the dominant statistical method applied in 51% of the cases within existing empirical studies on BDAC.

5. Results

Both first-order and second-order models were used in this study. The first-order model included one endogenous construct—PERF (business performance)—and seven exogenous constructs—TECH (big data analytics technology capability); MAN (big data analytics management capability); HUM (big data analytics human capability); INFR (big data analytics infrastructure capability); SUST (sustainability); DIGIT (digitalisation); and DSR (digital sustainability reporting). BDAC (Big data analytics capability) and DIGITTY (digitainability), were conceptualised as higher-order constructs within the second-order model.

5.1. Model 1

First-order model.

Figure 3 presents the first-order model. The model contained one reflective and eight formative constructs, and the following assessment criteria for validity and reliability were used to evaluate the model: outer weights and loadings (size and significance), composite reliability, average variance extracted (AVE; for convergent validity), and discriminant validity. R-squared and variance inflation factor (VIF) tests were performed, following [11] guidelines to evaluate validity for the formative constructs. The first-order model was constructed and estimated using SmartPLS 4. The impact of the control variables on the individual level—job position, gender, work area [13]—and on the organisational level—sector [9,75], size [9,76], age [9], iXBRL reporting experience [77], and year of the first sustainability reporting [1]—on the dependent variable, business performance, was examined using dummy variables.

Outer weights and loadings. Because the majority of the constructs are formative, outer weights were examined to assess whether the indicators have relative significance in their association with related constructs. For the indicators with the lowest outer weights, outer loadings were also checked. According to [90], formative constructs contain some indicators with low weights. They suggested that these could be retained if justified by the researcher. Thus, the author examined the statements and wording of the indicators that showed weights and loadings below the 0.5 threshold, and the following items were removed from the model: DIGIT8, 9, 11; SUST5, 6, 7; HUM1, 2, 11, 12; INFR4, 7, 8; MNG2, 3, 8, 10, 13; TECH1, 2, 4, 5, 8; PERF 2, 4, 6. Some were retained because of their critical importance for their distinct contribution to the model. Furthermore, the author explains the possible reasons for poor indicators and the logic behind removing the above-listed indicators (see the questionnaire in Table S1).

Omitting DIGIT8, 9, and 11 highlights the irrelevance of aspects related to the digitalisation of sustainability accounting processes and the non-appearance of their integration with financial accounting. This may be because the digitalisation policies of large Latvian organisations currently do not incorporate DSR-related accounting processes due to the emerging nature of the phenomenon and a lack of expertise in its underlying methodologies.

SUST 5, 6, and 7 illustrate the corporate sustainability framework from a compliance perspective, aligning it with regulations and best-practice standards. The low statistical significance of these measurements for sustainability construct may indicate that adherence to best-practice standards in environmental and social responsibility is not prioritised in Latvian organisations. This may be attributed to the low demand for such sustainable practices within the Latvian market and the lack of financial incentives to support their implementation.

HUM 1, 2, 11, 12, and MNG 2, 3 measurements are deleted, as they statistically do not contribute to the BDA human capability. Content-wise, they mainly refer to the data analytics unit’s staff, while other measurements that contribute to the construct relate to the analytical competencies of other organisational staff. Such discrepancy may be due to the fact that large Latvian organisations often do not have separate data analytics units within their organisational structures.

The indicators MNG 8, 13, and INFR4 relate to the understanding and control of organisational sustainability. The lack of their statistical contribution to the respective constructs may indicate that executives in Latvia are not yet prepared to implement sustainability frameworks at a level comparable to other areas of business operations. Thus, they are eliminated from the model.

The low significance of INFR 7 and 8, which focus on operational and sustainability risk management systems, may stem from the fact that many Latvian organisations lack well-integrated risk management systems, as it was revealed by the qualitative field study by [80]. Thus, INFR 7 and 8 do not contribute to BDA infrastructure capability and are omitted from further analysis of the model.

The poor construct fit of TECH 1 suggests a limited understanding among responders of XBRL-based systems. Hence, it is excluded. TECH2, 4, 5, and 8 also show statistical insignificance, indicating that they do not contribute to measuring the unity of the information systems as the key success factor of the BDA technology capability construct.

Notably, the exclusion of the reverse-coded item PERF2—“Management evaluates financial results separately from environmental and social results”—due to its lack of statistical significance in measuring business performance, suggests that executives in Latvia do not perceive environmental and social factors as integral to the business performance construct. This aligns with previous empirical findings from the Latvian market [80]. One possible explanation is that these factors are not seen as directly contributing to short-term financial outcomes, which are typically prioritised by management in large organisations. As a result, such indicators may not be considered valuable for assessing the dependent variable. PERF 4, 5, and 6, which attributed to the ROI, ROA, and ROE ratios, were eliminated as these measurements of business performance, in addition to poor weights and loadings, exhibited extremely high collinearity. This may be because the respondents were primarily not professionals with financial backgrounds and were, therefore, unable to differentiate between these ratios. However, measurement referencing overall financial performance was retained in the model.

DSR was examined as a reflective construct based on the loading with the same threshold, resulting in the elimination of DSR1, 2, and 7. Indicators that measure readiness for iXBRL reporting in the DSR construct, DSR8 (0.443) and DSR9 (0.464), were rounded to the threshold value of 0.5 and kept in the model. The bootstrapping algorithm was then applied to define statistical significance. The results showed that all loadings were statistically significant (p-value 0.000).

Collinearity statistics. Next, the variance inflation factor (VIF) was examined for formative constructs. A VIF greater than 10 suggests high multicollinearity. However, a more conservative approach to formative constructs involves applying a threshold of 5. All values in the first-order model were below 5, indicating the absence of multicollinearity.

Cronbach’s alpha. As shown in

Table 3. Constructs’ reliability and validity.

	Cronbach’s Alpha	Composite Reliability	Composite Reliability	Average Variance Extracted
DIGIT	0.826	0.833	0.868	0.456
DSR	0.828	0.922	0.861	0.482
HUM	0.887	0.893	0.911	0.564
INFR	0.851	0.867	0.886	0.530
MNG	0.830	0.851	0.875	0.507
PERF	0.829	0.843	0.881	0.600
SUST	0.892	0.913	0.917	0.617
TECH	0.769	0.805	0.843	0.522

Source: created by the author.

, all results for the constructs were above the minimum required threshold of 0.7 for Cronbach’s alpha. Therefore, the scale of measurement reliability in the model was internally consistent.

Convergent validity. The convergent validity of the constructs was evaluated using AVE, considering a threshold of 0.5. Table 3 shows that the digitalisation (DIGIT) value of 0.456 and DSR value of 0.482 are below the threshold. However, they could be rounded to 0.5 and accepted as valid in terms of convergent validity. The composite reliability values were higher than the minimum threshold of 0.7 Thus, according to [91], even if the AVE is less than 0.5, but composite reliability is acceptable, the convergent validity of the construct is considered acceptable as well. Therefore, all constructs in the model were treated as valid in terms of convergent validity.

Composite reliability. All constructs are above 0.8, which is above the threshold of 0.7, indicating sufficient composite reliability.

Discriminant validity. The heterotrait–monotrait ratio of correlations (HTMT) was used to assess discriminant validity [92,93]. The threshold <0.85 should be applied for conceptually distinct constructs, for example, in

Table 4. Discriminant validity—heterotrait–monotrait (HTMT) ratio.

	DIGIT	DSR	HUM	INFR	MNG	PERF	SUST
DIGIT
DSR	0.356
HUM	0.690	0.413
INFR	0.740	0.411	0.688
MNG	0.819	0.419	0.757	0.878
PERF	0.657	0.561	0.671	0.666	0.609
SUST	0.750	0.515	0.761	0.733	0.751	0.687
TECH	0.558	0.552	0.602	0.774	0.670	0.740	0.764

Source: created by author.

for BDA management capability and DSR (0.419) or digitalisation and business performance (0.657). However, for conceptually similar constructs, the threshold should be <0.9. Only MNG-INFR had a value of 0.878 and was considered to be below the limit of 0.9 applied to conceptually similar constructs. The HTMT values of all other constructs range from 0.411 for INFR-DSR to 0.819 for MNG-DIGIT, which were below the lower limit of 0.85. Thus, the validity of the measured constructs’ differentiability was confirmed.

The results support the appropriateness of all indicators for measuring their respective constructs and suggest that the measurements of the first-order model are valid for further analysis.

Validity of the first-order model. Following [11], the validity of the endogenous variables was evaluated based on the R-squared values. The R-squared criterion for SEM measures the predictive accuracy of the model [94]. The impact ranges between 0 and 1. The R-squared values of 0.25, 0.50, and 0.75 for endogenous variables indicate low, moderate, and high predictive capacity, respectively. As shown in

Table 5. First-order model R-squared.

	R-Square	R-Square Adjusted
DIGIT	0.534	0.501
DSR	0.318	0.279
PERF	0.726	0.656
SUST	0.659	0.634

Source: created by author.

, the adjusted R-squared demonstrate that all indicators provide valid representations of their respective constructs, with the highest predictive accuracy for business performance (0.656).

First-order model results.

Table 6. First-order model results.

	DIGIT		DSR		PERF		SUST
	Path Coefficient	p Value	Path Coefficient	p Value	Path Coefficient	p Value	Path Coefficient	p Value
DIGIT					0.155	0.212
SUST					−0.241	0.080
HUM	0.236	0.012	0.180	0.200	0.208	0.088	0.315	0.000
INFR	0.243	0.059	0.005	0.978	0.175	0.183	0.141	0.258
MNG	0.352	0.008	0.096	0.625	−0.087	0.560	0.117	0.308
TECH	0.007	0.953	0.377	0.007	0.225	0.058	0.262	0.037
DSR	−0.020	0.825			0.236	0.069	0.176	0.072

Source: created by author.

presents the path coefficient results for the first-order model. Using these effect sizes, users can ascertain whether the effects indicated by path coefficients are weak, medium, or strong. The recommended values are 0.02, 0.15, and 0.35, respectively [95].

The individual path coefficient data analyses revealed a substantial relationship between MNG and DIGIT (0.352), HUM and SUST (0.315), and TECH and DSR (0.377).

The insignificant values of the path coefficients between TECH and DIGIT (0.007), and DSR and DIGIT (−020) show that these relationships do not affect each other. Thus, BDA technology capability does not affect DSR. DSR has no impact on digitalisation; this is also supported by the poor p-value (0.825), proving the absence of a related effect. Thus, H6.2 is rejected.

The relationships between SUST and PERF and MNG and PERF had negative path coefficients and p-values higher than 0.05, indicating no connection between these constructs. All other constructs showed medium positive cause–effect relationships.

According to the literature, a p-value of 0.05 and lower indicates that the predictor variable relates to changes in the response variable. Table 6 shows the strong predictive power of the relationship between HUM, MNG, and DIGIT; TECH with DSR and SUST; and HUM with SUST. None of the constructs in the first-order model were below the 0.05 p-value threshold; thus, none had predictive power for business performance. However, the p-value for TECH was 0.058 (with a medium path coefficient of 0.225), which is close to the significance level and could potentially have some effect. Therefore, further investigation and discussion are required.

Finally, only sector (p = 0.017) and size (p = 0.044) significantly influenced business performance. Therefore, they were retained for control in the second-order model. Based on evaluating the path coefficients for control in the second-order model, the author also retained the variables with high path coefficients: work area (0.691) and iXBRL (0.359). Considering poor path coefficients and p-values, as well as Table’s 1 examination, other control variables (gender, position, age and CSRD) were excluded from further analysis.

5.2. Model 2

Second-order model. PLS enables the investigation of models with high abstraction levels. After the fit has been performed within analyses of the first-order model, a higher-order model using the hierarchical components approach was designed to test BDAC and Digitainability within the model relationships. The second-order constructs, BDAC and Digitainability, were directly measured using observed variables from the first-order constructs. First, in the second-order model, only BDAC, as a formative–formative general construct represented by the manifest constructs—TECH, HUM, MAN, and INFR—was introduced in the analysis to test the direct relationships of BDAC with sustainability and digitalisation (Figure 4). Digitainability was then added as a formative–reflective second-order construct in the model, represented by sustainability and digitalisation as its manifest variables (Figure 5).

The higher-order model results from the PLS analysis are summarised in

Table 7. Second-order model: Path coefficients, p-value, and R-squared.

	BDAC		DIGIT		SUST		DIGITTY		DSR		PERF
	Path Coef.	p Value	Path Coef.	p Value	Path Coef.	p Value	Path Coef.	p Value	Path Coef.	p Value	Path Coef.	p Value
R-squared adjusted	0.997		0.795		0.871		0.673		0.267		0.650
BDAC			0.739	0.000	0.664	0.000	0.760	0.000	0.528	0.000	0.391	0.003
DIGIT											0.130	0.243
SUST											−0.157	0.187
DIGITTY			0.893	0.000	0.934	0.000					0.003	0.983
DSR			−0.042	0.643	0.218	0.016	0.000	0.140			0.238	0.022 *
HUM	0.370	0.000
INFR	0.299	0.000
MNG	0.305	0.000
TECH	0.213	0.000

* DSR and PERF have positive p-value (0.22) only in Model 2. Model 3 demonstrated a p-value of 0.062, which is above the acceptable 0.05 value. Source: created by author.

. Based on the model shown in Figure 4, the author tested the hypotheses related to sustainability and digitalisation. Both SUST (0.871) and DIGIT (0.795) had excellent explained variance (R-squared) with high predictive capacity.

BDAC has a significantly positive effect on sustainability (path coef. 0.664; p-value 0.000), thus supporting H3.1. These empirical results align with the SLR findings of [54], who emphasised data openness and accountability in management control systems, staff analytical collaboration, centralised information sharing, and knowledge management as key factors in promoting sustainability. BDAC has also been proven to have a substantial positive impact on digitalisation (path coef. 0.739; p-value = 0.000), thus supporting H3.2. The hypotheses testing showed that higher BDAC improves organisational sustainability and digitalisation. The key findings of [54] also indicate that achieving all four BDAC elements for enhanced sustainability goals requires integrating digital tools and technologies, thereby fostering organisational digitalisation.

DSR has a moderately positive effect on sustainability (path. coef. 0.218; p-value 0.016). Thus, H6.1 is supported: organisations that introduce DSR positively influence their sustainability. However, a high p-value (0.643) in the second-order model for DSR-DIGIT supports the conclusion from the first-order model that H6.2 is rejected.

Neither sustainability nor digitalisation were found to influence business performance, leading to the rejection of H5.1. Aligned with the empirical qualitative research findings on the key challenges of preparing digital sustainability reports by large organisations in Latvia by [80], the rejection of sustainability as an influential aspect of business performance may be attributed to the fact that sustainability is not treated as a core component of the business model but rather as a stand-alone practice within the organisation. H5.2 is also rejected. Since digitalisation does not demonstrate a relation to business performance, the author argues that Industry 4.0 has become an essential standard for modern market players rather than a source of competitive advantage. This suggests that digitalisation alone is insufficient to enhance competitiveness and profitability unless complemented by strong BDAC. This is consistent with the findings reported by [96], who suggests that to achieve superior business performance results, organisations should move towards the integration of Industry 5.0.

5.3. Model 3

The DIGGITY construct was added to the second-order model, as presented in Figure 5. Following the recommendations of Sarstedt et al. (2022) [97], the author chose the final structural model from the PLS analysis with the best fit based on the Bayesian information criterion (BIC) results, with the aim of minimising BIC. In Model 3, the independent (BDAC) and dependent (PERF) BIC values were substantially away from zero (

Table 8. Bayesian information criterion (BIC) for Model 3.

	BIC (Bayesian Information Criterion)
	Model 2	Model 3
BDAC	−410.423	−414.306
DIGITTY	N/A	−660.375
DSR	−16.891	−16.627
PERF	−50.985	−52.640

Source: created by author.

). Thus, Model 3 has a better fit than Model 2 for testing the other hypotheses in further analysis.

The Model 3 results revealed that BDAC has a significant positive effect on business performance (path coef. = 0.391; p-value = 0.003), and the demonstrated validity of the construct was substantially high (R-squared = 0.997). Thus, H2 is supported. This aligns with [98], whose meta-analysis of 34 selected articles and over 60,000 observations provides strong evidence that enhanced sustainability reporting significantly improves economic performance, highlighting its crucial role in corporate success.

Additionally, the author compares the strong impact of BDAC on business performance with the results of the first-order model (Table 6), where none of the BDAC formative constructs (TECH, MNG, HUM, and INFR), when presented as independent variables, showed an effect on business performance. This indicates that BDAC positively impacts business performance (PERF) only when all first-order constructs are collectively present and accounted for in the model. TECH, MNG, HUM, and INFR were identified as key success elements of BDAC. Thus, it supports H1.

As shown in Table 7, DIGITTY is moderately strong (R-squared = 673), reinforcing its theoretical significance as an assemblage of sustainability and digitalisation within this model [63] in a reflective manner. As shown in Figure 5, Model 3 demonstrates a strong positive BDAC–DIGITTY relationship (path coef. 0.760; p-value = 0.000). Accordingly, H3 is supported.

Further, the author tests the relationship between digitainability and business performance. The Model 3 results could not establish this (path coef. 0.0.003; p-value = 0.983); thus, H5 is rejected.

Model 3 also shows a significant effect of BDAC on DSR (path coef. 0.528; p-value = 0.000). The R-squared for DSR (0.267) had the lowest value in the model but was still statistically significant. Thus, it supports H4. This indicates that the development of BDAC is essential for the successful execution of DSR. Furthermore, the author tested the relationship between DSR and business performance. Figure 5 shows the medium effect of DSR (path coef. 0.192) on business performance. The p-value was 0.062, slightly above the threshold of 0.05. Interestingly, in Model 2, the p-value of the aforementioned relationship was 0.022 (Table 7), which is below the threshold of 0.05. Additionally, in Model 2, DSR demonstrates a stronger impact on business performance, with a path coefficient of 0.238, compared to Model 3. Consequently, given the borderline p-value in Model 3, which does not definitively reject the hypothesis, the author attributes a better fit of the DSR and business performance relationships to the second-order model without digitainability (Figure 4). Therefore, H7 is supported. This phenomenon requires further investigation to gain deeper insight into its underlying mechanisms and implications. Furthermore, a comparison of the first-order and second-order model results, in alignment with the empirical findings of [33], confirms that DSR has a positive impact on sustainability and business performance only when the organisation possesses all four BDAC factors.

Furthermore, the author tested the impact of DSR on digitainability, and the results (path coef. 0.000; p-value = 0.140) did not reveal any relationship between these constructs. Thus, H6 is rejected. Thus, while DSR positively impacts sustainability, as supported in H6.1, it does not impact digitainability and digitalisation, as rejected in H6 and H6.2. The author suggests that, at this early stage of implementation, the approach to DSR remains primarily centred on sustainability, with little to no emphasis on broader organisational digitalisation aspects related to the new regulatory environment. This may indicate that the Latvian market has yet to fully recognise the necessity of establishing robust sustainability accounting and reporting systems to ensure robust DSR.

Furthermore, mediation tests were performed. The results of the mediation tests for Model 3 (Figure 5), shown in

Table 9. Mediation tests.

	Models	Path Coefficient	p-Value
BDAC→DSR→PERF	Model 2	0.126	0.024
	Model 3	0.101	0.064
BDAC→DIGITTY→PERF	Model 3	0.002	0.983
Total mediation: BDAC→PERF	Model 3	0.488	0.000

Source: created by author.

, strongly reject the mediating effect of digitainability on the relationship between BDAC and business performance. Therefore, H2.1 is rejected. However, the p-value in the relationship between BDAC and DSR and PERF was 0.62, which is close to the threshold of 0.5. Thus, the author analysed this mediation without a digitainability construct based on Model 2 (Figure 4). The results presented in Table 9 support the hypothesis, with a p-value of 0.024 and moderate path coefficient of 0.126, indicating a statistically significant mediation effect. Thus, H2.2 is supported. Interestingly, the total mediation effect between BDAC and business performance shown in Table 9, had the strongest correlation (path coef. 0.488; p-value = 0.000). This aligns with the survey results of [49], who found that BDA and corporate sustainable performance are strongly mediated by data-driven competitive sustainability.

The main results of the hypotheses testing from Models 2 and 3 are shown in Figure 6. In addition to confirming the strong direct relationship between BDAC and business performance, the results indicate that BDAC has a significant positive influence on DSR, sustainability, digitalisation, and digitainability. This provides compelling market evidence that enhancing BDAC—by strengthening its key factors—is essential for driving successful business development in the evolving DSR environment.

5.4. Model 4

Finally, to analyse the bidirectional relationships between DSR and digitainability and explore the mediating role of DSR in the relationship between digitainability and business performance, Model 4 was constructed and estimated, as presented in Figure 7.

Table 10. Model 4: Second-order model results.

	DSR		PERF
	Path Coefficient	p Value	Path Coefficient	p Value
DIGITTY	0.502	0.000
SUST	0.591	0.000
DIGIT	−0.057	0.587
DIGITTY→DSR			0.102	0.036
SUST→DSR			0.119	0.038
DIGIT→DSR			−0.011	0.634
BDAC→DIGITTY→DSR→PERF			0.083	0.042
BDAC→SUST→DSR→PERF			0.093	0.043
BDAC→DIGIT→DSR→PERF			−0.009	0.6

Source: created by author.

indicates that sustainability and digitainability have strong positive effects on DSR. Thus, H8 and H8.1 are supported. However, digitalisation and DSR have an insignificant path coefficient (−0.057) and a p-value of 0.587, indicating no effect between the constructs. Hence, H8.2 is rejected.

The mediation analyses revealed that DSR is a good mediator between digitainability and business performance and between sustainability and business performance. Therefore, H9 and H9.1 are supported. However, DSR does not mediate the relationship between digitalisation and business performance. Thus, H9.2 is rejected.

Findings. Table S2 summarises the hypothesis testing results. The analyses revealed a strong direct impact of BDAC on business performance. First, in line with [40], the author emphasises that successful BDAC that positively influences business performance requires not only a managerial focus on BDA technical aspects but also equal investments in management, human, and infrastructure capabilities.

Second, BDAC demonstrated the strongest positive correlation with digitainability, highlighting its essential role as an antecedent to establishing robust sustainability accounting. In turn, digitainability had a significant influence on DSR readiness. The results in Models 3 and 4 confirm that when BDAC is formed by its four elements, the integration of sustainability and digitainability through the preparation of DSR readiness demonstrates a weak but statistically significant effect on business performance (Table 10). While the total statistical effect of BDAC on business performance, considering all relationships, is substantially strong (Table 9), this underscores the importance of implementing all levels of the presented research model (Figure 2) to maximise business outcomes.

The results provide market-based evidence of the critical role of robust DSR readiness. Model 3 demonstrated the significant positive impact of BDAC on sustainability. However, Model 4 indicates a moderate impact of DSR readiness on sustainability. While none of the models demonstrate a direct impact of sustainability on business performance, Model 4 shows that sustainability has a positive impact on business performance through DSR readiness mediation.

5.5. Challenges to DSR Readiness

The survey included an open-ended question: “Please name the main challenge faced when preparing for digital sustainability (ESRS) reporting?”. All respondents provided an answer to this question. To summarise 75 answers and highlight the main challenges mentioned by the respondents, the author visualised the results using word cloud (Figure 8) and text blocks (Figure 9) techniques in the online Voyant text analysis tool [99]. The aim is to apply content analysis to the qualitative data in order to support and enrich the interpretation of the statistical results.

Figure 8 clearly illustrates the central and challenging role of data in the preparation process for Digital Sustainability Reporting (DSR). The word “data” was mentioned 40 times in respondents’ answers, making it the most frequently used term—significantly more than “report” (14 mentions) and “sustainability” (13 mentions). Figure 8 and Figure 9 together highlight a strong contextual emphasis on data-related challenges, particularly the recurring theme of its insufficiency.

Besides the lack of data, the main challenges faced by organisations related to sustainability data are its quality and availability (see Figure 9). Based on the links provided in Figure 8, respondents also outline the absence of the systems’ universalism, stating that data lacks consolidation, as it is located in different departments and systems or unavailable from the value chain. The respondents provided the following related answers about the challenges: “Correct data entry, data consolidation”, “Data gaps and quality of existing data in systems”, “Data extraction from the value chain, data interpretation, manual data processing”, “Data aggregation”, “Lack of unified data management in the group (directions, departments work in different systems, we also widely use Excel, but in different quality)”, “Receiving information from third parties; compiling and analysing existing data (different data formats, etc.)”. These responses underscore the critical role of integrated IT systems within organisations in ensuring data availability and quality, key prerequisites for developing BD technology capability that support robust sustainability reporting.

The connections illustrated in Figure 9 highlight an additional, distinct theme in the responses: the challenge of controlling data processing. This, in turn, influences data quality and reliability in DSR. For instance, the author retrieved the following statement: “Much of the data collection and processing process is manual—with high chances of human error. There is no synchronised single system that automatically reads in the sustainability data currently in the company…”. According to the literature, control over data is crucial for flexibility over sustainability reporting related processes [100]. That draws BDA infrastructure capability as the second crucial element of BDAC to be developed in the organisation for securing sufficient DSR readiness.

Additionally, respondents stress that environmental uncertainty around CSRD regulation brings challenges to understanding requirements for DSR. In line with that, as shown in Figure 8 and Figure 9, ten respondents said that the main challenge for their teams is “understanding the scope of ESRS reporting” or “insufficient staff resources that understand ESRS reporting, data point content…” This point highlights a key challenge related to organisational human analytical competencies within the context of DSR. Additionally, these competencies should be adaptable and capable of evolving in parallel with the rapidly changing regulatory environment. This indicates that the compliant and timely implementation of DSR depends on the proactive development of the BDA human capability.

Finally, Figure 8 shows the ultimate role of the management. Some respondents identified that “lack of understanding at the top management level, which limits funding to address technical and data collection issues for reporting purposes” or “management buyout” mentioned as the main challenge they have to achieve DSR readiness. In line with that, it is challenging to overcome “management complexity to meet strict regulatory standards and deadlines”. The respondents’ statements position top management at the forefront of the organisational changes compelled by the new SR regulation. Therefore, the dynamic capability for knowledge management must be embedded at the top management level to establish a coherent foundation for developing other elements of BDAC essential for preparing DSR. Considering the points outlined above, the presence of BDA management capability within an organisation is a critical element in establishing the artefacts and routines required for effective sustainability reporting.

The investigation conducted by the author in this section provides content-based evidence that BDA technology, management, human, and infrastructure capabilities are necessary elements for BDAC to have a substantial positive effect on DSR readiness. Thus, the results of the content analyses for the answers to the open-ended question from the survey visualised in Figure 8 and Figure 9 validate the approval of H1 and H4 (Table S2) based on the statistical hypotheses tests. However, results show that it is challenging for organisations to integrate BDAC elements to ensure proper DSR readiness.

Therefore, to address the gap, the author proposes step-by-step guidance to support organisations in identifying challenges and building the robust BDAC necessary for integrating DSR and enhancing business performance. A sustainable business development driven by BDAC and digitainability framework (Figure 10) was developed, drawing on an extensive literature review, the GST, OIPT and RBT theoretical framework, and empirical analyses. This framework outlines five optimal and coherent key steps comprising the main elements that organisations should adopt to enhance their business performance.

Based on the hypotheses testing, the author stresses that to achieve a positive impact on business performance, steps from the sustainable business development driven by BDAC and digitainability framework (Figure 10) should not be applied in a chaotic order, and none of the steps should be skipped or adopted partially. Only a consistent implementation of all the outlined elements in the framework will lead to an efficient and effective result.

6. Discussion

The widespread attention surrounding the sustainability agenda (SDG 2030), DSR (Omnibus package), digital transformation (Artificial Intelligence [AI]), and big data (iXBRL) is unprecedented. However, only few studies have examined these interconnected elements within a unified analytical framework [12]. This appears paradoxical, considering the abundance of articles published across academic literature and leading business outlets, such as the Harvard Business Review, Forbes, The Economist, and Fortune, which extensively discuss the transformative conditions, potential, and pitfalls of each phenomenon in isolation. This could be partly because of the lack of a unified integrated framework for practitioners and researchers. This may also be partly attributed to the substantial costs associated with implementing BDAC-driven DSR, which is intended to integrate seamlessly with the broader sustainability and digitalisation agenda, relying on robust sustainability accounting practices. However, the findings of this study demonstrate that introducing such practices poses significant operational challenges and may not always yield positive economic results. It is unclear if the market is ready to overcome these challenges. However, the initial CSRD framework was designed to guide organisations in developing such practices, overlooking administrative and cost-related burdens, as well as limited comprehension of executive management for this complex phenomenon due to the overall market’s low readiness for such a transformative paradigm shift. The empirical results in this study, covering a sample of 75 large Latvian organisations, highlight the interconnectedness of organisational subsystems and reveal the inherent fragility of relationships within a complex open organisational system. Hence, policymakers, by not providing robust competitive guiding frameworks for organisations, have underestimated the risk of first-mover failure in the independent integration of DSR.

The findings of this research offer valuable insights into the impact of DSR on performance within the context of unique pressure and the early challenges of transformative regulatory adoption. However, it is important to acknowledge the significant limitations imposed by the sample and contextual scope of this study. In particular, the focus on organisations in Latvia, characterised by a distinct economic and regulatory environment within the EU, restricts the generalisability of the findings. The findings may not be fully transferable to organisations in other EU member states that operate under different market conditions, possess varying degrees of digital maturity, or have diverse experiences with sustainability reporting. Furthermore, the timing of the study, which captured organisations in the preparatory phase of DSR rather than those with established practices, may reflect temporary states or intentions rather than stable capabilities and measurable outcomes. Additionally, the sample of 75 responses was statistically adequate to reflect the target population of 269 organisations, it remained relatively small for a complex PLS-SEM model comprising multiple constructs and paths. This limitation may lead to less stable parameter estimates and diminished statistical power, thereby affecting the robustness of effect detection and the reliability of model fit assessment. Altogether, these factors should be carefully considered when interpreting the conclusions and applying the proposed framework.

6.1. Theoretical Implications

From a theoretical perspective, the findings of this study contribute to the existing body of literature in several ways. First, it represents an initial effort to explore the intricate relationships among the crucial elements of complex organisational phenomena as an open system through the lens of GST. While the impact of BDAC on business performance has been extensively examined [12], this study is the first to adopt GST to address the shortcomings of existing literature on the mediating roles of DSR and digitainability in these relationships. Thus, this study contributes to the GST theoretical framework. Additionally, this study is the first to integrate OIPT and GST by proposing a framework that illustrates how internal organisational changes align with external environmental conditions. This alignment emerges from the interconnectivity of organisational subsystems, conceptualised within the organisation as an open and dynamic system.

This study contributes to the theoretical understanding of the relationship between sustainability, digitalisation, and business performance by finding no significant direct effect of either sustainability or digitalisation on business performance. This challenges assumptions commonly held in the literature and supports a growing body of research that questions the existence of straightforward causal links between these constructs [70,71]. By rejecting the direct-effect hypotheses (H5, H5.1, H5.2), this study highlights the limitations of models that fail to account for the complexities and dynamic external environment that shape how digital and sustainability initiatives translate into tangible business value. Several contextual factors may explain the absence of direct effects in the specific setting of large Latvian organisations. These include high implementation costs that may outweigh short-term benefits, a lack of financial incentives or insufficient market demand for sustainable practices, fragmented internal organisational processes, outdated or siloed digital infrastructures, and limited executive awareness or education regarding the strategic value of digital transformation or sustainability, particularly in relation to the triple bottom line framework. Specifically, the findings underscore the need to examine the mediating mechanisms, such as DSR readiness, that can bridge these initiatives with quantified performance outcomes. Second, this study empirically validated the sustainability reporting diamond logic by [46] that reinforces the critical role of all four BDAC elements: technology, management, human, and infrastructure capabilities. The results of this functional quantitative research underscore that the successful integration of all these elements, driven by their key success factors, namely unity of IT systems, dynamic capability of knowledge management, human analytical competencies, and flexibility, is essential for embedding BDAC within the organisation to facilitate enhanced business performance, viable sustainable and digital practices, and ensure robust sustainability reporting.

Furthermore, this study contributes significantly to the emerging literature on the concept of digitainability by empirically validating its reflective theoretical nature using a quantitative approach. Based on the empirically validated reflectiveness of the digitainability concept, the author emphasises the critical role of strategic integration of sustainability and digitalisation into one subsystem to ensure its viability.

Additionally, this study sheds light on the transformative role of BDAC in driving digitainability within large organisations. This study contributes to the theoretical framework of sustainability accounting by proving that digitainability is necessary for ensuring integrated, reliable, and insightful data in sustainability reporting. The analyses also reveal that digitainability has a positive effect on business performance only when mediated by DSR, providing a solid basis for future researchers to explore the underlying causes of this phenomenon.

Another key contribution of this study is the empirical validation of the theoretically grounded DSR construct [22], including the iXBRL format in its measurement scale. The DSR construct and its associated measures applied in the survey provide a basis for future studies investigating regulatory frameworks and digital data aspects within the development of sustainability reporting. Specifically, this study contributes to the body of iXBRL-related research by emphasising the potential positive impact of this open data format on the structure and reliability of sustainability reports and on market data availability.

6.2. Practical Implications

This study offers valuable practical implications for both executives and sustainability managers. Notably, the proposed sustainable business development driven by BDAC and digitainability framework offers a reliable roadmap for the management of large organisations to develop sustainable and competitive business practices. This is ultimately important in the context of political debates around the Omnibus package, driven by the narrative that CSRD-compliant sustainability reporting weakens the competitiveness of European businesses [7].

Therefore, considering the complexity and uncertainty surrounding CSRD requirements, to minimise business risks, organisational management should prioritise the development of BDAC. This will facilitate the effective implementation of sustainability accounting, an essential element of digitainability, ensuring accurate, efficient, and meaningful DSR, and ultimately driving positive economic outcomes. Consequently, this study provides empirical evidence supporting the justification for organisational investments in sustainability accounting and reporting, but only when BDAC is effectively established. The presented framework has great significance for top-level executives, because the complex relations of organisational subsystems in the high-uncertainty and high-pressure external regulative environments are often overlooked. However, addressing them swiftly has the potential to generate a competitive advantage, regardless of the outcome of the reporting regulation.

The results provide market-based evidence for managers on the critical role of robust DSR, which, through mediation, establishes a positive sustainability impact on business performance. Particularly in the case when Omnibus potentially decreases (by 80%) the scope of the entities required to comply with CSRD [6], leaving many executive managers alone with the decision of whether to initiate or continue sustainability reporting. The empirical findings show that, for organisations with sustainable practices, producing high-quality sustainability reports is essential for enhancing organisational efficiency and reinforcing economic performance. However, organisations that do not have established sustainability practices should first ensure that all key success factors from the sustainability reporting diamond [46] are fully integrated, and sustainability accounting is built to efficiently foster sustainability practices. The author suggests that managers should introduce sustainability reporting only when it is done to ensure that these practices have a positive effect on corporate finance. The author suggests that this is the most efficient pathway to follow, especially for organisations that are at a crossroads in their decision-making regarding whether to prepare a sustainability report.

Finally, managers of organisations in Latvia, as well as those operating in similar economic and regulatory environments, should not assume that investments in sustainability or digitalisation will automatically result in improved business performance. Instead, it is essential to understand the complex and often indirect nature of this relationship to maximise ROI from sustainability- and digitalisation-related initiatives. Managers shall also address key contextual challenges, such as organisational readiness, limitations in digital infrastructure, and gaps in leadership awareness. These findings highlight the strategic importance of integrated planning, capacity-building, and executive-level education to ensure that digital and sustainability initiatives are effectively aligned with performance-oriented objectives.

6.3. Policy Implications

This research has significant implications for regulators, policymakers, and standard setters, both within the EU and beyond, particularly in the refinement of mandatory sustainability reporting implementation strategies. The empirical findings of this research validate the pivotal role of DSR in enhancing the competitiveness of organisations with sustainability-oriented strategies. However, it also provides evidence that when organisational BDAC has gaps or deficiencies, DSR fails to deliver the anticipated positive impact on business performance and does not strongly boost sustainability. Instead, such organisations incur high costs and resistance from organisational elements to coercive change. This leads to a negative attitude towards sustainable practices in this part of the market. Thus, the author proposes a sustainable business development driven by BDAC and digitainability framework that can support policymakers in refining the CSRD after evaluating market maturity against each of the presented blocks. These findings highlight the need for a more flexible and phased implementation of DSR requirements. Following the logic presented in the framework, it is important to ensure that DSR brings its intended positive economic value to the market.

Moreover, as BDAC is found to be the main driver of the sustainable practices required to achieve SDG, in line with [54] the author suggests that policymakers and standard setters should encourage BDAC investments through policy incentives. Given BDAC’s essential role in enhancing digitainability, sustainability reporting and business performance, policymakers should also consider financial incentives, tax relief, or subsidies for organisations investing in BDAC elements and sustainability accounting systems. This support can help offset high upfront implementation costs and encourage long-term commitment.

This study highlights the importance of structured open-data formats (such as iXBRL) in ensuring DSR, thereby enhancing market data transparency, comparability, and usability. The author suggests that increased educational guidance from standard setters on these formats would facilitate market adoption of the standardised approach to sustainability reporting and foster acceptance of costs associated with high-quality iXBRL tools. This can serve as a bridge between sustainability efforts and performance outcomes, facilitating data-driven decision-making and transparency.

Policymakers shall promote collaborative platforms that bring together businesses, academia, and public institutions to facilitate the exchange of academic findings, best practices, case studies, and lessons learned from successful sustainability and digitalisation initiatives, and the integration of BDAC and DSR. Such collaboration would help accelerate learning, enhance knowledge diffusion, and reduce the risks associated with isolated, trial-and-error implementation efforts.

7. Conclusions

This study was motivated by academic and practical limitations in reasonably addressing the interrelationships among big data analytics, the global sustainability agenda, the Fourth Industrial Revolution, sustainability reporting, and iXBRL practices, particularly in the context of high environmental uncertainty and competitiveness constraints. While there are extensive ongoing market and political discussions about the Omnibus package aimed at significantly changing the CSRD, there is a lagging theoretical backing for these debates. This allows different interest groups to speculate on and manipulate information for their own interests. This study is built on GST, OIPT, and RBT and recent studies on BDAC, digitainability, sustainability accounting, and sustainability reporting. The empirical results highlight the importance of investing in BDAC elements and sustainability accounting before introducing DSR to ensure its efficacy. By doing so, organisations enhance their evolutionary fitness, as insights derived from BDAC facilitate sustainable business practices, while DSR ensures that these efforts are translated through evidence-based decisions into positive business performance. Thus, to address the research questions and help organisations overcome environmental uncertainty, the author introduced a novel empirically validated sustainable business development driven by BDAC and digitainability framework for approaching DSR implementation to secure advanced organisational competitive advantage.

Limitations and Future Research

Despite the theoretical and practical contributions of this study, it has several limitations that future studies can explore. First, however, the model is grounded in theory, the functional paradigm of the cross-sectional research design, applying the quantitative survey method, has limitations. These include restricted capacity for causality, limited depth of responses, as well as potential response bias. While these issues are partially addressed through procedural remedies for common method bias and content analysis of the responses to open-ended question, a more robust validation of the proposed causal pathways and mediating effects would require longitudinal research. Specifically, future studies should track changes in BDAC, DSR adoption, and business performance over time to strengthen the evidence for the framework’s sequential relationships. Future studies may also triangulate the results by conducting qualitative analyses based on expert interviews and case studies of organisations that have published DSRs. This would provide a more nuanced understanding of the cause-and-effect findings revealed in this study. Additionally, the focus on sustainability reporting managers narrows the perspective. Future studies should include managers from other functional areas to enable comparative analysis and provide a more comprehensive view of organisational readiness for DSR.

Second, the empirical data were obtained only from large Latvian organisations. Therefore, the findings have limited generalisability. Thus, in future, researchers could conduct similar surveys in another EU market, ideally with larger sample sizes to enhance the robustness of statistical results. This need is particularly relevant given the postponement of the DSR requirements for a substantial number of organisations under the EU Omnibus package, extending the timeline by two years. The results are also limited by the timeframe reflecting early evidence from the market of initial CSRD wave. Thus, future scholars can longitudinally explore how the evolution of the CSRD regulatory framework transforms organisational subsystems.

The practical applicability of the sustainable business development driven by BDAC and digitainability framework developed in this study is subject to limitations arising from the research methodology and design constraints. Therefore, future validation in independent real-world market settings is necessary to confirm its relevance and effectiveness.