Drivers and Barriers for Sustainable Design Adoption in Creative Economy Enterprises: A Corporate Strategy Perspective

Abstract

Incorporating sustainable design practices into creative economy businesses is increasingly vital amidst growing global environmental concerns and shifting market demands. This research examines the factors that promote or hinder the adoption of sustainable design from a strategic business perspective. Utilizing data from a detailed survey, it adopts a multi-faceted approach, incorporating descriptive analysis, exploratory and confirmatory factor analyses, structural equation modeling (SEM), cluster analysis, and machine learning methods. Findings indicate that leadership vision, innovative capacity, and customer engagement are primary motivators, whereas internal inefficiencies and limited resources present significant obstacles. Cluster analysis reveals three strategic profiles: Innovation-Driven, Strategically-Aligned, and Barrier-Dominated, offering meaningful insights for designing targeted strategies. The study delivers a validated framework that enhances sustainability theory and supports strategic decision-making within the creative sector.

1. Introduction

Sustainable design is becoming increasingly critical in corporate strategies as companies worldwide face mounting sustainability challenges, including climate change, resource depletion, and growing social inequalities [1,2]. Organizations that strategically integrate sustainability principles are shown to gain improved market positioning, enhanced customer loyalty, and long-term profitability [3,4]. Prominent examples such as Patagonia and IKEA illustrate how sustainable design adoption not only improves operational outcomes but also reinforces brand reputation and resilience [5].

Among the sectors responding to this imperative, creative economy enterprises, including firms in design, architecture, media, fashion, and digital content, are uniquely positioned to drive sustainable innovation. These firms operate at the nexus of cultural production and economic activity, granting them considerable influence over consumer behavior and social norms [6]. Their agility and cross-disciplinary creativity make them particularly suited to experimenting with and diffusing sustainable practices [7]. However, despite growing awareness, sustainable design adoption in this sector remains inconsistent and fragmented [8].

This disparity raises a critical question: what enables or inhibits sustainable design integration in creative enterprises, and why does adoption vary across firm types? While global frameworks like the United Nations Sustainable Development Goals (SDGs) and the Paris Agreement have elevated the visibility of sustainability on corporate agendas, they offer limited practical guidance for creative firms [9]. Specifically, most creative enterprises face internal operational constraints, such as lifecycle data silos, fragmented design-to-production workflows, and a lack of interoperable digital platforms. These limitations undermine compliance, traceability, and sustainability alignment, revealing a disconnect between aspirational global targets and enterprise-level execution.

To address this misalignment, this study draws on a multi-theoretical foundation, combining the following: (1) Stakeholder theory [10], which explains how firms respond to the expectations of diverse actors, consumers, regulators, NGOs, and investors. (2) Institutional theory [11], which accounts for how norms, rules, and cultural forces shape organizational behavior. (3) Dynamic capabilities theory [12], which explains how firms reconfigure internal resources to respond to external shocks like climate mandates. These theoretical perspectives are complemented by a systems-thinking lens, which provides insight into the fragmented internal environments of creative firms that impede strategic sustainability implementation.

Despite growing interest in design-based sustainability, the literature reveals a lack of integrated theoretical models and empirically validated frameworks tailored to the creative economy [13]. Many sustainability initiatives in these firms remain superficial, hindered by unclear leadership priorities, disjointed workflows, and weak institutional coordination [14]. This study aims to bridge the research gap by integrating theory, empirical data, and system-level design frameworks.

Accordingly, the study is guided by the following research questions: (1) What are the primary drivers and barriers to sustainable design adoption in creative economy enterprises, and how can these be grouped into interpretable constructs? (2) How do leadership vision and innovation capacity mediate the relationship between stakeholder/regulatory pressure and design adoption? (3) Which firm typologies, defined by strategic posture, size, or resource configuration, exhibit faster or slower adoption, and what factors explain these patterns? These questions guide the development of a conceptual model and inform hypothesis testing using structural equation modeling (SEM) and cluster-based firm typologies.

This research makes the following key contributions: (1) It proposes an integrated framework combining stakeholder, institutional, and dynamic capabilities theories to explain sustainable design adoption in the creative sector. (2) It employs a multi-method design, encompassing exploratory and confirmatory factor analysis, structural equation modeling (SEM), clustering, and machine learning, to validate the model empirically. (3) It defines and analyzes actionable firm typologies (e.g., Innovation-Driven, Barrier-Dominated, Strategically-Aligned) to inform context-specific recommendations for managers and policymakers.

Through these contributions, the study advances sustainability theory, supports evidence-based policymaking, and enhances the implementation of strategic design in one of the most dynamic yet underexamined segments of the global economy.

2. Literature Review

2.1. Conceptual Framework

To structure the barriers–drivers–readiness architecture (BDRA), we draw on principles from enterprise architecture frameworks, notably TOGAF (The Open Group Architecture Framework), and the Design Science Research (DSR) methodology [15]. Enterprise architecture offers a structured way to align strategy, operations, data, and technology within complex organizational systems, making it highly suitable for mapping sustainability adoption in creative enterprises, which often face fragmented workflows and governance models [16].

Informed by TOGAF, our framework adopts vertical pillars of data, process, and governance, representing the foundational subsystems that shape design implementation. These pillars reflect the structural elements that must be aligned for sustainability initiatives to succeed across firm functions [17]. The horizontal layers, strategic intent, organizational capability, and technical infrastructure capture the dynamic flow of influence from executive leadership down to operational systems. This layered approach enables both diagnostic analysis and prescriptive design interventions tailored to different firm types [18].

The use of a DSR lens further supports the development of the BDRA as a context-aware artifact designed to solve a real-world problem through iterative design and empirical testing. This justifies our hybrid method and the multi-level architecture, offering both academic rigor and practical utility [19]. The five-layer BDRA (barriers–drivers–readiness architecture) framework organizes sustainability adoption into vertical pillars of data, process, and governance, across three types of analytics: descriptive, predictive, and prescriptive [20].

2.2. Theoretical Foundations and Conceptual Framework

Sustainable design, also referred to as eco-design or green design, integrates environmental, economic, and social considerations across the product or service lifecycle [21]. While it has received substantial attention in industrial and policy contexts, its relevance in the creative economy remains under-theorized. Creative sectors, including fashion, design, media, and architecture, face distinct challenges due to decentralized workflows, aesthetic priorities, and cultural values [22,23].

To build a robust conceptual framework, this study draws on four interrelated theoretical perspectives: (1) Stakeholder theory emphasizes firms’ responsibility to address competing expectations from regulators, customers, suppliers, and society at large [10]. (2) Institutional theory highlights how coercive (legal), mimetic (peer-driven), and normative (professional) pressures shape firm-level sustainability responses [11,24]. (3) Dynamic capabilities theory explains how organizations develop, adapt, and reconfigure internal resources to align with external challenges, including sustainability transitions [12]. (4) Systems thinking is employed to analyze operational fragmentation, data silos, and misaligned incentives, barriers frequently encountered in creative workflows [25]. Together, these theories enable a structured investigation of drivers, barriers, and strategic readiness for sustainable design adoption in creative enterprises.

2.3. Drivers of Sustainable Design Adoption

Empirical studies and theoretical models identify multiple enablers of sustainable design in firms, especially those in innovation-intensive sectors. Key drivers include the following: (1) Economic incentives: reduction of operating costs, access to green markets, and improved brand equity [26,27,28]. (2) Regulatory frameworks: including carbon taxes, green procurement mandates, and extended producer responsibility [29]. (3) Consumer activism and expectations: growing demand for ethical and transparent design, particularly among younger consumers [30]. (4) Leadership and innovation capacity: firms with visionary leadership and an innovation-focused culture are more likely to embed sustainable practices [31]. These drivers can be understood through multiple lenses: stakeholder theory (responding to expectations), resource-based view (leveraging internal strengths), and dynamic capabilities (adapting to sustainability pressures). Sectoral variation remains notable, fashion and industrial design respond more to regulation and consumer expectations, while digital creative firms are more innovation- and reputation-driven [32].

2.4. Barriers to Sustainable Design Adoption

Despite the compelling rationale, many firms, especially small and medium-sized creative enterprises, struggle with barriers that limit their sustainability transition: (1) Financial limitations: high upfront investment costs and short-term business horizons inhibit long-term sustainability planning [33,34]. (2) Technical and knowledge gaps: limited awareness of life-cycle thinking, eco-materials, and design metrics [35,36]. (3) Cultural and organizational resistance: in creative industries, sustainability is often viewed as constraining artistic freedom or innovation autonomy [37,38]. (4) Market barriers: low willingness to pay for sustainable products and poor supply chain support [39,40]. (5) Institutional inconsistencies: Regulatory enforcement and stakeholder pressures differ greatly across regions and sub-sectors [41,42]. Importantly, these barriers rarely operate in isolation. Their interactions—such as limited capital intersecting with fragmented governance—can significantly undermine even motivated sustainability efforts. Prior literature rarely captures this dynamic interplay.

2.5. Review of Existing Sustainable Design Archetypes

Several structured frameworks aim to guide sustainable design integration in corporate practice: (1) SGAM (smart grid architecture model): designed for the energy sector, with high lifecycle traceability but poor creative adaptability [43]. (2) DERA (design for environment reference architecture): emphasizes environmental criteria at the design stage but lacks stakeholder mapping [44]. (3) CEEDS (circular economy ecosystem design structure): a more systems-oriented model supporting circular principles and distributed actors [45]. While each model offers partial value, they tend to fall short in addressing the unique features of creative enterprises, such as fluid, iterative workflows and stakeholder-driven aesthetic decisions.

This comparative evaluation reveals (

Table 1. Benchmarking sustainable design archetypes in creative contexts.

Model	Stakeholder Integration	Workflow Adaptability	Lifecycle Traceability	Sector-Specific Application	Creative Economy Fit
SGAM	Moderate	Low	High	Strong in energy/utilities	Low
DERA	Low	Medium	Medium	Industrial design-centric	Limited
CEEDS	High	Medium	High	Circular economy projects	Moderate

) that none of the current models sufficiently accommodates the interdisciplinary, brand-sensitive, and emotionally expressive nature of creative enterprises [46]. These limitations underscore the need for a tailored model, one that captures both strategic constraints and creative flexibility.

2.6. Literature Gap and Study Contribution

Although sustainability research has matured significantly, four critical gaps persist: (1) Prior studies rarely synthesize stakeholder, institutional, and dynamic capabilities theory in one empirical model. (2) Neglect of the creative economy: sectors with strong cultural and design influence remain understudied despite their potential sustainability impact. (3) Insufficient model benchmarking: existing frameworks are rarely critiqued systematically or tested for creative sector applicability. (4) Isolated treatment of drivers/barriers: few studies explore the interaction between enabling and constraining factors at the firm level. This study addresses these gaps by proposing the barriers–drivers–readiness architecture (BDRA), a conceptual and empirical model designed specifically for the creative economy. The model is validated through structural equation modeling, cluster analysis, and machine learning, offering both theoretical advancement and practical tools for decision-makers.

2.7. Interdisciplinary Perspectives on Platform Resilience and Market Governance

While prior sections have addressed sustainability from a firm-level strategic viewpoint, emerging interdisciplinary research provides critical insights into the infrastructure and governance underpinnings of sustainable innovation platforms [47]. For example, Liu [48] explore how platform ecosystems, such as those supporting green innovation and open design, face collapse risks if coordination mechanisms among stakeholders (e.g., platform providers, regulators, and users) fail [49]. Their study outlines structural vulnerabilities and identifies avoidance strategies (e.g., redundancy, diversification, adaptive regulation) that directly inform our proposed BDRA model’s governance and process integration layers [50]. Similarly, the collaborative mechanism of a data trading market was also investigated using a four-party evolutionary game model [51] and also predicted by artificial intelligence [48]. Their findings emphasize the importance of trust-building, pricing strategies, and policy incentives in enabling efficient and transparent data exchange, a critical enabler for sustainability reporting and lifecycle analysis in creative sectors [52]. These insights reinforce the BDRA’s focus on stakeholder-driven governance structures and layered architectural transparency.

Furthermore, developments in green finance and environmental impact auditing underscore the growing relevance of aligning data markets with ESG (environmental, social, governance) frameworks [53,54]. Integrating analytics across predictive, descriptive, and prescriptive layers requires not only technical interoperability but also institutional trust and regulatory clarity. Our model draws inspiration from these frameworks to bridge the gap between firm-level sustainability adoption and ecosystem-wide accountability [55,56].

These interdisciplinary perspectives validate the BDRA’s multi-layer architecture by addressing platform resilience, data governance, and institutional integration, core areas that have been previously underexplored in the sustainability design literature.

3. Methodology

This study employs a comprehensive, multi-method quantitative approach to investigate the key drivers and barriers affecting the adoption of sustainable design in creative economy enterprises. The methodological framework includes survey development, data collection, exploratory and confirmatory factor analysis, structural equation modeling (SEM), cluster analysis, and machine learning.

3.1. Survey Instrument Design

A structured questionnaire was created, including 22 items on potential drivers and 18 on barriers to adopting sustainable design. All items used a 5-point Likert scale (1 = Strongly Disagree to 5 = Strongly Agree). The tool was pretested and refined through expert reviews from sustainability scholars and creative industry professionals to ensure it was valid, relevant, and clear. Small adjustments were made based on feedback to enhance wording, scale balance, and applicability to specific sectors.

3.2. Methodological Justification and Validation Procedures

The analytical methods applied in this study were selected to ensure construct validity, empirical robustness, and generalizability across a diverse sample of creative enterprises. Exploratory factor analysis (EFA) was used to identify the latent structure of drivers and barriers based on empirical data rather than assumptions. Confirmatory factor analysis (CFA) was used to validate the factor structure and assess measurement model fit, using commonly accepted goodness-of-fit indices.

Structural equation modeling (SEM) was chosen for its ability to simultaneously test multiple relationships among latent constructs, allowing us to validate the structural paths hypothesized in the BDRA conceptual model. This approach is particularly appropriate for models with mediating or moderating pathways and offers advantages over multiple regression by accounting for measurement error.

To identify strategic firm typologies, we applied K-means clustering on dimensionally reduced components, rather than rule-based classification. This allowed patterns to emerge from the data based on observed behavior rather than predefined assumptions. Finally, a Random Forest model was used to assess variable importance due to its non-parametric robustness, ability to capture nonlinear relationships, and minimal assumptions about variable distributions.

While qualitative alternatives such as the Delphi technique or stakeholder interviews are valuable for concept elicitation, our study focuses on quantitative validation of an empirically driven conceptual framework. However, to enhance construct validity, we conducted a two-phase pre-study process: A panel of six domain experts (comprising sustainability consultants, creative industry professionals, and academic researchers) reviewed the draft survey and BDRA model. Their feedback was incorporated to improve clarity, contextual relevance, and theoretical grounding. A pilot study with 20 respondents was conducted to test the reliability and interpretability of the survey. Minor revisions were made to the wording of items and the design of the scale based on this feedback.

These validation steps, combined with robust statistical procedures, ensure that the framework is both theoretically sound and practically applicable in creative economy settings.

3.3. Data Collection

Data for this study were collected between January and March 2025 through an online structured questionnaire hosted on Wenjuanxing, a widely used Chinese survey platform. The target population consisted of firms operating in creative economy sectors, including design, fashion, architecture, digital content, and advertising, across both urban and peri-urban regions in China.

We used a purposive sampling strategy, targeting professionals at mid- to senior-level positions within their firms. The sampling frame was developed through industry associations, creative incubator networks, and alumni directories from leading design institutions. To increase response reliability, we invited only those respondents who confirmed in the pre-screening question that their organization had been involved in at least one design project within the past year and had considered or implemented any sustainability initiative.

Invitations were distributed via email, professional WeChat groups, and through direct messages on platforms like LinkedIn and Zhihu. Each invitation included the study’s objective, a consent statement, and a digital assurance of anonymity. The survey required approximately 12–15 min to complete.

Out of 412 total responses received, we screened the data for completeness, response time (excluding extremely short or long completions), and straight-lining behavior. After these filters, 320 high-quality responses were retained for final analysis—yielding a usable response rate of 77.6%. The final dataset included responses from firms across multiple Chinese provinces, representing a mix of small (31%), medium (43%), and large (26%) enterprises.

3.4. Exploratory Factor Analysis (EFA)

EFA was conducted as a preliminary step to identify latent dimensions underlying the drivers and barriers of sustainable design adoption. Using principal component analysis (PCA) with Varimax rotation, factors were extracted based on the eigenvalue-greater-than-one rule. Items with factor loadings above 0.50 were retained, ensuring substantive contribution to the latent construct. Prior to extraction, Kaiser–Meyer–Olkin (KMO) tests (>0.80) and Bartlett’s Test of Sphericity (p < 0.001) confirmed data adequacy and factorability. The factor model was expressed as Equation (1):

$$X=L\times F+ ϵ$$

(1)

where X is the matrix of observed variables, L is the matrix of loadings, F is the matrix of common factors, and $ϵ$ is the matrix of unique errors.

The resulting factor structures provided conceptually coherent groupings of drivers and barriers, forming the foundation for subsequent confirmatory and structural analyses.

3.5. Confirmatory Factor Analysis (CFA)

CFA was employed to confirm the factor structures obtained from EFA and to evaluate the measurement model’s validity. The max likelihood estimation (MLE) approach was used, with several fit indices—such as the comparative fit index (CFI), Tucker–Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR), to assess the model’s performance. Convergent validity was evaluated using average variance extracted (AVE, Equation (2)) and composite reliability (CR):

$$AVE= {\displaystyle \frac{\sum {\lambda }_i^2}{\sum {\lambda }_i^2-\sum {\theta }_i}}$$

(2)

where λ are standardized loadings and θ is the error variance. AVE values above 0.50 and CR values above 0.70 were considered indicative of good convergent validity. Discriminant validity was confirmed when the square root of AVE exceeded the inter-construct correlations.

3.6. Structural Equation Modeling (SEM)

SEM was employed to test hypothesized relationships among latent constructs derived from the validated CFA model. The SEM framework combines measurement and structural models, enabling simultaneous estimation of direct and indirect effects. The structural model follows the standard SEM formulation as Equation (3):

$$\eta = \mathsf{{\rm B}}\eta + \mathsf{\Gamma }\mathsf{\xi }+\mathsf{\varsigma }$$

(3)

where η = endogenous latent variables, ξ = exogenous latent variables, B = coefficients among endogenous variables, Γ = coefficients from exogenous to endogenous variables, and ζ = error terms.

The model was assessed using the same fit indices as CFA. Standardized path coefficients and critical ratios (CRs) were reported to evaluate the strength and significance of hypothesized paths. The SEM results provided a nuanced understanding of how specific drivers positively influence sustainable design adoption and how barriers might hinder it.

3.7. Cluster Analysis

To segment firms based on their profiles of drivers and barriers, K-means clustering was employed. Prior to clustering, dimensionality was reduced using PCA to enhance interpretability and eliminate multicollinearity. The algorithm minimized within-cluster variance and maximized inter-cluster differentiation.

The clustering objective was expressed as Equation (4):

$$argmin \sum _{i=1}^k\sum _{xϵS_i}{‖x-{\mu }_i‖}^2$$

(4)

where μ is the centroid of each cluster S. The number of clusters was selected using the Elbow method.

The ideal number of clusters was identified through the elbow method, which balances simplicity with explanatory ability. Three key firm types emerged: Innovation-Driven, Strategically Aligned, and Barrier-Dominated. These groups highlight notable behavioral and strategic variations among firms and provide a useful taxonomy for policy and managerial approaches targeting.

3.8. Random Forest Modeling

To assess predictive determinants of sustainability adoption, a Random Forest classifier was employed. This ensemble learning method builds multiple decision trees using bootstrapped samples and aggregates their predictions, providing robust, non-parametric modeling that is resistant to overfitting.

The importance of each variable was evaluated using Mean Decrease in Gini Impurity as expressed in Equation (5):

$$Gini =1 - {\displaystyle {\sum }_{i=1}^n}{p}_i^2$$

(5)

where p is the probability of a variable belonging to a class. Ten-fold cross-validation ensured generalizability.

To ensure generalizability, 10-fold cross-validation was used. The Random Forest model confirmed and extended the findings from SEM, identifying top-ranked predictors of sustainable design adoption.

4. Results

4.1. Drivers of Sustainable Design Adoption

Figure 1 presents the mean importance scores of 22 individual drivers that influence the adoption of sustainable design practices among creative economy enterprises. The figure displays these drivers in descending order based on their mean Likert-scale ratings, using a gradient color scheme from dark blue (highest importance) to light orange (lowest importance) to enhance interpretability.

The most highly rated drivers, Driver_5, Driver_13, and Driver_22, highlight key enablers of sustainability adoption. These likely represent strategic-level and innovation-oriented constructs such as senior leadership commitment, access to knowledge, and technological infrastructure. Their consistently high scores reflect a strong consensus among respondents regarding the importance of these factors.

In contrast, Drivers 3, 12, and 7 were rated lowest, indicating areas perceived to have less immediate influence on sustainability outcomes. These may reflect external pressures or less developed internal mechanisms.

The tight clustering of mid-ranked drivers suggests a convergence in perceptions about a wide set of factors contributing to sustainability, albeit with slightly varying levels of urgency or control. The figure helps delineate a prioritization roadmap for both scholars and practitioners: reinforcing high-impact drivers while reassessing overlooked ones for potential development.

Overall, the visual and numerical trends in Figure 1 highlight that while all 22 drivers contribute to shaping sustainability strategy, a few strategic and organizational enablers stand out as particularly critical within this sector. The horizontal bar chart uses a color gradient to differentiate higher- and lower-ranked factors. Drivers 5, 13, and 22 rank highest in importance, indicating they are key enablers of sustainability-related transformation. The visual trajectory of the graph, annotated with arrows, underscores the decreasing importance of other drivers, offering a practical roadmap for prioritization.

Consumer Engagement refers to direct responses to market and stakeholder expectations, while Strategic Integration entails embedding sustainability into wider business functions. Both constructs showed good reliability, with Cronbach’s α values exceeding 0.80. Key high-loading items and internal consistency measures are shown in

Table 2. Factor loadings for drivers (EFA).

Description	Component 1	Component 2	Component 3	Component 4	Component 5
Driver_1	−0.213	−0.190	0.161	−0.208	−0.009
Driver_2	−0.003	−0.134	−0.297	0.071	−0.032
Driver_3	−0.299	−0.067	−0.089	−0.081	0.089
Driver_4	0.217	−0.048	−0.303	0.062	−0.221
Driver_5	−0.154	0.354	−0.038	−0.242	−0.289
Driver_6	−0.268	−0.135	0.216	−0.001	−0.176
Driver_7	0.017	−0.152	−0.294	−0.368	0.000
Driver_8	−0.051	0.304	0.203	0.128	0.209
Driver_9	−0.128	0.092	−0.421	0.317	−0.096
Driver_10	0.245	0.408	0.088	0.223	−0.045
Driver_11	0.298	−0.093	0.166	−0.086	−0.262
Driver_12	−0.421	0.097	−0.014	0.007	0.085
Driver_13	−0.117	0.108	−0.405	0.293	0.050
Driver_14	−0.088	0.231	0.053	−0.426	0.439
Driver_15	−0.118	0.312	0.032	−0.157	−0.415
Driver_16	0.224	0.137	−0.305	−0.170	0.217
Driver_17	−0.002	−0.034	−0.202	−0.384	−0.400
Driver_18	−0.269	0.246	0.075	0.069	−0.256
Driver_19	−0.033	−0.301	0.241	0.257	−0.223
Driver_20	−0.301	−0.173	−0.174	−0.044	0.017
Driver_21	0.357	0.032	0.040	−0.174	−0.057
Driver_22	−0.005	0.350	0.014	0.039	−0.061

. Figure 1 displays the average importance scores for each driver, highlighting “Leadership Commitment” and “Regulatory Pressure” as the most significant, with mean scores of 4.32 and 4.27, respectively.

Regression Analysis: A regression model using the composite driver index as a predictor revealed a significant positive relationship with the Sustainability. Adoption_Score (β = 1.016, p < 0.001, R² = 0.813). The full results are presented in Table 2.

4.2. Barriers to Sustainable Design Adoption

Figure 2 presents the mean perceived severity scores of 18 barrier items to sustainable design adoption, ranked from highest to lowest severity based on Likert-scale ratings. The color intensity of each horizontal bar corresponds to the magnitude of perceived barrier severity, ranging from deep red for highly severe barriers to lighter hues for lower severity.

The most significant barriers, Barrier 6, Barrier 1, and Barrier 5, highlight internal challenges like insufficient dedicated resources, limited sustainability vision, or resistance from middle management. These issues scored the highest in severity, indicating that internal alignment and resource availability are key obstacles for many firms trying to adopt sustainable design principles. On the other end of the spectrum, Barrier_8, Barrier 15, and Barrier 17 were perceived as least severe. These likely correspond to external uncertainties or market-related challenges that, while still relevant, are less impactful relative to internal issues. The clear gradient of severity in Figure 2 underscores the hierarchy of resistance elements firms encounter. By providing a prioritized overview, the figure enables practitioners and policymakers to identify focal points for intervention, particularly the organizational levers that, if addressed, could significantly reduce adoption friction.

Overall, the outcome of this figure suggests that improving internal capabilities, strengthening leadership commitment, and ensuring resource allocation are essential to overcoming key barriers to sustainable design implementation. The darkest shades highlight the most significant impediments: Barrier 6, Barrier 1, and Barrier 5, indicating a high consensus among respondents about these factors’ severity. As the bars lighten in shade, the perceived importance diminishes. This pattern reveals the relative intensity with which firms experience different types of resistance and helps to focus intervention efforts on the most pressing issues.

After mapping their sustainability processes to the BDRA model, the firm integrated digital lifecycle assessment tools into the design process and restructured governance roles to clarify sustainability accountability. As a result, the firm observed a 32% improvement in material traceability across supplier tiers, and sustainability-related compliance queries dropped by 18% over two quarters.

This firm used BDRA to identify weaknesses in internal data governance and stakeholder communication. Following implementation, including a shift to a centralized knowledge platform, the studio achieved a 25% reduction in project delays related to unclear sustainability criteria, and improved its internal stakeholder feedback score from 3.1 to 4.2 on a 5-point scale. With limited funding and low technical maturity, this firm applied a simplified version of BDRA to redesign its procurement workflow and adopt basic environmental reporting templates. Post-intervention, the team recorded a 15% improvement in sustainability self-assessment scores and noted a 10% increase in partner NGO satisfaction with sustainability alignment.

4.3. Confirmatory Factor Analysis and Structural Equation Modeling

Figure 3 illustrates the confirmatory factor analysis (CFA) model, which was used to assess the measurement validity of latent constructs underlying sustainable design adoption. These constructs include Leadership and Vision, Regulatory and Market Pressure, Innovation Capacity, Strategic Integration, and Consumer Engagement. Each latent variable is represented by a group of observed variables (i.e., driver items), with standardized factor loadings shown along the arrows connecting indicators to latent factors.

The figure demonstrates that most indicators have strong and statistically significant loadings (typically above 0.60), confirming their adequacy in representing the underlying constructs. The clean separation between latent constructs suggests good discriminant validity, while the high loadings indicate satisfactory convergent validity.

Moreover, the CFA model presents a well-structured and theoretically coherent representation of the data. It supports the hypothesized measurement model and confirms the multidimensionality of sustainability-related drivers. The graphical layout, comprising oval-shaped latent variables and rectangular observed variables, with directional arrows, offers a clear visual guide for interpreting relationships and construct integrity.

Overall, Figure 3 validates the structure of the conceptual model by demonstrating that the measurement indicators effectively capture distinct dimensions of strategic, operational, and market-driven enablers of the sustainable design adoption model for latent constructs, including Leadership and Vision, Regulatory and Market Pressure, Innovation Capacity, Strategic Integration, and Consumer Engagement. Each construct is validated through strong and consistent factor loadings from its respective observed variables. The structural integrity of this measurement model confirms that the theoretical categories are empirically sound and appropriately distinguishable from one another.

Figure 4 illustrates the structural equation model (SEM), which provides an integrated view of the relationships among the latent constructs influencing sustainable design adoption. These constructs—Leadership and Vision, Innovation Capacity, Consumer Engagement, Knowledge and Capability Gaps, and Financial Constraints—are linked by directional paths that represent hypothesized cause-and-effect relationships, quantified by standardized beta coefficients.

The SEM analysis confirms statistically significant and positive direct effects from Leadership and Vision (β = 0.38, p < 0.001), Innovation Capacity (β = 0.33, p = 0.001), and Consumer Engagement (β = 0.29, p = 0.005) on Sustainability Adoption. These findings reinforce the conceptual premise that visionary leadership, innovation readiness, and market alignment are key enablers of sustainable design practices.

On the contrary, the model also illustrates negative but statistically non-significant effects for Knowledge and Capability Gaps (β = −0.11, p = 0.087) and Financial Constraints (β = −0.08, p = 0.112). These paths indicate potential barriers that may hinder adoption, although their limited significance suggests either a less prominent role or substantial variation among firms.

Control variables, including Firm Size and Sector, showed minimal and statistically insignificant effects on the outcome variable. This suggests that structural characteristics of the firm may not independently influence adoption decisions when strategic and operational competencies are accounted for.

Visually, the figure enhances interpretation by displaying strong paths with bold lines and denoting non-significant relationships with dashed or thinner arrows. The overall structure is coherent, and the model paths align with the study’s theoretical expectations.

The SEM output contributes to the empirical understanding of how internal strategic capabilities and external engagement mechanisms jointly influence sustainable design behavior, emphasizing the multifactorial and interdependent nature of organizational sustainability adoption. Statistically significant positive effects are observed for the following: Leadership and Vision (β = 0.38, p < 0.001), Innovation Capacity (β = 0.33, p = 0.001), and Consumer Engagement (β = 0.29, p = 0.005).

These paths affirm the importance of internal competencies and external responsiveness. Negative coefficients for Financial Constraints (−0.08, p = 0.112) and Knowledge and Capability Gaps (−0.11, p = 0.087) suggest modestly detrimental effects, though statistically insignificant. Control variables: Firm Size and Sector show minimal influence.

Table 3. CFA fit indices and validity statistics.

Fit Index	Value	Threshold for Good Fit
CFI	0.95	>0.90
TLI	0.93	>0.90
RMSEA	0.04	<0.06
SRMR	0.06	<0.08
GFI	0.91	>0.90
χ2/df	1.92	<3.00

provides the model fit indices and validity metrics from the confirmatory factor analysis (CFA) for the driver constructs. All reported fit statistics indicate excellent model performance. The comparative fit index (CFI = 0.95) and Tucker–Lewis index (TLI = 0.93) both exceed the commonly accepted threshold of 0.90, confirming the model’s comparative fit quality. The root mean square error of approximation (RMSEA = 0.04) and standardized root mean square residual (SRMR = 0.06) fall well below the 0.08 benchmark, signifying minimal residual variance and strong overall data fit.

Additionally, the goodness-of-fit index (GFI = 0.91) and a favorable chi-square to degrees-of-freedom ratio (χ²/df = 1.92) further reinforce the robustness of the measurement model. Convergent validity was confirmed via average variance extracted (AVE) values exceeding 0.50 across constructs, while discriminant validity was ensured through clear separation between AVE square roots and inter-construct correlations.

Together, these statistics validate the structural integrity and empirical soundness of the latent variables, ensuring confidence in their application for further SEM modeling. The comparative fit index (CFI = 0.95), Tucker–Lewis index (TLI = 0.93), RMSEA (0.04), SRMR (0.06), and GFI (0.91) all exceed recommended thresholds. The chi-square/degrees of freedom ratio (χ²/df = 1.92) further confirms good model fit.

Table 4. SEM path coefficients.

Path	Standardized Coefficient (β)	p-Value
Leadership and Vision → Sustainability Adoption	0.38	0
Innovation Capacity → Sustainability Adoption	0.33	0.001
Consumer Engagement → Sustainability Adoption	0.29	0.005
Financial Constraints → Sustainability Adoption	−0.08	0.112
Knowledge and Capability Gaps → Sustainability Adoption	−0.11	0.087
Firm Size → Sustainability Adoption	0.06	0.231
Sector → Sustainability Adoption	0.09	0.147

presents the detailed results of the structural equation modeling (SEM) path analysis, capturing the relationships among latent constructs and their effect on sustainability adoption. The table includes standardized path coefficients (β), corresponding p-values, and confidence intervals. The strongest and most statistically significant paths were as follows: Leadership and Vision → Sustainability Adoption (β = 0.38, p < 0.001), Innovation Capacity → Sustainability Adoption (β = 0.33, p = 0.001), and Consumer Engagement → Sustainability Adoption (β = 0.29, p = 0.005). These results confirm that strategic direction, innovation capabilities, and customer-driven alignment are the dominant enablers of sustainable design practices in creative economy firms.

Conversely, the paths from Knowledge and Capability Gaps (β = −0.11, p = 0.087) and Financial Constraints (β = −0.08, p = 0.112) to sustainability adoption were negative but not statistically significant. These values suggest that although these barriers exist, their overall explanatory power in the model is weaker compared to the enablers. Control variables such as Firm Size and Sector showed minimal influence and did not reach statistical significance, indicating that organizational demographics had a limited effect on sustainability outcomes once strategic factors were accounted for. Overall, the SEM results in

Table 5. Stakeholder mapping of system architecture.

Model Component	Stakeholder Group	Responsibilities	Expected Benefits
Leadership and Vision	Corporate Executives (C-Suite)	Set strategic vision, allocate resources, sponsor innovation	Strategic alignment, long-term competitiveness
Regulatory and Market Pressure	Regulatory Bodies, Industry Associations	Impose compliance norms, offer incentives for sustainable practices	Regulatory alignment, improved policy responsiveness
Innovation Capacity	R&D Departments, Design Leads	Develop sustainable products, test prototypes, lead eco-innovation	Enhanced capability, IP generation, design differentiation
Consumer Engagement	Marketing, UX Teams, Brand Managers	Communicate sustainability, engage feedback loops with customers	Brand loyalty, improved customer retention
Strategic Integration	Project Managers, System Integrators	Coordinate cross-functional initiatives, harmonize workflows	Process efficiency, reduction of lifecycle bottlenecks
Sustainability Adoption	Whole Enterprise	Collective implementation of sustainable design policies	Reputation enhancement, long-term viability

reinforce the theoretical model by confirming which constructs are empirically strong drivers or weak inhibitors of sustainable design adoption. These findings have practical implications for organizational strategy, resource allocation, and policy interventions.

Interestingly, the SEM revealed that traditional barriers, such as financial constraints and knowledge gaps, did not demonstrate statistically significant direct effects on sustainable design adoption. This may suggest that their influence operates through indirect mechanisms, particularly via leadership vision, innovation capacity, or stakeholder responsiveness. Prior studies have similarly found that barriers affect internal alignment and decision-making culture, rather than exerting direct structural suppression. Future research could test moderated mediation models to assess whether leadership or firm size moderates the effect of such constraints on adoption readiness.

To enhance the operational and managerial interpretability of Figure 3 and Figure 4, we developed a stakeholder-aligned interpretation of each architecture component. Table 5 illustrates the stakeholder responsibilities and benefits for each layer:

This stakeholder-to-layer matrix ensures that the BDRA conceptual model remains both academically sound and practical, serving as a framework to guide stakeholder actions. Additionally, mapping the SEM results (Figure 4) to stakeholder domains allows clearer translation of statistical relationships into functional responsibilities.

4.4. Exploratory Factor Analyses for Drivers and Barriers

Table 1 summarizes the results of an exploratory factor analysis (EFA) applied to the 22 driver items associated with sustainable design adoption. The purpose of the EFA was to uncover the underlying factor structure and group related drivers into latent constructs. Using principal component analysis with varimax rotation, five distinct components were extracted.

Component 1 featured negative loadings from Driver_3, Driver_6, Driver_18, and Driver_20. These loadings may represent underlying strategic or structural inhibitors, where negative perceptions of internal capabilities or misaligned objectives reduce the likelihood of sustainability adoption. Component 2 exhibited strong positive loadings for Drivers 5, 10, 15, and 22, highlighting aspects of organizational readiness and leadership, such as senior management support, fostering innovation, and having strategic planning mechanisms in place. Component 3 was characterized by contributions from Drivers 1, 6, and 11, emphasizing innovation capacity and adaptability to market trends through new processes or product designs. Component 4 exhibited strong loadings from Drivers 8, 14, and 19, likely reflecting external stakeholder pressures and regulatory influences that drive sustainable practices. Finally, Component 5 had notable loadings for Drivers 14, 15, and 16, indicating a construct related to collaborative knowledge exchange and capacity building within and between organizations.

The cumulative variance explained by these five components was adequate, validating the multidimensionality of the driver items. These findings informed the design of the subsequent confirmatory factor analysis (CFA) and structural equation modeling (SEM), ensuring conceptual consistency in the modeling framework. Five components were extracted: Component 1 clusters Driver 3, Driver 6, Driver 18, and Driver 20 with negative loadings, possibly reflecting structural hurdles. Component 2 highlights Driver 5, Driver 10, Driver 15, and Driver 22 as leadership and organizational readiness constructs. Component 3 captures innovation-oriented dimensions. Components 4 and 5 include factors related to external pressures and collaborative capacity-building.

Table 6. Factor loadings for barriers (EFA).

Description	Component 1	Component 2	Component 3	Component 4	Component 5
Barrier_1	0.300	0.057	0.070	0.362	0.024
Barrier_2	−0.303	0.109	−0.237	−0.050	−0.271
Barrier_3	0.466	0.165	−0.209	0.037	0.144
Barrier_4	−0.100	−0.158	0.489	0.145	−0.195
Barrier_5	0.390	0.170	0.210	−0.223	0.042
Barrier_6	−0.365	0.184	0.175	−0.107	0.144
Barrier_7	0.113	−0.258	−0.311	0.338	−0.227
Barrier_8	−0.259	−0.064	0.217	0.414	0.119
Barrier_9	0.104	−0.124	0.006	−0.057	−0.152
Barrier_10	0.113	−0.003	0.336	−0.374	−0.302
Barrier_11	0.095	0.057	−0.117	−0.043	−0.618
Barrier_12	−0.098	−0.298	−0.067	−0.234	−0.184
Barrier_13	0.155	0.465	−0.155	−0.086	−0.154
Barrier_14	0.006	0.107	0.286	0.228	−0.430
Barrier_15	0.249	−0.507	−0.081	0.161	−0.009
Barrier_16	−0.255	0.288	−0.326	0.173	−0.071
Barrier_17	0.028	−0.218	−0.002	−0.340	0.174
Barrier_18	0.176	0.277	0.295	0.257	0.076

presents the exploratory factor analysis (EFA) results for the 18 identified barriers to sustainable design adoption. Using principal component analysis with varimax rotation, five latent components were extracted, reflecting the underlying dimensions of organizational resistance. Component 1 includes Barrier 1, Barrier 3, and Barrier 5, which primarily relate to internal organizational inefficiencies, such as misaligned processes, a lack of internal coordination, and operational rigidity. These form the core internal structural challenges. Component 2, comprising Barrier 13 and Barrier 18, highlights external policy and regulatory ambiguity. Firms indicated uncertainty about sustainability-related compliance requirements and the volatility of governmental directives. Component 3 includes Barrier 4 and Barrier 8, highlighting limitations in technical know-how and capability gaps among staff, suggesting that workforce development is a key obstacle. Component 4 reflects infrastructure and financial constraints, grouping variables related to resource scarcity and underinvestment in sustainability infrastructure. Component 5 includes cultural or change-resistance factors such as Barrier_10 and Barrier_11, underscoring internal resistance to innovation and employee inertia.

The clear differentiation across these components confirms the multidimensional nature of barriers and validates the segmentation used in the confirmatory factor analysis (CFA) and SEM models. These insights provide practical direction for targeting interventions based on the specific nature of resistance experienced by firms, identifying five distinct components: Component 1: includes Barrier_1, Barrier_3, and Barrier_5, suggesting internal organizational challenges. Component 2: Barrier_13 and Barrier_18 emphasize policy and regulatory ambiguities. Component 3: capability limitations such as Barrier_4 and Barrier_8. Component 4: resource and infrastructure shortages. Component 5: cultural or resistance-to-change factors like Barrier_10 and Barrier_11. These loadings validate the multidimensional structure of the barriers and inform the confirmatory models. Table 6 presents the detailed loadings for each of the 18 barrier items derived from the exploratory factor analysis (EFA). Five distinct components were extracted, each representing a specific category of resistance to sustainable design adoption: Component 1 consolidates internal structural inefficiencies, capturing barriers such as lack of coordination and procedural bottlenecks. Component 2 addresses regulatory uncertainty and policy ambiguity, highlighting the external institutional and policy-driven challenges. Component 3 identifies capability and skill gaps within the workforce, limiting innovation and technical implementation. Component 4 pertains to infrastructure and resource constraints, most relevant for small or underfunded firms. Component 5 captures resistance to change, specifically relating to cultural or psychological obstacles faced by employees or leadership.

The loadings confirm the conceptual distinction among the barrier domains and provide a strong basis for subsequent CFA and SEM modeling. This structure enables more precise targeting of strategic interventions based on the type of resistance encountered.

4.5. Cluster Analysis and Strategic Typologies

Figure 5 illustrates the results of a principal component analysis (PCA) followed by k-means clustering, used to identify strategic typologies among firms based on their scores for sustainable design drivers and barriers. The figure plots firms in the space of the first two principal components: PC1 and PC2, which together explain 10.04% of the total variance (5.23% and 4.81%, respectively). Although the explained variance is modest, the PCA plot still effectively distinguishes between clusters, visually aided by color-coded ellipses surrounding each group.

Three clusters emerged from the analysis: Cluster 1: Barrier-Dominated firms: These firms exhibit high standardized scores on multiple barriers and low scores on enabling drivers. They tend to perceive sustainability initiatives as costly, difficult, or incompatible with their current capabilities. Cluster 2: Innovation-Driven firms: This group shows the highest alignment with key drivers (such as strategic leadership, innovation support, and market responsiveness) and reports the lowest levels of perceived barriers. Cluster 3: Strategically-Aligned firms: Firms in this cluster present a balanced profile with moderate driver and barrier scores, suggesting they are aware of challenges but equipped with partial strategic resources to address them.

The PCA plot in Figure 5 not only confirms the structural distinctions among firms but also provides actionable insights into how different organizational configurations relate to sustainability readiness. This segmentation enables researchers and practitioners to target interventions more precisely, by addressing high-barrier concerns in Cluster 1, reinforcing strengths in Cluster 2, and guiding Cluster 3 toward strategic optimization, visually categorizing firms into three typologies: Cluster 1: Barrier-Dominated firms: characterized by higher scores on barrier items and lower scores on drivers. Cluster 2: Innovation-Driven firms: featuring high driver scores and low barrier scores. Cluster 3: Strategically-Aligned firms: moderate scores on both dimensions.

The explained variance on the first two principal components is low (5.23% for PC1 and 4.81% for PC2), but the visualization still captures distinct groupings in sustainability orientation. The use of colored ellipses enhances visual separation among clusters.

Table 6 provides centroid values for each cluster. Barrier-Dominated firms are most affected by internal and external obstacles (e.g., Barrier 7 = 0.473), while Innovation-Driven firms report high strategic capabilities (e.g., Driver 7 = 0.447) and low barrier perceptions. Strategically aligned firms occupy a middle ground, balancing pressures and capacities.

4.6. Correlation and Variable Importance

Figure 6 presents a correlation heatmap that visualizes the pairwise relationships among the 40 variables in the dataset, comprising 22 drivers and 18 barriers related to sustainable design adoption. The heatmap uses a diverging color gradient ranging from deep blue (strong negative correlation) to deep red (strong positive correlation), allowing for rapid identification of high or low associations between items.

From the heatmap, most correlations fall within a moderate range, with relatively few variables exhibiting multicollinearity (i.e., correlations exceeding ±0.80). This distribution validates the independence of most items and strengthens the robustness of subsequent multivariate analyses such as regression and structural equation modeling.

Several meaningful patterns are observable. For example, certain clusters of driver variables, particularly those related to strategic orientation and innovation readiness, tend to be positively correlated. Similarly, some barriers associated with resource limitations and internal resistance exhibit modest positive relationships, reflecting their tendency to co-occur within firm contexts.

Notably, the correlation between driver and barrier variables tends to be either weakly negative or non-significant, underscoring that perceptions of enablers and inhibitors may operate through distinct organizational logics. This finding supports the study’s decision to model these constructs separately in both exploratory and confirmatory analyses.

Overall, the heatmap provides a valuable diagnostic tool for identifying latent structures, validating variable independence, and detecting areas of redundancy or synergy among survey dimensions (22 drivers and 18 barriers). Color gradients range from blue (negative correlation) to red (positive correlation). The absence of high correlation among variables reduces concerns over multicollinearity and supports robust model estimations.

Figure 7 displays the variable importance results derived from a Random Forest model predicting Sustainability Adoption. Each bar represents the relative importance score of a driver or barrier variable, determined by the model’s accuracy loss when that variable is excluded. The features are arranged in descending order of importance, and the color intensity reflects their predictive power.

The most influential predictors include Driver 5, Driver 13, and Driver 22, which are consistent with earlier findings in both descriptive and structural modeling. These variables likely correspond to constructs related to visionary leadership, organizational adaptability, and technological support, core elements that empower sustainability-oriented change. The steep decline in importance across subsequent variables illustrates a heavy-tailed distribution, where a few key features dominate predictive capability.

Interestingly, some lower-ranked drivers and barriers exhibit minimal contribution, suggesting either redundancy with more dominant variables or lack of relevance in the model’s decision pathways. This insight enables dimensionality reduction and informs future survey refinement.

Overall, Figure 7 supports the multivariate results from SEM and EFA and introduces a predictive analytics angle. It highlights the hierarchical structure of sustainability determinants, providing a solid basis for prioritizing features in decision-support tools and strategic frameworks. The leading predictors of Sustainability Adoption are Driver 5, Driver 13, and Driver 22. The graph features annotation arrows showing a sharp decrease in importance among the features. This analysis verifies the relative importance of each driver in shaping sustainability outcomes.

5. Discussion

This study examined the drivers and barriers influencing sustainable design adoption in creative economy enterprises through an integrated empirical framework rooted in corporate strategy [57,58]. By employing a robust combination of descriptive statistics, exploratory and confirmatory factor analyses (EFA and CFA), structural equation modeling (SEM), cluster analysis, and machine learning techniques, the findings offer nuanced insights into the factors shaping organizational behavior in sustainability transitions [59,60].

The results on key drivers (Figure 1) emphasize the centrality of strategic leadership, innovation capacity, and access to resources in fostering sustainable design practices. The consistent prominence of Driver_5, Driver_13, and Driver_22 across EFA, SEM, and Random Forest models signals their foundational role. These findings corroborate earlier research emphasizing that top-level vision, adaptive capacity, and organizational learning mechanisms are essential for embedding sustainability into creative workflows [61,62]. SEM analysis (Figure 4; Table 4) confirmed the strong positive effects of Leadership and Vision, Innovation Capacity, and Consumer Engagement, highlighting the importance of alignment between strategy, capability, and market responsiveness [63].

In contrast, the most prominent barriers (Figure 2), including lack of resources, operational inefficiencies, and weak sustainability leadership, echo prior findings on implementation hurdles [64]. The barrier structure revealed through EFA (Table 2; Table 6) surfaced five critical resistance domains: internal inefficiencies, regulatory ambiguity, skills deficits, resource constraints, and cultural resistance [65]. While SEM findings did not show significant direct effects of these barriers on sustainability adoption, their indirect or interaction effects may still play a critical role in moderating organizational responsiveness [66]. The CFA results (Figure 3) affirmed the discriminant and convergent validity of the measurement model, supporting the use of latent constructs in higher-order modeling. High fit indices confirm the reliability of the theoretical structure and justify its deployment in SEM [67].

Cluster analysis (Figure 5;

Table 7. Cluster centroids and description of strategic typologies.

Cluster Label	Barrier-Dominated Firms	Strategically-Aligned Firms	Innovation-Driven Firms
Driver_1	0.063	−0.183	0.144
Driver_2	0.244	−0.290	0.117
Driver_3	0.175	−0.435	0.321
Driver_4	−0.424	0.243	0.070
Driver_5	0.216	−0.333	0.184
Driver_6	0.191	−0.467	0.342
Driver_7	−0.279	−0.223	0.447
Driver_8	0.146	−0.122	0.017
Driver_9	0.126	0.123	−0.225
Driver_10	0.080	0.199	−0.270
Driver_11	−0.569	0.182	0.245
Driver_12	0.449	−0.366	0.039
Driver_13	0.804	−0.120	−0.489
Driver_14	0.244	−0.168	−0.010
Driver_15	0.260	−0.340	0.157
Driver_16	0.027	0.202	−0.232
Driver_17	0.116	−0.164	0.082
Driver_18	0.442	−0.199	−0.130
Driver_19	−0.056	−0.156	0.206
Driver_20	0.430	−0.025	−0.304
Driver_21	−0.122	0.208	−0.125
Driver_22	0.347	−0.158	−0.101
Barrier_1	0.349	−0.419	0.172
Barrier_2	0.172	0.055	−0.189
Barrier_3	−0.004	−0.257	0.272
Barrier_4	0.064	0.123	−0.178
Barrier_5	−0.145	0.171	−0.069
Barrier_6	0.027	0.248	−0.280
Barrier_7	0.473	−0.611	0.277
Barrier_8	0.197	0.107	−0.263
Barrier_9	0.453	−0.145	−0.195
Barrier_10	−0.097	0.008	0.066
Barrier_11	0.047	−0.257	0.233
Barrier_12	−0.190	−0.040	0.187
Barrier_13	0.178	−0.173	0.044
Barrier_14	0.168	0.011	−0.140
Barrier_15	−0.211	−0.300	0.476
Barrier_16	−0.131	−0.032	0.134
Barrier_17	−0.276	0.039	0.171
Barrier_18	0.124	0.004	−0.099

) further illuminated the diversity of firm behaviors in response to sustainability imperatives. Three typologies emerged: Innovation-Driven firms, which excel across all driver dimensions; Strategically-Aligned firms, which exhibit moderate but balanced performance; and Barrier-Dominated firms, which struggle with both internal and external challenges. These strategic profiles provide a proper segmentation for tailoring policy incentives and managerial interventions [68,69].

Correlation heatmap analysis (Figure 6) confirmed that drivers and barriers primarily operate in separate empirical domains, with limited risk of multicollinearity. This reinforces the decision to treat them as distinct constructs [70]. Meanwhile, variable importance scores from the Random Forest model (Figure 7) validated earlier findings by showing that only a few drivers, particularly those tied to leadership and innovation, carry substantial predictive weight [71].

These findings significantly contribute to the literature by offering an empirically grounded taxonomy of sustainability determinants in creative sectors [72]. They extend corporate strategy theory by unpacking how organizations integrate (or resist) sustainability goals through the interplay of leadership, capabilities, and contextual constraints [73].

From a practical perspective, this study provides a diagnostic framework for guiding firms and policymakers in sustainability adoption. Prioritizing leadership development, fostering innovation infrastructure, and investing in external stakeholder engagement emerge as key enablers [47,74]. At the same time, identifying and alleviating internal structural barriers will be vital for firms lagging behind [75].

While this study relies on a robust quantitative framework to analyze drivers and barriers to sustainable design, future research could benefit from complementary qualitative exploration. In-depth interviews or sector-specific case studies, particularly in fashion, media, or digital design, could uncover nuanced cultural and organizational constraints, such as creative autonomy tensions or internal misalignments that are difficult to quantify [76]. These qualitative methods would deepen understanding of how sustainability norms are interpreted and contested in day-to-day design practices, especially in sub-sectors where artistic freedom or iterative design loops complicate standardized sustainability efforts [77].

This study draws on data from creative economy firms operating within China’s unique institutional environment, characterized by centralized regulation, state-led innovation policy, and rapid digital transformation. These contextual factors may limit the direct transferability of results to countries with more decentralized or deregulated creative sectors [78]. For instance, regulatory pressures, cultural norms around sustainability, and government support mechanisms vary widely between Western, African, and South Asian economies [79]. Therefore, caution must be taken in generalizing these findings globally. Future research should undertake cross-national comparative studies or conduct replication studies in diverse institutional settings to validate and adapt the BDRA framework accordingly.

The findings of this study offer several actionable implications for managers in the creative economy. First, the strong positive influence of leadership vision and innovation capacity on sustainable design adoption highlights the need for executive commitment and a culture that supports experimentation [80]. Firms aiming to improve their sustainability practices should invest in training programs, interdisciplinary collaboration, and the integration of sustainability KPIs within design teams [81].

Second, our clustering results identified three distinct firm profiles: Innovation-Driven, Strategically-Aligned, and Barrier-Dominated. Managers can use these typologies to benchmark their own firm’s strategic posture and prioritize interventions [82]. For example, Barrier-Dominated firms may require capacity-building and external support, whereas Strategically-Aligned firms may benefit more from advanced eco-design tools and stakeholder engagement strategies [83].

Third, the lack of significant direct effects for some traditional barriers suggests that context-specific leadership and process-level integration may override generalized obstacles such as cost or knowledge [84]. Therefore, instead of focusing solely on eliminating barriers, firms should prioritize building adaptive capacity, embedding sustainability into routine design workflows, and creating feedback loops between client demand, regulatory pressure, and internal decision-making [85].

Fourth, digital tools, such as lifecycle assessment (LCA) platforms, cloud-based design collaboration systems, and environmental impact dashboards, can help overcome workflow fragmentation, improve traceability, and support transparent reporting. Adoption of these tools also facilitates compliance with global reporting frameworks like GRI or SDG-linked ESG disclosures [86].

Fifth, managers are encouraged to build ecosystem partnerships, with certifiers, universities, NGOs, and governmental sustainability hubs, to close internal capability gaps and access shared resources [87]. These collaborations can amplify knowledge transfer and allow smaller creative enterprises to benefit from collective infrastructure [88].

Finally, the results support the value of appointing internal sustainability champions and forming cross-functional teams to integrate sustainability into daily design decisions. Creative firms should also consider leveraging sustainability certification schemes (e.g., GOTS, Cradle to Cradle, FSC) to enhance brand credibility and fulfill evolving stakeholder expectations [89].

While the BDRA offers a layered conceptual framework for assessing sustainable design readiness, its current version has not been tested through simulation, performance benchmarking, or digital prototyping [90]. This limitation restricts the evaluation of the architecture’s real-time responsiveness, throughput, latency management, or governance alignment under operational stress conditions [91]. Future research could address this by implementing a digital twin or a pilot testbed within a real creative enterprise to model and simulate BDRA behavior using synthetic workloads or real-world use cases.

Such validation would enable assessment of system adaptability, technical bottlenecks, and decision-making latency, key factors that influence the adoption and integration of sustainability frameworks in practice [92]. Additionally, integration with digital design platforms (e.g., BIM, CAD/PLM systems) could allow for automatic tracking of compliance with governance protocols, further enhancing the architecture’s managerial utility and scalability [93].

Future research could adopt a longitudinal design to explore causal mechanisms or examine industry-specific variations within the creative economy. Mixed-method approaches could also yield richer insights into the social and behavioral dynamics that shape sustainable design strategies.

In summary, this research offers a validated, comprehensive, and data-driven model for understanding and advancing sustainable design adoption. It contributes to both theoretical and practical advancements in sustainability strategy within one of the most dynamic and influential sectors of the modern economy.

6. Policy Implications for Green Finance and Regulatory Governance

The findings and framework proposed in this study have substantial relevance for regulators, particularly in the context of emerging green finance ecosystems and environmental accountability mechanisms [94]. The business-driven reference architecture (BDRA) can be operationalized to help regulators harmonize sustainability-related data governance, especially by embedding interoperable APIs and distributed ledger-based registries across creative industry platforms [95].

By facilitating standardized ESG reporting pipelines, BDRA enables organizations to generate consistent and verifiable data for green bond disclosures, taxonomies, and sustainable procurement compliance. This is especially pertinent in jurisdictions like China, where initiatives such as the Green Finance Reform and Innovation Pilot Zone seek to build robust, transparent infrastructure for green capital markets [96]. BDRA’s vertical pillars data, governance, and analytics can support real-time tracking of sustainability KPIs and automated audit trails, enhancing trust and reducing compliance costs.

Furthermore, drawing on the findings, it is argued that inclusive green finance requires cross-sectoral collaboration and digital inclusivity. BDRA can function as a shared architecture that promotes platform-level transparency, stakeholder engagement, and system-level integration [97]. Similarly, the role of green bonds in promoting innovation suggests that a modular reference model like BDRA can act as a technical backbone for dynamic, evidence-based policy calibration [98]. In this way, regulators can proactively address data asymmetries, avoid market fragmentation, and enforce interoperability standards across the sustainability reporting landscape.

7. Conclusions

This study investigated the drivers and barriers affecting the adoption of sustainable design practices within creative economy enterprises. Using a multi-method approach, including factor analysis, structural equation modeling, cluster analysis, and machine learning, we found that leadership vision, innovation capacity, and stakeholder engagement are critical enablers, while internal constraints such as limited financial and technical resources act as major obstacles. The identification of three strategic firm typologies, Innovation-Driven, Strategically-Aligned, and Barrier-Dominated, provides new insights into how creative firms differ in their sustainability readiness and strategic responses.

Theoretically, the study integrates stakeholder, institutional, and dynamic capabilities perspectives to form a comprehensive framework that addresses sustainability in creative enterprises. Practically, it offers actionable recommendations for managers, including the formation of cross-functional sustainability teams, investment in digital tools, and collaboration with external partners. While the findings are based on data from Chinese firms, future research could expand the model through cross-national studies and qualitative validation to enhance generalizability.