Enterprise Architecture for Sustainable SME Resilience: Exploring Change Triggers, Adaptive Capabilities, and Financial Performance in Developing Economies

Abstract

Enterprise architecture (EA) provides a strategic foundation for aligning business processes, IT infrastructure, and organizational strategy, enabling firms to navigate uncertainty and complexity. In developing economies, small and medium-sized enterprises (SMEs) face significant challenges in maintaining financial resilience and sustainable growth amidst frequent disruptions. This study investigates how EA-driven change events affect SME financial performance by activating three key adaptive mechanisms: improvisational capability, flexible IT systems, and organizational culture. A novel classification of EA change triggers is proposed to guide adaptive responses. Using survey data from 291 Pakistani SMEs collected during the COVID-19 crisis, the study employs structural equation modeling (SEM) to validate the conceptual model. The results indicate that improvisational capability and flexible IT systems significantly enhance financial performance, while the mediating role of organizational culture is statistically insignificant. This study contributes to EA and sustainability literature by integrating a typology of EA triggers with adaptive capabilities theory and testing their effects in a real-world crisis context.

1. Introduction

In today’s dynamic and unpredictable business landscape shaped by rapid technological advancements, global disruptions and economic volatility, organizations face growing pressure to remain adaptable, resilient, and sustainable. These challenges are particularly more critical for small and medium-sized enterprises (SMEs) in developing economies, where limited resources and increased uncertainty threaten their survival and growth. Addressing such complexity requires robust frameworks that support not only operational alignment but also strategic responsiveness.

Enterprise architecture (EA) has emerged as a fundamental framework that provides structured methodologies to align business processes, IT systems, and strategic objectives. By identifying, representing and integrating core enterprise components, EA enhances decision making, streamlines processes, and fosters adaptive transformation. The ANSI/IEEE Standard 1471-2000 defines enterprise architecture (EA) as “the fundamental organization of a system, embodied in its components, their relationships to each other and the environment, and the principles governing its design and evolution” [1]. This definition underscores EA’s dual role as both a strategic blueprint and a dynamic instrument for facilitating organizational change.

However, traditional EA models emphasize standardization and long-term planning, often at the expense of flexibility and agility [2,3]. This rigidity limits their effectiveness in unpredictable settings, especially during sudden crises, such as the COVID-19 pandemic, where rapid responsiveness is essential.

Recent research calls for an evolution toward “agile EA”, which reconceptualizes EA as a dynamic, adaptive tool that enables real-time sensing and capability deployment in response to sudden change events [4,5]. Yet despite this conceptual shift, few studies offer systematic methods for identifying and classifying the types of events that trigger EA adaptation, or for empirically evaluating their implications for SME performance.

Although some studies have acknowledged that change events influence EA adaptation, they typically treat such events in an ad hoc manner without offering a systematic classification based on characteristics such as origin, predictability, or organizational impact. This limits organizations’ ability to proactively tailor EA responses to different types of disruptions. To address this critical gap, this study proposes a structured taxonomy of change events that enables organizations to tailor EA responses more effectively and enhance resilience. It further examines how this classification links to improvisational capabilities and their effect on SME financial performance under conditions of uncertainty.

This need for an agile, responsive architecture is especially relevant in the context of sustainable development. As SMEs are key drivers of employment, innovation, and inclusive growth, their ability to endure shocks and maintain continuity directly supports the sustainability goals [6,7]. In developing economies, SMEs often operate under heightened uncertainty, institutional voids, and resource scarcity. These conditions amplify the importance of recognizing both the source (internal vs. external) and urgency (sudden vs. gradual) of environmental disruptions. Classifying EA change triggers along these dimensions provides SMEs with a structured lens for prioritizing response strategies and aligning their architectural adaptations accordingly. By embedding flexibility into EA frameworks and tailoring responses based on change characteristics, SMEs can enhance their financial resilience, pursue innovation-driven growth, and contribute more effectively to long-term sustainability. SMEs in Pakistan, in particular, operate within environments marked by regulatory uncertainty, limited institutional support, inadequate digital infrastructure, and financial constraints. These challenges are exacerbated by a significant informal sector and constrained access to capital, which restrict long-term planning and technological investment. Such conditions make adaptive architectural responses even more critical, reinforcing the need for a structured EA change framework that can guide firms in managing disruptions efficiently and sustainably.

To advance this agenda, this study proposes a classification of change events that trigger EA adaptation. While our conceptual framework introduces a multi-dimensional classification of EA change events, this study empirically examines only one category, external, occurrence-based change events, with a focus on the COVID-19 pandemic as a representative trigger. It further investigates how improvisational capabilities, the ability to respond creatively and rapidly to unexpected disruptions, along with flexible IT systems and organizational culture, influence SME financial performance in uncertain environments. This design choice reflects the immediate and pervasive impact of the pandemic on SMEs and serves as a real-world case for evaluating adaptive capabilities.

Specifically, this study addressed four core research questions.

(1)

What types of change events are recognized within the context of enterprise architecture (EA) in SMEs and how can these events be systematically classified?

(2)

How do improvisational capabilities impact SMEs’ financial performance during times of environmental uncertainty?

(3)

What role do flexible IT systems play in mediating the relationship between improvisational capabilities and financial performance?

(4)

How does organizational culture mediate the impact of improvisational capabilities on SMEs’ financial performance?

By answering these questions, this study offers both theoretical contributions and practical guidance for building resilient, sustainability-oriented enterprises in emerging markets. The findings aim to inform managers, policymakers, and EA practitioners about how to embed agility, improvisation, and sustainability in EA design, ultimately supporting SMEs’ capacity to thrive in a turbulent world.

2. Literature Review

This literature review explores the evolving role of enterprise architecture (EA) in helping organizations adapt to uncertainty. It is organized into three key areas: Section 2.1 reviews EA in relation to change events and their classification; Section 2.2 examines EA-driven capabilities that support organizational adaptability, and Section 2.3 the role of EA in enabling sustainability outcomes.

2.1. Enterprise Architecture and Change Events

The evolution of enterprise systems is often triggered by environmental turbulence or internal change, and enterprise architecture (EA) is no exception. While various studies highlight how EA adapts to such triggers, a consistent and structured typology is still missing.

Event-based triggers prompted by sudden external disruptions are well documented. Abraham, Aier, and Winter [8] show that crises and regulatory changes often lead to EA adaptations in volatile contexts. Pavlou and El Sawy [9] emphasize that turbulence forces organizations to quickly reconfigure processes, supported by IT and sensing capabilities that EA can help structure. Likewise, Eisenhardt and Martin [10] identify high-velocity environments as drivers of dynamic capabilities, requiring fast architectural responses. Time-based triggers, such as scheduled reviews or milestone completions, are discussed by Buckl et al. [11], who differentiate between periodic and on-demand EA updates. Fischer et al. [12] proposed a federated approach to EA maintenance, suggesting that “special events” derived from external data sources can prompt updates but offer limited insight into the characteristics of those events. Data-driven triggers are highlighted by Mannapur [13], who underscores the role of event-driven architectures in detecting real-time anomalies, such as those from sensors or logs that can automatically prompt EA adjustments. Model-based internal triggers are explored by Dam et al. [14], who propose reusable change patterns to resolve inconsistencies within EA models. De Boer et al. [15] add to this by introducing semantic change impact analysis, which shows how structural changes in EA models can cascade across enterprise layers.

Although these studies address different types of change triggers, the literature remains fragmented. Most focus on individual mechanisms without integrating them into a cohesive framework. This gap limits EA’s strategic value, especially for small and medium-sized enterprises (SMEs), which face greater uncertainty and resource constraints. To fill this gap, the current study consolidates existing research and proposes a unified classification of EA change triggers based on two dimensions: origin (internal vs. external) and activation type (event, time, data, or model driven). This typology aims to improve EA’s adaptability and resilience in SMEs by providing a clear and replicable framework for managing architectural change.

Table 1. Proposed change events and relevant research.

Change Event	Description	Sources
Event-based	Triggered by events that occur outside the EA model.	Abraham, Aier, and Winter [8]; Buckl [11], Pavlou and Sawy [9]; Eisenhardt & Martin [10]; Winter [16].
Time-based	Triggered due to activities conducted after a predetermined amount of time.	Fischer, et al. [12], Buckl [11]
Data-driven	Triggered by data sources.	Mannapur [13]; Farwick et al. [17], Fischer, et al. [12].
Internal Model Events	Internally triggered by the EA model	Farwick et al. [17], Dam et al. [13], Lê, and Ghose [14], and de Boer et al. [15].

summarizes the foundational studies that have informed this classification, which aims to reinforce EA’s strategic role as a capability-enabling mechanism for resilience and sustainability under conditions of uncertainty.

2.2. EA-Driven Capabilities and Organizational Adaptability

As internal and external turbulence intensifies across global markets, enterprises face increasing pressure to reconfigure their structures, systems, and operational procedures. Effectively navigating such disruptions requires capabilities, broadly defined as a firm’s capacity to utilize, combine and integrate resources to achieve strategic objectives [18]. Within the context of enterprise architecture (EA), these capabilities are conceptualized as “the ability of firms to orchestrate and employ the firm’s resources using EA, while simultaneously aligning strategic goals with the use of IS/IT” [5].

EA-driven capabilities have garnered increasing scholarly attention for their role in enhancing organizational agility, adaptability, and innovation [19,20,21,22].

Enterprise architecture (EA) serves as both a structural and cognitive enabler of organizational adaptability. Through mechanisms such as shared architectural blueprints, resource modularity, and real-time information flows, EA makes it easier to coordinate and reorganize organizational resources in reaction to change. These characteristics of EA enable organizations to develop two distinct but interrelated adaptive capabilities: dynamic capability, which involves proactive and systematic adaptation to anticipated developments, and improvisational capability, which entails rapid and creative responses to sudden, unforeseen disruptions. As shown in Figure 1, EA-driven mechanisms translate architectural planning into both dynamic and improvisational capabilities.

The type of adaptive response an organization must adopt is largely determined by the characteristics of the environmental turbulence it faces, including its frequency, magnitude, and predictability. Dynamic capabilities are typically activated in response to anticipated or gradual environmental changes, whereas improvisational capabilities become critical during unexpected and high-velocity disruptions.

Table 2. A comparison between dynamic and improvisational capabilities. Developed by the authors based on interpretations from Refs. [8,9,23,24].

Criteria	Dynamic Capabilities	Improvisational Capabilities
Type of Event	Predictable, expected, planned environmental events (“waves”)	Unpredictable, novel, sudden environmental events (“storms”)
Organizational Response Required	Strategic, long-term reconfiguration efforts	Short-term, spontaneous adaptations using current resources
Process Structure	Systematic and guided by planned frameworks	Informal, often intuitive, unstructured.
Time requirements	Adequate time available for planning evaluation and execution.	Requires immediate response.

presents a comparison of these two adaptive approaches, highlighting their key distinctions and contextual relevance.

Although both capability types are vital, their relevance varies depending on the characteristics of the triggering event. Given this study’s classification of EA change events, it becomes critical to assess which form of capability is most effective in response.

While the literature has extensively examined EA-driven dynamic capabilities, particularly in relation to digital transformation [20,22], platform development [25], and process agility [3,26], research on improvisational capabilities remains relatively limited. Moreover, much of the current research has been conducted in advanced economies, such as the United Kingdom [27], South Korea [28], and China [29]. This leaves a significant geographic and contextual gap, particularly in developing nations such as Pakistan, where SMEs often encounter more frequent and sudden disruptions but operate with limited formal resources and infrastructure.

To address these gaps, this study investigates how EA-driven improvisational capabilities contribute to financial resilience under conditions of sudden change. By linking a structured classification of EA change events to adaptive capability responses, this study provides novel insights into how SMEs in emerging markets can leverage EA to enhance agility, maintain performance, and foster innovation when faced with uncertainty.

2.3. EA-Enabled Sustainability

As the global business landscape becomes increasingly complex and sustainability focused, organizations are expected not only to perform economically but also to operate responsibly within social and environmental contexts. This evolving expectation aligns with the triple bottom line framework, which defines sustainability as the integration of economic, environmental, and social value creation [30]. In this context, enterprise architecture (EA) has begun to emerge as a valuable enabler of sustainability-oriented transformation within organizations.

EA supports sustainability by promoting system efficiency, reducing operational redundancies, and aligning technology use with broader strategic and societal goals. Through improved transparency and integration across organizational units, EA enables firms to better manage resources, implement eco-efficient practices, and embed sustainability into decision-making processes [31,32]. These functions make EA instrumental in supporting green IT initiatives, energy informatics, and responsible digital transformation strategies [33].

Beyond environmental benefits, EA also contributes to organizational resilience and adaptability, key components of long-term economic sustainability [34,35]. By facilitating agile responses to disruptions and fostering innovation through reconfigurable architectures, EA strengthens a firm’s capacity to withstand shocks and evolve in line with changing societal expectations.

Despite these conceptual connections, empirical research explicitly linking EA to sustainability outcomes remains limited. Most existing studies focus on operational performance or IT alignment, with little attention to how EA enables sustainability at the organizational or ecosystem level, particularly in the SME sector within developing economies. Additionally, the role of EA-driven adaptive capabilities (such as improvisation) in supporting sustainable performance under uncertainty remains largely unexplored.

This study addresses these gaps by examining how EA-enabled improvisational capabilities contribute to financial resilience and sustainable value creation in SMEs. By positioning EA as both a strategic and adaptive foundation, the study contributes to the broader discourse on how architectural agility can support sustainable enterprise ecosystems, especially in the context of emerging markets, such as Pakistan.

While this section reviewed theoretical foundations and key constructs, the next section develops the conceptual model and hypotheses derived from this literature.

3. Conceptual Framework and Hypotheses Development

3.1. EA-Driven Change Events

To address the complexity and unpredictability of today’s business environment, this study introduces a comprehensive classification system for change events within the enterprise architecture (EA) framework. This system offers both theoretical contributions, enabling organizations to systematically identify, categorize, and respond to diverse types of change events. By doing so, it facilitates the adaptability and alignment of EA models with dynamic, real-world conditions. Changes, defined as “the alteration of one state to another” [36], are driven by both internal and external factors. These factors, known as change events, catalyze the EA model’s evolution and initiate maintenance processes crucial for retaining the model’s relevance [37]. This study categorizes EA-based change events into two main categories: (1) external change events and (2) internal change events.

Figure 2 presents a conceptual classification framework for change events, distinguishing external and internal triggers into specific subtypes. This detailed classification is subsequently elaborated upon in the following sections. While the framework encompasses a broad range of change events, the empirical model developed for this study (illustrated in Figure 2) specifically focuses on occurrence-based external events to analyze their impact on EA-driven capabilities and organizational financial performance.

3.1.1. External Change Events

External change events refer to disruptions originating outside the organization, often requiring rapid sensing and adaptive strategic responses. Addressing such events comprehensively necessitates considering environmental factors alongside technological influences. External change events, originating beyond the EA model’s internal systems, can be classified into two primary types: (a) occurrence-based events and (b) time-based events.

(a)

Occurrence-based events

• These events emerge due to external turbulence, including technological innovations, changing market landscapes, and global crises, such as the COVID-19 pandemic, compelling organizations to adapt swiftly. Such phenomena are often described as manifestations of environmental dynamism. The triggers for these adaptive responses can be systematically categorized according to various dimensions:
  • Frequency: Time span between event occurrences.
  • Amplitude: Magnitude of deviation from the initial conditions caused by the event.
  • Predictability: Degree of irregularity in the overall pattern of events.
  • Velocity: The rate of change to be adopted as a result of the event.

Such events can be manifested as “waves” or “storms” depending on their frequency, magnitude, and predictability [8,9,23], the details of which are given in

Table 3. Types of environmental change events.

Type of Environmental Turbulence	Frequency	Amplitude	Predictability	Velocity
Waves	High	High	High	Low
Storms	Low	High	Low	High

. “Waves” are anticipated occurrences that can be predicted, giving organizations enough time to prepare formally to deal with them. Meanwhile, “storms” are sudden, unpredictable events that require a quick response to survive.

Thus, external change events differ not only in their origin but also in the urgency and predictability of the responses they necessitate within EA frameworks, distinguishing planned adaptations from emergent improvisations.

(b)

Time-based events

• These triggers are activated at predetermined intervals aligned with the organization’s operational or strategic cycles. Often arising from scheduled activities, such as periodic strategic planning sessions, they systematically assess and realign the organization’s direction, serving as catalysts for change within the enterprise architecture.

3.1.2. Internal Change Events

Internal change events refer to disruptions originating within the technical components of the EA model itself. These events primarily focus on maintaining operational alignment, ensuring that the architecture accurately reflects the organization’s evolving infrastructure, processes, and capabilities. Internal change events can be further categorized into two primary types: (a) data-based events and (b) model-based events.

(a)

Data-based events

• These originate from data sources and necessitate updates to the EA model to reflect new or changed data. Following the framework by Farwick et al. [17], these events prompt manual, semi-automated, or fully automated updates.
  • Periodic/task-based reminders:
• These reminders are meant to prompt manual data collection from stakeholders at regular intervals in order to initiate the change process. A specific person is responsible for certain parts of the model. The frequency is also set to update the data and, if necessary, to check for model changes. Finally, the EA team or a supporting tool will trigger the change process at a predefined frequency.
  • Type-based reminders:
• This approach activates the change process through events identified within information systems. For instance, a change in architecture signaled by a project within a project portfolio management tool could be a catalyst. This is particularly valuable in scenarios where structured data sources are not suited to address EA modifications. This method proves beneficial in organizations where specific event sources are configured to detect and act upon EA-relevant changes, ensuring the architecture remains current and aligned with organizational objectives.
  • Structured data collection:
• The initiation of the change process occurs upon the commencement of importing data, facilitated either through a structured data source initiating a push or an EA tool executing a pull operation. This method leverages automated data collection mechanisms, utilizing sources such as enterprise service buses (ESBs) and configuration management databases (CMDBs), which furnish structured, EA-pertinent data directly applicable to the model. However, it is essential to note that the accessibility and applicability of such data sources might vary, rendering this approach infeasible for specific organizations due to their unavailability.

(b)

Model-based events

• These events are internally triggered by the EA model elements. Change is initiated in these situations under two circumstances.
  • Model Element Expires:
• EA artifacts possess varying lifespans. Identifying the impending expirations of model elements facilitates the generation of model expiration events. These events serve as triggers to commence the change cycle. Specifically, a model expiry event is activated when the elapsed time since a model element’s last review or modification surpasses its designated lifespan, signaling the need for updates or revisions.
  • Constraint Violation:
• This method involves applying specific constraints to model elements within the EA. Change events are triggered when any of these constraints are breached, effectively initiating the change cycle. This process ensures the EA model remains compliant and up-to-date by identifying and addressing violations promptly.

The key characteristics distinguishing each type of change event, including its origin, predictability, required organizational response, and typical examples, are summarized in

Table 4. Summary of EA change event types.

Event Type	Origin	Predictability	Nature of Response	Example
Occurrence-based	External	Low/Variable	Dynamic/Improvisational	Pandemic outbreak, sudden market shift, emergence of IoT/AI technologies
Time-based	External	High	Proactive and planned	Annual strategic planning cycle
Data-based	Internal	High/Moderate	Routine updating	Regular data input from CRM or ESB system
Model-based	Internal	High	Compliance driven updating	Expiry of an architecture element, constrain violation

.

In summary, distinguishing between external and internal change events and further classifying their subtypes provides a structured foundation for organizations to tailor their EA adaptation strategies. By systematically identifying the origin, predictability, and nature of change events, firms can deploy targeted and agile responses, enhancing both operational continuity and strategic resilience in uncertain environments. It is important to note that, while we present a comprehensive classification of EA change triggers, the current empirical study tests only the effects of external occurrence-based events. Internal, data-driven, or strategy-based triggers are conceptually relevant but fall outside the scope of this empirical investigation. Future studies may extend this framework to evaluate these additional trigger types.

The classification of EA-triggered change events proposed in this study contributes to the literature by offering a structured, event-based lens for understanding when and how organizations must adapt. Unlike previous EA studies that broadly reference change or disruption, this typology distinguishes between event origin (internal vs. external), nature (sudden vs. gradual), and intensity. This classification advances theory by linking specific types of change events, such as occurrence-based external shocks like pandemics, to targeted activation of adaptive capabilities, such as improvisation and IT flexibility. Importantly, the framework underpins the empirical model tested in this paper by operationalizing the concept of “environmental turbulence” and offering a replicable basis for modeling EA-driven adaptation across uncertain contexts. Thus, this contribution is not only theoretical but also instrumental in developing decision-making strategies for resilience in SMEs.

3.2. Adaptive Capabilities in the EA Context

EA enables firms to activate internal capabilities that respond to change events. Three key capabilities are central to this study:

(1)

Improvisational Capability: The ability of an organization to spontaneously develop novel responses under time pressure and ambiguity. It involves real-time learning and action without pre-existing procedures.

(2)

Flexible IT Systems: IT systems that are modular, scalable, and capable of supporting rapid reconfiguration when new environmental demands arise.

(3)

Organizational Culture: Shared values and behavioral norms that facilitate or hinder strategic responsiveness and employee-level improvisation.

3.3. Conceptual Model

Figure 3 presents the conceptual framework developed for this study, which links EA-triggered environmental change events to SME financial performance through three adaptive organizational mechanisms. These mechanisms, improvisational capability, flexible IT systems, and organizational culture, are theorized to mediate the relationship between change triggers and performance outcomes. The model positions externally driven, unpredictable change events as the starting point for organizational adaptation, such as those arising from crises such as the COVID-19 pandemic. Improvisational capability captures the organization’s capacity for real-time, novel responses; flexible IT systems represent the technological readiness to reconfigure operations under pressure. Organizational culture is included as both a direct and mediating factor, influencing how adaptive responses are enacted. This framework draws upon the theoretical foundations of dynamic capabilities and EA adaptation, forming the basis for the empirical hypotheses tested in the following section.

3.4. Hypothesis Development

Based on the conceptual framework outlined in Figure 3, this section develops the study’s hypotheses by drawing on relevant literature in enterprise architecture, dynamic capabilities, and organizational adaptation. The model posits that EA change events act as external triggers that activate specific internal capabilities, namely, improvisational capability. Improvisational capability is conceptualized as an exogenous (independent) variable, with organizational culture and flexible IT systems as mediating variables. The organization’s financial performance is treated as the endogenous (dependent) variable, thus establishing a framework to examine how improvisational capabilities, influenced by the supportive roles of organizational culture and IT flexibility, impact financial outcomes.

3.4.1. Improvisational Capability and Organization’s Financial Performance in Sudden Environmental Change

Improvisational capability is a philosophy that underscores the importance of rapid, innovative responses and creative problem solving with available resources in the face of unforeseen changes. The term ‘improvisation’ combines the essence of ‘proviso,’ indicating ‘planning,’ with the prefix ‘im,’ denoting ‘absence.’ Therefore, improvisation refers to acting without formal planning or adapting swiftly without predefined strategies [9]. When this concept is applied to organizational contexts, it transforms into ‘impact learning,’ a process through which organizations adapt and reconfigure to survive and secure a competitive edge in the face of abrupt and unforeseen challenges [9]. Organizational improvisation consists of three dimensions: “speedy response, reconfigurability, and novel solutions” [38]. Rapid response narrows the gap between planning and action; reconfigurability involves the agile reallocation of available resources to address urgent needs, and innovation in solutions encompasses crafting appropriate, unique responses divergent from routine practices.

In the current volatile business landscape, characterized by high turbulence and unforeseen challenges, the capacity for improvisation has emerged as a crucial factor for organizational survival and competitive differentiation. This capability has sparked significant research interest across various domains, including its influence on the digital economy [39], entrepreneurship [40], everyday problem solving and organizational learning [41], and product innovation and development [9,42]. However, a gap persists in our understanding of how improvisational capabilities can bolster organizational performance in the face of natural uncertainties. In this study, “natural uncertainties” refer to unpredictable external disruptions originating from environmental, technological, or societal sources, such as pandemics, market shocks, or sudden technological breakthroughs that significantly impact organizational operations without sufficient lead time for formal planning.

The recent COVID-19 pandemic provides a unique and timely scenario for exploring how improvisational capabilities impact organizational financial performance during times of such uncertainties. Previous research has hinted at the positive influence of improvisational capabilities in emergencies and uncertain situations. Previous research conducted on improvisational capability states that it plays a positive role in states of emergency and uncertainty; thus, for natural uncertainty, we also propose the following hypothesis:

H1: 

Improvisation capability has a positive influence on an organization’s financial performance during natural uncertainties.

3.4.2. Organizational Culture Mediates the Link Between Improvisation Capability and Organizational Financial Performance Under Natural Uncertainties

Implementing improvisational capabilities necessitates rapid, innovative responses to unforeseen changes [38]. This underscores the crucial role of an organizational culture that welcomes new ideas, encourages risk taking, and executes changes swiftly. As highlighted by Zeb et al. [43], such a culture significantly enhances an organization’s development and output.

In the face of unexpected change, the effectiveness of improvisational capability is closely linked with the organization’s strategy, structure, culture, and flexibility [44,45,46]. Kung and Kung [38] highlight that the skills necessary for leveraging improvisational capabilities, crucial for navigating sudden shifts, can be developed through targeted training, ongoing practice, and a supportive organizational culture. Importantly, they note a symbiotic relationship between improvisational capabilities and an innovative organizational culture. As they argue, this relationship mutually reinforces each other to drive superior performance. A positive culture that champions creativity and innovation, therefore, not only facilitates improvisation but also significantly outperforms more conventional approaches [43]. Orlikowski’s [46] improvisational change model further underscores the importance of the organizational context, including culture, structure, roles, and responsibilities, as a critical dimension in the efficacy of change processes. This model, importantly, emphasizes that an adaptable and responsive organizational environment is beneficial and fundamental for the successful implementation and effectiveness of improvisational capabilities.

H2: 

Organizational culture mediates the link between improvisation capability and organizational financial performance under natural uncertainties.

3.4.3. Flexible IT Mediates the Link Between Improvisation Capability and Organizational Financial Performance During Natural Uncertainties

Organizations are facing an era of ever-increasing turbulence and change. It is now necessary to analyze changing market conditions, make quick decisions, and react to the changing environment in the least possible time. A flexible IT system that supports business processes has a great influence on an organization’s capability to make changes [47]. It provides services, technical platforms, and resources to deal with unpredictable changes quickly, which also enhances the improvisational capabilities of the organization [38,48]. Pavlou and Sawy [9] argue that effectively leveraging IT systems is critical to fostering these adaptive traits, ultimately strengthening an organization’s ability to navigate uncertainty more efficiently. According to Kung and Kung [38] and Milliman et al. [49], flexible IT systems amplify these improvisational capabilities by making organizations more responsive, innovative, and adaptable in the face of environmental volatility.

Ahmadi [50] notes that flexible IT systems are the foundation for managing rapid environmental shifts. They foster system integration within and across organizational boundaries, facilitate information sharing, and support significant business process modifications in response to evolving needs. Ness and Fullerton [51] view these systems as central to making organizational functions more amenable to change, enabling quicker adaptation and, thus, empowering organizations to effectively deploy their improvisational capabilities. This, in turn, leads to enhanced performance amid sudden and unpredictable challenges.

H3: 

Flexible IT mediates the link between improvisation capability and organizational financial performance during natural uncertainties.

3.4.4. Role of Organizational Culture in Showing Better Financial Performance When Dealing with Environmental Change

Organizational culture in its simplest form can be defined as “what the organization stands for and what it believes in” [52]. It is a system of common assumptions, shared values, and principles that produce normative pressures on the originating members. Organizational culture has four dimensions: (1) mission, which refers to the goals or reason for the existence of any organization and how it plans to achieve them, (2) involvement, which is the authority given to the employees to decide, (3) consistency, which is the harmony among different parts of the integration within an organization and (4) adaptability, which is flexibility towards a changing environment [53].

Each organization’s unique culture is pivotal in determining its success, influencing factors such as employee commitment, satisfaction, operational efficiency, and financial performance. Coyne [54] posits that organizational culture is not just a lever for competitive advantage but also a powerful tool for overcoming barriers to change. The capacity for change within any organization is profoundly shaped by its culture, serving as a catalyst for transformation or a formidable barrier [55]. The influence of organizational culture on employee behavior and attitudes is significant, thereby affecting the organization’s overall performance and its dynamism in the face of environmental shifts, instilling confidence and reassurance [56,57].

An empirical study by Imran et al. [53] supports the notion that a culture fostering innovation correlates with superior organizational performance amidst changing conditions and brings about positive change. Cultures that prize flexibility encourage creative problem solving, meet client expectations, and foster overall growth and development [58]. This underscores the critical role of organizational culture in navigating the complexities of today’s ever-changing business landscape, offering a beacon of hope and optimism. Thus, we conclude the following:

H4: 

Organizational culture positively affects organizational financial performance.

3.4.5. Effect of Flexible IT Systems on Organization Financial Performance During Environmental Changes

IT flexibility is the capacity of an IT system to adapt beyond its standard operations for enhanced performance amidst environmental shifts. It allows an organization’s IT framework to respond swiftly and cost-effectively to changes in the business landscape [59,60]. The adaptability of an IT system is characterized by its modularity, connectivity, and compatibility [59]. Modularity refers to the ease of disassembling and reassembling system components. Connectivity denotes the system’s ability to link with external and internal components. Compatibility is the capacity for seamless information exchange across technological elements.

Davenport and Linder [61] highlighted IT flexibility as a fundamental organizational competency, emphasizing that the efficacy and resilience of an IT system are measured by its adaptability and robustness in accommodating changes. A flexible IT infrastructure ensures optimal effectiveness and alignment with business objectives in scenarios marked by complexity and flux. Goldman et al. [62] argued that flexible IT becomes critical for profitable operations amidst continuous change and uncertainty. It facilitates the swift delivery of results and underpins sustainable growth in volatile markets [63]. Conversely, lacking flexibility can hinder organizational performance, locking the system into a change-resistant framework [63]. Fiegenbaum et al. [64] posited that flexible IT systems empower organizations with operational efficiency, heightened productivity, and a competitive edge in uncertain and dynamic contexts. Thus, flexible IT infrastructure is regarded as pivotal for superior performance in challenging environmental conditions. Hence, several authors consider flexible IT systems as a basis for showing better performances in turbulent environmental conditions [62,65]. Thus, we propose the following hypothesis:

H5: 

Flexible IT practices play a positive role in improving an organization’s financial performance during environmental changes.

4. Empirical Study: Testing the Model Under COVID-19 Context

The COVID-19 pandemic, an unprecedented global crisis, has demonstrated the resilience of organizations in the face of adversity. Its unpredictability, sudden onset, and widespread impact forced global mobility to a halt, leading to widespread home confinement and significant disruptions in industries such as transportation and manufacturing. This crisis pushed organizations into a critical survival mode, necessitating profound changes to their operational frameworks. It shattered the notion of enterprises as static entities with unchanging procedures. To survive and thrive in this unpredictable environment, organizations must cultivate innovation, agility, and responsiveness, adeptly recognizing and adapting to changes.

To illustrate practical applications of this classification, the COVID-19 pandemic serves as an example of an unpredictable crisis. Within the context of the proposed classification framework, the COVID-19 pandemic is conceptualized as an occurrence-based external event. Originating beyond organizational boundaries, it is characterized by low predictability and high-magnitude disruptions across industries, aligning closely with the dimensions associated with storm-type environmental turbulence. The pandemic compelled organizations to engage in rapid, emergent responses rather than planned adaptations, thereby presenting an ideal context for examining the role of EA-driven improvisational capabilities. Accordingly, the COVID-19 case study is well situated within the category of occurrence-based, high-velocity change events identified in this study.

Therefore, this case study delves into the impact of EA-driven improvisational capabilities on organizational financial performance amidst the COVID-19 crisis, examining the supporting roles of flexible IT systems and organizational culture. A deductive quantitative method was employed to formulate the research model, with structural equation modeling (SEM) analysis applied to evaluate the proposed hypotheses.

4.1. Research Methodology and Sampling Approach

This study employs a quantitative research methodology, using survey data to examine the impact of change events on enterprise architecture (EA) in small and medium-sized enterprises (SMEs) in Pakistan. SMEs were selected as the primary focus due to their substantial economic contributions and vital role in employment generation within Pakistan’s developing economy. Contributing over 30% to the national GDP and employing approximately 78% of the non-agricultural workforce, SMEs are both economically significant and vulnerable to external disruptions. Given the dynamic, resource-constrained environment in which these firms operate, Pakistani SMEs provide a compelling context for analyzing adaptive capabilities, especially in light of recent economic fluctuations and global disruptions.

4.1.1. Sample Demographics

The study gathered data from a sample of 291 small and medium-sized enterprises (SMEs) across diverse sectors, including manufacturing, services, and technology, ensuring a comprehensive representation of industries impacted by change events. Respondents primarily comprised business managers and senior IT executives, recognized in the literature as key informants with a well-rounded perspective on both operational and IT-related challenges within their organizations [66,67]. A summary of respondents’ demographic characteristics is presented in

Table 5. Respondents’ demographic characteristics summary.

Category	Subcategory	Number of Employees	Percentage (%)
Years in Operation (Years)	10	154	52.92%
	15	106	36.43%
	20 and above	31	10.65%
Firm Size (Employees)	>50	58	46.05%
	>100	53	19.93%
	>150	46	18.21%
	>200	121	15.81%
Employee Experience (Years)	5~10	121	41.58%
	11~15	60	20.62%
	16~20	71	24.40%
	Above 20	39	13.40%
Educational Level	Higher secondary school (12 years)	105	36.08%
	Bachelor’s degree (14 years)	115	39.51%
	Master’s or higher degree	71	24.39%

.

4.1.2. Data Collection and Analysis

Data were collected targeting senior managers and executives of SMEs across Pakistan. The survey was distributed via two channels: email invitations sent to verified SME contacts using industry directories and professional networks and direct in-person engagement facilitated by academic–industry collaboration at our university. Many SME professionals regularly engage with the university as members of industrial advisory boards, guest lecturers in postgraduate programs, or participants in strategic consultation sessions. During these interactions, eligible respondents were invited to participate in the study. To ensure response quality, we applied role-based screening, attention consistency checks, and filters based on response duration. Covariance-based structural equation modeling (CB-SEM) was employed to analyze both direct and indirect relationships among latent variables. It was chosen as the primary analytical method due to its strong theoretical underpinnings and its ability to test complex causal relationships involving latent constructs. Unlike regression techniques, CB-SEM accounts for measurement error and enables simultaneous estimation of both direct and indirect effects. Given our objective to evaluate a theory-driven model involving mediation pathways and reflective constructs, CB-SEM was methodologically appropriate and aligned with best practices in structural modeling [68]. The Preacher and Hayes [69,70] bootstrapping approach was applied using 5000 samples at a 95% confidence interval. The significance of indirect effects was determined by examining whether the bias-corrected confidence intervals excluded the value of zero. Corresponding p-values were also reported. All analyses were performed using IBM SPSS AMOS version 23.

4.1.3. Variable Measurement

The measurement instrument used in this study was adapted from the validated scale developed by Kung and Kung [38], which was originally constructed through an extensive literature review and empirical testing. The instrument comprehensively captures constructs related to improvisational capability, IT flexibility, and organizational culture. Minor contextual modifications were made to tailor the items to the SME context and to address the specific operational challenges presented by the COVID-19 pandemic.

All items were rated on a 5-point Likert scale ranging from 1 = strongly disagree to 5 = strongly agree.

The details of each construct are provided below:

(a)

Independent variable

• Improvisational capability was measured using a seven-item scale reflecting three interrelated dimensions:
  • Speedy response: the ability to minimize the time lag between planning and execution.
  • Reconfigurability: the capacity to rapidly recombine and reuse available resources.
  • Novel solutions: the ability to generate creative, context-specific responses to unpredictable disruptions.
• These items capture the organization’s capacity to act flexibly and effectively during high-uncertainty conditions.

(b)

Mediating variable

• Two mediating variables, organizational culture and flexible IT systems, were considered in the analysis.
• Flexible IT systems were measured using a seven-item scale reflecting the following dimensions:
  • Modularity: the ease of decoupling and recombining digital infrastructure.
  • Connectivity: the extent to which IT systems can communicate and exchange data across functional boundaries.
  • Compatibility: the alignment between IT systems and operational processes.
• Organizational culture was measured using an eight-item scale comprising three dimensions:
  • Top Management Support: leadership commitment to responsiveness and innovation.
  • Innovation Orientation: organizational openness to novel ideas and experimentation.
  • Employee Learning and Training: the cultivation of skills that support adaptability.

(c)

Dependent Variable

• Organizational financial performance was treated as the dependent variable, measured through four items designed to assess the construct. These items were designed to evaluate the firm’s self-reported financial outcomes during the COVID-19 period, emphasizing stability, profitability, and goal attainment. The final items were as follows:
• OP1: Our financial performance has remained stable during the pandemic.
• OP2: Our profitability has been higher than that of our competitors.
• OP3: Our organization achieved its revenue targets during the pandemic.
• OP4: Our sales and profitability levels met or exceeded expectations.

These items were selected to reflect short-term financial resilience during crisis conditions.

4.2. Analysis and Results

The validity and reliability of the measurement instrument were rigorously tested, with findings detailed in

Table 6. Results of the measurement model.

Latent Variable	Indicator	F-L	AVE	CR
Improvisational Capability (IC)	IC1	0.763
	IC2	0.907
	IC3	0.747
	IC4	0.699	0.515	0.860
	IC5	0.603
	IC6	0.591
	IC7	0.539
Organization IT architecture flexibility (OIT)	OIT1	0.549
	OIT 2	0.761
	OIT 3	0.744
	OIT 4	0.632	0.5068	0.8769
	OIT 5	0.613
	OIT 6	0.760
	OIT 7	0.821
Organizational Culture (OC)	OC1	0.787
	OC 2	0.735
	OC 3	0.835
	OC 4	0.826
	OC 5	0.652	0.537	0.902
	OC 6	0.622
	OC 7	0.610
	OC 8	0.793
Organizational Performance (OP)	OP1	0.863
	OP 2	0.626	0.601	0.856
	OP 3	0.798
	OP 4	0.763

F-L: Factor Loading; AVE: Average Variance Extracted; CR: Composite Reliability.

. While most indicators exceeded the recommended threshold of 0.7 as suggested by Lohmöller [71], a few items exhibited loadings between 0.5 and 0.7, which are considered acceptable. According to Hair et al. [72], standardized factor loadings above 0.5 are acceptable for convergent validity, while Stevens [72] suggested that loadings greater than 0.4 are adequate for interpretation purposes. Similarly, Cheung et al. [73] emphasized that items with loadings above 0.5 may be retained if they are theoretically important and if the construct as a whole demonstrates acceptable convergent validity. Following these guidelines, all items were retained to preserve the theoretical comprehensiveness of the constructs.

While most indicators exceeded the recommended threshold of 0.7, per Lohmöller [71], a few items exhibited loadings between 0.5 and 0.7, which are considered acceptable. According to Hair et al. [74], standardized factor loadings above 0.5 are acceptable for convergent validity, while Stevens [72] suggested that loadings greater than 0.4 are adequate for interpretation purposes. Similarly, Cheung et al. [73] emphasized that items with loadings above 0.5 may be retained if they are theoretically important and if the construct as a whole demonstrates acceptable convergent validity. Following these guidelines, all items were retained to preserve the theoretical comprehensiveness of the constructs.

In this study, structural equation modeling (SEM) was employed as a statistical technique to test the proposed hypotheses. SEM follows a two-step approach, consisting of confirmatory factor analysis (CFA), also known as the measurement model, and the structural model. CFA, an integral part of SEM, was conducted prior to testing the structural model to evaluate the accuracy of the measurement properties within the hypothesized model using fit indices. Specifically, CFA was utilized to assess unidimensionality, following the guidelines of Joreskog and Sorbom [75]. The favorable fit indices presented in

Table 7. Model fit results of the confirmatory factor analysis.

χ2	DF	χ2/DF	GFI	AGFI	CFI	RMSEA	RMR
420.545	240	1.745	0.901	0.89	0.907	0.072	0.704

indicated that the measurement model was well suited for the data, thereby justifying the progression to structural model testing using covariance-based structural equation modeling (CB-SEM).

Discriminant validity was evaluated employing the Fornell and Larcker [76] criterion, with findings detailed in

Table 8. Discriminant validity of constructs.

Variables	IC	OIT	OC	OP
Improvisational Capability (IC)	0.718
Organization IT architecture flexibility (OIT)	0.570	0.711
Organizational Culture (OC)	0.554	0.698	0.733
Organizational Performance (OP)	0.616	0.540	0.463	0.775

. This involved comparing the square root of the average variance extracted (AVE) for each construct (displayed on the diagonal) against the correlation coefficients (off diagonal) corresponding to each construct across the relevant rows and columns. The analysis confirmed that all diagonal values exceeded those in the corresponding rows and columns below them, thus satisfying the requirements for discriminant validity.

The second component of SEM is the structural model, which represents a set of dependency relationships among the variables examined in this study. The purpose of the structural model is to determine the relationships that exist between the study’s constructs [74]. In this study, SEM was employed to test the hypotheses developed based on the theoretical framework.

As presented in

Table 9. Testing hypotheses H1, H4 & H5.

Hypothesis	Relationship	Beta-Value	S.E.	Z2	p-Value	Remarks
H1	Imp capability → Org financial Performance.	0.758	0.226	3.361	0.001	Supported
H4	Org Culture → Org financial Performance.	0.101	0.131	0.766	0.444	Not-Supported
H5	Flexible IT systems → Org financial Performance.	0.614	0.131	4.672	p < 0.001	Supported

, improvisational capability (β = 0.758, p = 0.001) and flexible IT systems (β = 0.614, p < 0.001) demonstrated a significant and positive effect on firm organizational performance. In contrast, the effect of organizational culture (β = 0.101, p = 0.444) was not statistically significant. Consequently, Hypotheses H1 and H5 were supported while Hypothesis H4 was not.

To test Hypotheses H2 and H3, we examined the mediating effects of organizational culture and flexible IT systems on the relationship between improvisation capability and organizational financial performance. Specifically, these hypotheses investigate whether the presence of a supportive organizational culture and flexible IT systems strengthens or weakens the impact of improvisation capability on financial outcomes.

Table 10. Testing hypotheses H2 & H3.

Hypothesis	Relation	Indirect Effect	95% Bootstrapped CI	p-Value	Remarks
H2	Imp capability → Org Culture → Org financial Performance	0.260	−0.041, 0.511	0.112	No-mediation
H3	Imp capability → Flexible IT → Org financial Performance	0.03	0.011, 0.064	0.018	Partial mediation

presents the results of mediation testing for H2 and H3, including indirect effect estimates, bias-corrected bootstrapped 95% confidence intervals, and p-values. A mediation effect is considered statistically significant if the confidence interval does not include zero.

The analysis of Hypothesis H2 reveals that organizational culture does not mediate the relationship between improvisational capability and organizational financial performance. Although the total effect is significant (p = 0.01), the indirect effect is non-significant (p = 0.260), indicating that organizational culture does not act as a mediator in this relationship. In contrast, the analysis of Hypothesis H3 shows that a flexible IT system exhibits partial mediation in the relationship between improvisational capability and financial performance. The total effect (p = 0.001), direct effect (p = 0.001), and indirect effect (p = 0.03) are all statistically significant, suggesting that flexible IT systems partially mediate the impact of improvisational capability on organizational financial performance. These results indicate that while organizational culture does not play a mediating role, the presence of flexible IT systems enhances the effectiveness of improvisational capability in improving financial outcomes.

5. Discussion

The findings of this study offer important insights into how enterprise architecture (EA)-driven capabilities influence SME financial performance under conditions of environmental uncertainty. Improvisational capability and flexible IT systems exhibited strong positive effects on organizational financial performance, supporting H1 and H5. These results underscore the critical roles of agility, responsiveness, and technological flexibility in SMEs operating in volatile environments.

However, not all the relationships were equally strong. Specifically, the role of organizational culture (H4) was not supported, exhibiting a low standardized regression coefficient (β = 0.101, p > 0.05). One possible explanation is that during periods of acute environmental turbulence, immediate operational flexibility and adaptive capabilities may play a more critical role in financial resilience than underlying cultural factors. Under highly uncertain conditions, firms may prioritize rapid decision making and resource reconfiguration over embedded cultural norms, thereby reducing the observable impact of organizational culture on short-term financial outcomes.

This finding contrasts with several studies that highlight the enabling role of organizational culture in supporting adaptability and performance, especially in dynamic environments [77,78]. Our study suggests that while culture is theoretically important, its influence may be temporally contingent and less visible during immediate crises and more active during periods of long-term recovery or transformation. This context-specific insight adds to the literature by illustrating that cultural mechanisms may not be uniformly activated across all types of uncertainty.

Moreover, the mediation hypothesis involving organizational culture (H2) was also not supported, indicating no significant mediating effect between improvisational capability and financial performance. Although organizational culture is often recognized as an enabler of dynamic capabilities, this study suggests that in highly volatile and disruptive environments, cultural norms may not be sufficiently activated to influence immediate organizational outcomes. During critical incidents, the urgency for rapid action may override slower, deeply embedded cultural routines, causing improvisational responses to bypass formal cultural pathways. Consequently, culture may play a more supportive or long-term role rather than directly mediating short-term financial resilience. Future research could further explore this phenomenon through qualitative interviews or case studies, examining how different types of organizational culture (e.g., hierarchical vs. adhocracy) influence improvisational capabilities under conditions of extreme uncertainty.

6. Limitations and Future Research

Although this study provides important insights into the role of EA-driven capabilities in fostering financial resilience among SMEs, several limitations should be acknowledged. First, the sample is geographically limited to SMEs operating in Pakistan, which may constrain the generalizability and external validity of the findings to SMEs in other developing or developed economies. Future research could expand this investigation to different regional, cultural, and economic contexts to assess the broader applicability of the classification framework.

Second, while AMOS-SEM was employed for hypothesis testing, future studies could apply PLS-SEM to validate the robustness and predictive relevance of the proposed model. Additionally, post hoc analyses could be conducted to examine potential moderating effects that might explain variations in the strength of relationships across different organizational settings.

This study focuses on a single category of change triggers. Future research should aim to test the broader typology proposed here, including internal and digital-event-driven changes across different organizational contexts and time horizons.

Finally, given the unexpected non-significant role of organizational culture in mediating financial performance, qualitative research approaches, such as case studies or interviews, could offer deeper insights into how different types of organizational culture interact with improvisational capabilities during environmental turbulence.

By addressing these avenues, future research can build a more comprehensive understanding of how enterprise architecture enables organizational resilience and performance under conditions of uncertainty.

7. Conclusions

Overall, this study advances the understanding of how enterprise architecture (EA)-driven capabilities, particularly improvisational capability and flexible IT systems, contribute to financial resilience in SMEs facing environmental uncertainty. By introducing a structured classification of EA change events and linking these to adaptive capability responses, the study makes both theoretical contributions and practical implications for sustainable organizational development.

While the literature has extensively examined EA-driven dynamic capabilities, particularly in the contexts of digital transformation [20,22], platform development [25], and process agility [3,26], research specifically focused on improvisational capabilities remains relatively underdeveloped. Moreover, most prior studies have concentrated on advanced economies, such as the United Kingdom [27], South Korea [28], and China [29], overlooking the unique challenges and adaptive responses of SMEs in developing nations such as Pakistan. This study addresses that gap by showing that, in resource-constrained and disruption-prone environments, improvisational capability and IT flexibility are more influential in enhancing financial performance than formalized cultural systems.

The findings challenge conventional assumptions about the universal applicability of organizational culture as a mediator in adaptive processes, emphasizing the importance of contextual sensitivity in EA research. For practitioners, this underscores the need to prioritize flexible infrastructure and real-time responsiveness. For scholars, this work opens avenues for further exploration, particularly through qualitative research into culture-adaptation dynamics and comparative studies across regions and industry sectors. As organizations worldwide face increasing uncertainty, designing EA models that enable agile, capability-driven adaptation will be essential for long-term sustainability.