Ambidextrous Market Orientation and Digital Business Model Innovation

Abstract

With accelerating digital transformation, firms must renew how they create, deliver, and capture value to remain competitive and to advance sustainable competitiveness. This study examines how ambidextrous market orientation drives digital business model innovation (DBMI) through the mediating role of digital resource bricolage and the moderating effect of environmental turbulence. Using survey data and structural equation modeling (SEM), we find that both proactive and responsive market orientations positively affect DBMI. Digital resource bricolage partially mediates both relationships, with a stronger mediation effect for responsive orientation. Environmental turbulence strengthens the association between ambidextrous market orientation and digital resource bricolage. Complementing variable-centric tests, fuzzy-set qualitative comparative analysis (fsQCA) identifies three configurational pathways sufficient for high DBMI, revealing alternative routes to business-model renewal under different contextual conditions. The findings extend ambidextrous market orientation research to digital contexts, enrich the resource-recombination perspective on DBMI, and provide actionable guidance for firms seeking to orchestrate data, platforms, and legacy assets to reconfigure activity systems. By clarifying when and how market sensing and shaping translate into effective digital recombination, this study informs strategies for sustainable competitiveness in turbulent environments.

1. Introduction

Digital technologies are reshaping how firms create, deliver, and capture value by introducing a new organizing logic of innovation grounded in layered, modular, and recombinable digital architectures [1,2]. The widespread application of emerging technologies such as big data, artificial intelligence, and the Internet of Things has not only driven the reconstruction of business model innovation but also brought unprecedented opportunities and challenges to firms. Digital business model innovation (DBMI) has become a primary vehicle for strategic renewal and for advancing sustainable competitiveness when markets and technologies shift rapidly [3]. DBMI denotes digitally enabled, novel reconfigurations of a firm’s activity system that change the mechanisms of value creation and appropriation [4,5]. DBMI is strategically salient, because business models are central levers of competitive advantage and can complement or substitute for product and process innovation when technologies and markets shift rapidly [4,6]. In the context of digital transformation, digital technologies comprehensively reconfigure an organization’s value creation, delivery, and capture, making DBMI a central vehicle for institutional renewal and the attainment of sustained competitive advantage [7,8,9]. Recent reviews further indicate that business-model research has become a central focus; however, the specific characteristics of digitalization necessitate theoretical frameworks that integrate both demand-side and resource-side mechanisms to explain sustainability-relevant outcomes in turbulent environments [6]. Consequently, identifying when and how firms intentionally redesign their business models through digital means remains a critical question for both scholars and managers.

Against this backdrop, the market-orientation literature has examined how market-facing capabilities stimulate DBMI and related innovation outcomes, beginning with foundational definitions of market orientation as either an organization-wide process of generating, disseminating, and responding to market intelligence or as a culture that prioritizes superior customer value creation [10,11]. A large body of evidence links market orientation to innovation and performance across industries and contexts [12]. Within market orientation, scholars distinguish between responsive market orientation (addressing expressed needs) and proactive market orientation (anticipating latent needs), both of which shape innovation, with proactive market orientation often providing stronger impetus for novelty [13]. Emerging work further explores the ambidextrous deployment of proactive and responsive market orientation logics, documenting their joint nonlinear effects on innovation outcomes [14]. Taken together, these insights suggest a plausible pathway from market sensing and shaping to DBMI, yet the underlying mechanisms remain underexplored in digital contexts.

Three gaps motivate this study. First, the applicability of traditional market-oriented theories in digital contexts has yet to be fully validated. Second, although ambidextrous market orientation—the balanced and concurrent deployment of proactive and responsive market orientations—has been shown to enhance innovation and performance, the specific relationship between ambidextrous market orientation and DBMI has received little empirical attention, as most prior work examines product or general innovation rather than DBMI [13,14]. Filling this gap advances theoretical understanding by positioning ambidextrous market orientation as a demand-side microfoundation that complements resource-side perspectives in explaining how DBMI reconfigures the mechanisms of value creation and capture at the business-model level—the central lever of competitive advantage [4,15]. In practical terms, clarifying how proactive and responsive market orientations can be orchestrated to drive DBMI offers firms actionable guidance for redesigning business models and sustaining organizational renewal in turbulent, digitally enabled environments [7,8]. Third, we lack evidence on how ambidextrous market orientation yields DBMI outcomes via resource-side mechanisms that are consistent with digital recombination and how these relationships unfold under environmental turbulence. To address these gaps, we conceptualize ambidextrous market orientation as the simultaneous and balanced pursuit of proactive and responsive market orientations and investigate their comprehensive effects on DBMI, explicitly incorporating resource-side processes and contextual turbulence into an explanatory framework [13].

Building on a resource-based view, we posit digital resource bricolage as the resource-side capability that mediates the effect of ambidextrous market orientation on DBMI, thereby integrating the demand-side market-sensing/shaping role of ambidextrous market orientation with a resource-side recombination mechanism [16]. Bricolage, i.e., “making do” by recombining resources at hand, has been shown to spur novelty under constraints; in digital contexts, it entails flexibly recombining data, analytics, platforms, and legacy assets to craft new value propositions and revenue architectures [17,18]. Aligned with contemporary views of digital innovation being generative and recombinatorial, digital resource bricolage provides a microfoundation for orchestrating digital and non-digital resources into novel activity-system designs [2]. We further argue that environmental turbulence—shifts in technologies, competitive intensity, and customer preferences—moderates the relationship between ambidextrous market orientation and digital resource bricolage [19,20]. To capture causal complexity beyond variable-centric tests, we complement structural equation modeling (SEM) and fuzzy-set qualitative comparative analysis (fsQCA) and identify multiple pathways to high DBMI that combine ambidextrous market orientation, digital resource bricolage, and environmental turbulence in different ways [21,22].

As an initial exploration into the effect of market orientation on DBMI, this study enriches the literature in four ways. First, it extends the application of ambidextrous market orientation to digital contexts by theorizing and empirically distinguishing the comprehensive and differential roles of proactive and responsive orientations in shaping DBMI. Moving beyond single-facet treatments of market orientation, this research provides new theoretical support for its role in business model innovation. Second, we open the “black box” linking market orientation to DBMI by introducing digital resource bricolage as a mediating capability. Our evidence extends the research boundaries of resource bricolage theory and enriches a resource-orchestration view of DBMI. Third, we theorize and test environmental turbulence as a boundary condition that shapes returns to ambidextrous market orientation. Environmental turbulence significantly moderates the relationship between ambidextrous market orientation and digital resource bricolage and thus conditions when demand-side sensing and shaping translate into effective digital recombination, refining contingency perspectives in the market-orientation literature. Fourth, we identify three different configurations that are sufficient for high DBMI under different contextual conditions. This neo-configurational evidence documents causal complexity and offers context-sensitive routes to business-model renewal. Through these contributions, this study aims to provide new theoretical perspectives and practical insights for both academia and industry, helping firms innovate and upgrade their business models in the digital era and ultimately achieve sustainable competitiveness.

2. Theoretical Background and Hypotheses

2.1. Ambidextrous Market Orientation and DBMI

DBMI refers to digitally enabled reconfigurations of a firm’s activity system that alter the mechanisms of value creation and capture, building on the business-model lens and the distinctive, layered, recombinable nature of digital innovation [2,6]. Prior work shows that industry-wide technological shifts and the rise of platform-based ecosystems reshape complementarities and governance, thereby opening/closing spaces for new business-model designs [23,24]. Systematic reviews of the business-model innovation literature also identify environmental dynamism, regulation, and competitive intensity as salient external antecedents that condition the choices and outcomes of business-model innovation/DBMI [15]. At the micro level, DBMI depends on firms’ dynamic capabilities—sensing, seizing, and reconfiguring—to orchestrate digital and nondigital assets into new value architectures [25]. Digital/IT capabilities and their strategic alignment with the business foster organizational agility that facilitates DBMI under uncertainty [7,26]. Managerial cognition and organizational ambidexterity further shape when and how firms shift logics and recombine resources to renew their business models in digital contexts [27,28]. Recent years of research have consolidated DBMI as a distinct construct and explicated its underlying mechanisms. Conceptual advances clarify DBMI’s attributes and its boundaries in relation to digital transformation [29]. Other contributions map the linkages among digital technologies, organizational capabilities, and ecosystems and outline priorities for future research [30,31]. Empirically, industry-level digital shifts open new design spaces for value creation and capture [32], whereas firm-level configurations of sensing and seizing capabilities produce alternative DBMI pathways, including consolidating business model innovation [33]. Micro-evidence from digital start-ups shows that DBMI components translate into superior performance through identifiable mediating processes [34]. Together, these macro- and micro-level perspectives provide complementary foundations for understanding the drivers and mechanisms of DBMI and motivate the linking in this study of market-sensing/shaping capabilities to digitally enabled business-model redesign.

Having established the theoretical background of DBMI, we now turn to market orientation, which may trigger digitally enabled business-model redesign. The theory of market orientation was first proposed in the late 1980s by Kohli and Jaworski (1990) [10], and Narver and Slater (1990) [11]. It has since been widely applied in the fields of corporate decision-making and innovation, evolving into one of the core theories in marketing research. Kohli and Jaworski defined market orientation as an organization’s ability to generate, disseminate, and respond to market intelligence, emphasizing firms’ sensitivity and responsiveness to market dynamics. Market sensing is the capability that triggers strategic change and is linked to business-model redesign, while market shaping captures deliberate efforts to alter market structures and rules [35,36]. Sensing matters when coupled with seizing in digital pathways that reconfigure business models [33]. Platform governance and network centrality enable actors to steer ecosystems, linking sensing to shaping [37]. Actionable strategies and effective configurations span offerings, pricing, exchange arrangements, representations, and norms [35,36,38]. Emerging technologies provide platforms for shaping [39]. Narver and Slater further introduced a three-dimensional model of market orientation—customer orientation, competitor orientation, and interfunctional coordination—and argued that market orientation is a key determinant of firm profitability. They stressed the importance of understanding customer needs, competitor behavior, and the market environment holistically, enabling firms to adjust strategies in a timely manner to meet market demands. As research progressed, many scholars further refined the conceptualization of market orientation.

The theory of ambidexterity was proposed by March (1991) [40], emphasizing that firms must possess the capability to both explore new resources (exploratory activities) and exploit existing ones (exploitative activities). Recent empirical work deepens evidence on balancing exploration and exploitation. Temporal analyses show that the speed of switching between exploratory and exploitative R&D shapes performance and is beneficial only under technological dynamism [41]. Meta-analytic evidence in SMEs indicates that focusing on exploration or exploitation can outperform ambidexterity, with institutional conditions moderating this effect [42]. Industry comparisons reveal contingent gains to ambidextrous versus specialized R&D [43]. Micro-foundations link CEO cognitive flexibility to ambidextrous innovation through information search [44]. Narver et al. (2004) [13] introduced the concept of ambidextrous market orientation into the marketing domain, categorizing market orientation into proactive and responsive orientations based on the explicitness of customer needs. Proactive market orientation focuses on identifying and fulfilling customers’ latent needs, while responsive market orientation aims to satisfy their expressed needs. The two approaches differ significantly in their strategic goals and implementation paths. Narver et al. (2004) [13] argued that relying solely on one type of market orientation may limit a firm’s strategic flexibility; thus, both orientations should be viewed as being equally important in practice. Building on this view, ambidextrous market orientation denotes the joint deployment of proactive and responsive orientations that govern how firms sense, interpret, and act on market intelligence at the customer–competitor interface [45]. Organizational ambidexterity, in contrast, is the firm-level capability to pursue exploratory and exploitative activities simultaneously, enabled by managerial microfoundations such as cognitive flexibility and information search [44]. Accordingly, market orientation operates as a domain-specific antecedent/mechanism that can feed ambidextrous business-model change rather than being equivalent to it [46].

Linking these insights to DBMI, prior evidence indicates that ambidextrous market orientation is particularly valuable for firms seeking innovation in dynamic market environments. In their study on the transformation toward cloud computing business models, Khanagha et al. (2014) [47] examined how firms leverage ambidextrous capabilities to update their business models in the face of technological shifts. They argued that ambidextrous market orientation is key to achieving a balance between exploration and exploitation during digital business model innovation. By dynamically adjusting organizational structures and strategic intentions, firms can effectively explore new business models in uncertain environments while simultaneously utilizing existing resources and capabilities to drive transformation.

Based on the above analysis, we propose the following hypotheses:

H1a. 

Proactive market orientation is positively associated with digital business model innovation.

H1b. 

Responsive market orientation is positively associated with digital business model innovation.

2.2. The Mediating Role of Digital Resource Bricolage

In the resource-based view (RBV) pioneered by Barney (1991) [16], a firm’s sustainable competitive advantage originates from its possession of valuable, rare, inimitable, and non-substitutable resources and capabilities—known as the VRIN framework. This theory highlights the critical role of a firm’s unique resources in creating and sustaining competitive advantage. However, as market environments have become increasingly dynamic, the limitations of a relatively static interpretation of RBV have become apparent. In response, Teece et al. (1997) [48] proposed the dynamic capabilities theory as a complement to RBV. This theory emphasizes a firm’s ability to integrate, build, and reconfigure internal and external resources in rapidly changing environments, enabling firms to identify market opportunities in a timely manner and respond to environmental changes. Eisenhardt and Martin (2000) [25] further extended this theory by suggesting that dynamic capabilities are not rigid routines but consist of identifiable and repeatable organizational and strategic processes, such as speedy decision-making, resource reconfiguration, and organizational learning.

In recent years, dynamic-capabilities scholarship has sharpened how capabilities are created and exercised in turbulent, digital contexts. It reconceptualizes capability building to include purposeful learning from rivals—showing when imitation and counter-imitation become mechanisms for sensing, seizing, and reconfiguring, rather than the mere erosion of rents [49]. Recent information-systems research specifies digital dynamic capabilities in platform and ecosystem settings. It shows how firms use modular digital resources that are addressable via application programming interfaces and strategic initiatives as the vehicles that enact sensing, seizing, and reconfiguring [50]. Complementing the dynamic capabilities view, resource orchestration theory explains how managers structure, bundle, and leverage assets across boundaries to generate innovation outcomes [51]. Recent studies extend this logic to digital innovation. Information technology is shown to enable the acquisition and configuration of external and internal knowledge for open innovation—an orchestration role that connects resource portfolios to novel value creation [52]. At the ecosystem level, aligning complementarities, governing participation, and deliberately shaping co-evolution are core orchestration practices that speed business-model renewal when uncertainty is high [53].

Although a resource-based view, dynamic capabilities theory, and resource orchestration theory have provided important theoretical explanations for firms’ competitive advantages, they still fall short in explaining how firms succeed under conditions of extreme resource scarcity. Addressing this gap, Baker and Nelson (2005) [17] observed that, despite severe constraints, some entrepreneurs create value by “making do” with the resources at hand. Traditional organizational theories and RBV struggle to explain this value creation from scarcity; thus, Baker and Nelson introduced “bricolage,” a process in which entrepreneurs creatively leverage and recombine available resources to seize opportunities.

With the advancement of digital technologies, Nambisan (2017) [54] argue that digitalization opens new domains for entrepreneurship research and practice and call for incorporating digital-technology perspectives, concepts, and frameworks into entrepreneurship studies. Building on this view, Autio et al. (2018) [55] highlight how digital technologies relax spatial constraints yet interact with spatially grounded elements—such as geographic proximity, cluster-based trust, and institutional legitimacy—that shape entrepreneurial ecosystems. This interplay has, in turn, given rise to the notion of digital resource bricolage.

We conceptualize digital resource bricolage as an emergent, making-do mode of digitalization in which organizations creatively combine readily available apps, data, and platform services. Recent work has defined digital resource bricolage, distinguishing it from entrepreneurial/IT bricolage and showing that while it enables short-term innovation under scarcity, overreliance can produce a “digital-bricolage trap” [56]. In contrast, traditional bricolage emphasizes making do by applying resources at hand to new problems in locally embedded, low-resource contexts [57]. Digital resource bricolage is distinctive because digital resources possess modular designs, encapsulated value, and application programming interfaces, enabling fine-grained recombination, reuse, and orchestration at relatively low coordination cost [50]. Within digital ecosystems, a web of application programming interfaces function as structuring devices that allow distributed actors to unbundle/rebundle functionalities and interoperate across organizational boundaries, thereby widening the searchable design space for novel configurations [58].

The theory of ambidextrous market orientation offers a new perspective for understanding resource bricolage. Atuahene-Gima et al. (2005) [59] proposed that ambidextrous market orientation is highly compatible with the concept of resource bricolage. Through ambidextrous market orientation, firms can better guide the direction of resource bricolage, thereby achieving innovation and value creation under resource-constrained conditions. Proactive market orientation broadens firms’ search horizons and triggers experimentation and problem reframing; through sensing, seizing, and reconfiguring routines, it mobilizes novel combinations of digital resources to address latent needs, thereby catalyzing digital resource bricolage, which feeds business-model renewal [46,56]. In contrast, responsive market orientation routinizes the market sensing and cross-functional translation of expressed needs into rapid adjustments in offerings and pricing, strengthening capabilities that leverage and recombine existing resources—an efficiency-oriented pathway that is consistent with digital resource bricolage [56,60]. Together, proactive market orientation and responsive market orientation provide complementary triggers for digital resource bricolage in DBMI. Proactive market orientation drives exploratory recombination to create new value configurations, while responsive market orientation enables the swift reassembly of extant resource sets to deliver timely solutions. This dual balance enables firms to flexibly adjust their resource bricolage strategies in dynamic environments, allowing them to better respond to market changes.

In the context of DBMI, digital resource bricolage plays a critical role. Bricolage enables firms to achieve digital transformation and innovation even under resource-constrained conditions. Digital resource bricolage advances DBMI through concrete recombination and orchestration mechanisms. First, from the perspective of innovation-driven processes, Senyard et al. (2014) [18] suggest that resource bricolage promotes business model innovation through low-cost, rapid, trial-and-error approaches. Digital resource bricolage leverages the modularity and programmatic interfaces of digital resources to rapidly reconfigure value creation/capture with low marginal integration costs; the recombinability of such resources enables the quick prototyping of new activity-system configurations [50,61,62]. Second, digital resource bricolage operationalizes the sensing-seizing link by translating discovered opportunities into implementable design moves, thereby converting market intelligence into business-model change [33]. Third, digital resource bricolage orchestrates complements across platform ecosystems, i.e., reusing external data/services and mobilizing partners, so firms can iterate faster and align offerings with ecosystem dynamics, especially under turbulence [37]. Taken together, these mechanisms explain how digital resource bricolage converts readily available digital building blocks into novel combinations of offerings, processes, and exchange arrangements, thereby improving DBMI outcomes while economizing on time and resources.

Based on the above analysis, we propose the following hypotheses:

H2a. 

Proactive market orientation is positively associated with digital resource bricolage.

H2b. 

Responsive market orientation is positively associated with digital resource bricolage.

H3a. 

Digital resource bricolage mediates the relationship between proactive market orientation and digital business model innovation.

H3b. 

Digital resource bricolage mediates the relationship between responsive market orientation and digital business model innovation.

2.3. The Moderating Role of Environmental Turbulence

Emery and Trist (1965) [63] introduced the notion of the “causal texture” of environments, identifying four types, i.e., placid-randomized, placid-clustered, disturbed-reactive, and turbulent, with the turbulent environment being characterized by high uncertainty and complexity, thereby increasing demands on organizational adaptability. In such contexts, the ability to balance exploration and exploitation became more valuable for firms seeking renewal, consistent with ambidexterity logic.

Building on the foundational characterization of task environments along with their munificence, dynamism, and complexity [64], the marketing literature subsequently operationalized environmental turbulence along two distinct dimensions—market turbulence, capturing demand instability, and technological turbulence, capturing the rate of technological change [65]. This two-dimensional operationalization extended the broader taxonomy and has become a standard way to model contextual boundary conditions in studies of market orientation and innovation.

Regarding market orientation, Narver et al. (2004) argue [13] that a responsive market orientation alone is insufficient for new-product success, and that incorporating a proactive market orientation substantially increases explanatory power. Extending this argument, Bodlaj et al. (2012) [66] note that market and technological turbulence conditions these relationships, i.e., proactive market orientation is associated with higher innovation and market success, whereas the effect of responsive market orientation is positive and significant only under high market turbulence. Complementary evidence indicates that market and technological turbulence can amplify the benefits of proactive/ambidextrous logics for innovation outcomes.

From a technological perspective, research on ambidextrous organizations suggests that firms have to accommodate both evolutionary and revolutionary change under high turbulence, reinforcing the need to balance exploratory and exploitative activities [67]. This logic is consistent with the idea that environmental turbulence strengthens the impetus for resource recombination activities such as digital resource bricolage. Moreover, evidence on environmental moderators shows that environmental dynamism and competitiveness shape the performance consequences of exploratory and exploitative innovation, supporting the broader expectation that turbulence may magnify the effects of ambidextrous market orientation on resource recombination [68]. At the market level, Jaworski and Kohli (1993) [65] conceptualized and measured market and technological turbulence and argued that heightened turbulence increased the need for strong market-oriented capabilities to respond effectively, which aligns with the view that turbulence amplifies the pathway from ambidextrous market orientation to digital resource bricolage. Review articles on organizational ambidexterity identify environmental dynamism as a key contextual variable that shapes the payoffs to ambidextrous capabilities and thereby reinforces the role of turbulence as a boundary condition [69].

Taken together, market and technological turbulence are expected to strengthen the association between proactive/responsive market orientations and digital resource bricolage.

Accordingly, we hypothesize:

H4a. 

Environmental turbulence moderates the relationship between proactive market orientation and digital resource bricolage.

H4b. 

Environmental turbulence moderates the relationship between responsive market orientation and digital resource bricolage.

In summary, this study explores market strategies in the context of digital transformation—specifically, how ambidextrous market orientation influences digital business model innovation. The theoretical framework is illustrated in Figure 1.

3. Research Methodology and Data

3.1. Sample and Data

To test the research hypotheses, we collected primary data through a questionnaire survey. To lower participation barriers and foster an optimal response environment, we devoted substantial time to briefing respondents on the study’s background and objectives. We implemented rigorous data-cleaning procedures, including a screening rule that excluded questionnaires showing inconsistent answers on two consecutive items. The survey was administered online and distributed via SMS, email, and WeChat to students and alumni of the MBA program at a leading national “985” (Double First-Class) comprehensive university in China. This program enrolls participants from enterprises and government agencies, covering both SMEs and large firms across the manufacturing, information technology, service, and finance sectors.

Data were gathered in two phases over a two-month period. In Phase 1, firm managers assessed their firms’ proactive market orientation, responsive market orientation, and digital resource bricolage. In Phase 2, following the same procedures, managers reported their firms’ DBMI and perceived environmental turbulence. In total, 200 questionnaires were distributed and 152 responses were received (response rate = 76%); after removing incomplete or invalid cases, 150 valid observations remained for analysis (valid response rate = 75%).

Table 1. Sample Characteristics.

Questionnaire Items	Content	Sample Size	Percentage (%)
Gender	Male	30	20.00
	Female	120	80.00
Educational Background	High school and below	1	0.7
	Associate Degree	6	4.0
	Bachelor’s Degree	101	67.3
	Master’s Degree	40	26.7
	Doctoral Degree	2	1.3
Firm Age	Within 3 years	3	2.0
	3–5 years	9	6.0
	5–8 years	25	16.7
	8–10 years	40	26.7
	More than 10 years	73	48.7
Firm Size	50 people or fewer	6	4.0
	51–100 people	10	6.7
	101–250 people	40	26.7
	More than 250 people	94	62.7
Industry	Manufacturing	55	36.7
	Services	30	20.0
	Information Technology	33	22.0
	Retail	6	4.0
	Biopharmaceuticals	2	1.3
	Others	24	16.0

reports the sample characteristics.

3.2. Variable Measurement

The study employed established scales adapted from prior research. All items were translated from English into Chinese following the Brislin back-translation procedure, which involves independent forward translation, blind back-translation, and reconciliation by a bilingual panel [70]. Responses were recorded on a five-point Likert scale (1 = “strongly disagree”, 5 = “strongly agree”). To align with contemporary best practice, we also followed widely cited guidelines for the cross-cultural adaptation of self-reporting measures [71].

Proactive and responsive market orientation were measured with items adapted from Narver et al. (2004) [13], capturing both proactive and responsive dimensions. Example items for proactive market orientation included “We help customers anticipate changes in the market” and “We actively establish connections with potential customers to understand their future needs”; an example for responsive market orientation was “We react more quickly than our competitors to customer needs.”

Digital resource bricolage was measured with eight items adapted from Senyard et al. (2014) [18], grounded in the conceptualization of Baker and Nelson (2005) [17], and tailored to digital resources.

Environmental turbulence was modeled as a two-dimensional construct comprising market turbulence and technological turbulence. Following Jaworski and Kohli (1993) [65], market turbulence captures instability in terms of the composition of customers and their preferences, whereas technological turbulence captures the rate of technological change. All items were adapted to our digital context, translated using a forward-backward procedure, and pretested; the final wording appears in

Table 2. Measurement Items.

Variables	Items
Proactive Market Orientation	1. We help customers anticipate changes in the market.
	2. We continuously discover potential or latent needs that customers have not yet realized.
	3. Even if innovation might render our own products obsolete, we still pursue it.
	4. We explore how customers use our products and services.
	5. We actively establish connections with potential customers to understand their future needs.
	6. We target untapped or emerging market segments.
	7. We collaborate closely with lead users who are ahead of the majority in identifying needs.
Responsive Market Orientation	1. We continually monitor and respond to customer complaints and concerns.
	2. Our competitive advantages lie in our understanding of customers’ current needs.
	3. Our management often analyzes customer satisfaction data.
	4. We react more quickly than our competitors to customer needs.
	5. Our company has implemented standardized and periodic customer service practices.
	6. We believe that the primary purpose of this company’s existence is to serve its customers.
	7. We freely share both successful and unsuccessful customer experience information across all business functions.
	8. Customer satisfaction data is regularly disseminated across all levels of our business unit.
Digital Resource Bricolage	1. We are confident in finding feasible solutions to new challenges by utilizing our existing digital resources.
	2. We are willing to take on a broader range of challenges than others who possess similar digital resources.
	3. We utilize any available digital resources that appear useful for addressing new problems or opportunities.
	4. We address new challenges by combining existing digital resources with other low-cost, readily available digital resources.
	5. When dealing with new problems or opportunities, we take action under the assumption that a feasible solution will emerge.
	6. By combining existing digital resources, we have taken on a surprisingly wide range of new challenges.
	7. When facing new challenges, we patch together feasible solutions from existing digital resources.
	8. We combine digital resources to address new challenges that these resources were not originally designed to handle.
Environmental Turbulence	Market turbulence:
	1. Customer preferences in our markets change rapidly.
	2. Changes in customer preferences are difficult to predict.
	3. The composition of our customers changes quickly.
	4. Customers in our markets frequently seek new products/services.
	Technological turbulence:
	1. Technology in our industry changes rapidly.
	2. Technological changes create many new product/service opportunities in our industry.
	3. Many new product/service ideas are enabled by technological breakthroughs in our industry.
Digital Business Model Innovation (DBMI)	1. In the context of digital technology adoption, our business model offers new combinations of processes, products, services, and information.
	2. In the context of digital technology adoption, our business model has attracted many new customers.
	3. In the context of digital technology adoption, our business model has attracted many new suppliers and other business partners.
	4. In the context of digital technology adoption, our business model connects internal and external stakeholders in novel ways.
	5. In the context of digital technology adoption, our business model is transforming the way business transactions are conducted.
	6. In the context of digital technology adoption, we frequently introduce new ideas and innovations into our business model.
	7. In the context of digital technology adoption, we frequently introduce new processes, routines, and standards into our business model.
	8. In the context of digital technology adoption, our business model is pioneering.
	9. Overall, in the context of digital technology adoption, our business model is novel.

. Confirmatory factor analyses supported the two-factor structure.

DBMI was measured using nine items adapted from Soluk et al. (2021) [72]. Respondents were asked to assess their firm’s capabilities regarding DBMI. An example item read: “In the context of digital technology adoption, our business model provides new combinations of processes, products, services, and information.” To reduce bias, we included both individual-level and firm-level control variables, such as managers’ gender and educational background, as well as firm age, size, and industry.

3.3. Common Method Bias Test

Since all data were collected from a single questionnaire, there was a potential concern regarding common method bias. To mitigate this issue, we implemented several procedural remedies. First, during the scale design, we balanced the order of items and avoided ambiguous or double-barreled questions. Second, we ensured respondent anonymity and clarified that the data would be used exclusively for academic research purposes. Additionally, we conducted Harman’s single-factor test to statistically assess common method bias. The results showed that the first unrotated factor accounted for 19.7% of the total variance, which was well below the critical threshold of 40%. This indicated that common method bias was not a serious concern in this study.

3.4. Reliability and Validity Analysis

Before estimating structural relations, we conducted confirmatory factor analyses (CFA) to establish the convergent and discriminant validity of the multi-item constructs and to confirm that the proposed factor structure fit the data adequately. Estimating the CFA with a robust maximum-likelihood procedure accommodated mild non-normality and helped ensure that subsequent inferences about structural paths were not confounded by measurement misspecification. As several items exhibited mild non-normality, we used maximum likelihood robust (MLR) in Mplus 8.0 for both CFA and SEM. For robustness, models re-estimated with MLM (Maximum Likelihood Mean-adjusted) yielded substantively similar conclusions; the fit indices reported below were based on MLR.

Following conventional criteria, RMSEA (Root Mean Square Error of Approximation) ≤ 0.08 and SRMR (Standardized Root Mean Square Residual) ≤ 0.08 indicate acceptable fit, whereas CFI (Comparative Fit Index)/TLI (Tucker-Lewis Index) ≥ 0.90 indicate acceptable fit and ≥0.95 indicate good fit. Although the TLI of Model 2 (0.859) and Model 3 (0.898) were marginal, other indices met the recommended thresholds, suggesting an overall acceptable model fit. Cutoffs were in line with guidelines; model fit was judged holistically across indices given model size and estimation method [73,74,75].

Because several constructs contained many items, we estimated three measurement submodels to avoid an over-parameterized model.

Table 3. Confirmatory Factor Analysis.

Model	Dimensions	RMSEA	CFI	TLI	SRMR
Model 1	Ambidextrous Market Orientation—Digital Resource Bricolage	0.027	0.990	0.974	0.037
Model 2	Ambidextrous Market Orientation—Digital Business Model Innovation	0.067	0.943	0.859	0.049
Model 3	Turbulence-Digital Resource Bricolage—Digital Business Model Innovation	0.068	0.952	0.898	0.038

reports the results; all RMSEA and SRMR values were within acceptable ranges and most CFI/TLI values approached or exceeded recommended cutoffs, confirming satisfactory reliability and validity.

4. Empirical Analysis

4.1. Correlation Analysis

Table 4. Correlation Analysis.

Variables	1	2	3	4	5
1. Responsive Market Orientation	1
2. Proactive Market Orientation	0.626 **	1
3. Digital Resource Bricolage	0.638 **	0.622 **	1
4. Environmental Turbulence	0.561 **	0.537 **	0.436 *	1
5. Digital Business Model Innovation	0.616 **	0.672 **	0.712 **	0.466 **	1

Note: Entries are bivariate Pearson correlation coefficients (two-tailed). * p < 0.05; ** p < 0.01.

presents the correlation matrix for the study variables. The coefficients are significant at conventional levels (p < 0.05; p < 0.01), indicating meaningful associations consistent with expectations. Following the guidance of DE Winter et al. (2016) [76] on when to use Pearson versus Spearman correlation, we report bivariate Pearson correlation coefficients, which are appropriate for assessing linear associations between two continuous variables [77].

To assess potential multicollinearity, we computed variance inflation factors (VIFs). All VIFs were below 3.0 (see

Table 5. Multicollinearity Test.

Variables	Collinearity Statistics
	Tolerance	VIF
Responsive Market Orientation	0.45	2.224
Proactive Market Orientation	0.439	2.279
Digital Resource Bricolage	0.417	2.398
Environmental Turbulence	0.627	1.596
Digital Business Model Innovation	0.394	2.535

), suggesting that multicollinearity was not a concern.

4.2. Hypothesis Tests

We used ordinary least squares (OLS) as the baseline tool to test the linear associations between proactive and responsive market orientations and DBMI. The focal constructs were multi-item, Likert-based composite scales that were treated as approximately continuous. Our objective was to estimate the partial effects net of standard firm-level controls (firm age, size, and industry), yielding interpretable coefficients and model-fit diagnostics. Estimating proactive market orientation and responsive market orientation in comparable specifications also allowed us to assess their incremental explanatory power, directly addressing H1a and H1b. We examined the relationships between proactive market orientation and DBMI, between responsive market orientation and DBMI, between proactive market orientation and digital resource bricolage, and between responsive market orientation and digital resource bricolage. Models 1-0 and 2-0 included only controls. Model 1-1 showed that proactive market orientation was positively associated with DBMI (β = 0.486, p < 0.001), supporting H1a. Model 1-2 indicated that responsive market orientation was also positively associated with DBMI (β = 0.481, p < 0.001), supporting H1b (

Table 6. Regression Analysis of DBMI.

Variables	DBMI
	Model 1-0	Model 1-1	Model 1-2
Firm Age	−0.047	−0.015	−0.052
Firm Size	0.058	0.037	0.094
Industry	−0.166 *	−0.111	−0.099
Proactive Market Orientation		0.486 ***
Responsive Market Orientation			0.481 ***
R-squared (R2)	0.032	0.265	0.258
Adjusted R-squared	0.012	0.244	0.237
F	1.622	13.04 ***	12.583 ***

Note: * p < 0.05, *** p < 0.001.

).

Turning to digital resource bricolage, Model 2-1 showed a positive association between proactive market orientation and digital resource bricolage (β = 0.616, p < 0.001), supporting H2a, while Model 2-2 showed a positive association between responsive market orientation and digital resource bricolage (β = 0.639, p < 0.001), supporting H2b (

Table 7. Regression Analysis of Digital Resource Bricolage.

Variables	Digital Resource Bricolage
	Model 2-0	Model 2-1	Model 2-2
Firm Age	−0.157	−0.124 *	−0.051 *
Firm Size	−0.007	0.026	0.057
Industry	−0.101	−0.102	−0.131 *
Proactive Market Orientation		0.616 ***
Responsive Market Orientation			0.639 ***
R-squared (R2)	0.033	0.41	0.426
Adjusted R-squared	0.013	0.394	0.41
F	1.65	25.228	26.887

Note: * p < 0.05, *** p < 0.001.

).

We examined whether environmental turbulence strengthened the links from ambidextrous market orientation to digital resource bricolage using hierarchical OLS with mean-centered predictors and their interaction terms. Model 3-0 reported the main effects. Model 3-1 included the interaction between proactive market orientation and environmental turbulence; the coefficient on this product term was positive and statistically significant (b = 1.899, p < 0.05), indicating that higher turbulence amplified the positive association between proactive market orientation and digital resource bricolage, supporting H4a. Model 3-2 likewise included the interaction between responsive market orientation and environmental turbulence; the coefficient on this term was positive and statistically significant (b = 1.681, p < 0.05), indicating that turbulence amplified the association between responsive market orientation and digital resource bricolage, supporting H4b.

All moderation models included the full set of controls discussed in Section 3.2, namely, respondent gender and educational background, as well as firm age, firm size, and industry fixed effects. For parsimony, only the coefficient for educational background is displayed in

Table 8. Moderation Effect Analysis.

Variables	Model 3-0		Model 3-1		Model 3-2
	Standardized Coefficient (β)	Significance (p)	Coefficient (b)	Significance (p)	Coefficient (b)	Significance (p)
Educational Background	0.07	0.304	0.099	0.15	0.089	0.197
Proactive Market Orientation	0.272	<0.001	1.347	<0.001
Responsive Market Orientation	0.234*	0.005			1.331	0.005
Environmental Turbulence	0.21	0.012	1.471	0.003	1.197	0.004
Proactive Market Orientation × Environmental Turbulence			1.899	0.014
Responsive Market Orientation × Environmental Turbulence					1.681	0.027
F	20.756	<0.001	20.008	<0.001	18.39	<0.001
R-squared (R2)	0.364	/	0.356	/	0.337	/
Adjusted R-squared	0.347	/	0.338	/	0.318	/

Note: All models include the following controls: respondent gender, respondent educational background, firm age, firm size, and industry fixed effects. For brevity, only educational background is displayed; other control coefficients were estimated but suppressed. Model 3-0 reports standardized coefficients (β); Models 3-1 and 3-2 report unstandardized coefficients (b). Two tailed p values are reported, and predictors were mean centered prior to creating interaction terms.

, whereas the coefficients for the remaining controls were included in the estimations. The magnitude and significance of the focal effects were unchanged when the full set of control estimates is reported; the complete results are available upon request.

We aggregated market turbulence and technological turbulence into a single environmental turbulence composite by z-standardizing each subscale and averaging the standardized scores to keep the regression specification parsimonious. This approach captured the overall level of environmental turbulence, reduced collinearity when forming interaction terms, and facilitated the interpretation of the moderation effect in the linear models.

Because the core constructs were measured with multiple items, we employed SEM to examine the mediating effect of digital resource bricolage. SEM jointly estimates the measurement model and the structural paths, explicitly accounts for measurement error, and yields comparable standardized coefficients. This approach allowed us to test indirect effects via bias-corrected bootstrapping within a unified framework and to assess overall model fit. The structural model achieved adequate fit, shown as follows: χ²/df = 2.61, RMSEA = 0.068, CFI = 0.923, TLI = 0.912, and SRMR = 0.056. Mediation was tested with bias corrected bootstrapping based on 5000 resamples following Hayes and Scharkow (2013) [78], and a mediating effect was inferred when the 95 percent confidence interval excluded zero. Results are reported in

Table 9. Mediation Effect Test Using Bootstrap Method.

Mediation Effect Test			95% Confidence Interval
	Effect Size	Boot SE	Lower Bound	Upper Bound	Significance
Total Effect
Proactive Market Orientation—Digital Business Model Innovation	0.647	0.059	0.531	0.763	Significant
Responsive Market Orientation—Digital Business Model Innovation	0.622	0.065	0.493	0.751	Significant
Direct Effect
Proactive Market Orientation—Digital Business Model Innovation	0.359	0.065	0.231	0.487	Significant
Responsive Market Orientation—Digital Business Model Innovation	0.276	0.072	0.133	0.419	Significant
Indirect Effect
Proactive Market Orientation—Digital Business Model Innovation	0.288	0.078	0.159	0.467	Significant
Responsive Market Orientation—Digital Business Model Innovation	0.346	0.098	0.188	0.569	Significant

.

For proactive market orientation, the total effect on digital business model innovation was 0.647 with a 95 percent confidence interval of 0.531–0.763. Controlling for the mediator, the direct effect remained significant at 0.359, with a 95 percent confidence interval of 0.231–0.487 indicating partial mediation. The indirect effect was 0.288 with a 95 percent confidence interval of 0.159–0.467. The mediated proportion was 44.5 percent, thereby supporting H3a.

Similarly, for responsive market orientation, the total effect was 0.622, with a 95 percent confidence interval of 0.493–0.751, and the direct effect was 0.276, with a 95 percent confidence interval of 0.133–0.419, again indicating partial mediation. The indirect effect equaled 0.346, with a 95 percent confidence interval of 0.188–0.569, yielding a mediated share of 55.6 percent, thereby supporting H3b.

Taken together, it was found that digital resource bricolage partially mediated both relationships. The mediated share was larger for responsive market orientation than for proactive market orientation, which was 55.6 percent versus 44.5 percent, thereby supporting H3a and H3b while indicating a comparatively stronger mediating channel for the responsive logic.

A plausible explanation for the larger mediated share via digital resource bricolage for responsive market orientation is the closer alignment between the emphases of these constructs. Responsive market orientation focuses on addressing customers’ expressed needs and typically leverages existing processes and assets, whereas digital resource bricolage denotes “making do” by recombining resources at hand, making it a low-cost, improvisational response that exploits current resource endowments. This conceptual fit can render the pathway from responsive market orientation to digital resource bricolage and, in turn, to DBMI, particularly efficient in terms of translating market signals into digitally enabled recombinations that support business-model change.

By contrast, proactive market orientation emphasizes latent needs and more exploratory research, which often requires new knowledge acquisition and investment beyond the recombination of extant resources. Proactive market orientation targets latent and emergent needs and, by design, pushes firms beyond the frontier of their current resource endowments. To convert weak signals into DBMI, firms typically undertake boundary-spanning research. They secure new resource inflows, such as novel data streams and advanced analytics capabilities. They integrate external application programming interfaces and platform services that are not yet part of their technology stack and form alliances with technology providers. They also recruit specialized digital talent and fund exploratory pilots, as well as research-and-development projects. These actions reconfigure the architecture of value creation rather than merely recombining what is at hand; they often entail building new modular components, redesigning data pipelines, and integrating previously absent complements.

Therefore, responsive market orientation tends to rely more on the reconfiguration and recombination of existing resources, whereas proactive market orientation is more likely to depend on the acquisition of external resources or the pursuit of breakthrough innovation.

5. Configuration Analysis

We employed fuzzy-set qualitative comparative analysis (fsQCA) to identify configurational pathways sufficient for high DBMI. This study treats proactive market orientation (PMO), responsive market orientation (RMO), digital resource bricolage (DRB), and environmental turbulence (ET) as antecedent conditions and digital business model innovation (DBMI) as the outcome. Raw Likert scores were calibrated into fuzzy sets using the direct method. Anchors were set on theoretical and distributional grounds—full membership, crossover point, and full non-membership—and were ordered to respect the monotonic interpretation of each construct. In the configuration notation below, “~” denotes low membership in a calibrated set and “*” denotes logical conjunction.

The calibration anchors were defined as follows. For PMO, the full membership anchor was 5.0, the crossover point was 4.5, and the full non-membership anchor was 3.8. For RMO, the full membership anchor was 4.9, the crossover point was 4.2, and the full non-membership anchor was 3.7. For DRB, the full membership anchor was 4.8, the crossover point was 4.0, and the full non-membership anchor was 3.5. For ET, the full membership anchor was 3.3, the crossover point was 2.9, and the full non-membership anchor was 1.0. For DBMI, the full membership anchor was 3.3, the crossover point was 2.6, and the full non-membership anchor was 2.4. Anchors were chosen with reference to the scale ranges, and empirical percentiles and sensitivity checks using nearby anchors produced substantively similar coverage and consistency. Using the Quine-McCluskey algorithm with a frequency threshold of 2 and a consistency threshold of 0.90, three configurations sufficient for high DBMI were identified (

Table 10. Configuration Analysis of High-Level DBMI.

	Solution
Configuration of Conditions	1	2	3
High-Level Proactive Market Orientation (PMO)
High-Level Responsive Market Orientation (RMO)			●
High-Level Digital Resource Bricolage (DRB)
High-Level Environmental Turbulence (ET)
Consistency	0.900	0.991	0.983
Coverage	0.792	0.661	0.426
Unique Coverage	0.112	0.065	0.065
Solution Consistency	0.898
Solution Coverage	0.864

Note: 1. ● indicates the presence of a condition; indicates the absence of a condition; blank cells indicate “don’t care” conditions. 2. Data source: Quine-McCluskey algorithm; frequency threshold = 2, consistency threshold = 0.90.

). Because QCA evaluates set-theoretic sufficiency and allows equifinality, cases may have been covered by more than one configuration. These findings complement the variable-centric analyses in Section 4 rather than contradict them.

Configuration 1: Focused specialization path (~PMO * ~DRB).

This configuration indicates that low membership in the high-PMO set together with low membership in the high-DRB set constituted a sufficient path to high DBMI (consistency = 0.900, coverage = 0.792). It characterized firms that compete in well-defined niches where a concentrated strategic focus and established routines substitute for broad exploration. ET did not enter this configuration, suggesting broad contextual adaptability.

Configuration 2: Technology-centered path (~PMO * ~RMO * ~ET).

Low membership in the high-PMO set, together with low membership in the high-RMO set and low environmental turbulence, constituted a sufficient path to high DBMI (consistency = 0.991, coverage = 0.661). This configuration typically characterizes R&D-intensive firms operating in relatively stable environments, where attention and resources are concentrated on cumulative technological development. Under such conditions, deemphasizing market-facing activities can conserve attention and resources for longer-term technological accumulation. Excessive market sensitivity in such settings may fragment attention and slow deep technological exploration. This result should be interpreted in terms of configurational sufficiency rather than as a general negative main effect of market orientation.

Configuration 3: Market-responsive under stability path (RMO * ~DRB * ~ET).

High membership in the RMO set together with low membership in the DRB and ET sets constituted a sufficient configuration for high DBMI (consistency = 0.983, coverage = 0.426) and is frequently observed among established retail and service firms. It underscores the value of timely market responsiveness, i.e., when environmental turbulence is muted while ad hoc resource bricolage may be unnecessary or even counterproductive relative to systematic resource management.

Overall, the three configurations exhibited high solution consistency (0.898) and substantial solution coverage (0.864), underscoring causal complexity and equifinality. To reconcile these findings with the variable-centric results, it is essential to note the different logics of inference. Our OLS regressions for the direct effects indicated that both proactive and responsive market orientations were, on average, positively associated with DBMI. By contrast, fsQCA evaluates set-theoretic sufficiency, allows equifinality and causal asymmetry, and therefore can identify sufficient configurations in which a given condition. Against this backdrop, the three solutions are theoretically coherent. The focused specialization path describes firms competing in well-defined niches that achieved DBMI through concentrated strategic foci and digital streamlining of established routines; broad exploratory search and ad hoc recombination are not required for business-model renewal in such settings. The technology-centered path characterizes R&D-intensive firms in relatively stable environments; here, attention and resources were concentrated on cumulative technological development and architecture redesign; deemphasizing market-facing activities helps conserve focus for deep technological accumulation. The market-responsive under stability path showed that when turbulence is muted, disciplined responsiveness to expressed needs—combined with systematic, rather than improvised, resource management—can suffice for high DBMI.

6. Discussion

6.1. Theoretical Implications

Existing scholarship has increasingly focused on the compelling phenomenon of DBMI [1,7,79]. Drawing on analyses of e-business cases and samples of publicly listed firms, prior research found that firms can realize DBMI by reconfiguring the mechanisms of value creation and appropriation in different ways: first, by integrating the four value creation drivers unique to digital connectivity and interactivity, i.e., efficiency, complementarities, lock-in, and novelty; second, by implementing a digital business strategy defined as the fusion of IT and business strategy that coordinates platforms analytics and agile processes in terms of enterprise scope and speed; and third, by building modular platforms application programming interfaces, recombinable digital components, and so on. Consistent with this line of inquiry, the present study examines the effects of ambidextrous market orientation on digital business model innovation and the contextual conditions under which these effects occur. Our results show that both proactive and responsive market orientations exert significant positive influences on DBMI. Digital resource bricolage plays a partial mediating role in both relationships, and environmental turbulence significantly moderates the association between ambidextrous market orientation and digital resource bricolage.

This study makes four principal theoretical contributions. First, the literature has undertaken initial inquiries into how market orientation shapes new product development, value creation processes, and DBMI, drawing on perspectives and theoretical logics related to innovative service formats pricing systems logistics models and market positioning strategies [80]. Recent research has refined what market orientation and ambidextrous market orientation actually do. Disaggregating market orientation, studies show that responsive and proactive facets channel through distinct capabilities: in exporting SMEs, responsive market orientation and proactive market orientation differentially build pricing and product-development capabilities that raise differentiation advantage and export performance [60]. At the competitive interface, “completing the market orientation matrix” finds that proactive competitor orientation catalyzes innovation (directly and via technology orientation), whereas responsive competitor orientation lifts performance through learning orientation [45]. At the product level, recent work has examined how market orientation and technological opportunity relate to new-product innovation performance, discussing absorptive capacity as a key intermediary capability [81]. Process evidence also links market orientation to ambidextrous business-model renewal, showing how firms combine market-driving and market-driven logics to achieve ambidextrous business model innovation [46]. Building on this trajectory, our study tests how ambidextrous market orientation fosters DBMI under digital conditions, thereby corroborating market orientation theory’s applicability in digital contexts and adding a new perspective on the outcomes of ambidextrous market orientation.

Second, by examining the substance and patterns of business model innovation in the digital era, this study deepens our understanding of DBMI. As high-technology digitalization accelerates, scholars have increasingly investigated how firms craft strategic positions and redesign their business models, revealing distinctive behaviors, pathways, and processes [82,83]. Along this line, our findings show that ambidextrous market orientation enables firms to reconfigure value creation and to shape core competitive advantages by effectively orchestrating and leveraging resources under conditions of scarcity. The strategic integration of ambidextrous market orientation thus has the potential to enhance resource acquisition and operating efficiency while reshaping the mechanisms of value creation [84]. Given the central role of market orientation in enabling operations, innovation, competitive renewal, and adaptation to rapidly changing environments, purposeful deployment of ambidextrous market orientation—together with disciplined orchestration of the resources at hand—is especially consequential for firms [10,13,85,86]. In today’s business-model environment, persistent uncertainty, dynamism, and complexity prevail. This implies that in highly competitive markets, operating business models through traditional templates and routines faces serious challenges to survival and growth. Our empirical results show that the effective deployment of both proactive and responsive market orientations has become critical for DBMI, with responsive market orientation being particularly salient. These findings advance our understanding of DBMI and help identify a complementary pathway for strengthening it.

Third, we investigated the antecedents, consequences, and mediating role of digital resource bricolage within the framework of ambidextrous market orientation. Our evidence shows that both proactive and responsive market orientations increase firms’ engagement in digital resource bricolage, which, in turn, strengthens DBMI. This inquiry adds a new dimension to the literature on bricolage. Recent work positions digital resource bricolage as both enabling and constraining. A multi-stakeholder study showed that micro-enterprises ‘make do’ with available digital tools to achieve short-term innovation and survival, yet overreliance can hinder integration and long-term value creation [56]. Within established firms, middle-manager bricolage facilitates business-model innovation during omni-channel transformation by mobilizing bottom-up recombination practices [87]. In B2B markets, salesperson bricolage fosters service–sales ambidexterity and supports value co-creation, linking improvisational resource use to outcomes [88]. The importance of digital resource bricolage is increasingly being recognized. As a further development of resource orchestration theory, digital resource bricolage is expected to shape DBMI substantially, yet this relationship has not been examined systematically. At the same time, the formation mechanisms and potential consequences of digital resource bricolage remain insufficiently understood. Building on prior research, we therefore analyzed the outcomes of digital resource bricolage and, from the perspective of ambidextrous market orientation, identified managerial antecedents that may lead firms to adopt this approach. In addition, environmental turbulence enhances the benefits that ambidextrous market orientation confers on DBMI. Accordingly, our study advances the contingency literature by clarifying how firms should configure their strategic market orientations under varying levels of turbulence, thereby helping organizations better adapt to environments with different degrees of turbulence.

Fourth, using fsQCA, we identified three configurational pathways that are sufficient for high levels of DBMI, thereby offering guidance for configuring DBMI under different contextual conditions. High-level DBMI enables firms to convert pervasive digitization into defensible value by recasting how they create, deliver, and capture value across customers and partners. It is the linchpin of effective digital transformation, aligning data, platforms, and ecosystem governance with market evolution to sustain advantage. Firms that continuously renew their business models around digital capabilities and customer journeys are better positioned to outperform as competitive boundaries and revenue logics shift [7,8]. Across recent configurational studies, high DBMI emerges through multiple sufficient pathways, rather than a single best practice. Firms can combine distinct bundles of digital innovation attributes—such as platform openness, data-analytics intensity, and ecosystem collaboration—to reconfigure value creation and capture [89]. Under environmental shocks, temporary but coherent packages of changes to the value proposition, value creation processes, and revenue model provide additional viable routes to business model renewal [90]. Existing work has approached high-DBMI configurations from a configurational view of digital innovation attributes and IT infrastructure capabilities, as well as through a temporary business-model innovation lens in crisis contexts, showing that multiple bundled pathways can lead to strong business-model outcomes. Building on this foundation, we identified three sufficient configurations for achieving high DBMI as follows. First, low membership in the high proactive market orientation set, together with low membership in the high digital resource bricolage set, constitutes a sufficient path to high DBMI. Second, low membership in the high proactive market orientation set, together with low membership in the high responsive market orientation set and low environmental turbulence, constitutes a sufficient path to high DBMI. Third, high membership in the responsive market orientation set, together with low membership in the digital resource bricolage and environmental turbulence sets, constitutes a sufficient configuration for high DBMI. These findings broaden the avenues through which high levels of digital business model innovation can be realized and provide new insights into configurational pathways of digital business model innovation.

6.2. Practical Implications

The findings of this study have several practical implications for firms and management practitioners.

First, both proactive and responsive market orientations have significant positive effects on DBMI, indicating that in digital contexts, firms must balance the capabilities of opportunity exploration and demand fulfillment when redesigning business models. To this end, firms should consider instituting governance and incentive systems that deliberately balance proactive and responsive routines—for example, forming cross-functional teams to sense latent needs while responding to expressed requirements and tracking dual metrics for exploration and exploitation. Firms seeking sustainable competitiveness should institutionalize dual market-sensing routines, invest in data literacy and platform governance to mobilize and recombine resources at speed, and calibrate exploration–exploitation balances as turbulence rises.

Second, digital resource bricolage plays a partial mediating role between ambidextrous market orientation and DBMI, implying that the flexible recombination of existing resources can effectively propel business model innovation, particularly when it is driven by a responsive market orientation. This insight offers a viable approach for firms operating under resource constraints, as engaging in digital resource bricolage enables swift responses to market shifts and agile adjustments to resource portfolios, thereby achieving the renewal and upgrading of business models in dynamic environments.

Third, environmental turbulence encompassing both market turbulence and technological turbulence significantly strengthens the effect of ambidextrous market orientation on digital resource bricolage, indicating that the strategic value of ambidexterity becomes more salient under pronounced uncertainty. This finding provides guidance for strategic choices in turbulent contexts, suggesting that firms should emphasize the balanced deployment of proactive and responsive market orientations and flexibly recalibrate market strategies and resource configurations to accommodate environmental change and pursue sustainable development.

Finally, there are three pathways to high DBMI, i.e., the Focused Specialization path, the Technology-Centered path, and the Market-Responsive under Stability path, indicating that firms can attain innovation objectives through differentiated strategies under varying contextual conditions. This evidence moves beyond the notion of a single innovation trajectory by advancing a context-contingent framework for selecting innovation modes and offers concrete guidance for advancing sustainable competitiveness in practice, whereby firms may choose the path that is most consistent with their industry characteristics, resource endowments, and the degree of environmental turbulence.

At the same time, the findings of this study offer several practical implications for policymakers.

Policymakers should build interoperable digital and data infrastructure that lowers the friction of resource recombination. Governments should invest in high-quality connectivity, common data standards, and governance frameworks that enable fair access to and the reuse of industrial data across sectors, thereby reducing the transaction costs of digital resource bricolage and accelerating firms’ business-model redesign. Governments should strengthen the digital enabling environment by expanding high-quality connectivity, cloud access, and interoperable data infrastructures while advancing trustworthy data governance and cross-border data flows. Coherent policies create the conditions under which firms can orchestrate digital resources and reconfigure business models at scale. Complementary measures that enhance access to data and the sharing of data further lower coordination costs and accelerate recombination across public- and private-sector datasets.

Governments should shift support from technology acquisition alone to capability development for market sensing and orchestration. Because our evidence shows that digital resource bricolage partially mediates the effect of ambidextrous market orientation on DBMI, policies should fund programs that build firms’ complementary managerial and organizational capabilities, e.g., cross-functional market intelligence, data literacy, platform governance, and agile processes, especially for small and medium-sized enterprises that face adoption and skills gaps. The binding constraint in many firms is not technology per se but the capability to mobilize, recombine, and orchestrate resources at speed. Policy should shift from subsidizing tools alone to building organizational capabilities through programs that fund analytics and data-literacy training, cross-functional market-intelligence routines, platform-governance skills, and advisory support for small and medium-sized enterprises, thereby converting digital inputs into sustained business-model innovation.

7. Limitations and Future Directions

This study deepens our understanding of how ambidextrous market orientation shapes DBMI, but it has several limitations. These limitations suggest avenues for future research. First, the sample size (N = 150) may raise concerns about statistical power and generalizability. Second, large firms were overrepresented, which may have limited the applicability of the findings to smaller firms. Third, while we have reported bias-corrected bootstrapped confidence intervals for the mediation effects, for parsimony we did not extend bootstrap intervals to all structural paths or conduct additional robustness checks. Future research should replicate the model with larger, more balanced samples and use complementary robustness procedures. Longitudinal panel data or experimental designs would further corroborate our conclusions.

In sum, as one of the first studies to emphasize, from a resource perspective, how ambidextrous market orientation influences DBMI, this research highlights the importance of digital resource bricolage and environmental turbulence. A key insight is that activities centered on ambidextrous market orientation can generate competitive advantage by effectively mobilizing recombining and orchestrating resources at hand. This study aimed to provide more comprehensive and in depth theoretical support and practical guidance for the field of DBMI, thereby helping firms achieve sustainable competitiveness in the digital era. We look forward to more scholars jointly advancing an in depth exploration of the resource perspective on ambidextrous market orientation and business model innovation in digital contexts.