The Agile PMO Paradox: Embracing DevOps in the UAE

Abstract

This study investigates how Development and Operations (DevOps) practices impact Project Management Office (PMO) governance within the technology sector of the United Arab Emirates (UAE). It addresses the need for agile-aligned governance frameworks by exploring how DevOps principles affect traditional PMO structures. A quantitative cross-sectional survey was conducted, and data was collected from 321 DevOps and PMO professionals in UAE organizations. The analysis, using Partial Least Squares Structural Equation Modelling (PLS-SEM), revealed a moderate positive correlation between specific DevOps practices—such as microservices, Minimum Viable Experience (MVE) culture, continuous value streams, automated configuration, and continuous delivery—and effective PMO governance. The study’s novel theoretical contribution is the integration of the Dynamic Capabilities Framework (DCF) with the Agile DevOps Reference Model (ADRM) to examine this alignment, bridging strategic agility and operational execution. This research offers actionable insights for UAE organizations and policymakers seeking to enhance governance and digital maturity.

1. Introduction

Recent years have ushered in unprecedented technological advances that have indelibly changed how project development and management function, particularly in software engineering. The transformation that has led to DevOps presenting itself as a key driver has accelerated between the years 2012 and 2017 [1,2]. While DevOps enables efficiency and responsiveness in software development [3], it may face challenges when integrated with traditional project management structures like PMOs [4].

Traditionally, PMOs have been responsible for project governance, coordination, and portfolio management [5,6], but they often struggle to work with the agile and incremental nature of DevOps and may need to change their frameworks [7,8]. There is a need to re-evaluate PMO governance to accommodate the pace and fast iteration of DevOps practices [9,10,11]. Integrating DevOps into traditional PMO structures can be likened to rewiring an organization’s nervous system—shifting from rigid, top-down control to adaptive, real-time responsiveness. Much like the transformation from a compass to a GPS, this evolution reflects the move from static governance to dynamic coordination in high-velocity environments [12]. Despite these challenges, the successful integration of DevOps into PMO governance brings several benefits, including increased adaptability, enhanced communication, and faster project delivery [13,14].

The UAE represents a distinctive and highly relevant environment for studying this integration. Economically, it has positioned itself as a global technology hub, with strategic national agendas such as UAE Vision 2031 and the Centennial 2071 initiative explicitly prioritizing digital transformation and innovation. These initiatives make the UAE a model for other emerging economies and justify its relevance to the international community. Socio-culturally, the UAE’s workforce is highly diverse, with over 70% expatriates [15], creating both opportunities for cross-cultural innovation and challenges in communication and collaboration. Furthermore, the government actively promotes technology adoption through initiatives like Digital Dubai and the UAE AI Strategy, making it an ideal setting to examine agile governance adaptation.

Although there is abundant research on DevOps and its implications for PMOs, little attention has been paid to finding ways of working out DevOps within the modern traditional PMO structure. In the UAE, this gap is especially important given its high pace of technological advancement and a competitive landscape where organizations must take agile approaches to remain leaders [16,17]. With the UAE seeking to take on global technological leadership [18,19], IT-oriented organizations should understand the intricate cooperation between DevOps and PMO governance. This study addresses this gap by exploring the following research questions:

RQ01: How do DevOps practices impact traditional PMO governance practices within the UAE technology sector?

RQ02: What are the key challenges and opportunities PMOs face when adopting DevOps practices?

RQ03: What strategies can PMOs implement to effectively integrate DevOps practices into their governance frameworks?

These research questions are directly linked to the methodology by which we test our hypotheses, which are derived from the theoretical frameworks. The quantitative data collected, analyzed through PLS-SEM, directly provides data-driven answers to these questions in the Results and Discussion sections, particularly concerning the impact of specific DevOps practices on PMO governance.

2. Theoretical Framework

To conceptualize how organizations can effectively adapt their governance mechanisms in response to rapid technological change, this study employs the Dynamic Capabilities Framework (DCF) as a strategic foundation. As illustrated in Figure 1, the DCF, first introduced by Teece, consists of three interrelated dimensions—Sense, Seize, and Transform—that collectively enable firms to reconfigure their competencies and maintain agility in volatile environments [20,21]. The Sense capability refers to the identification of emerging technological opportunities and external shifts; Seize focuses on designing and refining business models, allocating resources, and anticipating competitive reactions; while Transform emphasizes the reconfiguration of internal structures, including governance, to sustain innovation. This framework offers a macro-level lens to interpret how DevOps practices, when aligned with PMO governance, can enhance an organization’s ability to respond to and capitalize on technological disruptions [22,23].

Building on this strategic foundation, the study integrates the Agile DevOps Reference Model (ADRM) to examine the operational dimensions of DevOps-PMO alignment. The ADRM consists of four interconnected layers—Values, Principles, Dimensions, and Practices—that together provide a comprehensive framework for embedding DevOps across both cultural and technical domains. This layered structure enables organizations to systematically translate agile philosophies into executable actions. Focusing particularly on practices such as microservices, automated configuration, and continuous delivery, the ADRM allows for a granular assessment of how DevOps capabilities influence governance outcomes such as visibility, accountability, and strategic alignment. In this way, the ADRM complements the DCF by offering an operational roadmap for executing agile transformations, thus forming a cohesive framework that links high-level strategic adaptation with on-the-ground agile implementation [23,24].

Together, the DCF and ADRM form a cohesive analytical framework—while the DCF offers a strategic scaffold for understanding how organizations develop, orchestrate, and reconfigure capabilities in dynamic project ecosystems, the ADRM provides the operational detail necessary to assess how DevOps principles are enacted within PMO governance structures. This dual-framework approach enables a comprehensive examination of both the strategic intent and the practical execution of agile transformations, particularly in rapidly evolving technology environments such as the UAE [22,23,24].

The main purpose of this study is to examine the link between various DevOps practices (e.g., microservices architecture, MVE culture) and efficient PMO governance. Additionally, it endeavors to investigate how modern technological trends, such as CI/CD methodologies, automation, machine learning, and artificial intelligence, are used to forge contemporary software development methods with governance models [2,5,20]. Considering the relevance of these objectives, this research bridges the knowledge gaps that exist currently and provides valuable insights into how PMOs in the UAE technology industry adapt and promote DevOps adoption in the industry.

This study provides both theoretical and practical implications. Accordingly, it advances the theory of DevOps and PMO governance by providing theory (gaps in understanding) regarding the interplay of these two fields. In addition, it contributes to the current debate related to agile project management adoption [25,26].

3. Literature Review

3.1. Related Work on DevOps and PMO

The integration of DevOps and Project Management Office (PMO) governance is a developing field, with much of the existing literature focusing on each domain individually rather than their intersection. Prior research has predominantly explored the benefits and challenges of adopting DevOps practices in various contexts. Studies by Anjaria & Kulkarni, Gwangwadza & Hanslo, and Hamza, provide systematic reviews of DevOps adoption, outlining key success factors, challenges, and benefits like improved efficiency and faster delivery times [8,27,28]. For instance, some research highlights that the cultural shift required for DevOps can be a major barrier. Other work, such as that by Alenezi in the Saudi software sector, points to a lack of organizational sponsorship and excessive administrative constraints as key hindrances to DevOps adoption [5].

Concurrently, a separate body of literature examines the evolution of PMOs from traditional, bureaucratic structures to more agile and adaptive forms. Researchers have explored the concept of the Agile PMO, where the PMO’s role shifts to training, empowering agile teams, and implementing flexible procedures to support project objectives [29,30]. However, this work often focuses on general agile methodologies rather than the specific, technical practices of DevOps. For example, studies discuss balancing flexibility with strategic alignment and navigating external pressures, but they do not provide a detailed, operational framework for this integration [31].

Despite this foundational research, a significant gap remains. The existing literature lacks a comprehensive, empirically backed model that connects specific DevOps practices directly to their impact on PMO governance within a dynamic, high-growth technology environment [29,30,31]. While some research notes the importance of adapting frameworks and architectures for DevOps, and others acknowledge the need for leadership and government support, these studies do not provide a structured, testable framework for the interplay between DevOps practices and governance outcomes. This study fills this gap by integrating the Dynamic Capabilities Framework (DCF) with the Agile DevOps Reference Model (ADRM) to provide a robust conceptual and analytical framework. We empirically test how practices detailed in the ADRM (Microservices, Minimum Viable Experience culture, Continuous Value Stream Integration, Automated Configuration, and Continuous Delivery) act as the “how” for achieving the strategic “what” of the DCF, thus providing a concrete, measurable link to effective PMO governance in the UAE technology sector.

3.2. DevOps in Current Application Trends

DevOps is a word derived from the union of two terms which are development and operations, and it is multidisciplinary strategy to facilitate working integration and cooperation between various software developers and IT operations. This methodology focuses on automation, integration and delivery of software to improve the process of software development [28]. The principles of DevOps are focused on the value of interacting teams, more productivity, and faster delivery time, accompanied by increased software quality [32].

The subjects that make up the DevOps knowledge domain include automation technologies, Continuous Integration and Continuous Delivery (CI/CD), Infrastructure as Code (IaC), and monitoring and logging. These components are useful for enabling high-speed, efficient, and safe delivery of software without the compromise of quality of the software [33]. But some issues are typical for organizations that implemented DevOps, such as culture change, personnel scarcity, and problems with the transition to new tools [34].

However, DevOps has some serious barriers to face when being implemented. One of the five challenges outlined is the unwillingness of organizations to foster a culture that would embrace DevOps [28]. Moving from a functional organizational structure, the transition to cross-functional teams needs a lot of work on change management due to the employees’ resistance to such changes based on previous norms/standards. Moreover, another challenge is the lack of enough people with DevOps skills and skills that meet organizational needs. Research shows that only a mere 28–40% of companies face issues with implementing DevOps into their operations [5]. On the other hand, the implementation of security in the integration of application development and deployment, known as DevSecOps, adds several more complications, as security should be taken into consideration in each stage of SDLC (System Development Life Cycle) [35].

In addition, DevOps organizations in the world are investing more in automation and adopting artificial intelligence and machine learning into their practices. IT operations have been transformed into a powerful tool known as AIOps that helps organizations use data for analytics, especially for the predictive cycle, including maintenance and incidents. It reduces time spent on technical issues, which in one way or another reduces the quality of services [29]. Moreover, the microservices architectural pattern is inevitably emerging, with the ability to build and deploy components with application independence, in harmony with the fundamental concept of DevOps [36].

The current body of literature on Agile PMOs often emphasizes global best practices but lacks focused attention on emerging markets such as the UAE. To address this gap, this manuscript integrates a more critical analysis of foundational studies (e.g., high-impact works on Agile maturity models and DevOps practices) and evaluates their relevance in dynamic, innovation-driven economies.

The UAE, with its strategic focus on digital transformation, innovation, and national initiatives such as UAE Vision 2031, provides a unique context where Agile PMOs can significantly influence organizational agility, technology adoption, and value-driven delivery. By situating the research within the UAE’s socio-economic and technological landscape, this study extends existing theory while offering practical insights for technology governance and portfolio management.

Currently, the UAE is experiencing a trend toward increasing the use of DevOps since companies are striving for digital transformation. The UAE government is paving this way by promoting innovation and technology and has been calling on organizations within the dev op sectors, both from the public and private domain [36]. A survey conducted in early 2022 shows that more than 60% of the organizations based in UAE are either already using or are planning to implement DevOps in the coming year, reaffirming the organizations’ focus on organizational and operational improvements [10]. In addition, the adoption of cloud computing with DevOps is on the rise in UAE organizations since they understand the benefits of cloud environments. This is in line with the shift in organizations towards multi-cloud solutions, whereby a company integrates cloud services from different providers to gain better efficiency and relay [1].

Despite the vast opportunities available for improving application development with DevOps and optimizing business operations, there are certain drawbacks to its execution. Challenges arise based on issues such as embracing changed culture, lack of skills, and challenges when there is a need to integrate this new technology into the existing infrastructure [28,33,37,38]. However, UAE’s active step towards the enhancement of its strong DevOps culture makes it ready to reap the benefits of this dynamism methodology in response to the fluctuating global market. Further research should be conducted to better understand how organizations can implement DevOps as well as what success factors can be used to measure the implementation of DevOps practices, with emphasis being placed on technology-based economies such as the UAE.

3.3. DevOps in UAE

From the existing literature on DevOps in the UAE, it was established that there are complex interactions between technology, culture, and human capital systems informing its implementation. This analysis shows that different opportunities and threats arise from the UAE’s socio-cultural and technological environment.

3.3.1. Diversity as a Double-Edged Sword

Due to an extremely high percentage of foreign workers, over 70% in the UAE case [15], enhanced workforce management remains a major contradiction when adopting DevOps. While it encourages revenue generation through cross-cultural invention, it also has brought concerns about communication gapping. Intercultural communication is therefore significant in DevOps since a lack of efficiency might hinder the expected teamwork in a culturally diverse environment.

3.3.2. Integrating DevOps into Education for Workforce Readiness

Mason et al. [39] pointed out that DevOps principles should inform educational frameworks, notably those based on Agile. This is particularly important in UAE because to resolve conflicts that arise with the different teams, a skilled and competent workforce that is well equipped with modern software development practices is required. Additionally, they mentioned the need for Adaptive Management Frameworks like Agile Scaling Technology (AST), where most of the teams are either remote or geographically distributed as they are in the UAE.

3.3.3. Adapting Frameworks and Architectures for DevOps

Ghantous and Gill [26] suggest a reference framework by which to implement Agile-DevOps in education while noting its importance as a systems approach. Despite this, several issues are noted concerning aligning the fluidity associated with DevOps with the more formal structures seen in IT service provision, as discussed by Maroukian & Gulliver [40]. This tension is best illustrated in the UAE context, where invention is always met with a restriction made by the organization structure, legal framework, or culture.

3.3.4. Navigating Obstacles and Embracing Cultural Shifts

In a study of the Saudi software sector done by Alenezi [5], some barriers that are apt in the case of the UAE include the absence of organizational sponsorship and excessive administrative restraints. These negatives undermine the responsiveness and elasticity that DevOps embraces. On the other hand, Gwangwadza and Hanslo [27] spoke of cultural adaption in DevOps, which corresponds with UAE’s diverse workforce and increasing need for enhanced teamwork in organizational institutions.

3.3.5. Prioritizing Technological Innovation and Security

According to Bildirici and Akdemir [13], effective software release management practices are necessary to avoid potential risks, and the UAE underlines them. Hossain Tanzil et al. [31] rightly point out that automation and technical adaptability are imperative for organizations that must sustain competitive advantage in a dynamic market, propounded in the context of the UAE technology plan. Moreover, security is a crucial factor. Morais et al. [41] show that it is becoming incorporated with DevOps, which is known as DevSecOps. It makes sense with the ongoing emphasis on security in the UAE, as it seeks to protect its economy and digital systems.

3.3.6. The Role of Government Initiatives and Leadership

Karampatsis et al. [42] explain that enthusiasm in leadership with the vision and sponsorship of technology has been greatly propelling advancement in the UAE. Likewise, engagements such as Digital Dubai and the UAE AI Strategy show that the government is working towards using DevOps-compatible frameworks to transform the public sectors. These initiatives offer logical support for the adoption of DevOps by making it easier for organizations to obtain enhanced tools and industry standards.

3.3.7. Addressing Skills Shortage and Investing in Human Capital

The skills gap threat is global, and the UAE remains committed to addressing it. To address this problem, it is essential to fund reskilling activities, including education and training, which are tailored to produce the desired relevant skills. Schtein [1] has also argued that the creation of a learning culture can be regarded as the most obvious step in preparing a workforce for operations in a DevOps environment.

The decision to adopt DevOps in UAE is therefore a complex one, exercised by various factors including diversity of the workers, education and training, technology and policies. Obstacles, such as global regulation, skills deficit, and economic volatility, are always there; however, UAE leaders showed succession to innovation and strategic development to harness the development opportunities offered by DevOps adequately.

3.4. The Agile Transformation of PMO Governance

The adoption of Agile practices has changed the structure of PMO greatly and shifted it from bureaucratic and big practice to lean practice. This evolution focuses on the changes that are characterized by flexibility, constructiveness and learning to meet contemporary project challenges.

3.4.1. Evolving Governance Roles and Structures

Lowrance [43] first introduced the idea of Agile PMOs, where the PMO trains the employees, empowers the Agile teams, and adopts flexible procedures for fulfilling the objectives of Agile governance. This marked work established the truth concerning Agile methodologies in relation to customers and the time to market, as well as the foundations for alteration of the governance activities. Khan et al. [30] developed this view further in their study while perceiving PMOs as the change enabler which is intrinsically flexible. They stressed the role of the PMO framework functions in terms of the emerging objectives of modern business contexts in relation to projects.

3.4.2. Balancing Flexibility and Strategic Alignment

Gómez González and Vargas [22] developed small and medium enterprises’ Agile PMO framework and mentioned that PMOs should be both embedded and emergent. This strategy is characterized by sustainability and market orientation. Philbin [44] emphasized that it is not effective to impose strict guidelines starting from the top level of the organization; at the same time, they also stated that an association should have its’ distinct strategy it would like to follow whilst being as flexible as possible at the same time.

3.4.3. Navigating External Pressures and Cultural Shifts

Though Agile encourages the devolution of power, in teams having a low maturity level of Agile, authors Nkukwana and Terblanche [45] found they use bureaucratic structures of governance more often. This leads to the rather significant issue of whether team orientation or better controlling mechanisms should be chosen. They accepted the objective benefits of Agile in terms of project delivery but also pointed out the difficulties that many organizations face in incorporating the iterative characteristic of Agile into their current PMO paradigms.

3.4.4. Adapting to Future Trends and Maintaining Momentum

To achieve this, Menon et al. [46] examined the relationship between Agile and software architecture, calling particular attention to architecture decisions. In their study, Fawzy et al. [25] pointed out that real-time analytics and cloud solutions play a crucial role in Agile data governance. Collectively, these studies show that PMO governance continues to transform to address Agile principles, which emphasise teamwork, creativity, and flexibility.

For agile methodologies to be implemented effectively within organizations there is need to transform the governance structure of PMO. Adopting flexibility, decentralisation, and the focus on delivering and maximising value, PMOs can mitigate the threats resulting from contemporary projects and create consistent value for an organisation based on its objectives.

3.5. Synergizing the Dynamic Capabilities Framework and Agile DevOps Reference Model

The Dynamic Capabilities Framework (DCF), in combination with the Agile DevOps Reference Model (ADRM), enables organizations to act accordingly to tackle dynamic contexts [47,48]. Agreeably, the DCF emphasizes organizational change correspondingly with Agile DevOps principles, enhancing flexibility and innovation [49]. This synergy starts with what is referred to as “sensing,” where organizations generate and gather knowledge not only on internal operational changes but also on the market environment. Agile DevOps further refines this by achieving comprehensive performance monitoring systems that are vitalizable [42]. By using dashboards as well as tools that indicate access to a particular system, every stakeholder can see trends and anomalies. The integrated strategy where developers study logs and support teams handle user complaints helps in a total evaluation of system function [30,50]. A non-centralized plan to monitor mass infrastructure guarantees the ongoing insight of the system state irrespective of the failures. This resilience is necessary for monitoring and ensuring functionality and for identifying points of possible problems.

Combining product performance with great options for monitoring the system grants accurate readings from the system. Scalability means that no factors limit performance, and flexible tools enable users to retrieve the data that they require in their work or study [50]. High-velocity teams also share information at a very fast pace and can make changes in the environment to develop new systems for measuring data or even improve on the current ones available. Monitoring methods remain dynamic because they are both deployable and modifiable throughout their use in different settings [37].

The usefulness of monitoring systems can reach its peak only in case the possibility of rigorous testing and inclusion of automation is investigated. This process of calibration of all the monitoring equipment guarantees that the alarms provided are accurate and that the measurements attained are genuine [38]. Regular checking of sensing systems optimizes their practical functionality and reduces the risk of failure that may compromise other significant activities. In addition, organizations are also using AI-aided analytical applications and real-time data warehouses to gain insights to analyze and possibly flag their outliers [31].

Thus, organizations that integrate these various systems achieve better sensing. This integration allows for an increased range of available data, makes it possible to operate in conditions other than optimal, and guarantees the ability to handle disruptions and shocks [51]. Combined, DCF and ADRM help organizations control available resources as they adapt to change occurring in business environments. For that reason, this research underlines the relevance of Agile DevOps to enhance and sustain dynamic capability. This can be seen in

Table 1. Integration of Dynamic Capabilities Framework (DCF) with Agile DevOps Reference Model (ADRM).

Sensing (DCF)	Seizing (DCF)	Seizing (DCF)	Seizing (DCF)	Transforming/ Reconfiguring (DCF)
Dimension (ADRM)	Principle (ADRM)	Practice (ADRM)	Culture	Value (ADRM)
Ease & Simplicity	Microservices Approach			Collaboration [Improved Pace of Delivery]
Removal of SPOF
Availability & Capacity
Tailor-Made Functionalities
Team Velocity
Agility			Minimum Viable Experience (MVE) Culture	Adaptation [Continuous Improvement]
Learning Curve
Knowledge Management
Expanding Capabilities
Customer Experience (CX)
Dynamic Coding		Continuous Value Stream Integration		Value Delivery [Enhanced Quality & Reliability]
Central Repository
Lean Time Metrics [Value Added (VA) & Lead Time (LT)]
Completion & Accuracy [%Complete/Accurate (%C/A)]
Lean
Programmability		Automated Configuration [Infrastructure as Code (IaC)]		Automation [More Efficient & Effective Operations]
Idempotence
Version Control
Standardized Patterns
Performance Measurement
Deployability		Continuous Delivery		Outcome-focused [Deployment Artifact (Build, Test, Release)]
Modifiability
Testability
Automated Testing
Emerging Technology Adoption

.

3.6. Research Framework

To examine the PMO governance within the DevOps environment, this research adopted the Dynamic capabilities framework (DCF), integrating the concepts culled from the Agile DevOps reference Model (ADRM). Sensing, seizing, and transforming/reconfiguring the framework in the research hypothesis gives some insights about organizational changes in the governance that support the essence and success of DevOps.

Framework Constructs

This study examines the three core dynamic capabilities—sensing, seizing, and transforming—within the context of DevOps and PMO governance:

Sensing Capabilities: Sensing denotes the ability of an organisation to detect opportunities and threats within its internal and external context [52]. This capability leverages some key constructs of the Agile DevOps Reference Model (ADRM) that include ease & simplicity, no single point of failure, agility & customer experience (CX). These factors help to get information regarding changes exhibited by technologies, customers’ requirements or even working processes. For instance, willingness to change as a key attribute embrace, emerging technologies need executives to prepare to innovate while knowledge management guarantees the perpetual flow of new information.

Seizing Capabilities: Opportunities refer to available resources that should be harnessed for the achievement of organisational objectives, seizing entails the optimal utilisation of these opportunities. This stage is driven by five key constructs: microservices, the Minimum Viable Experiment (MVE) culture, Continuous Value Stream Integration (CVSI), automated configuration and finally following methodologies; CD. These concepts of ADRM and DCF are based on automation, modularity and agile deployment life cycle to make it more operational for heterogeneous environment [52].

Transforming/Reconfiguring Capabilities: Transforming relates to the realigning of resources to sustain competitive advantage [52]. Five ADRM outputs are collaboration, adaptation, value delivery, automation and outcome-focused metrics that show culture and operation adaptability. These factors allow organizations to build a flexible environment through which they can always be implementing new changes and remodifying governance structures.

Building on the integration of the Dynamic Capabilities Framework (DCF) [52] and the Agile DevOps Reference Model (ADRM) [48], this study conceptualizes five hypotheses to examine the relationship between DevOps practices and PMO governance effectiveness. DCF provides a strategic lens through which organizations can sense, seize, and transform their capabilities into rapidly changing environments. The ADRM complements this by detailing operational practices that align with Agile and DevOps principles. Together, these models form a robust foundation for analyzing how DevOps practices influence PMO governance.

4. Method and Instruments

This research engaged a positivist paradigm to employ quantitative research to assess the level of PMO governance in DevOps projects. The research employs mostly developed hypotheses from both the Dynamic Capabilities Framework (DCF) and Agile DevOps Reference Model (ADRM) hypothesis testing. Particularly, a survey strategy was employed because, during this process, data can be collected from numerous and heterogeneous groups of respondents, and it is appropriate for quantitative research. The cross-sectional approach assists in developing a snapshot look at the current samples and identifying the factors that impact on the use of PMO governance in the technology industry within the UAE.

4.1. Data Collection Techniques and Procedures

This study relied on both primary and secondary data. Primary data were gathered using a questionnaire obtained from 321 participants consisting of DevOps practitioners and PMO professionals, who were purposively identified due to their job relevance and experience. The survey link was shared with 856 potential respondents using LinkedIn and email, and 321 completed the survey, giving a 34.26% response rate. Participants were recruited through pre-survey screening to ensure they met specific criteria, including their professional roles, industry experience, and their affiliation with organizations that adopt DevOps. Follow-up questions were offered to clarify complex questions and ensure the validity of the structured questionnaire answers. These follow-ups were open-ended and designed to gather qualitative data. Their purpose was to clarify complex responses and validate the data from the Likert-scale questions, providing deeper context for the findings. The follow-up questions sought to:

• Provide Specific Examples: Instead of just rating their agreement with a statement, participants were asked to share concrete examples from their work experience. For instance, after rating a question about the benefits of microservices, a follow-up asked them to describe a specific time it improved a project’s speed.
• Explore Challenges and Context: The questions were used to probe the “why” and “how” behind a response. For example, if a participant indicated that their organization’s culture was a barrier, a follow-up was to elaborate on the specific cultural issues they faced.
• Connect Perceptions to Practice: The follow-ups helped bridge the gap between a participant’s subjective perception and the tangible reality of their organization’s DevOps and PMO practices. This was expected to help the researchers understand the practical implications of the survey findings.

Secondary data consisted of a systematic analysis of academic articles, business reports, and influential models. This systematic analysis involved a targeted literature search using keywords related to DevOps, PMO governance, and the theoretical frameworks. A model was considered “influential” based on its citation count in peer-reviewed academic databases and its repeated use in a wide range of academic and industry publications. This multi-faceted data collection strategy provided maximum and diverse approaches to the analysis of the research problem.

4.2. Participant Profile

The survey included demographic questions to create a detailed participant profile. Respondents were required to have at least two years of experience in PMO or DevOps roles within UAE-based technology organizations. The data collected included their current role (e.g., DevOps Engineer, Project Manager, IT Director), years of experience, and the industry sector (e.g., Finance, Telecommunications, Government). This information was crucial for contextualizing the findings and allowing other researchers to replicate the study.

4.3. Instrumentation and Framework Constructs

The survey consisted of five structured sections: (1) Introduction and consent statement; (2) Demographics; (3) Likert-scale items (1 = Strongly Disagree to 5 = Strongly Agree) assessing constructs from the DCF and ADRM frameworks; (4) Open-ended questions on organizational DevOps adoption challenges; and (5) Closing remarks.

Questionnaire Design

The questionnaire was designed based on the theoretical foundations of the DCF and ADRM frameworks. The frameworks themselves do not provide pre-existing questionnaires; therefore, the survey items were carefully developed to operationalize the key constructs. Each question was crafted to measure a specific facet of a construct, ensuring that the collected data accurately reflected the theoretical model. For example, questions related to Automated Configuration (AC) were designed to capture the participant’s perception of the level of automation within their organization’s infrastructure. Similarly, questions related to Continuous Delivery/Deployment (CD) assessed the perceived speed and frequency of their software release cycles. The open-ended questions were grounded in a review of existing literature on common challenges in DevOps adoption, allowing participants to share specific experiences that complemented the quantitative data.

This study examines five core constructs based on the DCF and ADRM frameworks:

• MS (Microservices): This construct measures the adoption of a microservices architecture.
• MVE (Minimum Viable Experience) Culture: This measures the cultural emphasis on rapid, iterative learning and experimentation.
• CVS (Continuous Value Stream) Integration: This assesses the alignment of development and delivery pipelines with business value.
• AC (Automated Configuration): This measures the use of automated configuration, including Infrastructure as Code.
• CD (Continuous Delivery/Deployment): This assesses the adoption of practices for rapid and frequent software releases.

Table 2. Measurement Items and Sources.

Theme	Variable	Adapted Source, Reference	Indicator	Questions
DevOps’ Microservices approach (MS1)	Ease & Simplicity	[53,54,55,56,57]	MS1.1	The modular architecture of ‘microservices’ in DevOps eases the scaling up of the system and since each service is designed, developed, and deployed independently, a seamless functioning and update process of the software system is enabled, thus resulting in easier and quicker deployment.
	Removal of SPOF		MS1.2	The culture of DevOps helps in improving fault tolerance by minimizing the risk of single point of failure (SPOF) in project management of DevOps-based projects.
	Availability & Capacity		MS1.3	Microservices in DevOps bring resilience to projects by ensuring high availability and efficient capacity management.
	Tailor-Made Functionalities		MS1.4	The modular nature of microservices supports implementing customized functionalities and, hence, offers high levels of flexibility.
	Team Velocity		MS1.5	The microservices approach helps boost team velocity in project management as and when progress is made during the stages of the Software Development Life Cycle (SDLC).
DevOps’ MVE Culture (MVE2)	Agility	[37,58,59,60,61,62]	MVE2.1	The MVE culture in DevOps methodology helps add agility to project management practices employed for DevOps-based projects.
	Learning Curve		MVE2.2	With continuous learning, the MVE culture helps in optimizing the learning curve of project teams.
	Knowledge Management (KM)		MVE2.3	In order to target ‘Kaizen’, knowledge management becomes critical when the culture of MVE is introduced by a team/organization for its product/service.
	Expanding Capabilities		MVE2.4	The MVE culture helps expand capabilities through continuous improvement.
	Customer Experience (CX)		MVE2.5	The MVE culture enhances customer experience (CX) in DevOps-based projects.
DevOps’ Continuous Value Stream Integration & Testing (CVS3)	Dynamic Coding	[63,64,65,66]	CVS3.1	Dynamic programming amidst continuous value stream integration and testing helps in significantly enhancing the quality of the product.
	Central Repository		CVS3.2	Having a centralized and shared repository plays a key role in the process of continuous integration by helping enrich quality, reliability, and precision.
	Lean Time Metrics [Value Added (VA) & Lead Time (LT)]		CVS3.3	Monitoring Lean time metrics in DevOps-based projects for continuous integration/continuous development can have an auspicious effect on the efficiency of the development and operations’ teams.
	Completion & Accuracy		CVS3.4	The percentage of time a task is completed accurately on the first attempt increases through CI and CT, thus resulting in higher precision and reliability for DevOps-based projects.
	Lean		CVS3.5	With CI, developers gradually foster a stable system by operating through a Lean theme, i.e., concise batches and short cycles. This enables project teams to work on shared code, which increases the visibility into the development and quality of the system.
DevOps’ Automated Configuration (AC4)	Programmability	[67,68,69,70]	AC4.1	The provision of programmability in automated configuration helps yield improved efficiency and control for DevOps-based projects.
	Idempotence		AC4.2	Idempotent configuration management provides system administrators with the ability to repeatedly carry out a set of actions to the same result. This becomes vital for organizations that move towards a DevOps approach based on continuous delivery (CD), as it results in safer processes and more reliable results.
	Version Control		AC4.3	Version control for configuration management is critical for automation in DevOps-based projects, as it offers significant advantages such as disaster recovery, auditability, and visibility.
	Standardized Patterns		AC4.4	Because they are defined by code in Infrastructure as Code (IaC), infrastructure and servers can quickly be deployed using standardized patterns, updated with the latest patches and versions, or duplicated in repeatable ways, thus resulting in more efficiently delivered projects.
	Performance Management		AC4.5	From an IaC perspective, performance monitoring is an indirect assessment of the provisioned infrastructure and helps improve project performance visibility.
DevOps’ Continuous Delivery/Deployment (CD5)	Deployability	[71,72,73,74]	CD5.1	Deployability is critical to the CI/CD stage in DevOps-based projects since it results in enhanced velocity on account of automated testing and deployment.
	Modifiability		CD5.2	In DevOps-based projects, having version control in place allows increased modifiability that further results in highly sustainable software solutions.
	Testability (QA/QC)		CD5.3	For DevOps-based projects, continuous deployment results from increased testability, as testability improves due to smaller and more specific changes.
	Automated Testing		CD5.4	Automation of testing through continuous deployment leads to growth in velocity and productivity.
	Emerging technology adoption		CD5.5	Container technologies (containerization) make it possible to deploy an application consistently on any computing environment, whether on-premises or cloud-based, thus boosting project velocity.
PMO Governance (PMOGov6)	Refined Collaboration & Pace of Delivery (Collaboration)	[75,76,77,78,79,80,81]	PMOGov6.1	How critical is the selection of an appropriate agile framework (like SAFe, LeSS, etc.) for improvised collaboration and pace of delivery in successful DevOps adoption?
	Continuous Improvement (Adaptation)		PMOGov6.2	How imperative is the principle of Kaizen towards continuous improvement in the adapting to the changing requirements in DevOps?
	Enriched Quality, Precision & Reliability (Value Delivery)		PMOGov6.3	How significant is the selection of a quality & reliability testing platform in delivering data-driven value with respect to. DevOps?
	Increased Efficiency, Control & Visibility (Automation)		PMOGov6.4	How cardinal is the choice of an IT service management framework (like ITIL) for delivering and maintaining efficient and effective automated operations in a DevOps culture?
	Enhanced Velocity, Productivity & Sustainability (Outcomes)		PMOGov6.5	How crucial is the aspect of a dynamic product tracking tool like KANBAN for monitoring the outcomes from a DevOps initiative?

illustrates the questionnaire constructs, indicators, and items.

4.4. Nonresponse Bias and Common Method Variance

To address the potential for nonresponse bias, a standard check was performed by comparing the responses of early respondents (the first 25% of the sample) with those of late respondents (the last 25%). No statistically significant differences were found in the key demographic variables or the primary study constructs, suggesting that nonresponse bias is not a major concern. To mitigate the risk of common method variance (CMV), which can inflate correlations when a single instrument is used, Harman’s single-factor test was performed. The results indicated that no single factor accounted for a majority of the variance, suggesting that CMV is not a significant threat to the validity of the findings.

4.5. Data Analysis Procedures

The collected data were analyzed through Partial Least Squares Structural Equation Modelling (PLS-SEM) with ADANCO software version 2.4.1. The analysis proceeded in two phases: first, the Measurement Model Evaluation, to assess indicator reliability, internal consistency reliability, convergent validity, and discriminant validity; and second, the Structural Model Evaluation, to analyze the hypothesized relationships between the constructs. Additionally, a bootstrapping approach was used to confirm the model and explore the importance of the tested path coefficients. The full survey instrument, scale sources, and all cross-loadings are provided in the Appendices [Appendix A and Appendix B] to enhance transparency and reproducibility.

5. Hypothesis Development

Building on the integration of the Dynamic Capabilities Framework (DCF) [52] and the Agile DevOps Reference Model (ADRM) [48], this study conceptualizes five hypotheses to examine the relationship between perceptions of DevOps practices and perceptions of PMO governance effectiveness. The wording of the hypotheses reflects that the data were collected based on the participants’ subjective perceptions of these variables. One-tailed critical values were used for hypothesis testing, as the theoretical foundations of the study—namely, the established positive relationships between agile practices and project success in existing literature—led to a priori directional expectations for all proposed relationships.

5.1. Microservices (MS) Architecture and PMO Governance

The microservices architecture is a core component of modern DevOps practices. It allows for modular, independent deployment of software components, thereby increasing system agility, fault isolation, and scalability [63]. In a governance context, microservices enhance visibility and control over project components, aligning with PMO goals for standardization and performance tracking. This modularity supports the “sensing” and “seizing” functions of the DCF, enabling PMOs to respond swiftly to environmental changes.

H1: 

The perception of adopting a microservices architecture in DevOps is significantly correlated with the perception of effective PMO governance practices.

5.2. Minimum Viable Experience (MVE) Culture and PMO Flexibility

Minimum Viable Experience (MVE) culture emphasizes rapid prototyping, experimentation, and iterative learning. This mindset aligns with Agile values of responsiveness and customer-centric design and plays a critical role in promoting flexibility within PMO structures [46]. By fostering a culture of experimentation and continuous learning, organizations can more effectively “seize” opportunities and adjust governance processes to emerging requirements.

H2: 

The perception of a culture that embraces Minimum Viable Experience (MVE) is significantly correlated with the perception of the flexibility of PMO governance within a DevOps environment.

5.3. Continuous Value Stream Integration and Governance Reliability

Continuous Value Stream Integration (CVS) focuses on aligning development and delivery pipelines with customer value and business outcomes. By continuously integrating and delivering increments of value, CVS ensures that project outputs are closely aligned with strategic goals [23]. This practice supports PMO governance by enabling consistent measurement, accountability, and transparency—key attributes of high-performing governance structures. It also contributes to the “seizing” capabilities described by Teece, where timely decision-making is enhanced by real-time insights [21].

H3: 

The perception of Continuous Value Stream Integration (CVS) is significantly correlated with the perception of the quality and reliability of PMO governance practices.

5.4. Automated Configuration and Governance Efficiency

Automated configuration, often enabled through Infrastructure as Code (IaC), significantly reduces manual intervention, minimizes human error, and enhances repeatability across environments. These attributes directly improve the efficiency and reliability of PMO operations [82]. From a DCF perspective, automated configuration enables the “transforming” capability, as organizations can reconfigure their resources dynamically in response to evolving project needs and market conditions [49].

H4: 

The perception of Automated configuration (AC) in DevOps is significantly correlated with the perception of the efficiency of PMO governance.

5.5. Continuous Delivery/Deployment (CD) and Adaptive Governance

Continuous Delivery/Deployment (CD) is foundational to DevOps, emphasizing rapid release cycles, automation, and customer feedback loops. These capabilities necessitate that PMO governance becomes more adaptive and responsive to fast-paced project iterations [2,83]. CD aligns closely with the “seizing” and “transforming” dimensions of the DCF by enabling rapid resource reallocation, performance monitoring, and feedback integration into governance structures.

H5: 

The perception of Continuous Delivery/Continuous Deployment (CD) practices is significantly correlated with the perception of the purpose-driven nature of PMO governance.

5.6. Conceptual Framework

Based on the hypotheses and the theoretical models, the following conceptual framework illustrates the relationships between the independent variables (DevOps practices) and the dependent variable (PMO governance).

6. Results

6.1. Descriptive Statistics

A total of 321 professionals participated in the survey. The data collected from the demographic section of the questionnaire provided a descriptive profile of the sample, including age, years of experience in the industry, and job role (e.g., DevOps Engineer, Project Manager, IT Director). On average, participants reported a high level of familiarity with both DevOps and PMO practices. The raw Likert-scale data for each construct were also analyzed to provide a general overview of the participants’ perceptions. For example, the mean scores for the PMO Governance construct were high, indicating a strong positive perception of its effectiveness in the organizations surveyed. The full descriptive results are provided in the Appendix.

6.2. Measurement Model

The measurement model defines the relationships between latent constructs and their corresponding observed indicators [39]. To ensure the quality of the measurement model, several key aspects were evaluated, including reliability and validity. The constructs are defined as follows: Microservices (MS1), Minimum Viable Experience Culture (MVE2), Continuous Value Stream Integration (CVS3), Automated Configuration (AC4), Continuous Delivery/Deployment (CD5), and PMO Governance (PMO Gov6).

6.2.1. Model Fit

Table 3. Goodness of Fit.

Metric	Value	HI95	HI99
SRMR	0.0597	0.0448	0.0500
dULS	1.6563	0.9339	1.1630
dG	1.0717	0.5337	0.6255

presents the values of the Standardized Root Mean Residual (SRMR) and two additional fit indices (HI95 and HI99), which are commonly employed to evaluate the goodness-of-fit of structural equation models (SEM). These indices provide insights into how well the model represents the observed data. This is particularly true when the primary goal is to test specific, priori hypothesized relationships rather than to exhaustively explain all possible covariance.

The SRMR value falls below the commonly accepted threshold of 0.08 [84], indicating a good fit for the model. This suggests that the discrepancies between the observed correlations and the correlations predicted by the model are minor. The dULS and dG values exceed their confidence intervals, which suggests the model may not be fully capturing all the complex relationships in the data. However, in PLS-SEM, the focus is on predictive power, and a good SRMR value is often considered sufficient to proceed. This is particularly true when the primary goal is to test specific, a priori hypothesized relationships rather than to exhaustively explain all possible covariance.

6.2.2. Construct Reliability

Table 4. Construct Reliability.

Construct	Dijkstra-Henseler’s rho (ρA)	Jöreskog’s rho (ρc)	Cronbach’s Alpha (α)
MS1	0.8638	0.8610	0.8624
MVE2	0.9140	0.9034	0.9048
CVS3	0.8795	0.8776	0.8781
AC4	0.8755	0.8661	0.8680
CD5	0.8610	0.8574	0.8584
PMO Gov6	0.8849	0.8826	0.8825

presents the construct reliability for the six constructs included in the model. Both Cronbach’s alpha and composite reliability (rho) values are reported, and all constructs demonstrate acceptable to satisfactory levels of reliability. These coefficients range from 0 to 1, with values between 0.6 and 0.7 generally considered acceptable and values between 0.7 and 0.9 indicating satisfactory reliability. Values exceeding 0.95 are not desirable, as they suggest excessive correlation among indicator variables, potentially indicating redundancy. The results in Table 4 confirm that the constructs in this study exhibit adequate reliability.

6.2.3. Convergent Validity Using AVE

Table 5. Convergent Validity using AVE.

Construct	Average Variance Extracted (AVE)
MS1	0.5543
MVE2	0.6549
CVS3	0.5897
AC4	0.5671
CD5	0.5472
PMO Gov6	0.6014

displays the Average Variance Extracted (AVE) values for all six constructs in the model. Convergent validity, which assesses the degree to which a measure correlates with other measures of the same construct, is deemed sufficient when the AVE exceeds 0.5 [84]. As shown in Table 5, all six constructs surpass this threshold, confirming the convergent validity of the measurement model.

Convergent validity is typically evaluated by examining indicator loadings and AVE [85]. A general guideline is that a latent variable should explain at least 50% of the variance in each of its indicators. This implies that the variance explained by the construct should be greater than the measurement error variance. Consequently, indicator loadings should ideally be above 0.708, which is the square root of 0.5.

While lower loadings may be acceptable in certain cases, such as newly developed scales in social sciences, researchers should carefully consider the implications of removing indicators with loadings below 0.7 on the composite reliability and content validity of the construct [46]. In this study, indicators MS1.1, MVE2.4, AC4.1, AC4.4, and CD5.3 were retained despite their proximity to the 0.7 threshold due to their theoretical importance and contribution to their respective constructs. All other indicators exhibit loadings above 0.7.

Based on the AVE values and indicator loadings, the model demonstrates satisfactory convergent validity.

6.2.4. Discriminant Validity

Discriminant validity, which assesses the extent to which a construct is distinct from other constructs in the model, was evaluated using multiple methods. The cross-loadings analysis (Appendix B) provides initial support for discriminant validity. Furthermore, the Fornell-Larcker criterion [38] was applied, which compares the square root of each construct’s AVE with its correlations to other constructs. Discriminant validity is established if the square root of the AVE for each construct is greater than its correlations with all other constructs. The analysis of AVE values indicates that this criterion is met.

Additionally, the squared correlations between constructs (

Table 6. Discriminant Validity using Fornell–Larcker Criteria.

Construct	MS1	MVE2	CVS3	AC4	CD5	PMO Gov6
MS1	0.5543
MVE2	0.3596	0.6549
CVS3	0.2463	0.2357	0.5897
AC4	0.2222	0.2844	0.3720	0.5671
CD5	0.2190	0.2029	0.3166	0.2901	0.5472
PMO Gov6	0.4174	0.4634	0.4665	0.5571	0.4405	0.6014

Squared correlations, AVE on the diagonal.

) and the Heterotrait–Monotrait Ratio (HTMT) values (

Table 7. HTMT correlation values.

Construct	MS1	MVE2	CVS3	AC4	CD5	PMO Gov6
MS1
MVE2	0.5999
CVS3	0.4937	0.4774
AC4	0.4705	0.5338	0.6019
CD5	0.4644	0.4419	0.5584	0.5267
PMO Gov6	0.6427	0.6739	0.6760	0.7437	0.6651

) were examined. All values are below the recommended threshold of 0.85, further confirming the discriminant validity of the model [38]. These results demonstrate that the constructs in the model are distinct and measure different concepts.

6.2.5. Indicator Multicollinearity

Multicollinearity, which occurs when independent variables are highly correlated, can inflate the standard errors of regression coefficients and obscure the true significance of variables. To assess multicollinearity, Variance Inflation Factor (VIF) values were examined. Appendix C presents the VIF values for all constructs in the model. As all VIF values fall within the acceptable range of 1 to 5, it can be concluded that multicollinearity is not a concern in this model. This ensures that the estimated relationships between constructs are not distorted by high correlations among the independent variables.

6.3. Structural Model

Following the confirmation of the measurement model’s reliability and validity, the focus shifted to evaluating the structural model. This involved assessing the predictive capabilities of the model and examining the hypothesized relationships between constructs. Partial Least Squares Structural Equation Modelling (PLS-SEM) was employed using ADANCO 2.4.1 software. A bootstrapping procedure with 5000 resamples was implemented to estimate the statistical significance of the path coefficients. t-values and p-values were calculated for each path coefficient to determine their significance. Figure 2 provides a visual representation of the structural model, including the estimated path coefficients.

This figure displays the results of the structural model analysis, showing the relationships between the independent variables (MS1, MVE2, CVS3, AC4, CDS) and the dependent variable (PMO Gov6). Path coefficients are shown on the arrows, and the R-squared value for PMO Gov6 is also indicated.

6.3.1. Coefficient of Determination (R²)

Table 8. Coefficient of Determination (R²).

Construct	Coefficient of Determination (R2)	Adjusted R2
PMO Gov6	0.7652	0.7615

shows the high R² value of 76.5% in this study demonstrates the robustness of the PMO governance adoption model.

6.3.2. Assessment of the Hypotheses and the Path Coefficients

This study tested five hypotheses to determine the significance of the relationships between constructs. Specifically, the goal was to assess whether the path coefficients in the population are significantly different from zero. The null hypothesis of no effect (i.e., a zero-path coefficient) was rejected when the calculated t-value exceeded the critical value. One-tailed tests were employed, with critical values of 2.33, 1.65, and 1.28 corresponding to significance levels of 1%, 5%, and 10%, respectively [85].

Table 9. Significance levels: t-values and p-values for a one-tail test.

t-Values	p-Values	Significance
t < 1.28	p > 0.10	Not significant
1.28 < t < 1.65	0.10 > p > 0.05	Moderate
1.65 < t < 2.33	0.05 > p > 0.01	Significant
t > 2.33	p < 0.01	Very significant

defines significance levels, including t-values and p-values for each hypothesized relationship [38].

Table 10. Total Effects Inference.

Effect	Original Coefficient	Standard Bootstrap Results					Percentile Bootstrap Quantiles
		Mean Value	Standard Error	t-Value	p-Value (2-Sided)	p-Value (1-Sided)	0.5%	2.5%	97.5%	99.5%
MS1 → PMO Gov6	0.1750	0.1845	0.0696	2.5125	0.0121	0.0061	−0.0025	0.0362	0.3153	0.3694
MVE2 → PMO Gov6	0.2244	0.2194	0.0622	3.6072	0.0003	0.0002	0.0414	0.1011	0.3422	0.3856
CVS3 → PMO Gov6	0.1696	0.1700	0.0660	2.5719	0.0103	0.0051	0.0027	0.0354	0.2953	0.3352
AC4 → PMO Gov6	0.3286	0.3249	0.0737	4.4585	0.0000	0.0000	0.1308	0.1775	0.4657	0.5072
CD5 → PMO Gov6	0.2083	0.2103	0.0618	3.3727	0.0008	0.0004	0.0387	0.0875	0.3401	0.3715

depicts the total effects inference for the study as elaborated below:

The analysis of the structural model revealed significant positive relationships between several key DevOps practices and effective PMO governance within the technology industry:

• Microservices (MS1): Organizations adopting microservices demonstrate a stronger inclination towards implementing effective PMO governance practices (β = 0.1750, t = 2.5125, p < 0.01). This finding supports H1.
• Minimum Viable Experience (MVE2): Organizations leveraging minimum viable experiences are more likely to perceive and realize the benefits of PMO governance (β = 0.2244, t = 3.6072, p < 0.01), confirming H2.
• Continuous Value Stream (CVS3): Prioritizing continuous value streams is positively associated with the adoption of effective PMO governance (β = 0.1696, t = 2.5719, p < 0.01), supporting H3.
• Automated Configuration (AC4): Organizations utilizing automated configuration exhibit a stronger tendency towards implementing robust PMO governance practices (β = 0.3286, t = 4.4585, p < 0.01), confirming H4.
• Continuous Delivery/Deployment (CD5): Prioritizing continuous delivery and deployment is positively linked to the adoption of effective PMO governance (β = 0.2083, t = 3.3727, p < 0.01), supporting H5.

These findings underscore the importance of these DevOps practices in facilitating and enhancing PMO governance within technology-driven organizations.

6.3.3. Results & Summary of Hypotheses Analysis

Results

The Partial Least Squares Structural Equation Model (PLS-SEM) evaluates the influence of five core DevOps capabilities—Microservices (MS1), Minimum Viable Experience Culture (MVE2), Continuous Value Streams (CVS3), Automated Configuration (AC4), and Continuous Delivery (CD5)—on PMO Governance (PMO Gov6). All constructs demonstrated strong reliability and convergent validity, with item loadings exceeding the 0.70 threshold. The model explains a substantial proportion of variance in PMO governance (R² = 0.765), indicating robust explanatory power. Path coefficients reveal that Automated Configuration (β = 0.329, p < 0.01) and Continuous Delivery (β = 0.208, p < 0.01) are the most significant predictors of governance effectiveness. Moderate but significant effects were also observed for MVE Culture (β = 0.224, p < 0.01) and Microservices (β = 0.175, p < 0.05), while Continuous Value Streams showed a weaker effect (β = 0.170, p < 0.10). These results support the model’s validity and highlight the importance of technical automation and cultural adaptability in aligning agile DevOps practices with PMO governance functions.

The results reveal that all five hypotheses under study received the expected support, indicating that these practices have a positive impact on the PMO governance outcomes in technology organizations.

Summary of Hypotheses Analysis

The analysis of the hypothesis tests to compare the specific Agile DevOps with the PMO governance is given in

Table 11. Summary of Hypotheses Analysis.

Code	Relationship	Type	β-Value	t-Value	Supported?
H1	Microservices → PMO Governance	Direct	0.175	2.5125	Yes
H2	MVE Culture → PMO Governance	Direct	0.2244	3.6072	Yes
H3	Continuous Value Stream → PMO Governance	Direct	0.1696	2.5719	Yes
H4	Automated Configuration → PMO Governance	Direct	0.3286	4.4585	Yes
H5	Continuous Delivery/Deployment → PMO Governance	Direct	0.2083	3.3727	Yes

below.

The analysis of the structural model revealed significant positive relationships between several key DevOps practices and the perception of effective PMO governance. The findings confirm all five of the proposed hypotheses, as summarized below:

• H1 (Supported): Organizations with higher perceptions of microservices adoption also have higher perceptions of effective PMO governance (β = 0.1750, t = 2.5125, p < 0.01).
• H2 (Supported): Organizations with higher perceptions of a culture that embraces MVE also have higher perceptions of enhancing the flexibility of PMO governance (β = 0.2244, t = 3.6072, p < 0.01).
• H3 (Supported): A higher perception of continuous value stream integration is positively associated with the perception of effective PMO governance (β = 0.1696, t = 2.5719, p < 0.01).
• H4 (Supported): Higher perceptions of automated configuration are strongly linked to the perception of robust PMO governance (β = 0.3286, t = 4.4585, p < 0.01).
• H5 (Supported): Higher perceptions of continuous delivery and deployment are positively linked to the perception of effective PMO governance (β = 0.2083, t = 3.3727, p < 0.01).

Inferences from Two-Tailed Hypothesis Testing

For a two-tailed test, the critical p-value for a 95% confidence level is 0.05. This means a p-value less than 0.05 indicates a significant relationship [38].

Thus, below are the inferences drawn from Table 10.

• H1 (MS1 → PMO Gov6): The p-value for the two-tailed test is 0.0121. Since this is less than 0.05, the relationship between perceived microservices architecture and perceived PMO governance is statistically significant.
• H2 (MVE2 → PMO Gov6): The p-value is 0.0003. This is far below 0.05, so the relationship between perceived MVE culture and perceived PMO governance is highly significant.
• H3 (CVS3 → PMO Gov6): The p-value is 0.0103. This is less than 0.05, so the relationship between perceived continuous value stream integration and perceived PMO governance is statistically significant.
• H4 (AC4 → PMO Gov6): The p-value is 0.0000. This is well below 0.05, indicating a highly significant relationship between perceived automated configuration and perceived PMO governance.
• H5 (CD5 → PMO Gov6): The p-value is 0.0008. This is also well below 0.05, confirming a highly significant relationship between perceived continuous delivery/deployment and perceived PMO governance.

In conclusion, even with the more stringent two-tailed test, all five of the proposed hypotheses are supported. The relationships between all five DevOps practices and PMO governance remain statistically significant, reinforcing the primary findings of the study. This provides a stronger, more robust validation of the model’s results.

These recommendations provide direction for technology organizations that embark on or seek to expand DevOps practices. Through the adoption of these five best practices, organizations can improve the management of DevOps projects and the utilization of resources, and reduce risks, which in turn increases the success of organizational performance practices. Notably, the study underscores the importance of coming up with DevOps appropriate PMO governance models since the conventional linear models cannot be suitable for DevOps.

7. Discussion

This research focuses on the impact of DevOps on PMO governance in the technology industry of the UAE. It fills a gap in the current literature by discussing the application of DevOps within the context of established PMO frameworks. The study also has high practical implications for the UAE’s strategy, specifically where innovation and digital development are critical for growth.

The theoretical framework, based on the DCF and the ADRM, offers a rich conceptual perspective that emphasizes the strategic activities of sensing, seizing, and transforming. While the DCF helps manage the uncertainty of the socio-cultural and regulatory environment of the UAE, the ADRM provides a more granular view.

The findings, derived from data representing the perceptions of 321 professionals, suggest appreciable positive associations between specific DevOps practices and subcategories of PMO governance. The positive coefficient (β = 0.1750) for modularity as an attribute of microservices architecture confirms that it enhances agility and flexibility, as prior studies have postulated [48]. Likewise, the high level of MVE is positively related to the companies’ governance flexibility (β = 0.2244), which supports the idea of constant iteration of processes [86,87,88,89].

The study also focuses on CVS, which stresses continuous enhancement and improvement of the PMO’s efficiency (β = 0.1696). From the analysis, automated configuration was established to be the most impactful variable (β = 0.3286), supporting literature that suggests that automation of software delivery is key to dealing with governance issues in the DevOps space [90,91,92,93]. The positive correlation of CD with the adaptive governance models is evident with a value of β = 0.2083; consequently, there is a need for PMOs to adapt to the rapid deployment capabilities, with support from agile PM concepts [83]. The mixed fit indices, particularly the dULS and dG values, indicate that the model may not capture every subtle complexity in the relationships. However, the strong SRMR value and the statistical significance of all hypotheses provide robust evidence for the core relationships proposed, confirming the predictive power of the model.

Although these findings align with prior research, the current study offers important novel information for the socio-cultural context of the UAE. The observation of the positives and negatives of workforce diversity on DevOps is consistent with our findings from larger-scale DevOps studies; workforce cultural sensitivity remains an essential factor in DevOps implementation [86].

7.1. Theoretical and Practical Implications

This study’s most significant theoretical contribution is its novel integration of the Dynamic Capabilities Framework (DCF) with the Agile DevOps Reference Model (ADRM). While microservices and continuous delivery have been present in software engineering for many years, their role as mechanisms for enacting dynamic capabilities in a governance context is an emergent theoretical insight. DCF provides a strategic lens, focusing on how organizations sense opportunities, seize them, and transform their structures to adapt to dynamic environments. While DCF is a powerful conceptual tool, it can be abstract. The ADRM, in contrast, provides a granular, operational roadmap of specific practices (like microservices and automated configuration). The core theoretical insight of this paper is that the ADRM’s operational practices are the tangible mechanisms that enable an organization to enact DCF’s strategic capabilities.

From a practical standpoint, the findings offer UAE-based organizations a roadmap for adapting PMO governance to successfully integrate DevOps practices. Specifically, the results emphasize the importance of automation, repeatability, and continuous value stream optimization—providing actionable insights for IT leaders and transformation teams navigating agile transitions [94,95,96,97,98]. The study highlights that effective DevOps-PMO integration hinges on embracing not only tools and technologies but also cultural shifts toward transparency, iterative learning, and shared accountability [99,100,101,102,103,104]. These insights can help organizations reduce delivery friction, improve risk visibility, and align project outcomes with strategic business goals.

7.2. Limitations

Despite its valuable contributions, this study has some limitations. The research response rate of 34.26% may pose a threat to response bias limiting the generalizability of the research findings [41]. Moreover, the use of qualitative data and qualitative analysis could be included to give more information regarding the model and make it more extensive. It should also be noted that since participants expressed their perceptions in the instrument, a more quantitative study using objective metrics would be needed for a more robust analysis of the impact of DevOps on PMO governance.

8. Conclusions

This study establishes that the adoption of DevOps practices is exerting a positive impact on the governance of PMOs in the technology sector of the UAE. The research findings suggest that DevOps requires a culture shift from a governance-driven physical model of governance to agile, cyclical and value based. This change is to attain modularity, automation, and to deliver value to the customers on a continuous basis as promoted by DevOps. Aspects such as microservices and automated configuration make it imperative for PMOs to address the aspect of flexibility while making decisions and allocation of resources considering the fast-changing technological environment.

The research also establishes that unlike other structures that are rigid and have well-defined structures and interfaces, PMOs operating within a DevOps environment require flexibility and the ability to respond to change especially within the way they deliver their governance mechanisms. This flexibility is crucial owing to the nature of continuous delivery and deployment processes that define DevOps and are founded on agile frameworks. The dynamics experienced in today’s global market require organizations to adapt to remain relevant and achieve their set goals—a reality that cannot be underemphasized more so in the UAE which is leveraging on technology in its progress.

Overall, this work provides practical insights for policymakers and organizations on how the DevOps and PMO governance supply chain can be formulated with an understanding of the key components to learn from. This framework provides teams with the tools they need to tackle the challenges that are inherent in the process of introducing and deploying DevOps and achieving strategic objectives.

For future research, it should be added that since participants expressed their perceptions, a more quantitative study is needed. Future research could extend the analysis by replicating this study in other geographical locations, different industries, and organizational situations to check the cross-sectional validity and discover any potential qualifications. Furthermore, longitudinal research could document the development of such practices over time, or studies could explore the effects of DevOps practices, such as DevSecOps or AIOps, on the operations of the PMO and the organization generally.