Data Governance as the Digital Backbone of Proactive Obsolescence Management: A Design Science Case Study in Asset-Intensive Industries

Abstract

Background: The service life and availability of electronic components are steadily declining, whereas the operational lifespan of industrial devices that incorporate such components often extends over several decades. This disparity creates a mismatch between the durability of individual components and the longevity of the overall systems in which they are embedded. Obsolescence Management (OM) addresses this issue by establishing a structured and controlled process aimed at anticipating and mitigating the impacts of component and product obsolescence. As defined by the international standard International Electrotechnical Commission [IEC] 62402:2019, obsolescence refers to the transition of an (electronic) item from availability to unavailability by the manufacturer, in accordance with the original specification. To implement proactive OM, obsolescence managers require data that are comprehensible, accurate, complete, trustworthy, secure, and discoverable. In this context, Data Governance (DG) offers a promising approach to enhance data literacy and intelligence within OM. Methods: This study employed a sequential mixed-methods design, integrating qualitative and quantitative approaches including a Systematic Literature Review (SLR), Expert Interviews (EIs), Focus Groups (FGs), Content Analysis (CA), and Workshops (WKSHs), within a case study informed by Design Science Research (DSR). Results: This paper proposes a DG structure tailored to support OM through data integration and business intelligence methods, drawing on established DG reference frameworks within an SME. The proposed structure encompasses a set of processes and knowledge domains recognized as best practices in the field. Furthermore, we present a model designed to facilitate the implementation of DG in OM and to assess the quality of the data required. This enables more reliable obsolescence processes across key functional areas such as product management, procurement, and product development, ultimately supporting data-driven and accurate decision-making.

1. Introduction and Theoretical Background

For several years, there has been an ongoing societal discourse regarding whether products are deliberately designed for obsolescence—either by being manufactured with limited longevity or by misleading consumers into believing they are durable, while in reality they incorporate intentional design weaknesses that lead to premature failure. This practice is presumed to stimulate repeat purchases and thereby increase corporate revenue (Poppe & Longmuß, 2019). Nevertheless, it is broadly acknowledged that no technical product is intended to function indefinitely under conditions of regular use (Hess, 2017, p. 18; Deutsches Institut für Normung [DIN], 2017).

Various factors contribute to the phenomenon of product obsolescence. Notably, general technological advancement (Bartels et al., 2012; Romero Rojo et al., 2010; Ates & Acur, 2022a; Salas Cordero et al., 2020; Zaabar et al., 2021), the natural aging of installed components (Sandborn & Singh, 2002; Bellmann, 1990, p. 25; Krumme, 2019; Ates & Acur, 2022b), and a decline in market demand (Bartels et al., 2012, p. 8 f) play central roles. Moreover, regulatory frameworks and restrictions (Wilkinson, 2015, p. 14)—such as the Restriction of Hazardous Substances (RoHS, 2011), the Registration, Evaluation, Authorization and Restriction of Chemicals (REACH, 2006), and the Waste Electrical and Electronic Equipment Directive (WEEE, 2012)—as well as liability risk mitigation and societal trends associated with the “throwaway society” (Paech et al., 2020; Bartels et al., 2012), further compel manufacturers to replace products on a regular basis (Bartels & Poppe, 2019), thereby perpetuating obsolescence. Additionally, in the high-tech, service-oriented society of the 21st century, such products exert a considerable environmental burden (Prakash et al., 2016).

In the industrial sector, obsolescence frequently occurs when individual components of a system exhibit shorter service lives and market cycles than the overarching system in which they are integrated (Bartels & Poppe, 2019; Boissie et al., 2022). This discrepancy necessitates a proactive approach to Obsolescence Management (OM) that extends beyond purely technical considerations (Moon et al., 2022). Effective OM relies on data that are comprehensible, accurate, complete, trustworthy, secure, and traceable. In this context, Data Governance (DG) emerges as a strategic framework for enhancing data competence and data intelligence in support of robust OM practices.

The primary objective of this study is to develop a DG structure for OM within a small and medium-sized enterprise (SME), drawing upon established DG frameworks. The concept of DG can be analyzed across three distinct levels: macro, meso, and micro (He et al., 2019). At the macro level, DG is approached from a top-level design perspective, establishing the overarching system framework. The meso level focuses on the formulation of implementation mechanisms, encompassing planning and execution strategies that address DG from a specific functional or organizational dimension. The micro level, in turn, is concerned with practical principles, detailing concrete strategies, procedures, and operational measures applied to individual elements of the system. The actual implementation of the proposed DG structure will be addressed in a subsequent study.

1.1. Product Data

Data can be defined as symbols or signs that are aggregated for the purpose of processing and serve to represent information—specifically, information about facts and processes—within a known or assumed contextual framework (Siepermann, 2018).

In this context, product data are defined as the “representation of information about a product in a formal manner suitable for communication, interpretation, or processing by human beings or computers” (International Organization for Standardization [ISO], 2021, Chapters 3.1.29, 3.1.50, 3.1.51).

In the business context, product data are categorized as master data and exist in various formats. They may be stored electronically—for instance, within Enterprise Resource Planning (ERP) systems—or in physical form, such as product brochures used in sales and marketing, or printed information sheets utilized in logistics. Electronic data are typically centralized, easily accessible, and straightforward to update, thereby facilitating efficient processing, integration, and analysis. In contrast, physical data are location-dependent, less transferable, and more susceptible to inconsistencies and obsolescence. The reliance on physical formats also entails greater manual effort and increases the risk of outdated or contradictory information.

Alongside customers and suppliers, master data represent one of a company’s core business objects, and their quality and effective management are critical determinants of business success (Otto, 2011c). In an increasingly globalized and digitalized business environment, standardized product data formats play a vital role in facilitating both internal and external exchanges of product information (International Organization for Standardization [ISO], 2021; International Organization for Standardization, 2014).

1.2. Product Lifecycle

Every technical product undergoes a distinct life cycle. The concept of the product life cycle was introduced by German economist Theodore Levitt, who published his model in the Harvard Business Review in 1965 (Lewitt, 1965). Although developed decades ago, this model remains widely used today. It describes the progression of a product through five distinct phases (Solomon et al., 2000; Romero Rojo et al., 2009, 2010). In the first phase—development—the company begins to generate awareness of the new product through advertising and public relations; sales begin to rise, but profitability has not yet been achieved. The second phase—growth—marks the point at which the product becomes profitable and begins to attract the attention of competitors. During the third phase—maturity—the product reaches its peak market share and profitability, though it faces increasing competition from similar offerings. Typically, this is followed by the saturation phase, in which both sales and profits begin to decline. Finally, in the decline phase, the product experiences continued loss of market share, negative growth, and falling profits; unless a relaunch occurs, the product is ultimately withdrawn from the market. In essence, the product life cycle encompasses the entire duration from the initial idea to the final, whether planned or unplanned, market withdrawal of the product.

1.3. Obsolescence

According to IEC Standard 62402:2019 (Internationale Elektrotechnische Kommission [IEC], 2019), obsolescence is defined as the transition of an (electronic) item from availability to unavailability by the manufacturer in accordance with the original specification (Internationale Elektrotechnische Kommission [IEC], 2019). This typically refers to the natural aging of a product, which results from material degradation and use-related quality losses, ultimately leading to diminished functionality, reduced performance, or total failure of the product to fulfill its intended purpose (Hübner, 2013; Krajewski, 2014). This form of obsolescence must be distinguished from artificially induced obsolescence—often referred to as planned obsolescence—a concept first introduced by Bernard London in 1932 (London, 1932). Planned obsolescence occurs when a product becomes non-functional or undesirable before the end of its natural lifespan, either due to intentional design limitations or because newer products or technologies render the existing, still-functional product obsolete in the eyes of the user (Hübner, 2013; Krajewski, 2014; Zallio & Berry, 2017).

Obsolescence can have significant consequences for manufacturers. If a critical component becomes obsolete and the manufacturer is unable to fulfill demand or delivery obligations, this may result in reputational damage, loss of revenue, and, in extreme cases, legal claims for breach of contract (Meyer et al., 2003). Thus, obsolescence affects not only end users but also producers. Proactive OM is therefore essential to prevent such risks or to mitigate their impact.

1.4. Obsolescence Management

OM refers to the systematic identification, analysis, and monitoring of critical parts, components, raw materials, and software throughout their lifecycle (Internationale Elektrotechnische Kommission [IEC], 2019; Joint Service Publication [JSP] 886, 2016). Its primary objective is to ensure the long-term availability of spare parts for industrial systems, to respond proactively to product discontinuations, and to develop mitigation strategies along the trajectory of technological development. These strategies aim to safeguard the manufacturing, operation, and maintenance of machinery and industrial plants. The assessment of obsolescence risk should be guided by the criteria of cost, potential impact, and likelihood of occurrence (Romero Rojo et al., 2012).

Operators of capital-intensive goods typically have three conventional strategies for addressing obsolescence: stockpiling through the storage of critical components, raw materials, and software; procuring replacement parts with similar specifications; or reengineering existing machinery and systems. Additional strategies include refurbishing and rebuilding components in accordance with original manufacturer specifications. The selection of an appropriate course of action is a core responsibility of OM and must be made with careful consideration of contextual factors such as storage capacity and associated costs, the complexity and expense of redevelopment, and the technical requirements for integrating replacement components. Regardless of the chosen strategy, effective OM depends on the availability of data that are comprehensible, accurate, complete, trustworthy, secure, and discoverable.

In this regard, DG provides a viable framework to enhance data competence and data intelligence in support of informed and reliable OM decision-making.

1.5. Data Governance

The term DG remains fragmented in both academic research and practical application, rendering it conceptually ambiguous. Initially conceived as an extension of IT governance, the scope of DG has since expanded to encompass a wide range of topics, including data quality, data management, business intelligence and analytics, big data, cloud computing, data protection, and data security. However, this broadening of the field has not led to a unified or standardized scientific definition. Some scholars conceptualize DG primarily in terms of the allocation of decision-making rights over data assets, emphasizing organizational structures and authority (e.g., Wende & Otto, 2007; Otto, 2011a, 2011b, 2012; Weber et al., 2009; Weber, 2009). In contrast, more technically oriented researchers focus on the operational implementation of DG, particularly through available software solutions (e.g., Lee et al., 2018). Additionally, there are integrative perspectives that draw from both organizational and technical dimensions, as seen in the work of institutions such as the MDM Institute (2016) and authors like Newman and Logan (2006) and Gregory (2011).

Despite varying emphases in the literature, there is broad consensus among researchers that DG can be understood as the exercise of decision-making authority and control over all aspects related to organizational data (Thomas, 2006). It involves the formal coordination of processes, people, and technologies—conceived as a human-task-technology system (Heinrich, 1993, p. 173)—and is guided by overarching strategies aimed at enabling organizations to treat data as a strategic asset (MDM Institute, 2016). DG is not a standardized or “out-of-the-box” solution; rather, it must be tailored to the specific organizational culture and context in which it is implemented (Aisyah & Ruldeviyani, 2018). In this regard, DG frameworks serve as valuable tools to support organizations in the systematic design and implementation of DG initiatives.

Figure 1 depicts a two-tier conceptual framework for this study that links strategy with data governance. It is elaborated as a three-level architecture.

The upper tier, “Strategy,” nests the organization’s corporate objectives and strategy together with IT and business strategies and culminates in a dedicated data strategy. This placement signals that data strategy is not an isolated artifact but an integrating layer that aligns business aims with technological direction. The circular arrows indicate iterative alignment: strategic intent informs governance activities, while lessons from governance cycle back to refine strategy. This tier builds the normative level of the frame, where overall policies, rules, and target states are set.

In the framework’s lower tier, DG comprises three mutually reinforcing pillars: processes representing the principles, guidelines, and standards that govern data handling and control. This forms the task level of the frame, where principles, guidelines, and standards are set, which define how data is handled and controlled. The personnel/staff part clearly defined roles and responsibilities that ensure accountability (e.g., data owners, stewards, and custodians), and technology involves the tools and applications that operationalize and monitor those processes. Altogether, this forms the resource level of the frame, where concrete assets (systems, datasets, and people) reside. Dynamic feedback among these pillars, and between them and the overarching data strategy, positions governance as a continuous, adaptive capability rather than a static, one-time design.

The dotted vertical lines signify traceability across levels: specific data objects, data users, and data storage resources (as shown in the legend) are consistently governed from policy through process to implementation. In combination, the two panels communicate that effective data governance requires top-down strategic alignment, bottom-up feedback, and end-to-end traceability from normative prescriptions to operational resources.

1.6. Data Governance Frameworks

A DG framework serves as a logical structure for classifying, organizing, and communicating the complex activities involved in making decisions about, and taking action on, enterprise data. Such frameworks facilitate structured thinking and dialog around otherwise complex or ambiguous concepts, while also offering practical, actionable mechanisms for engaging stakeholders across the organization in the collective goal of transforming data into a strategic asset (Osu & Navarra, 2022). When implementing DG, each organization must critically assess and select the framework that best aligns with its specific context, needs, and strategic objectives.

A wide range of DG frameworks can be found in both the academic literature and practical applications. In this study, three of the most widely recognized models were analyzed in order to examine their structure and components and to evaluate their suitability for integration into an OM framework.

1.6.1. The Data Management Body of Knowledge

Data Management International (DAMA)—formerly known as the Data Administration Management Association—is a non-profit, vendor-neutral global association of business and technical professionals committed to promoting the principles and practices of effective information and data management (DAMA, 2017). In 2017, DAMA published the second edition of the Data Management Body of Knowledge (DAMA-DMBOK), a comprehensive reference framework that consolidates best practices, processes, and guidelines for each key area of data management. Within this edition, DG is positioned as a central coordinating function among ten core data management knowledge areas, which are visually represented in the widely recognized “DAMA Wheel” (see Figure 2).

The eleven knowledge areas outlined in the DAMA-DMBOK framework—including DG—are structured according to a standardized schema comprising seven components: business objectives, guiding principles, key concepts, core activities, tools and techniques, implementation guidance, and associated metrics. These elements are further detailed across 31 defined capabilities and 106 sub-capabilities. While the framework is notably comprehensive, its practical application requires careful adaptation to the specific context, needs, and maturity level of each organization.

1.6.2. IBM Data Governance Unified Process

Industrial Business Machines Corporation (IBM) presents its approach to DG through the “IBM Data Governance Unified Process” (see Figure 3). This framework comprises 14 main steps—ten of which are considered essential and four optional—designed to guide the implementation of an effective DG program. The model is supported by IBM’s proprietary software tools and best practices, providing practical guidance for operationalizing DG within organizational contexts (Soares, 2010).

The ten required steps of the “IBM Data Governance Unified Process” (indicated in blue) form the foundational elements necessary for establishing an effective DG program. Following this, organizations may choose to implement one or more of the four optional focus areas (indicated in green): Master DG, Analytics Governance, Security and Privacy, and Information Lifecycle Governance. To ensure sustained effectiveness, the DG Unified Process must be regularly evaluated through defined performance metrics, and the results should be communicated to key stakeholders to maintain organizational support.

As with other established frameworks, the IBM process operates as a continuous improvement cycle, emphasizing iteration and long-term integration within organizational practices.

1.6.3. Informatica Holistic Data Governance Framework

Informatica, a U.S.-based provider of data integration software, offers enterprise-wide solutions for data integration and data quality management. In 2012, the company introduced its “Holistic Data Governance Framework”, which has since been further developed and refined (Informatica, 2017, 2023). The framework is structured around ten interrelated and complementary dimensions, designed to provide a comprehensive approach to implementing DG across organizational functions (see Figure 4).

The Informatica “Holistic Data Governance Framework” begins with the formulation of a clear vision (1), from which a defined end goal is derived. A corresponding business case outlines the pathway for achieving this goal. The dimensions People (2) and Tools and Architecture (3) represent the key resources required, depending on the degree of automation involved in executing the designated DG tasks. At the task level, the framework includes Policies (4), Organizational Alignment (5), Performance Measurement (6), Change Management (7), Dependent (8) or explicitly defined DG Processes (10), and Program Management (9), all of which collectively support the implementation and sustainability of DG practices (see Figure 1).

1.6.4. Applying Data Governance in Organizational Contexts

Based on the findings of the literature review, a limited number of studies in recent years (see

Table 1. Previous research studies (Author’s elaboration).

Reference(s)	Study Title	Research Method	DG Frameworks
Arinanda (2010)	Designing Data Governance Structure to Establish Data Quality Management Strategy: A Case Study in Directorate General of Taxes	EIs	COBIT 4.1
Herdiyanto (2017)	Designing Data Governance Structure Using the Data Management Body of Knowledge (DMBOK) Guidelines: A Case Study in the Ministry of Research, Technology and Higher Education	CS	DAMA-DMBOK
Aisyah and Ruldeviyani (2018)	Designing Data Governance Structure Based On Data Management Body of Knowledge (DMBOK) Framework: A Case Study on Indonesia Deposit Insurance Corporation (IDIC)	EIs; CA	DAMA-DMBOK
Barrenechea et al. (2019)	Data Governance Reference Model to streamline the supply chain process in SMEs	CS	DAMA-DMBOK
Castillo et al. (2017)	Information Architecture Model for Data Governance Initiatives in Peruvian Universities	CS; OS	Data Governance Institute (DGI); Kalido; IBM
Romero et al. (2019)	Data Governance Reference Model under the Lean Methodology for the Implementation of Successful Initiatives in the Peruvian Microfinance Sector	CS	DGI; DAMA-DMBOK; IBM
Bowo (2020)	Data governance design using Data Management Body of Knowledge (DMBOK) guide: case study at PT JAS	EIs; SLR; CA	DAMA-DMBOK
Putra (2021)	Data governance design using data management body of knowledge guide: A case study at PT. Angkasa Pura I (persero)	CS	DAMA-DMBOK
Anandya (2022)	Designing Data Governance based on Data Management Body of Knowledge (DMBOK): A Case Study of Indonesia Central Securities Depository	CS	DAMA-DMBOK

Note: EIs = Expert Interviews; CA = Content Analysis/Document Observation; SLR = Systematic Literature Review; OS = online survey; CS = case study.

) have examined the application of DG in real-world corporate settings. The purpose of this compilation is to illustrate how various research methodologies and DG frameworks have been employed to develop concrete solutions tailored to specific organizational environments.

However, the comparability of these studies is significantly constrained due to differing contextual factors such as country-specific regulations, market conditions, and legal frameworks. Despite these limitations, the studies collectively demonstrate that the DG frameworks discussed in Section 1.6 provide a sound foundation for the practical implementation of rules, roles, and responsibilities related to data management within organizations.

This study aligns well with the existing body of research, as it similarly applies established DG frameworks and recognized research methodologies, which are presented in detail in the following section.

1.7. Positioning Obsolescence Management Factors Alongside Data Governance Domains

There are some OM factors that line up with DG domains.

OM policy and ownership → Data governance policy, roles, and stewardship. IEC 62402 (Internationale Elektrotechnische Kommission [IEC], 2019) says an OM program must define a policy, an organization/infrastructure, and a plan. DG frameworks, e.g., DAMA-DMBOK (DAMA, 2017), require the same “who decides/owns what” foundation.

Bill of Materials (BOM) accuracy → Data quality and master data. Proactive DMSMS/obsolescence (Livingston, 2000) hinges on a complete, accurate BOM. Studies and guides repeatedly call the BOM “indispensable.” Service/asset BOM quality work (e.g., at ASML, a Dutch multinational corporation and the world’s leading supplier of photolithography machines used in the semiconductor industry) shows why—bad BOM data block early risk detection.

Configuration control/traceability → Metadata and lineage. Avionics guidance stresses configuration control and the cost of losing it; in governance terms, that is data lineage/provenance for parts and configurations.

Continuous monitoring of external signals → Data acquisition and integration. Mature OM programs monitor PCNs/EOL notices, market supply, and supplier status and keep metrics, e.g., DDI DI-MGMT-82275 (DID Group, 2023); governance covers how those data are sourced, integrated, and trusted.

Risk assessment and KPIs → Governance of risk and performance. IEC 62402 (Internationale Elektrotechnische Kommission [IEC], 2019) requires risk-based approaches and “measuring and improving” outcomes; ISO/IEC 38505 (ISO, 2017) centers governance on responsibility, performance, and conformance.

Lifecycle planning (from design to operate) → Information lifecycle management. IEC 62402 (Internationale Elektrotechnische Kommission [IEC], 2019) spans all lifecycle phases; good governance sets rules for lifecycle, retention, and re-use of asset/part data.

Standards, templates, and common data environments (CDE) → Architecture and interoperability. Built-asset standards, e.g., ISO 19650 (ISO, 2020) formalize information requirements/CDEs so asset and parts data stay consistent across parties—exactly a governance concern.

Asset-intensive industry studies show the same pattern. Case studies and sector guidance consistently find that better data governance (quality, lineage, ownership, and information requirements) improves obsolescence and lifecycle outcomes (

Table 2. Case studies in asset-intensive industries (Author’s elaboration).

Industry	Description	Reference(s)
Utilities (water/wastewater)	Sector studies and ISO 55000 (ISO, 2024) guidance emphasize that asset decisions depend on quality, reliable data; integrating disparate systems/governance drives better maintenance/renewal planning—directly reducing “unknown” obsolescence risk.	CWN (2018); Aveva (2021)
Power companies	Research on asset data management frameworks in power utilities frames governance of asset data as foundational to effective asset management and decision-making.	Gavrikova et al. (2022)
Oil and gas/ heavy industry	A proactive OM framework validated on a 35-year-old refinery shows effectiveness when reliability/maintenance data are standardized and governed.	Ghaithan et al. (2024)
	Data quality programs (e.g., automated cleansing for 500k asset records) cut analysis time from months to minutes, enabling better renewal/OM decisions.	Trillium Software (2011)
High-tech manufacturing	The ASML study on service BOMs shows that data quality alerts materially improve BOM correctness—key for proactive obsolescence analysis.	Stip and van Houtum (2020)
Aerospace/defense	DMSMS best-practice guides highlight four success factors—accurate BOM, management commitment, predictive tools, and a team program—all of which depend on governed data and defined roles.	Livingston (2000)
Built environment	ISO 19650 case material (e.g., UK Environment Agency) demonstrates that setting explicit information requirements and a governed CDE improves whole-life asset data quality—supporting decisions on replacements and upgrades.	1Spatial (2025)
Rail	Formal OM Plans and IRIS/IEC alignment improve lifecycle risk control—data structure and configuration traceability are highlighted prerequisites.	IRQB (2022)
Automotive	Simulation/digital-twin-driven OM reduces downtime and cost; effectiveness hinges on the fidelity and governance of asset and spare-parts data.	Cosmotech (2025)
Cross-industry	IEC 62402 makes data monitoring/policy foundational for OM; ISO 14224/asset data standards repeatedly link governed, structured data to better maintenance and obsolescence outcomes.	Ciliberti (2018); (Internationale Elektrotechnische Kommission [IEC], 2019)
	2024 research connects obsolescence/shortages to RAM metrics, reinforcing the need for high-quality, governed reliability data.	Karaani et al. (2024)

Note: ISO = International Organization for Standardization; MRO = Maintenance, Repair, and Operations; BOM = Bill of Materials; LTB = Last time buy; OM = Obsolescence Management; DMSMS = diminishing manufacturing sources and material shortages; CDE = Common Data Environment; IRIS = International Railway Industry Standard; IEC = International Electrotechnical Commission; RAM = reliability, availability, and maintainability.

).

Wherever OM succeeds, you will find strong DG practices underneath—especially around data quality, e.g., ISO 8000/14224 (ISO, 2016), timeliness/change control, lineage across the digital thread, and clear stewardship. Case studies in rail, oil and gas, automotive, and utilities all report similar patterns: govern the data well, and OM risk, cost, and downtime all improve.

2. Methodology and Materials

2.1. Research Approach

This study employed a sequential mixed-methods research design, integrating both qualitative and quantitative approaches to data collection and analysis. The methodology included a Systematic Literature Review (SLR), Expert Interviews (EIs), Focus Groups (FGs), Content Analysis (CA), and Workshops (WKSHs), all conducted within the framework of a CS informed by the principles of Design Science Research (DSR). This combination of methods enabled a comprehensive exploration of the research problem from both theoretical and practical perspectives.

2.2. Research Method

The research process comprised four phases (Figure 5).

Phase 1—the knowledge discovery phase—consisted of a SLR conducted using the search terms “obsolescence” and “obsolescence management”, including their intersection with the term “data governance”. In this context, an SLR is understood as a structured process of identifying, reviewing, and synthesizing existing scholarly work on a defined topic, with the aim of developing a comprehensive understanding of prior research in relation to the study’s research questions (Creswell, 2009; Leavy, 2017; Saunders et al., 2019, p. 72).

In Phase 2, EIs and FGs were conducted with the individual responsible for OM within the SME, as well as with representatives from the relevant business departments. Additionally, a CA was carried out on existing OM documentation. CA is an analytical technique used to systematically categorize and code textual, verbal, or visual data using a predefined coding framework, thereby enabling both qualitative and quantitative interpretation (Saunders et al., 2019, p. 573). The EI method involves purposeful conversations between two or more individuals to collect valid and reliable data from practitioners (Saunders et al., 2019, p. 434). FG refers to moderated group discussions with actual or potential users of a product or process, aimed at generating insights through interaction (Saunders et al., 2019, p. 467). The combined use of these methods supported the identification of key focus areas for the development of a tailored DG approach for OM within the SME. Results are presented in Section 2.3.

In Phase 3, a DSR approach was applied to process and synthesize the collected data using the DG frameworks and reference models described by DAMA (Figure 2), IBM (Figure 3), and Informatica (Figure 4). The DSR methodology was selected due to its focus on solution-oriented research aimed at generating actionable knowledge that supports practitioners in addressing real-world business challenges (Van Aken, 2005). In addition to incorporating strategic and procedural dimensions, the designed DG structure also encompasses clearly defined responsibilities, roles, decision-making domains, and technical support mechanisms. Results are presented in Section 3.1, Section 3.2, Section 3.3, Section 3.4, Section 3.5, Section 3.6, Section 3.7, Section 3.8 and Section 3.9. The resulting DG structure was subsequently validated through a workshop (WKSH) involving key stakeholders from the case organization (see Table 2).

Phase 4 comprises the lessons learned phase, in which the experiences and insights gained from the preceding activities are systematically reflected upon and evaluated. A WKSH served as the primary instrument for this phase. According to Stahlknecht and Hasenkamp (2005, p. 234), WKSHs are structured, interactive events designed to facilitate knowledge transfer, problem-solving, or the collaborative development of new ideas and concepts. They are characterized by a high level of participant engagement, cooperative working formats, and a clearly defined methodological structure. Effective knowledge transfer in a lessons learned phase depends on a blend of human behavior (willingness to share, communication skills, experience and expertise level), organizational culture (blame or learning culture, knowledge-sharing values, recognition and incentives), structured processes (standardized methodology, timing, integration into workflows), supporting technology (collaboration tools, knowledge management systems), and the context of the project itself (complexity). If any of these are weak (e.g., no trust, poor tools, or lessons not reused), knowledge transfer breaks down.

Results are presented in Section 3.10.

2.3. Use Case Environment

The proposed DG model was applied within a real-world use case environment of a globally operating SME to evaluate its practical applicability and scalability.

The core of the SME’s product portfolio consists of solutions for risk mitigation, complemented by a range of related services. Despite having well below 1000 employees (headcount between 350 and 750 employees), the SME’s value chain encompasses all key operational areas, including procurement, production, warehousing, sales and marketing, research and development, and finance and controlling.

The company’s products incorporate various electronic components—such as transistors, resistors, inductors, capacitors, and microprocessors—which are integrated into complex assemblies. These components and assemblies are subject to significant operational and strategic risk, necessitating targeted management. This need arises primarily from the SME’s commitment to ensuring plant reliability and uninterrupted operation. Additionally, several external risk factors contribute to the vulnerability of these components, including market-driven supply constraints from electronic component manufacturers, extended lead times or supply disruptions, communication dependencies with suppliers, technological obsolescence, challenges associated with storing obsolete components, and unforeseen events such as supplier insolvencies.

The company’s environment is influenced by a tight interplay of technological trends, supply chain fragility, regulatory frameworks, economic volatility, and sustainability concerns (

Table 3. Circumference factors (Author’s elaboration).

Factor Group	Factor(s)
Technological factors	Rapid innovation cycles in semiconductors, microprocessors, and electronics force continuous R&D investments.
	Miniaturization and integration trends (e.g., SoCs, IoT) increase design complexity.
	Dependence on advanced manufacturing (like photolithography, nanotechnology) makes sourcing highly specialized.
Supply chain and geopolitical factors	Global supply shortages (e.g., semiconductor crisis since 2020) affect procurement.
	Geopolitical tensions (US–China, EU export controls) restrict access to critical tech or raw materials.
	Supplier concentration risk: key components may be sourced from few regions (Taiwan, South Korea).
Economic factors	Price volatility in raw materials (copper, rare earths) influences costs.
	Global economic cycles affect demand for consumer electronics and industrial equipment.
	Inflation and currency fluctuations impact margins and purchasing power.
Regulatory and legal factors	Compliance with safety standards (e.g., CE, FCC, UL) increases testing and certification needs.
	Export restrictions and tariffs may limit market entry or increase costs.
	Environmental regulations (RoHS, WEEE, REACH) restrict use of hazardous materials and enforce recycling obligations.
Operational and strategic factors	Component obsolescence: short product lifecycles mean constant redesign pressure.
	Quality assurance risks: failures in microprocessors or capacitors can cause recalls.
	Cybersecurity vulnerabilities: as assemblies integrate microprocessors and connectivity, exposure grows.
Market & competitive pressures	Consumer expectations (performance, miniaturization, low energy consumption) push innovation.
	Intense global competition drives down prices, pressuring margins.
	Customer dependency: losing a few key OEM clients can create disproportionate impact.
Environmental and sustainability pressures	Sustainable sourcing: growing need for traceability of rare earth elements and conflict-free minerals.
	Energy efficiency expectations: both in products (low power chips) and in production processes.
	E-waste management: companies face pressure to design for recyclability.

Note: R&D = Research and development; SoCs = System on a Chip; IoT = internet of things; US = United States of America; EU = European Union; CE = Conformité Européenne, the EU safety and environmental compliance; FCC = Federal Communications Commission, the U.S. radio/electromagnetic compliance; UL = Underwriters Laboratories, independent safety certification (mainly U.S., globally recognized); RoHS = Restriction of Hazardous Substances Directive; WEEE = Waste Electrical and Electronic Equipment Directive; REACH = Registration, Evaluation, Authorization and Restriction of Chemicals Regulation; OEM = Original Equipment Manufacturer.

).

These surrounding (“circumference”) factors amplify both operational risks (supply, quality, compliance) and strategic risks (innovation, market positioning, resilience).

Prior to the implementation of the DG approach, an existing OM system was already in place. This system has since been refined and enhanced through the integration of DG activities, with the aim of leveraging recognized efficiencies and improving overall effectiveness.

As indicated above, an initial CA was conducted. Based on existing documentation—such as component lists, supplier information, lifecycle data, risk assessments, and escalation procedures—a range of questions was addressed. These included (a) whether the data sources used are reliable and regularly maintained; (b) whether the OM data is complete or shows gaps, for instance in identifying discontinued or at-risk components; (c) which criteria are applied to identify obsolescence risks and how transparently these are defined; (d) how the existing OM system handles historical data, particularly in terms of tracking past obsolescence cases and the resulting mitigation measures; and (e) whether responsibilities for data management have been clearly assigned.

In the following phase, one-hour EIs were carried out with stakeholders directly involved in the OM process. A guideline was developed for the EIs, consisting of the phases (a) opening, (b) eliciting personal circumstances of the people involved, (c) developing a common understanding of OM, and (d) discussing the findings from CA. A diverse group of participants was selected, including individuals of different genders, with varying professional experience, values, and backgrounds (see

Table 4. Group of participants in EIs (Author’s elaboration).

No.	Gender	Position	Job Experience	Degree
1	Male	Obsolescence manager	>5 years	Engineering (B.Sc.)
2	Male	Product manager	>15 years	Engineering (M.Sc.)
3	Female	Strategic Procurement manager	>15 years	Economics (Business Economist’s degree)
4	Male	Business intelligence architect	>20 years	Business informatics (PhD)
5	Female	Controlling senior expert	>10 years	Economics (Business Economist’s degree)
6	Male	Principal developer	>10 years	Engineering (M.Sc.)
7	Female	Maintenance and repair senior expert	>5 years	Engineering (B.Sc.)
8	Female	Quality manager	>10 years	Engineering (M.Sc.)

Note: B.Sc. = Bachelor of Science; M.Sc. = Master of Science; PhD = Doctor of Philosophy.

), in order to capture the broadest possible spectrum of perspectives (Saunders et al., 2019).

All EIs were conducted during March 2025, using a combination of video conferencing and face-to-face formats. The EIs were subsequently transcribed and validated by the respective participants. Relevant content for the evaluation of the model was then extracted through thematic coding, following the approach outlined by Braun and Clarke (2006).

As a complement to the individual EIs, a FG was conducted to capture collective perspectives and interactions among stakeholders involved in OM, as well as to discuss preliminary research findings. The FG consisted of the aforementioned representatives and served as a platform for joint reflection on challenges, data- and interface-related issues, and potential areas for improvement within the existing OM system. The moderated discussion facilitated the identification of diverse viewpoints and interdependencies that were only partially evident in the individual interviews. The results of the focus group provided valuable input for the further development of an integrated and holistic OM approach.

3. Results and Discussion: Data Governance in Obsolescence Management

The team adopted a structured approach to designing the DG structure within the OM framework of the SME. The point of departure was an analysis of OM from the perspective of decision- and system-oriented business administration, which led to the development of an overarching architectural framework (see Section 3.1). Building upon this foundation, the team investigated which factors contribute to the success of OM or exert a lasting influence on its effectiveness (Section 3.2). Among these success factors, the availability of high-quality data emerged as particularly critical, thereby highlighting the necessity of a DG framework (Section 3.3). To determine an appropriate DG approach, the team explored whether data-related challenges exist within the current OM system and, if so, what specific issues could be identified (Section 3.4). Based on these insights, suitable DG solution approaches were identified (Section 3.5), followed by the development of a concrete DG model (Section 3.6). Subsequently, key elements such as processes (Section 3.7), organizational structures (Section 3.8), and technological support mechanisms (Section 3.9) were defined.

3.1. A System-Oriented Perspective of Obsolescence Management

To manage complex business systems, the system is hierarchically decomposed into subsystems whose components can be recursively subdivided. This hierarchy separates an external from an internal perspective: externally by interfaces and observable behavior (the outside view) and internally by structure and subsystem behavior (the inside view). Such a multi-level description is a fundamental device for reducing complexity in terms of system theory.

In this case, OM can be understood as an open, dynamic business system that engages in continuous interaction with its external environment through various forms of transactions (Ferstl & Sinz, 1997). It operates on the basis of coordination principles such as feedback control and negotiation. The openness of OM is evidenced by the transgression of its system boundaries, as seen in its interactions with external environmental entities such as suppliers—e.g., through the strategic procurement of critical components—and external service providers such as SiliconExpert, which supports the acquisition of information regarding manufacturers and the lifecycle specifications of electronic components.

The behavior of the OM system is guided by a set of objectives comprising both main and formal targets. Main objectives refer to the intended outcomes of OM, such as the reduction or mitigation of business risks associated with electronic components and assemblies. Formal objectives, by contrast, define the standards of performance expected after the implementation of OM, typically expressed in terms of time, cost, and quality—for instance, the minimization of costs through efficient and effective OM processes. The execution of OM tasks is performed collaboratively by personnel and machines (as actors of the tasks) and is increasingly supported by data, together constituting a socio-technical system characterized by synergistic interaction.

The conceptualization of OM—described as an open and goal-oriented system above—refers to an external perspective of the OM as a business system (see above). From an internal perspective (see above), OM can be described as a distributed system composed of autonomous and loosely coupled components (Ferstl & Sinz, 1993, 1995; Hartmann & Wolf, 2016). These components correspond to the business processes of strategic, proactive, and reactive OM, which operate in coordination to achieve the overarching objectives of the system. Depending on the level of automation, these processes draw on both human and technological resources. This dual perspective culminates in the development of an architectural framework for OM—referred to as the OM architecture—which structures and aligns the various system elements in support of efficient and effective OM (see Figure 6).

The layered framework for OM links strategic intent to organizational execution through two vantage points. At the top sits the OM strategy layer, which articulates the organization’s long-term posture toward anticipating, mitigating, and responding to obsolescence risks across products, components, and suppliers. These represents an “outside perspective” on OM. Beneath this, the “inside perspective” structures how the strategy is realized through processes and the resources that enable them.

The process layer decomposes OM into three complementary modes—strategic, proactive, and reactive—and organizes work into three process classes. OM processes provide governance, coordination, and performance control, OM core processes deliver the primary analyses and interventions (e.g., lifecycle forecasting, risk assessment, mitigation planning), and OM support processes supply enabling services such as procurement support, knowledge management, and supplier liaison. The tripartite (strategic–proactive–reactive) view clarifies time horizons and triggers: strategic sets direction, proactive monitors and prevents, and reactive addresses emergent obsolescence events.

The resource layer enumerates the inputs required to operate these processes: labor, technology, data, and other resources. The highlighting of data underscores its pivotal role as a cross-cutting asset that informs every mode and process class by providing evidence for forecasting, decision support, and performance feedback. Overall, the diagram conveys a coherent alignment: strategy guides processes, processes consume and transform resources, and data act as a unifying resource that enhances the effectiveness and traceability of OM activities. Object “data in OM”—as the digital footprint of all OM-related activities in the context of digital transformation—serves as the point of departure for implementing DG. The following section presents the key success factors for effective OM.

3.2. Success Factors in Obsolescence Management

Success is generally understood as the positive outcome of efforts or the realization of an intended effect. However, this broad definition does not specify which goals are pursued or which factors influence success in OM, where the absence of critical electronic components can often result in significant cost implications (Barthels, 2018).

As a result of the team’s discussion, ten success factors (SFs) were identified (see

Table 5. Success factors in OM (Author’s elaboration).

Success Factors (SFs)		Architecture Level 1							Reference(s)
UID	Description	S	P	R			Named in		Source(s)
				L	T	D	SLR	EI
SF1	Systematic management of electronic components (as business-critical objects) and maximization of the level of component monitoring	x	x	x	x	x	x	x	(Barthels, 2018)
SF2	Ensuring the flow of information throughout the supply chain	x	x	x	x	x	x	x	(Barthels, 2018)
SF3	Automated monitoring of (supply) bottlenecks		x		x	x		x
SF4	Establishment of secondary supplier sources and long-term integration of partner suppliers	x	x			x		x
SF5	Obsolescence analysis, goods evaluation, and risk assessment (identification, analysis, evaluation) on the basis of high-quality data and defined KPIs	x	x	x	x	x	x	x	(Barthels, 2018; Ferreira et al., 2019)
SF6	Implementation of OM as a management function	x	x				x	x	(Barthels, 2018)
SF7	Establishment of OM as a value (open communication of opportunity costs)	x	x				x		(Barthels, 2018)
SF8	Definition of roles and responsibilities for processes and data	x	x				x	x	(Barthels, 2018)
SF9	Implementation of an OM team and continuous knowledge process			x			x	x	(Meyer et al., 2003)
SF10	Modern information landscape (technology) for the maintenance of business-critical data objects in OM and support in the evaluation of this data				x			x

Note: UID = Unique Identifier; S = Strategy; P = Processes; R = Resources; L = Labor; T = Technology; D = Data; SLR = Systematic Literature Review; EI = Expert Interviews/group discussion; ¹ = see Figure 6.

) and assigned to the three model levels of the OM architecture: strategy, processes, and resources (Section 4.1). SFs represent key variables that determine the sustainable performance of operational systems, both holistically and within specific domains such as OM. When these factors are adequately addressed, the system—either in its entirety or within the defined area—is likely to achieve success. Conversely, deficiencies in these factors can have a direct negative impact on partial or overall outcomes (Szczutkowski, 2018).

Some of the identified success factors are derived from the previously published literature and were partially validated by the experts. Others emerged exclusively from the expert discussions. It became evident that data plays a particularly critical role, especially in light of the continuously increasing degree of digitization in operational systems.

3.3. Reasons for Need for Data Governance

The need for the introduction and implementation of DG in OM was determined on the basis of internal documents and interviews as part of the as-is analysis, the results of which are presented in

Table 6. Reasons for need for DG in OM (Author’s elaboration).

Reasons for Need (RfNs)		Success Factors (SFs, see Table 5)
UID	Description	SF1	SF2	SF3	SF4	SF5	SF6	SF7	SF8	SF9	SF10
RfN1	Constant changes in safety requirements, regulations, provisions and conditions	x	x	x		x		x
RfN2	Complex dependencies in the procurement process due to market allocations among suppliers	x	x		x			x	x
RfN3	Lack of alternatives for second and third suppliers in procurement	x			x
RfN4	Complicated demand planning for parts and components due to distributed organizational responsibility	x	x	x	x	x		x	x
RfN5	Complex management of product life cycles due to distributed data silos	x	x	x	x	x	x	x	x
RfN6	Expected discontinuation of parts in the next few years with an expected volume in the tens of millions	x	x		x
RfN7	Lack of or insufficient IT support for OM data in IT systems	x	x	x		x			x		x
RfN8	Complex approval procedure for the procurement of new parts in purchasing	x	x	x	x	x			x
RfN9	Incomplete identification and description of relevant business processes in OM (auditable process quality)	x		x	x	x			x
RfN10	Incomplete coverage of key KPIs in definition and implementation	x				x	x		x
RfN11	Distributed data sources	x				x					x
RfN12	Lack of or insufficient data skills in the organizational units involved	x				x				x

Note: UID = Unique Identifier; RfNs = Reasons for Need; SFs = Success Factors.

. They were compared with the success factors identified in a previous step (Section 3.2). Reasons for need are reasons for the desire for a successful OM, experienced as a lack by the economic subjects acting in an operational system. They concern overcoming barriers in OM.

In addition to the constantly changing regulatory requirements for the security of electronic systems and the complex dependencies in the company’s own business processes, organizational hurdles in the form of complicated approval procedures have also been identified as barriers. These reasons have an indirect effect on the “data” object. Furthermore, typical barriers were identified in connection with the “data” object itself. These include data silos, incomplete coverage of monitoring using key performance indicators (KPIs), and insufficient data excellence in the organizational units involved. They all have an (indirect or direct) impact on the success of the OM and must therefore be addressed.

3.4. Challenges in the Data

Treating relevant data as an operational asset is a mandatory prerequisite for successful OM. This requires comprehensible, accurate, complete, trustworthy, secure, and retrievable data. In this context, various challenges were identified in the SME, which were related to the described reasons for needing DG (see

Table 7. Challenges in OM’s data (Author’s elaboration).

Challenges (CHs) in OM’s Data		Reasons for Need (RfNs) of DG (See Table 6)
UID	Description	RfN1	RfN2	RfN3	RfN4	RfN5	RfN6	RfN7	RfN8	RfN9	RfN10	RfN11	RfN12
CH1	… in the quality of the data due to a lack of or insufficient data maintenance		x	x	x	x		x		x
CH2	… in the use of the data.					x						x
CH3	… in the responsibilities for the data.					x			x
CH4	… in the data infrastructure and data management due to conflicts of interest between the units involved in data and information.					x				x
CH5	… in business rules, standards and processes…		x	x	x	x	x		x		x
CH6	… in compliance with legal requirements and internal compliance.	x									x
CH7	… in personnel/labor.												x
CH8	… in technology											x

Note: UID = Unique Identifier; CHs = Challenges; RfNs = Reasons for Need.

).

Data infrastructure and data management (CH4) constitute the foundational elements upon which a successful operations management (OM) system must be established. Legal provisions—such as data protection, security, and regulatory compliance—and adherence to these requirements (CH6) provide the framework for responsible and legally sound data handling. The quality of the data (CH1) directly influences the extent to which it can be utilized effectively and purposefully (CH2). The coordination of data processes necessitates clearly defined business rules and standards (CH5), along with the corresponding responsibilities for data maintenance (CH3). A thorough understanding of appropriate technologies—such as business intelligence and artificial intelligence—and their respective areas of application (CH8) is essential for leveraging digital tools in conjunction with OM’s digital infrastructure in a goal-oriented manner. Employees must be equipped with the necessary data competencies and supported in integrating data into their daily work to enhance performance. This entails a fundamental transformation in the use of data, as well as a shift in mindset and organizational identity—commonly referred to as data culture.

The SFs (Table 5), the RfN of DG (Table 6), and the recognized CHs related to data (Table 7) collectively form the basis for effectively addressing the tasks of DG and its specific functions.

3.5. Data Governance Solution Process

To address the identified challenges in data relevant to OM (see Table 7) and thereby target the associated success factors (see Table 5), appropriate DG solutions (SOLN) were sought. The DG frameworks outlined in Section 1.6 are intended to serve this purpose.

The team did not rely on a single framework. The rationale for combining the aforementioned frameworks lies in the fact that each, when considered individually, presents certain limitations (Castillo et al., 2017), which can be mitigated through their integration. As demonstrated in

Table 8. DG solution procedures for overcoming identified data challenges (Author’s elaboration).

Solutions (SOLNs) to Alleviate or Overcome Recognized Data Challenges		Challenges (CHs) (See Table 7)	DG Frameworks (See Section 1.6)
UID	Description		DAMA	IBM	Informatica
SOLN1	Data quality	CH1	x	x
SOLN2	Data Utilization	CH2	x
SOLN3	Roles and responsibilities	CH3		x	x
SOLN4	Data management	CH4	x
SOLN5	Business rules, standards and processes	CH5		x	x
SOLN6	Legal requirements and compliance	CH6		x
SOLN7	Personnel/Labor	CH7			x
SOLN8	Technology	CH8			x

Note: UID = Unique Identifier; CHs = Challenges; SOLNs = Solutions.

, the combined application of all three frameworks provides comprehensive support for DG within OM.

3.6. Data Governance Framework

Our model (Figure 7) adopts both a management- and technology-oriented interpretation of DG and encompasses six key domains (DG Scope), each aligned with one of the six identified success factors of OM.

The model’s core components are derived from the generic DG model (Figure 2) and are informed by established frameworks, including DAMA-DMBOK, IBM, and Informatica.

The resource model—comprising human and application system resources—was subsequently specified as a sub-model within the internal perspective of OM at the task owner level. The resources required for executing tasks in OM are represented by the “OM team” and the “Analytical OM system,” which together form a coherent socio-technical system. These components are described in greater detail in the sections that follow.

3.7. Data Governance Integration into the Process Organization

For DG to be effective within OM, it must be integrated into existing process organization—across all levels of the process hierarchy. The foundation for this integration is the classical differentiation of business processes into management, core, and support processes, as applied to the context of DG (Dittmar & Fürber, 2020). Management processes (e.g., data strategy) establish the framework conditions necessary for DG. Core or service processes (e.g., data quality management, master data management) represent the actual value-creating activities related to data, while support processes (e.g., controlling) facilitate the efficient execution of the core processes.

Figure 8 illustrates an example of an OM process—“Not Recommended for New Designs” (NRND)—as a representative segment of the overall OM process landscape.

The diagram couples an OM workflow (upper pane) with the supporting DG pipeline (lower pane). The process begins when an obsolescence alert, specifically a “not recommended for new design” (NRND) notice, is received. Components flagged in the enterprise resource planning (ERP) system are screened, and two assessments are performed: (a) their functional use based on the engineering part list, and (b) their commercial significance based on the purchase part list. Two decision points follow. First, if a component is strategic, the organization immediately searches for and qualifies second- or third-party suppliers. If it is non-strategic, the case is monitored, and the team waits for a formal notice of discontinuation. Second, the procurement value determines the intensity of action: low-value parts (e.g., <EUR 2500) are monitored, while mid-value parts (≥EUR 2500 and <EUR 20,000) and high-value parts (≥EUR 20,000) trigger an investigation of the manufacturer’s discontinuation strategy, supported by external intelligence (e.g., SiliconExpert). All outcomes are consolidated in a result documentation step that produces a report and updates the relevant data repositories.

The lower pane explicates the DG backbone that enables these decisions. It depicts a linear but iterative pipeline—fetch, transform, analyze, and evaluate the data—that ingests information from operational systems (ERP, part lists) and external sources (e.g., SiliconExpert).

By structuring the data work in this way, the model ensures traceability, data quality, and repeatability of the analyses that inform each decision node in the obsolescence management workflow.

3.8. Data Governance and Organizational Structure: Obsolescence Management Team

At the organizational level, the core element of DG—the role model—was defined (Figure 9). The starting point is the identification of abstract roles (WHO?), which are responsible for and carry out the tasks associated with specific DG fields of action (Weber & Klingenberg, 2020, p. 48). Once these typical roles are defined, their corresponding competencies and responsibilities (WHAT?) are specified. In a final step, specific individuals are assigned to the respective roles, thereby operationalizing the role model within the organization.

The figure depicts an organizational design that integrates OM with DG across the firm’s primary functions involved in OM—purchasing, logistics, product management, and development. A joint board, representing OM and DG management, signals a shared, cross-functional authority that aligns policy, priorities, and oversight for both disciplines rather than leaving them to operate in separate silos.

The zoomed section clarifies the DG layers through which this authority is exercised within each function. From top to bottom, a managing board sets direction, an OM board provides cross-functional governance and escalation, an OM core team executes day-to-day coordination, and an extended OM team connects additional stakeholders as needed. Running in parallel. The DG hierarchy assigns accountability for information used in OM decisions: a client (business owner) mandates outcomes, a data lead orchestrates data policy and quality, data stewards (business and IT) bridge business requirements and technical implementation, and operational roles—users, data producers, and data custodians—consume, generate, and safeguard data, respectively.

For the concrete OM implementation within the SME, it was decided that all roles below the managing board (client) would be assigned in addition to their operational responsibilities. This approach ensures the continuous retention of expertise related to business processes and applications. The client represents the organizational and financial sponsorship of OM at the highest management level of the SME. At the level of the OM board, the role of data lead was established, holding central responsibility for all data management activities within OM and coordinating the various DG roles. Specialist data stewards are responsible for ensuring the quality of relevant OM data objects from a domain-specific perspective, while technical data stewards provide technical support in managing data. Users interact with OM data within their respective process areas or utilize it for decision-making purposes. Data producers are responsible for maintaining OM data in accordance with established data management standards and policies. Finally, data custodians are tasked with implementing business requirements within the relevant information systems.

3.9. Data Governance and Technology: Obsolescence Management Analytical System

To support information technology within OM, the decision was made to implement a business intelligence (BI) system, called the OM Analytical System—OMAS (Figure 10).

Enterprise Resource Planning (ERP) and Supply Chain Management (SCM) systems function as internal data sources, while SiliconExpert Technologies—a platform comprising over 20,000 electronics distributors and suppliers, and the leading provider of software for managing electronic component data and parts in the electronics industry (https://www.siliconexpert.com, accessed on 10 April 2025)—serves as an external data source. All data are consolidated and stored in a centralized knowledge database (data warehouse), where they are processed and transformed into actionable information for OM. These refined data serve as the foundation for various reports and analyses, including the OM cockpit, product lifecycle planning, assortment analyses, risk monitoring of electronic components, and other decision-support tools.

3.10. Validating the Effectiveness of the Proposed Data Governance Structure

The validation of the proposed DG structure for OM is conducted through two complementary approaches: First, cross-sectional-validation is carried out using defined key performance indicators (KPIs) related to data quality, user satisfaction, compliance rates, and decision-making efficiency. Second, longitudinal (temporal) validation is performed through regular audits, continuous stakeholder feedback, and periodic alignment with the established organizational management (OM) objectives.

Table 9. Cross-sectional validation KPIs (Author’s elaboration).

Group	Used KPIs
Data quality metrics	Accuracy (percentage of data entries that correctly reflect real-world values); Completeness (proportion of required data fields that are filled in); Consistency (degree to which data is uniform across systems, e.g., no conflicting entries); Timeliness (time lag between data creation and availability for use); Validity (percentage of data that conforms to defined formats, rules, or standards); Uniqueness (number of duplicate records in the dataset); Integrity (percentage of records with valid relationships due foreign keys correctly linked);
Satisfaction metrics	User Satisfaction Score (survey-based rating of user satisfaction with data availability, quality, and usability); Net Promoter Score (measures willingness of users to recommend the data systems/tools to others); Issue Resolution Time (average time taken to resolve data-related user requests or problems); Data Accessibility Rating (user feedback on ease of accessing required data for tasks and decisions); Training and Support Effectiveness (user ratings of DG training and support services); Adoption Rate of DG Tools/Processes (percentage of users actively using governed data tools or following defined data processes); Feedback Participation Rate (proportion of users who provide regular feedback, indicating engagement with the DG process); number of defined data owners/stewards; proportion of documented data objects; implementation time for data policies.
Compliance rate metrics	Policy Compliance Rate (percentage of data processes or records that adhere to defined DG policies and standards); Regulatory Compliance Rate (percentage of processes or datasets that meet external legal and regulatory requirements, e.g., GDPR, ISO); Data Stewardship Task Completion Rate (proportion of assigned governance tasks completed on time by data stewards); Audit Finding Rate (number or percentage of compliance issues identified during internal or external audits (lower is better); Exception Rate (percentage of data entries flagged for violating business rules or governance controls); Training Completion Rate (percentage of relevant personnel who have completed mandatory DG or compliance training); Access Control Compliance (proportion of data access permissions aligned with role-based access policies).
Decision-making efficiency metrics	Decision Cycle Time (average time taken from data request to final decision); Data Availability Rate (percentage of decisions supported by timely and complete data); User Confidence Score (survey-based metric on users’ trust in data quality and reliability for decision-making); Rework Rate (percentage of decisions that had to be revised due to data errors or inconsistencies); Use of Data-Driven Decisions (proportion of decisions explicitly based on data insights or analytics); Decision Accuracy (post-decision analysis showing how often data-informed decisions led to desired outcomes); Time to Insight (time from data collection to generation of actionable insights);

Note: DG = Data Governance; GDPR = General Data Protection Regulation; ISO = International Organization for Standardization.

presents examples of KPIs used for cross-sectional-validation.

To support visualization, a dashboard was designed. The dashboard serves the ongoing monitoring and evaluation of the effectiveness and efficiency of the proposed DG structure through clearly presented KPIs. Figure 11 presents a mock-up of the used dashboard.

The introduction of the DG structure led to measurable improvements across multiple dimensions of data management and usage. While the average resolution time for data quality issues remains relatively high at six working days, it was reduced by 50%, indicating a clear positive trend. Notably, 80% of surveyed data stewards reported greater clarity in data-related roles and processes, which results in faster data problem solving. The data availability rate increased from 68% to 85%, enabling more timely and informed decision-making. The share of data-driven decisions rose from 54% to 72%, reflecting a growing reliance on analytical insights. User trust in the operational model improved significantly, with the user trust index rising from 2.2 to 3.9 on a five-point scale. The rate of post-decision rework due to data issues dropped from 18% to 8%, suggesting enhanced data quality. Compliance with internal standards and policies rose from 48% to 73% following the definition of clear data responsibilities, while access control compliance increased from 69% to 95% through the implementation of automated permission checks. These exemplary results collectively indicate a substantial positive impact of the DG framework on data quality, governance, and decision effectiveness.

Regarding financial benefits, the proposed DG structure provided crucial product and component information to help the OM team proactively, reactively, and strategically handle OM cases and avert scenarios that could turn out to be costly for the company, e.g., suppliers discontinuing parts used in critical SME products. With the OMAS, the OM team could see the affected list of products and their consumption and estimate future consumption in order to buy large quantities of discontinued parts, ultimately avoiding a product redesign that usually costs hundreds of thousands of euros. Having all information in one OMAS saves time and minimizes efforts to gather data, because in critical OM cases, the time to prepare an OM case is very crucial, with thousands of euros at stake, and the time to buy parts from the supplier is also limited. The OMAS enabled seamless OM processes, which led to the timely preparation of crucial OM cases and saved capital.

4. Conclusions

4.1. Summary of the Key Findings

This study presents the development of a DG structure for OM in a SME. Guided by DSR, this CS used a sequential mixed-methods approach that combined SLR, EIs, FGs, CA, and WKSHs.

Based on a system-oriented perspective on OM, success factors of OM, reasons for needing DG, and challenges with data in existing OM were identified. The results formed a basis for the construction of a DG structure with concrete implementation of processes and subsequent definition of organizational structure and technological support through BI.

Implementing the DG framework halved issue resolution time (12 → 6 days), raised data availability (68 → 85%), data-driven decisions (54 → 72%), user trust (index 2.2 → 3.9), and compliance with standards (48 → 73%) and access controls (69 → 95%), while cutting rework due to data errors (18 → 8%) and giving 80% of data stewards clearer roles. Taken together, the findings show that the DG framework markedly improves data quality, strengthens governance practices, and boosts decision effectiveness.

4.2. Theoretical Implications

A mature data governance framework provides the informational backbone of proactive OM. By reducing information asymmetries, rendering lifecycle risks quantifiable, and embedding accountability in clearly defined roles, it enables organizations to identify obsolescence earlier, act strategically rather than reactively, and sustainably mitigate cost drivers such as emergency procurement, production stoppages, and redesigns.

Table 10. Theoretical implications (Author’s elaboration).

DG Lever	Theoretical Implications for OM	Theoretical Framing
Data quality and integrity	Clean, complete, and valid datasets enable early identification of components at risk of discontinuation and improve the accuracy of obsolescence forecasts.	Information-quality theory: higher data quality reduces decision uncertainty.
Lifecycle orientation	DG controls data “from cradle to grave,” providing an unbroken history for long-lifecycle products (e.g., spare-part traceability) on which OM relies.	Product-Lifecycle Management and ISO 14224/IEC 62402 (Internationale Elektrotechnische Kommission [IEC], 2019): comprehensive data trails underpin lifecycle decisions.
Roles, rights, and responsibilities	Clear owner/steward models prevent “orphan data”; every item has a responsible party who evaluates discontinuations and initiates countermeasures.	Principal-agent theory: unambiguous accountability lowers coordination and agency costs.
Interoperability and metadata standards	Standardized interfaces, APIs, and metadata integrate data flows from engineering, procurement, and service—the foundation for holistic obsolescence analytics.	Systems/network theory: interoperability reduces system friction and helps avert technological obsolescence.
Risk and compliance management	DG establishes controls (e.g., data-retention and archiving rules) that make regulatory discontinuation risks (REACH, RoHS, etc.) transparent.	Enterprise risk management: a single, trusted data source mitigates compliance and supply risks.
Analytics and early-warning systems	Curated master data and failure histories allow ML models to estimate time-to-obsolescence, size last-time-buy orders, and simulate retrofit scenarios.	Resource-based view: data as a strategic asset confers predictive and cost advantages.
Knowledge preservation and organizational learning	Versioned data objects, lineage information, and change logs prevent knowledge erosion; OM lessons feed back into new design cycles (“design for obsolescence”).	Knowledge-management theory: DG functions as an organizational memory.
Cost/benefit balance	Higher DG maturity incurs start-up costs but markedly lowers the total cost of obsolescence (emergency procurement, redesigns, production stops)	Transaction-cost economics: standardized data processes minimize opportunistic costs under crisis conditions.
Culture and decision capability	Governance fosters a data-driven mindset; OM decisions become evidence-based rather than experience-based, reducing investment risk.

Note: DG = Data Governance; OM = Obsolescence Management; ISO = International Organization for Standardization; REACH = Registration, Evaluation, Authorisation and Restriction of Chemicals Regulation; RoHS = Restriction of Hazardous Substances Directive; API = Application Programming Interface.

summarizes the theoretical implications.

4.3. Practical Implications

The integration of DG and OM can create significant synergies as both areas focus on the effective management and use of data and information based on it. A well-established DG structure can support OM by ensuring that relevant data on the lifecycle of components and technologies are available, accurate, and accessible. DG makes OM faster, more accurate, and more cost-effective.

Table 11. Practical implications (Author’s elaboration).

Group	Practical implications
Real-time transparency over Bills of Materials and spares	Unified data catalogs merge engineering, procurement, and inventory data, exposing obsolete items and duplicate safety stocks instantly.
Automated EOL alerts from suppliers	Standardized interfaces push discontinuation notices directly into the DG system, opening Last-Time-Buy windows and scheduling redesigns on time.
Data-driven procurement and inventory strategies	Integrated usage histories and demand forecasts optimize order quantities and stock levels.
Accelerated release and change workflows	DG routings distribute engineering change notifications automatically to purchasing, service, and quality.
Digital thread as closed feedback loop	Continuous data lineage from design to disposal feeds field-failure data back to engineering and supply-chain planning.
Streamlined audit and compliance evidence	DG logs every access, decision, and data change, enabling one-click generation of RoHS/REACH documentation.
Productivity boost within OM teams	Role-based access, curated metadata, and self-service catalogs replace ad hoc data hunting.
Quantified risk and cost reduction	Earlier detection of obsolete parts avoids unplanned downtime, emergency procurement, and redesign surcharges.

Note: DG = Data Governance; EOL = end-of-life; RoHS = Restriction of Hazardous Substances Directive; REACH = Registration, Evaluation, Authorisation and Restriction of Chemicals Regulation; OM = Obsolescence Management.

summarizes the practical implications.

By combining DG and OM, companies can optimize their processes, extend the life of their products, and reduce the costs and risks associated with obsolescence. This integrated approach enables proactive and data-driven decision-making, which ultimately helps increase efficiency and competitiveness.

4.4. Boundaries of the Study

One boundary is regarding the conceptual scope: the study focusses on information-centric mechanisms (data quality, roles, and standards) and does not analyze engineering design alternatives (e.g., modularity) that also influence obsolescence. Empirical data are drawn from a single company; the case involves a single SME. The study also reflects practices within EU and evidence relies on self-reported research methods. Benefits are measured over a short observation time period.

4.5. Areas for Future Inquiry

Future research in the field of OM will focus on various key topics. One important area is the development of advanced predictive models that use artificial intelligence (AI) and machine learning to make more accurate predictions about the lifetime of components and technologies. Further research could focus on creating standardized frameworks and methodologies for OM in different industries to improve the comparability and transferability of best practices. Another field of research could include the investigation of the environmental and economic impact of obsolescence and the development of sustainable strategies to reduce e-waste. In addition, the integration of OM into the early stages of product design and development could play an important role in creating products that are better prepared for the challenges of obsolescence from the outset. And for all these key topics—as a result of increasing digitalization—data are the fuel, the oil that keeps everything “going”. First and foremost, data are not just a technological issue, as the success of this technical implementation and technical measures taken in OM also depend on the people involved. Here, DG, in particular, and data culture, in general, can provide solutions to pave the way for more data competence, convictions, values, and sustainable behavior in OM.