Abstract
Organizations adopt process automation software to support digital transformation, but leaders managing diverse Information Systems (IS) face challenges in developing automation capabilities that deliver both strategic advantage and operational resilience. This study examines how organizations build and sustain process automation competencies through effective Information Technology (IT) governance, strategic alignment, and systematic capability development. Drawing on Technology-to-Performance Chain (TPC) theory and the Dynamic Capabilities framework, this research analyses IS management practices to demonstrate how process automation initiatives can evolve from tactical implementations into sustainable and resilient sources of competitive advantage. A multi-method research design was employed, combining survey data from 286 IT and business leaders across manufacturing, retail, logistics, financial services, and healthcare organizations in the Nordic region, together with 18 qualitative interviews with Chief Information Officers (CIOs), IT directors, and automation leaders. The study identifies pathways for developing three critical capability dimensions: (1) technical capabilities through platform selection and integration practices, (2) governance capabilities through portfolio management and performance measurement mechanisms, and (3) organisational capabilities through change management and continuous learning. The analysis shows that governance capabilities explain 61% of return on investment variance, while balanced capability development across all three dimensions produces returns 2.5 times higher than single-dimension focus. Organizations with mature capabilities establish governance before scaling these initiatives, adopt hybrid governance models, and invest approximately 35% in governance, 35% in technical, and 30% in organisational capabilities. The study provides a capability development framework, empirical evidence of development and performance outcomes, as well as practical guidance for building automation capabilities.
Introduction
In the contemporary digital business era, organizations face unprecedented pressure to develop information systems (IS) capabilities that simultaneously drive operational efficiency and maintain resilience during disruptions. Process automation software, encompassing Robotic Process Automation (RPA), Business Process Management (BPM), and Low-Code Development Platforms (LCDPs) with capabilities of Artificial Intelligence (AI), has emerged as a critical component of enterprise IT portfolios, with global spending expected to reach $25.9 billion by 2027. 1
While technical implementation guides are widely available, organizations continue to struggle to develop the broader IS management capabilities necessary to transform automation investments into sustainable competitive advantage.2,3 The COVID-19 pandemic provided strong evidence that organizations with mature automation capabilities rapidly adapted their operations, while those with immature or poorly governed initiatives experienced severe disruptions. 4 This highlights a critical insight: successful process automation extends beyond technology deployment to encompass sophisticated IS management capabilities spanning governance, strategic alignment, and organizational development.5,6 Despite the strategic importance of process automation, limited research has examined how organizations develop and sustain these essential capabilities, particularly in volatile business environments. Recent topic modelling of IS research indicates that while traditional topics such as IS development and IT adoption remain central, emerging areas of digital transformation and organizational capabilities represent important frontiers for the discipline 7
From an IS management perspective, process automation presents unique challenges that distinguish it from traditional IT implementations. Unlike conventional enterprise systems that require substantial infrastructure changes, automation software operates at the presentation layer, enabling rapid deployment but demanding new governance approaches. 8 CIOs and IT leaders must navigate complex decisions regarding platform selection, vendor management, integration architectures, and capability development while ensuring alignment with business strategy and maintaining operational resilience.9,10 Furthermore, automation initiatives often span organizational boundaries, requiring IS leaders to orchestrate cross-functional capabilities that traditional IT governance frameworks inadequately address.
The Nordic region provides an instructive context for examining automation capability development. Nordic organizations demonstrate advanced digital maturity, with 78% having formal digitalization strategies compared to 54% globally. 11 These organizations operate in highly regulated environments demanding robust governance, while cultural factors emphasizing collaboration and continuous improvement offer insights into capability development practices. Moreover, Nordic organizations’ experiences navigating recent disruptions, from supply chain crises to geopolitical uncertainties, reveal how mature automation capabilities contribute to organizational resilience.
This study addresses these gaps by examining how organizations develop IS management capabilities for resilient process automation. Three theoretical streams of debate motivate the research questions. First, IT governance literature emphasizes that decision rights and accountability structures determine whether IT investments deliver business value,
9
yet governance mechanisms for rapidly deployed, business-led automation remain underspecified.12,13 Second, dynamic capabilities theory suggests that sensing, seizing, and reconfiguring routines enable adaptation,14,15 but empirical evidence on how these capabilities develop in automation contexts, and which routines matter most at different maturity stages, is limited. Third, while complementarity theory indicates that IT value emerges from bundles of reinforcing capabilities,
16
optimal investment patterns and capability configurations for process automation have not been established. Building on these debates, we pose the following research questions: RQ1: How do organizations develop governance, technical, and organizational capabilities for process automation? RQ2: What capability development strategies and investment patterns enable organizations to achieve both operational efficiency and resilience during disruptions?
First, RQ1 responds to calls for research on complementary IT capabilities5,17 and addresses the gap between technology-centric automation studies and IS research emphasizing managerial and organizational factors.16,18 Second, RQ2 aims to situate dynamic capabilities theory 14 to automation contexts and addresses debates on IT investment allocation 19 and organizational resilience.20,21
This research aims to make three contributions to IS management literature and practice. First, it is to develop a comprehensive capability framework that extends technical implementation to encompass the full spectrum of IS management practice for automation success. Second, it is to provide empirical evidence from multiple industries demonstrating how capability maturity influences both automation performance and organizational resilience. Third, it shall offer prescriptive guidance for IS leaders, including capability assessment tools, development pathways, and governance templates that bridge the theory–practice gap.
Literature background
Process automation in information systems management
Process automation has evolved from a basic efficiency tool to a strategic IS capability requiring sophisticated management approaches. Robotic Process Automation (RPA) represents a paradigm shift in enterprise automation, using software robots to execute rule-based tasks through existing user interfaces without deep system integration.22,23 Unlike traditional workflow automation, these systems operate at the presentation layer, allowing rapid deployment but introducing distinct governance and integration challenges.
Recent IS research has begun to explore automation beyond technical implementation. Studies identify governance issues, organizational transformation, and the emergence of intelligent automation that combines RPA with AI.24–26 The integration of machine learning and natural language models has expanded automation toward cognitive tasks, previously requiring human judgment.27,28 Yet, while technical sophistication advances rapidly, IS management scholarship lags in explaining how organizations develop governance structures and organizational capabilities to sustain these technologies. 29 From an enterprise architecture perspective, automation platforms must coexist with legacy systems, cloud services, and emerging digital platforms, creating new demands for platform selection, vendor management, and architectural governance. 30
IT governance and strategic alignment
IT governance provides the decision-making structures and accountability mechanisms that ensure IT investments deliver business value. 9 Traditional governance models, however, require adaptation for process automation, given its rapid deployment cycles, business-led initiatives, and cross-functional impacts. 12 Recent studies suggest hybrid governance models, combining centralized oversight with distributed execution, are most effective. 13
Strategic alignment is equally critical. Unlike traditional IT projects with fixed scopes, automation portfolios evolve continuously as organizations identify new opportunities. This dynamic nature demands new alignment mechanisms, such as automation roadmaps, benefit realization frameworks, and cross-functional steering committees.10,31
The establishment of Centres of Excellence (CoEs) has emerged as an important governance mechanism, balancing standardization with flexibility. CoEs consolidate expertise while allowing business units autonomy, enabling both strategic alignment and scalability. 32 Yet empirical evidence remains limited on how CoE structures influence automation outcomes and capability development. 33
Organizational resilience and dynamic capabilities
Organizational resilience, defined as the ability to anticipate, respond to, and recover from disruptions, increasingly depends on IS capabilities. 34 Recent work on resilience and sustainability highlights how capability development in both domains can be mutually reinforcing, identifying congruent capabilities such as collaboration, agility, and transparency. 35 The COVID-19 pandemic highlighted how digital capabilities, particularly automation, enabled organizations to adapt and maintain continuity.36,37 In automated environments, resilience requires not just technical redundancy but also organizational capabilities for rapid reconfiguration and adaptation. 38
Dynamic capabilities theory provides a useful lens for understanding automation capability development. Teece 14 identifies three classes of capabilities: sensing, seizing, and reconfiguring. Applied to process automation, these manifest as scanning for automation opportunities, orchestrating resources for implementation, and continuously improving processes. 15 IS scholars have extended this thinking to digital contexts, emphasizing digital dynamic capabilities 39 and digital resilience. 40 These frameworks underscore the interdependence of technical and organizational investments, suggesting that resilient automation requires complementary governance, technology, and human capital development.5,41
Technology-to-Performance chain theory
Technology-to-Performance Chain (TPC) theory 42 provides a theoretical foundation for examining how technology characteristics and task requirements interact to influence performance outcomes (Figure 1). In process automation, task characteristics include process structure, volume, rule clarity, and exception frequency. 8 These differ significantly from traditional IT implementations, since automation operates at the interface rather than integration layer.
Technology to performance chain theory model (adopted from. 42 .
Technology characteristics in automation contexts are rather standardized across major platforms, enabling meaningful comparisons across implementations. Utilization, however, extends beyond individual adoption to encompass organizational patterns of use, governance mechanisms, and capability development trajectories. 43 Performance impacts also go beyong individual productivity, manifesting in organizational outcomes such as efficiency, adaptability, and resilience. By reconceptualizing utilization and performance at the organizational level, TPC theory provides a robust lens for analysing how IS capabilities enable automation success in turbulent environments.
The development of IS capabilities has long been central to IS research.17,44 Recent frameworks specific to digital transformation emphasize leadership, engagement, and operational dimensions. 45 However, these lack specificity for process automation, particularly regarding governance and resilience.
Capability maturity models offer a developmental perspective. Drawing on Capability Maturity Model Integration (CMMI), automation maturity can be viewed as progressing from ad hoc experimentation to optimized, continuously improving processes. 46 Yet empirical evidence on how organizations advance through these stages, and how non-technical capabilities evolve alongside technical ones, remains limited.
Research design
Research approach and sample
This exploratory study adopted a mixed-method design to examine IS management capabilities for process automation. Combining quantitative and qualitative methods enabled both breadth and depth of understanding, while also providing triangulation across data sources.47,48 Given the emerging nature of automation capabilities and the limited prior research, an exploratory design was particularly appropriate for theory development. 49
The Nordic region (Denmark, Finland, Norway, and Sweden) provided a suitable context for this investigation. Organizations in this region are digitally mature, with well-established IT governance practices and significant automation investments. 11 The region's transparent business environment and accessible data further supported comprehensive data collection.
To ensure cross-industry relevance, the study included organizations from five sectors: manufacturing (28%), financial services (24%), retail (20%), healthcare (16%), and logistics (12%). Selection criteria required that organizations: (1) employed at least 200 staff to ensure formal IT governance structures, (2) had active process automation initiatives, and (3) operated in Nordic markets. This purposive sampling ensured that participants had direct automation experience while representing diverse organizational contexts.
Data collection
The study began with a quantitative phase. An online survey was developed using IS capability measures adapted for automation contexts. Items were drawn from IT governance research,9,50 automation studies,43,51 and dynamic capabilities literature.15,52 The complete survey instrument is provided in Appendix.
The survey captured multiple dimensions of automation capability development and outcomes. Technical capabilities were assessed through items on platform selection, integration, and architecture (15 items). Governance capabilities were measured through decision structures, portfolio management, and performance mechanisms (18 items). Organizational capabilities were examined via change management, skill development, and knowledge sharing (16 items). Automation outcomes included operational performance, resilience, and strategic benefits (12 items). All items used seven-point Likert scales, complemented by demographic and maturity indicators.
Pre-testing with five IT executives and three academics ensured clarity and construct validity. The survey was distributed in two waves between Q3 2020 and Q3 2021. Invitations were sent to 1042 IT and business leaders identified via professional associations, LinkedIn networks, and Nordic company databases. A total of 286 usable responses were received (27.4% response rate), including CIOs/IT Directors (31%), automation leads (26%), business executives (23%), and IT managers (20%).
The second phase was qualitative. 53 Semi-structured interviews provided deeper insights into capability development processes and contextual factors shaping automation success. Participants were recruited from survey respondents who had indicated willingness to participate further. Theoretical sampling ensured representation across industries, maturity levels, and organization sizes.
Eighteen interviews were conducted via online meetings in December 2021, lasting between 19 and 41 min (average 26 min). The interview protocol focused on automation journeys, capability evolution, governance structures, and resilience practices. Participants reflected on success factors, challenges, and lessons learned. All interviews were recorded, transcribed, and anonymized.
Participation in both survey and interviews was voluntary, and all respondents provided informed consent. Data were anonymized and handled in accordance with institutional ethical guidelines on research with human subjects. Table 1 presents detailed participant profiles:
Participants with anonymized data (source: self-study).
Participants with anonymized data (source: self-study).
Quantitative data were analysed using SPSS. Reliability was assessed through Cronbach's alpha, with all constructs exceeding the 0.70 threshold. 54 Exploratory factor analysis confirmed the three-dimensional capability structure, explaining 67.3% of variance. Multiple regression tested relationships between capabilities and automation outcomes, controlling for size, industry, and maturity. Mediation analysis further examined whether governance capabilities mediated links between technical capabilities and resilience. To address potential common method bias, Harman's single-factor test was conducted, showing that no single factor accounted for the majority of variance.
Qualitative data were analysed using thematic analysis. 55 Initial coding identified 147 first-order concepts, which were grouped into 28 s-order themes and mapped onto the three capability dimensions. NVivo 14 supported coding and theme development. To ensure rigor, an external researcher independently coded three transcripts, with discrepancies resolved through discussion. 56 Member checking with five participants validated interpretations.
Integration of methods occurred through triangulation. 57 Quantitative analysis established relationships between capabilities and outcomes, while qualitative data provided contextual explanations. Convergent findings reinforced conclusions, while divergent insights prompted deeper examination. This integration produced a comprehensive framework grounded in empirical evidence.47,58
Research findings
The analysis revealed three interdependent capability dimensions that determine how organizations realize value from process automation: technical capabilities, governance capabilities, and organizational capabilities. While each dimension contributes uniquely, their interaction proved important for enabling both operational efficiency and organizational resilience.
Capability dimensions
Technical capabilities form the foundation of automation success. Regression analysis showed that organizations with strong competencies in platform selection, system integration, and architectural management reported significantly higher operational performance (β = 0.42, p < 0.01). This confirms that technical underpinnings remain essential for moving automation beyond isolated initiatives and embedding it into wider organizational processes.
As shown in Figure 2, governance capabilities emerged as the most critical dimension, with the highest mean importance rating (6.21), strongest correlations with both ROI (0.61***) and resilience (0.67***), and the highest percentage of organizations (84%) citing it as critical. Technical and organizational capabilities, while important, showed lower correlations with outcomes. All correlations were statistically significant at p < 0.001.
Interview evidence reinforced this finding. As one IT director explained, “The robots work fine individually, but the real challenge was embedding them into our architecture, so they became part of the overall system rather than just stand-alone fixes.”
These results resonate with prior IS research that identifies IT infrastructure as a critical enabler of organizational performance.2,59,60 In the context of automation, the role of infrastructure extends further, requiring interoperability across legacy systems, cloud platforms, and emerging AI tools. 61
Technical capabilities
Technical capabilities encompassed three critical sub-dimensions: platform assessment and selection, integration architecture, and technical resilience.
In terms of platform assessment and selection, successful organizations developed systematic approaches for evaluating automation platforms that extended beyond vendor demonstrations. Key assessment criteria are reported in Table 2. Security features, integration capabilities, and scalability emerged as the most important, while criteria such as AI/ML capabilities and user interface design were generally seen as secondary.
Capability dimensions and automation outcomes: mean importance ratings (7-point scale), correlations with return on investment (ROI) and resilience, and percentage of organizations citing each dimension as critical (N = 286). *p < 0.001.
Distribution of platform selection criteria priorities: percentage of organizations rating each criterion as high, medium, or low priority during automation platform evaluation (N = 286).
As one enterprise architect from a retail organization explained: “Platform selection isn't about choosing the most feature-rich solution. It's about understanding architectural constraints, integration requirements, and long-term scalability needs. A weighted scoring matrix was developed that saved expensive platform migrations later.” This finding reinforces prior IS research emphasizing the importance of structured IT evaluation and portfolio decision-making in achieving alignment and avoiding costly lock-in.2,62
With regards to integration architecture, organizations with mature technical capabilities adopted API-first architectures that enabled seamless integration of automation platforms. Rather than treating RPA as standalone tools, successful organizations embedded automation into their enterprise architectures. As a technology director observed: “Automation platforms are treated as another system requiring proper integration standards, not as tactical workarounds for system limitations.” Key architectural approaches are summarized in Table 3.
Adoption and outcomes of integration architecture approaches: percentage of organizations using each approach, benefits cited, and implementation challenges (N = 286).
These findings reflect broader IS research emphasizing digital platform architectures and modularity as key enablers of agility and scalability.61,63 In automation contexts, API management and microservices help firms integrate new tools with legacy systems while avoiding fragmented architectures.
Finally, technical resilience mechanisms were crucial for sustaining automation at scale. Organizations relied on multiple safeguards. Real-time monitoring dashboards were the most widely implemented (83%), ensuring ongoing stability. Redundancy protocols such as backup bots and failover mechanisms were used by 71% of respondents, while version control systems (64%), particularly Git-based repositories, provided reliable mechanisms for updates and rollbacks. Automated testing frameworks were also adopted by 56% of organizations to validate bot updates before deployment. These practices demonstrate how resilience in automation requires a layered approach, echoing IS insights into the importance of IT reliability, monitoring, and recovery capabilities for business continuity.3,64
Governance capabilities emerged as the strongest predictor of automation success, spanning decision rights, portfolio management, and performance measurement. Prior IS work shows that clear governance of IT decision rights and standards is foundational for realizing business value.9,10 In the context of automation, hybrid arrangements that mix centralized oversight with distributed execution are particularly effective.12,13
Comparison of automation governance models: characteristics, adoption rates, average return on investment (ROI), and time to scale across three governance approaches (N = 286).
Comparison of automation governance models: characteristics, adoption rates, average return on investment (ROI), and time to scale across three governance approaches (N = 286).
The Head of IT Governance (Informant 7) advocated the hybrid model: “Pure centralisation creates bottlenecks, whilst complete decentralisation leads to chaos. A CoE was established for complex, cross-functional processes whilst empowering business units to automate simple, department-specific tasks within the governance framework.” This pattern aligns with research on ambidextrous IT governance and platform ecosystems, which balance standardization with local innovation.12,30
Regarding portfolio management and prioritization, mature organizations treated automation as an investment portfolio and applied consistent evaluation criteria before funding. As shown in Table 5, proposals were screened on strategic alignment, technical feasibility, business impact, and risk.
Automation portfolio evaluation framework: assessment dimensions, weighting, key metrics, and approval thresholds used by organizations with mature governance capabilities (N = 52 high-maturity organizations).
The IT Portfolio Manager (Informant 14) described the approach: “Every automation opportunity is evaluated using a standardized business case template. This ensures automation targets strategic value, not just easy wins.” Treating automation opportunities as a managed portfolio reflects IS guidance on benefits management and value realization.18,65
For performance measurement and benefits realization, organizations with mature governance implemented comprehensive, multi-category dashboards. As reported in Table 6, operational, financial, and quality metrics were widely tracked, with growing, though still uneven, attention to strategic and employee outcomes.
Automation performance measurement practices: metric categories, specific measures, percentage of organizations tracking each category, and average performance improvements reported (N = 286).
These practices mirror IS research emphasizing explicit measurement and governance as mechanisms linking IT investments to firm performance.3,10 In automation programs, disciplined portfolio governance and benefits tracking help avoid fragmented initiatives and ensure scaling aligns with enterprise objectives.
Organizational capabilities encompassed human capital development, change management, and knowledge management practices that are essential for sustaining automation over time. Prior IS research highlights that realizing value from IT depends not only on technical investments but also on building complementary human and organizational capabilities.3,20,44
With respect to workforce capability development, successful automation required systematic programs that addressed both technical and business skills. Figure 3 summarizes adoption rates, typical investments, and perceived effectiveness. Citizen-developer programs and innovation workshops were widely used; cross-functional rotations, although adopted less frequently, were rated as highly effective. These patterns align with work on knowledge integration and boundary-spanning in IS projects, where cross-functional exposure accelerates assimilation and routinization of new digital practices.66,67
Workforce development initiatives for automation capabilities: adoption rates, investment levels, and effectiveness ratings across different training and development approaches (N = 286).
As the Digital Transformation Lead (Informant 4) emphasized, “Automation isn't about replacing people; it's about augmenting human capabilities. This requires comprehensive reskilling programs that build both technical competence and business acumen.”
Organizations with stronger change-management capabilities achieved markedly higher automation adoption rates (2.3×). The most frequently cited success factors were visible executive sponsorship (89%), consistent communication of updates and success stories (78%), early stakeholder engagement (71%), and proactive management of job-displacement concerns (65%). This pattern is consistent with IS research on user resistance and organizational change, which underscores the role of leadership support, communication, and participation in overcoming barriers to technology assimilation.68,69
Knowledge management and learning practices provided a backbone for scaling. Mature organizations reported comprehensive documentation standards (76%), best-practice repositories (68%), communities of practice for peer learning (54%), and formal lessons-learned reviews after go-lives (61%). These mechanisms echo IS scholarship on knowledge management and organizational learning as drivers of sustained IT value.70,71
The analysis also revealed significant interactions among capability dimensions. Organizations that developed balanced capabilities across technical, governance, and organizational domains outperformed those that concentrated on a single dimension. Figure 4 compares performance outcomes across five capability maturity profiles, revealing that balanced development strategies substantially outperform single-dimension approaches.
Performance outcomes by capability maturity profile: comparison of average ROI, resilience scores (10-point scale), recovery time (days), and employee satisfaction (10-point scale) across five capability development patterns (N = 286).
As shown in Figure 4, the “Balanced High” profile achieved the highest ROI (52%), resilience score (8.4/10), employee satisfaction (8.3/10), and fastest recovery time (2.1 days). In contrast, organizations focusing on a single dimension achieved substantially lower outcomes across all metrics. Notably, even “Balanced Medium” organizations (31% ROI) outperformed single-dimension focused organizations, suggesting that balanced development matters more than achieving high maturity in any single dimension. These complementarities reflect the broader IS view that IT value emerges from reinforcing bundles of capabilities rather than isolated investments.2,17 The Strategy Director (Informant 18) summarized: “Technical capabilities get you started, governance capabilities ensure sustainability, but organizational capabilities determine whether automation becomes a competitive advantage or an expensive experiment.”
Finally, recent disruptions provided a natural experiment for assessing resilience. Table 7 shows that high-maturity organizations were substantially more capable of rapid reconfiguration (within 48 h), maintaining process continuity, managing remotely, handling volume spikes, and deploying new automated processes. These findings align with IS research linking digitally enabled agility to faster response and recovery during environmental jolts.20,21
Crisis response capabilities by automation maturity level: percentage of organizations demonstrating specific response capabilities during COVID-19 disruption, compared across high, medium, and low maturity levels (N = 286).
As the Head of Automation (Informant 2) noted, “The pandemic validated capability investments. Our competitors struggled with manual tasks; automated processes continued operating, and actually improved as new bots were quickly deployed for emerging needs.”
Governance and is capabilities
The three-dimensional capability framework challenges traditional IS capability conceptualizations and extends existing literature in several ways. Figure 5 presents the IS management framework for building resilient process automation, integrating task and technology characteristics with technical, governance, and organizational capabilities, organizational-level utilization, outcomes, and dynamic capabilities. While prior research has emphasized technical infrastructure and IT resources as primary sources of advantage,17,44 the results in this study indicate that governance capabilities are the strongest predictor of automation success, explaining 61% of ROI variance. Rather than positioning technology as sufficient, this pattern is consistent with integrative reviews that link IT value to complementary managerial and organizational mechanisms16,18,72 and with work that stresses decision rights and accountability in realizing value from IT investments. 9 In the specific case of process automation, the findings suggest that formalized decision structures, portfolio oversight, and performance management play a central role in converting technical potential into outcomes.
These results stand in contrast to technology-centric perspectives on automation adoption, and they are more closely aligned with arguments that governance is foundational for digital initiatives. 73 They also accord with research on digital transformation showing that organizational factors often outweigh purely technical ones, 74 while differing from reviews of RPA that foreground platform capabilities over managerial complements. One interpretation is that as automation has evolved from a tactical efficiency tool to a strategic capability embedded in enterprise portfolios, governance has become the primary coordinating mechanism for scale, risk management, and benefit realization. 13 The findings further relate to work on information and process management capabilities. Prior studies report that such capabilities mediate the relationship between IT investments and firm performance by shaping data quality, process discipline, and benefit tracking.3,75 The evidence here is consistent with that view: technical strength appears necessary but not sufficient, and governance capabilities provide the structure through which automation platforms are prioritized, integrated, and measured. In this sense, governance does not replace technical or organizational capabilities; rather, it orchestrates them so that complementary investments cohere into sustained performance effects.
Finally, the observed primacy of governance lays groundwork for the theoretical extensions developed in the subsequent subsections. At the organizational level, governance structures shape patterns of utilization, that is, how automated processes are selected, deployed, and adapted, and thereby influence performance understood as both efficiency and resilience. This perspective is compatible with calls to broaden usage and success constructs in IS76,77 and sets up the extension of Technology-to-Performance Chain thinking to organizational and temporal dynamics developed next.
Is management framework for building resilient process automation.
This study extends the Technology-to-Performance Chain 42 by considering utilization and performance at the organizational level and by incorporating temporal dynamics under disruption. In our context, utilization is not only the use of a specific tool by an individual; it also includes organization-level patterns shaped by governance structures, portfolio decisions, Centres of Excellence routines, standards, training, and monitoring. Performance likewise extends beyond task efficiency to include adaptability and resilience during shocks. This broader view responds to calls for richer usage constructs and more comprehensive success models in IS.76,77
The evidence suggests an organizational “resilient fit,” where the alignment between tasks, automation technologies, and governance adapts over time. Rather than a one-off matching exercise, alignment operates as a dynamic process, influenced by evolving portfolios, architectural choices, and capability development. This perspective adds nuance to debates on alignment and performance under change 78 and complements work on organizational fit and embedded structures. 79 It also addresses critiques that static fit models understate environmental turbulence and temporal effects in digital settings. 21
Figure 6 presents the extended TPC developed in this study. The model places capability-enabled, organization-level utilization between technology and outcomes, with explicit links to both efficiency and resilience. Governance mechanisms (such as decision rights, portfolio management, performance tracking) steer how automation opportunities are sensed, selected, and scaled; organizational capabilities (such as change management, skills, knowledge sharing) influence adoption and adaptation; and technical capabilities (such as integration architecture, reliability practices) shape feasibility and stability. Together, these elements help explain why organizations differ in their ability to sustain performance during disruptions, even when using similar platforms.
Extended technology to performance chain theory model for process automation, showing how capability dimensions enable organizational-level utilization patterns that drive both efficiency and resilience outcomes (source: study).
Dynamic capabilities offer a useful lens for interpreting how automation capabilities evolve and how organizations respond under disruption. Following Teece, 14 sensing, seizing, and reconfiguring describe routines for identifying opportunities, mobilizing resources, and adapting configurations. In this study, early-stage progress appeared more closely tied to sensing, structured scanning for automatable processes, discovery pipelines, and rapid evaluation, whereas performance during shocks depended more on reconfiguring, redeploying bots, reprioritizing queues, and modifying integrations to new requirements. Quantitatively, sensing accounted for a large share of early-stage success variance (48%), while reconfiguring explained a larger share of crisis resilience variance (62%).
These patterns are compatible with prior work that conceptualizes dynamic capabilities as patterned processes rather than one-off events 80 and with IS operationalizations that link such routines to digitized processes and performance. 15 They also align with research on digital capabilities that emphasizes the interplay of technology, governance, and human capital in value creation. 5 In practical terms, sensing was evident in formal intake and assessment mechanisms; seizing in business cases, funding gates, and release management; and reconfiguring in playbooks for rapid bot changes, rollback procedures, and post-incident learning.
The results further suggest sequencing. Early on, organizations benefit from building sensing and seizing routines to create a reliable pipeline and codify deployment practices; as portfolios scale, reconfiguring routines become more salient for resilience. This view is consistent with capability lifecycle arguments 81 and with stage-based observations in digital transformation that different capabilities peak in salience at different maturity points. It also complements the Technology-to-Performance Chain extension developed above: dynamic capabilities shape organization-level utilization (what is automated, how, and how quickly it is adapted), which in turn relates to both efficiency and resilience. 82
Moreover, the evidence here indicates that governance acts as an orchestration mechanism for dynamic capabilities. Decision rights, portfolio management, and performance tracking help translate sensing into prioritized opportunities, seizing into disciplined delivery, and reconfiguring into systematic adaptation. In that sense, governance mediates how technical investments and human skills become outcomes,16,18 providing a pathway from dynamic capability routines to the performance effects observed in this study.
In the empirical data, organizational adaptability accounted for a larger share of resilience variance (54%) than technical redundancy (31%). Formal playbooks and governance routines were associated with a 68% reduction in recovery time, and high-maturity organizations reported reconfiguring 73% of affected processes within 48 h during COVID-19. These patterns align with e-resilience perspectives that emphasize adaptive capacity over redundancy 83 and suggest that properly governed automation can enhance, rather than constrain, flexibility, differing from early accounts of inherent rigidity (for example, Willcocks et al. 23 ) and consistent with emerging work on automation agility. 43
Hybrid governance and observed outcomes
In this study, hybrid governance, that is, central standards with distributed execution, was associated with higher ROI (52% vs. 34% for centralized) and faster time to scale (9 vs. 18 months), as reported earlier. While these associations do not imply causality, they are consistent with the idea that combining decision rights and standards with local experimentation can reduce duplication and fragmentation while preserving agility. Prior IS work similarly links clear decision rights and accountability to value realization 9 and shows that platform-style governance can enable innovation at the “edges” without eroding coherence at the core. 30
A practical mechanism for achieving this balance in automation programs is the Centre of Excellence (CoE), which concentrates methods, tooling, and guardrails while allowing business units to own domain-specific automation. The interview evidence in this study pointed to CoEs as reducing rework, improving reuse, and accelerating onboarding, in line with research on ambidextrous IT governance that balances exploration and exploitation 12 and with accounts of CoE practices in automation. 13
These results also help reconcile competing prescriptions in the literature. Bimodal views emphasize separation between traditional and digital initiatives, 84 whereas digital-backbone perspectives stress integration. 61 The hybrid pattern observed here can be read as a contingent synthesis: shared architecture, standards, and portfolio controls where interdependence is high, and business-led execution where local knowledge dominates. This interpretation aligns with adaptive governance arguments that advocate tailoring structures to context and uncertainty. 85
Finally, the governance patterns observed here connect directly to the TPC extension developed above. Hybrid arrangements appear to shape organization-level utilization, what gets automated, how consistently, and how quickly work is adapted, thereby relating to both efficiency and resilience outcomes. The evidence suggests that governance does not substitute for technical or organizational capabilities; rather, it orchestrates them so that complementary investments cohere during both routine operations and disruption.
Investment patterns and complementarities
In this study, organizations that balanced investments across governance, technical, and organizational capabilities, approximately 35–35–30, were associated with stronger outcomes than technology-heavy profiles. This pattern contrasts with vendor guidance that favors predominantly technical spending 86 and aligns with frameworks that foreground complementary investments and organizational factors.16,87 The 2.5× ROI improvement observed for balanced portfolios is consistent with complementarity theory, which predicts higher returns when reinforcing assets are co-invested, and it extends IT asset-allocation research by indicating that automation may require different spending patterns than traditional IT portfolios. 19 The data also suggest maturity-contingent shifts: a relatively larger technical share early (≈45%) to establish integration and reliability, tapering later (≈25%) as governance and organizational capabilities become more salient for scale and adaptation, nuance largely absent from static allocation advice (for example). 88
These allocation patterns are consistent with the outcome profiles reported in Figure 4, where the Balanced High capability configuration exhibits the strongest ROI and resilience with shorter recovery times. Interviewees described fewer stalled initiatives and higher reuse of components under balanced spending, attributing this to clearer decision rights, common standards, and coordinated change management.9,12 The reuse effect aligns with research on modular platforms and digital backbones that enable component sharing and scale.61,63 In the extended TPC, such balanced investments shape organization-level utilization by determining what gets automated, how reliably it integrates into enterprise architecture, and how effectively people and processes adopt and evolve it.76,77 Practically, this points to sequencing rather than a fixed formula: build technical foundations to make automation feasible and safe, then shift marginal resources toward governance and organizational capabilities to keep portfolios coherent and adaptable as they scale.16,18,19
Implications for financial, manufacturing and healthcare industries
The results here differ from the compliance-centric perspective common in the literature. While Kokina and Blanchette 8 emphasize technical accuracy for accounting automation and Warren et al. 89 focus on audit trails, governance capabilities explained 54% of success variance in this sample, compared with 31% for technical accuracy. This suggests that accounting-specific findings may not generalize across broader financial-services automation, where decision rights, portfolio discipline, and performance tracking appear more consequential. 9 The relatively low priority given to regulatory features in platform selection, ranked fifth of seven criteria (Table 2), aligns with Gomber et al., 90 who emphasize business-model innovation, and differs from assumptions that compliance is the primary driver. 91 Instead, strategic alignment and scalability dominate, consistent with viewing FinTech as strategic transformation rather than regulatory response ().
The findings in the manufacturing contexts point to greater complexity than early automation studies suggest. Asatiani and Penttinen 26 report that simple RPA can suffice for back-office processes, yet 76% of manufacturing respondents here cited integration complexity as the primary challenge, versus 34% in other industries, reflecting movement toward production-adjacent automation and OT–IT convergence. This pattern is consistent with Industry 4.0 perspectives 92 and with platform/modularity research that helps explain why microservices and API management aided scale and resilience.61,63 The result that organizational capabilities constrained success more than technical integration is in line with observed digital-transformation barriers. 93 The adoption of hybrid automation models by 71% of manufacturers extends Kusiak's 94 smart-manufacturing concepts to process-automation contexts.
Healthcare organizations showed distinct patterns, with 89% emphasizing validation protocols versus 56% industry average. While Agarwal et al. 95 stress patient-centric transformation, the evidence here suggests process efficiency is a primary adoption driver. The strong emphasis on change management (highest among industries) is consistent with IS studies that highlight participation and communication to mitigate resistance.68,96
Methodological comparisons and boundary conditions
The mixed-method approach addresses several limitations in prior automation research. The survey sample of 286 responses substantially exceeds typical study sizes in this stream, 97 enabling statistical tests of relationships that earlier work largely theorized or illustrated in cases. The cross-industry coverage further supports pattern detection beyond a single sector or firm, while the interview evidence adds contextual explanation for observed effects.
Integration of survey and interview data follows calls for methodological diversity and triangulation in IS.47,48,98 The design combined Yin's replication logic for qualitative analysis 53 with multivariate techniques for the survey, 99 providing triangulation that single-method studies typically lack. 58 Procedures for rigor and validity, pretesting, reliability and factor analysis, a common-method bias check, and member checking, align with IS guidance. 54 This comprehensiveness responds to critiques about limited empirical evidence 46 and calls for quantitative validation of qualitative insights in automation research. 100
The framework's applicability depends on several contextual factors that merit explicit acknowledgement. Organizational size emerged as a moderator: smaller organizations (fewer than 200 employees) reported comparable outcomes through informal coordination in place of formal governance. This finding supports SME digitization research and differs from claims that formal IT governance is essential regardless of size. 101 Cultural context also matters. The Nordic setting's emphasis on trust and collaboration 102 may facilitate capability development differently than more hierarchical environments; adaptations may be required in higher power-distance contexts. 103 Finally, regulatory intensity appears to shift optimal allocations: in highly regulated sectors, respondents indicated 45–50% governance investment versus a baseline 35–40%, consistent with the idea that compliance complexity raises coordination and oversight needs. 104
Conclusions
This study examined how organizations develop information systems management capabilities for resilient process automation. Drawing on 286 survey responses and 18 executive interviews across five Nordic industries, it developed an empirically grounded framework comprising three capability dimensions: technical, governance, and organizational, and analysed how these interact to shape outcomes.
The study contributes to IS management in three ways. First, it shows that governance capabilities are the strongest predictor of automation success in this sample, shifting emphasis from technology alone toward complementary managerial mechanisms.16,18,72 Second, it extends Technology-to-Performance Chain theory by reframing utilization and performance at the organizational level and by incorporating temporal dynamics under disruption, thereby linking capability-enabled utilization to both efficiency and resilience.76,77 Third, it operationalizes dynamic capabilities for automation, sensing, seizing, and reconfiguring, and shows how their salience varies with maturity, providing a pathway from routines to outcomes.14,15
Several practice-oriented implications follow. Governance should be established before scaling: organizations that put basic governance in place first scaled roughly 3.2× faster than those that attempted to retrofit control later. Hybrid governance, central standards with distributed execution, was associated with higher ROI (52% versus 34% for centralized) and faster time to scale, offering a workable balance between consistency and local innovation.9,13 Automation portfolios should be prioritized as investments, not as isolated projects, with decisions reflecting a balanced allocation across governance, technical, and organizational capabilities (approximately 35–35–30), rather than vendor-leaning technology spend.
Resilience should be designed in from the outset. In this study, organizational adaptability explained more variance in resilience than technical redundancy, and formal playbooks and governance routines were associated with shorter recovery times. Design choices that emphasized modularity, visibility/monitoring, rapid reconfiguration routines, and post-incident learning supported faster adaptation during shocks. 20 Finally, continuous learning mechanisms, communities of practice, documentation standards, and regular post-implementation reviews, were linked to faster capability development, reinforcing the need to institutionalize both single-loop and double-loop learning.
The primacy of governance in this setting invites renewed theorizing about IT governance for emerging automation technologies whose deployment cycles are rapid, and business led. Existing frameworks, developed around monolithic enterprise systems, may not fully account for cross-functional impacts and portfolio-level adaptation. The organizational-level extension of TPC suggests a “resilient fit” perspective that integrates temporal dynamics and environmental turbulence into alignment research. In addition, the configurational nature of results, where balanced bundles outperform single-focused profiles, supports a shift toward configurational methods (for example, QCA) to examine how capability combinations vary across contexts.2,17
Several limitations of our study must also be noted. The Nordic context, characterized by high digital maturity and collaborative culture, may limit generalizability; replication in other regions and regulatory environments is needed. The cross-sectional survey design supports association, not causality; longitudinal designs could track capability trajectories and inflection points. The focus on established automation technologies (RPA, BPM, low-code) leaves open how generative AI and autonomous agents may alter capability requirements, particularly around ethical governance and human–AI collaboration. Finally, the sample emphasized larger organizations (200 + employees). Studies of SMEs could explore alternative strategies, ecosystem partnerships or cloud-based “automation-as-a-service”, that enable capability development under tighter constraints.
Overall, the evidence indicates that automation success is less about any single platform and more about how organizations combine technical, governance, and organizational capabilities over time. By situating automation within an extended TPC and linking dynamic capabilities to organization-level utilization and resilience, the study offers a framework that IS leaders can use to sequence investments, design for adaptability, and evaluate progress as portfolios scale.
Footnotes
Acknowledgments
The author thanks all survey and interview participants for their time and insights. I am also grateful for constructive comments from colleagues during early drafts.
Ethical considerations
This study involved humans (survey and interviews). All participants were informed about the study purpose, data handling, and their rights; participation was voluntary and informed consent was obtained prior to data collection. Data were anonymized prior to analysis.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
An anonymized replication package containing the survey instrument, codebook, and analysis scripts will be made available upon request for the purposes of peer review.
Any other identifying information related to the authors and/or their institutions, funders, approval committees, etc, that might compromise anonymity.
To preserve double-anonymized review, all identifying details about author, institutions, funders, ethics committees, participating organizations, and locations have been removed or generalized.