A semiotics-informed conceptual maturity model for industrial IoT integration

Industrial Internet of Things (IIoT), the industrial counterpart of the Internet of Things (IoT), is recognised as a key enabling technology of digital transformation (DT) across industries (Ahmed et al., 2023). By connecting physical assets, digital platforms, and enterprise information systems, IIoT technologies allow organisations to collect real-time data, monitor operational processes, and automate decision making. In industrial environments, IIoT technologies enable the integration of cyber-physical systems, predictive analytics, and intelligent automation to improve operational efficiency and create new service models (Khan et al., 2020).

Despite these opportunities, the integration of IIoT technologies into enterprise environments remains complex. Organisations must coordinate heterogeneous technological infrastructures, enterprise systems, data architectures, and organisational capabilities to deploy IIoT solutions effectively. These technological and organisational interdependencies can hinder organisations from assessing their capabilities and planning effective IIoT-enabled transformation (Benotmane et al., 2023; Pino et al., 2024b).

Maturity models (MMs) support these transformations by assessing organisational capabilities and guiding their development. They represent increasing sophistication across technological, organisational, and managerial capabilities and help organisations identify capability gaps and improvement pathways (Benotmane et al., 2023; Mettler, 2011).

Reflecting the growing importance of IIoT technologies, several related MMs have been proposed in recent years. These models typically conceptualise maturity as a staged progression across several types of capabilities (Klisenko and Serral Asensio, 2022; Benotmane et al., 2023; Serral et al., 2020). However, several limitations remain in the current literature. First, many models emphasise broad organisational, technological, and human capabilities common to generic IT (Information Technology) MMs, while offering limited insight into how IIoT reshape business operations (Stoiber and Schönig, 2022). Second, existing models vary significantly in how they conceptualise maturity, resulting in heterogeneous frameworks that lack conceptual consistency across studies (Klisenko and Serral Asensio, 2022; Parab and Deshmukh, 2023). Third, several models remain context-specific, limiting their applicability across different organisational settings (Pino et al., 2024b).

These limitations indicate that existing research provides only partial guidance on how IIoT capabilities evolve within enterprise environments. Limited theoretical guidance exists for systematically conceptualising and assessing the technical and operational elements of IIoT integration. This study, therefore, develops a MM for assessing the integration of IIoT technologies within enterprise environments. It uses the Semiotic Ladder (Stamper, 1993), an artefact of organisational semiotics, as a structuring and interpretive lens for organising the elements involved in IIoT integration. The study addresses the following research question: How can the maturity of IIoT integration within enterprise environments be systematically conceptualised and assessed? Thereby developing a theory-informed conceptualisation of IIoT integration maturity.

The remainder of this article is structured as follows. Section 2 presents the key concepts related to the study, including IIoT integration, MMs, and the theoretical foundations. Section 3 describes the research approach. Section 4 presents the proposed IIoT MM. Section 5 discusses the model's interpretation, limitations, challenges, contextual factors, trade-offs, research agenda, and contributions. Section 6 provides concluding remarks. Appendix A provides additional methodological details supporting the development and evaluation of the MM, while Appendix B extends the current work by presenting a preliminary self-assessment grid for applying the proposed model.

This section establishes the conceptual and theoretical foundations for the study by first defining IIoT integration maturity, then examining relevant maturity models (MMs) across information systems (IS), digital transformation (DT), and IIoT. It then introduces the Semiotic Ladder as the theoretical lens used to structure and interpret the proposed MM.

IIoT integration is central to industrial DT (Stoiber and Schönig, 2022, 2024) and can be understood through several related literature streams. Socio-technical perspectives conceptualise integration as the alignment between technologies, people, organisational routines, and work practices (Iden and Bygstad, 2025). Data ecosystem perspectives emphasise the sharing, governance, and use of data across multiple actors (Bartolomucci et al., 2026; Kolagar, 2024). Platform-based views understand integration as the coordination of interactions, services, and innovation across organisational boundaries (Li et al., 2024; Cuel et al., 2024). Taken together, these perspectives show that integration is not limited to technical connectivity, but also involves organisational, data-driven, and ecosystem-level coordination. This broader view is particularly relevant to IIoT, where physical assets, data flows, operational processes, and digital services must work together to create business value.

Accordingly, IIoT integration should be viewed as a coordinated organisational effort rather than an isolated technical implementation. IIoT enhances industrial operations through the automation and control of machinery and infrastructure (Moustafa et al., 2018). Interconnected devices, assets, and machines enable communication and coordination across industrial settings (Stoiber and Schönig, 2022). The real-time data they generate further supports visibility, monitoring, and data-driven decision-making (Khasawneh et al., 2025). When effectively integrated, these technologies can reshape and optimise industrial processes. In retail, for example, IIoT improves order tracking, inventory management, and customer data collection (Alaa et al., 2021). IIoT can also be embedded in service offerings to enhance service delivery. In healthcare, remote monitoring technologies enable timely and high-quality telemedicine services (Bhuiyan et al., 2021). More broadly, IIoT enables service and business model innovation through data-driven, service-centred, and platform-based value creation (Endres et al., 2024).

Within this broader view, IIoT integration maturity concerns the degree to which IIoT technologies are embedded within business operations in a coordinated and value-oriented manner. Within this study, it specifically refers to the level of coordination achieved across physical artefacts, data flows, industrial processes, and services to support process improvement, service enhancement, and value creation (Stoiber and Schönig, 2024; Endres et al., 2024; Gökalp and Martinez, 2021; Hortovanyi et al., 2023).

MMs have been widely used in IS research and practice to assess organisational capabilities and guide systematic improvement. Conceptually, MMs describe capability development through maturity levels that represent increasing sophistication in processes, technologies, and governance structures (Mettler, 2011). They therefore enable organisations to assess their current state and identify areas for improvement.

MM thinking has its foundations in the Capability Maturity Model (CMM), originally developed to assess and systematically improve software development processes within organisations (Paulk et al., 1993). This foundational approach has since been adapted and extended across diverse organisational and technological domains; including business process management (Flechsig et al., 2022), enterprise architecture (Proença and Borbinha, 2017; Rod and Vomlel, 2023), digital transformation (Suprun et al., 2024; Han et al., 2025), and information governance (Proença et al., 2016). Collectively, these domains highlight the role of IT-enabled organisational integration. Prior research on enterprise systems and cross-functional coordination similarly emphasises the alignment of processes, information, and organisational structures (Davenport, 1998; Rai et al., 2006). As organisations increasingly rely on complex digital infrastructures, MMs have become important instruments for structuring organisational development and managing technology-enabled transformation.

MMs typically comprise maturity levels, which represent progressive states of maturity, and capability dimensions, which define the areas across which that maturity is assessed (Mettler, 2011; De Bruin et al., 2005).

Within the MM literature, three primary purposes are commonly distinguished. Descriptive MMs assess current capabilities by positioning an organisation within defined maturity levels. Prescriptive MMs recommend pathways for progressing between levels, while benchmarking MMs compare organisational maturity with industry standards or peers. These purposes position MMs as diagnostic and developmental tools for organisational learning, strategic planning, and capability improvement (De Bruin et al., 2005; Klötzer and Pflaum, 2017).

MMs are commonly implemented using either “staged” or “continuous” representations. Staged models describe progression through predefined maturity levels and typically assign an overall maturity level to the organisation or domain assessed. Continuous models assess capability areas separately, recognising that they may develop at different rates. Staged models provide a clear improvement path and facilitate benchmarking, whereas continuous models accommodate differences in organisational contexts, priorities, and capability trajectories (De Bruin et al., 2005; Mettler, 2010).

The MM developed in this study is descriptive, designed to support current-state assessment and interpretation of IIoT integration maturity rather than prescribe pathways for advancement. It adopts a continuous assessment approach; whereby individual dimensions and capabilities can be assessed at different maturity levels rather than aggregated into a single overall maturity level. This recognises that IIoT capabilities may develop unevenly across dimensions and follow different trajectories depending on organisational context.

Within DT contexts, MMs are particularly valuable because organisations must simultaneously manage technological innovation, organisational change, and strategic adaptation (Gökalp and Martinez, 2022). Recent studies show that digital maturity depends not only on technology adoption but also on the integration of data, processes, information management capabilities, and organisational practices across functions (Suprun et al., 2024; De Carolis et al., 2025). MMs can therefore support the assessment and development of organisational capabilities for complex digital technologies.

Despite their usefulness, MMs have also been subject to several critiques. Prior research suggests that many MMs oversimplify complex organisational transformations. They often assume linear development trajectories that do not capture the dynamic and iterative nature of organisational change (Mettler, 2011). They are also criticised for weak theoretical foundations and reliance on practitioner insights or expert opinion (Mettler, 2011; De Bruin et al., 2005). Without such foundations, MMs risk becoming ad hoc assessment frameworks.

These challenges are particularly salient in IIoT contexts, where organisations must coordinate heterogeneous technological infrastructures with operational and data capabilities (Pino et al., 2024a; Jia et al., 2025). IIoT integration may also extend beyond organisational boundaries. Recent platform ecosystem literature shows that DT is often shaped by platform-mediated relationships among multiple actors (Li et al., 2024; Cuel et al., 2024). This perspective is important for IIoT, where data, services, and innovation capabilities may be coordinated across organisational and ecosystem boundaries. Accordingly, IIoT integration maturity involves both internal technology readiness and the coordination of IIoT-enabled assets, data flows, processes, and services across these boundaries. However, existing IIoT MMs provide only partial insight into how this integration is embedded within business operations.

With the growing importance of IIoT technologies in DT initiatives, several MMs have been proposed to assess IIoT readiness and guide capability development. These models typically conceptualise IIoT maturity as a staged progression reflecting increasing levels of technological capability, data integration, organisational adoption, and DT readiness.

Recent reviews show that IIoT-related MMs vary considerably in their maturity levels, dimensions, capability structures, and intended application domains (Benotmane et al., 2023; Pino et al., 2024b). This diversity demonstrates the growing relevance of maturity thinking in IIoT research but also highlights inconsistent conceptualisation and operationalisation across existing models. In many cases, existing models provide useful inventories of organisational and technological capabilities but offer less explanation of how these capabilities interact to support IIoT-enabled operational integration.

Three limitations are particularly relevant to the present study. First, many models adopt a broad, all-encompassing perspective, incorporating organisational (e.g., strategy, governance, leadership, culture, ethics), technological (e.g., infrastructure, connectivity, standardisation), and human dimensions (e.g., people, skills, change management). While these dimensions are important for understanding IIoT adoption, they often overlap with generic IT and DT MMs. This shifts attention toward general readiness rather than IIoT-specific integration, making it difficult to examine IIoT integration as a distinct operational phenomenon. For instance, some models consider data management, data analytics, and business processes as relevant to IIoT integration (Benotmane et al., 2023; Stoiber and Schönig, 2022; Klisenko and Serral Asensio, 2022). However, these elements are typically subsumed within broader organisational or technological dimensions rather than examined as core constructs of IIoT integration. Consequently, existing models provide limited insight into how IIoT technologies are embedded in operational processes and how this supports value-generating activities such as process optimisation, improved service delivery, and new business model development (Stoiber and Schönig, 2022; Tortorella et al., 2023).

Second, existing MMs remain conceptually fragmented. Different frameworks define maturity dimensions, capability structures, and maturity levels in different ways, making comparison difficult and limiting a coherent understanding of IIoT maturity. For instance, even when similar dimensions are used, the underlying capabilities often differ across models (Serral et al., 2020; Klisenko and Serral Asensio, 2022). This fragmentation reflects the different organising structures used to define, group, and assess IIoT maturity components (Parab and Deshmukh, 2023). Consequently, there remains a need for a more coherent structure to organise the core elements of IIoT integration and clarify their relevance to business operations.

Third, many existing MMs are developed for particular organisational contexts, sectors, or application domains, including B2C environments, retail, and food supply chains (Serral et al., 2020; Klisenko and Serral Asensio, 2022; Pino et al., 2025; Visamitanan et al., 2025). While such models are valuable within their intended contexts, their transferability may be limited for organisations seeking to understand IIoT integration across multiple operational domains. This highlights the need for a generic MM that can examine IIoT integration across different organisational and sectoral contexts.

Taken together, these limitations suggest that existing IIoT-related MMs provide valuable, yet narrow or incomplete perspectives on organisational IIoT integration. This points to the need for a more coherent and transferable conceptual structure for examining IIoT integration. In response, this study develops a MM informed by organisational semiotics for analysing IIoT integration in business operations. The proposed model differs from existing models through its semiotics-informed theoretical framing, specific focus on integration, and generic structure for examining IIoT integration across business contexts.

Given the limited explicit theoretical framing in existing IIoT MMs, theoretical lenses used in MMs from related domains, such as Industry 4.0 and DT, were examined. For example, Senna et al. (2023) and Ojubanire et al. (2025) applied the Technology-Organisation-Environment (TOE) framework to structure the dimensions of their models. Kayikci et al. (2022) drew on systems theory to simplify complex organisational structures. Gökalp and Martinez (2022) adopted broader IS perspectives, such as the resource-based view and dynamic capabilities theory, to explain infrastructural evolution, organisational integration, and capability development.

These lenses are mainly used to structure model dimensions and explain their rationale. However, they tend to address broader organisational and contextual aspects, rather than the components specific to IIoT integration. IIoT integration requires the alignment of physical devices, data flows, business processes, and service offerings. Therefore, a theoretical lens was sought that could bring together these heterogeneous components and clarify their connections.

The Semiotic Ladder (SL), an artefact of Organisational Semiotics (Stamper, 1993), fulfils this role. It conceptualises the socio-technical components, across multiple layers. Accordingly, the SL was adopted as the theoretical lens for the proposed MM. It was used to structure and interpret the literature-derived IIoT components as maturity dimensions. Section 3.2 details how it informed the research process.

Semiotics, or the theory of signs, is based on the principle that information is transmitted through signs (Stamper, 1993). A sign is an entity or symbol that stands for something other than itself (Mingers and Willcocks, 2017). In organisations, information is processed and communicated through the generation, transmission, and use of signs (Liu, 2000). By connecting human intentions, meanings, communication, data, and organisational activities, signs support a broader understanding of information and related systems (Stamper, 1993; Beynon-Davies, 2009). Semiotic analysis has been applied across areas such as knowledge management, software design, construction informatics, social media data analysis, and product architecture development (Desouza and Awazu, 2004; Schiffel, 2009; Luo and Liu, 2009; Collinge et al., 2009; Mikhaeil and Baskerville, 2019; Hu et al., 2013).

Organisational semiotics applies semiotic concepts and methods to organisational phenomena (Collinge et al., 2009). A key framework within organisational semiotics is the Semiotic Ladder (SL), which provides a structured way to examine phenomena across different levels of abstraction, from technical artefacts to organisational meaning and action (Nicastro et al., 2015; Arantes, 2013). In this study, the SL is used as a classificatory and interpretive lens to organise IIoT-related capabilities across these levels. It is not used as a process or causal theory of organisational transformation.

The SL comprises six hierarchical layers: physical world, empirics, syntactics, semantics, pragmatics, and social world. The lower three layer; physical world, empirics, and syntactics, focus on the physical properties, transmission, and formal organisation of signs, while the upper three; semantics, pragmatics, and social world, address their meaning, purposeful use, and broader organisational implications (Stamper, 1993; Nicastro et al., 2015; Luo and Liu, 2009). Table 1 summarises the focus of each layer and illustrates their application in prior studies (Arantes, 2013; Paim and Prietch, 2020; Luo and Liu, 2009; Mingers and Willcocks, 2017; Nicastro et al., 2015; Liu, 2000).

The layered logic of the SL is relevant to IIoT integration because IIoT extends across physical objects, signals, data, interpretation, and organisational use. These elements reflect different levels of abstraction. The SL therefore provides a conceptual structure for classifying and interpreting these heterogeneous technical and operational elements. On this basis, the SL is adopted as the theoretical lens for structuring and interpreting the dimensions of IIoT integration maturity developed in this study.

This study adopted a Design Science Research (DSR) approach (Hevner et al., 2004; Peffers et al., 2007), to develop and iteratively refine the MM artefact. The following sections describe the overall model development approach, design cycles, and interview data analysis with additional details provided in Appendix A.

To develop the IIoT MM, prior MM development approaches were reviewed, including Becker et al. (2009), De Bruin et al. (2005) and Mettler (2010). From these approaches, common development activities were identified and synthesised into four steps: problem identification, scope definition, model development, and model evaluation. Although presented as distinct activities, model development and evaluation were undertaken iteratively through two design cycles. This enabled progressive refinement of the MM artefact. The synthesis of the existing frameworks and the resulting process adapted in this study, are summarised in Appendix A- Part A (see Table A.1). Figure 1 summarises the two Design Cycles.

During problem identification, the IIoT integration and MM literature was analysed to identify limitations in existing MM building approaches. These included insufficient attention to operational integration and limited theoretical framing. These gaps motivated the development of the proposed MM, as outlined in Sections 1 and 2.

During scope definition, the model's objective, purpose, target users, and application boundaries were established. The objective was to develop a theory-informed IIoT MM for assessing the integration of IIoT technologies into business operations. Following the decision criteria proposed by De Bruin et al. (2005) for defining the model focus, the targeted IIoT Integration MM model was positioned as a generic and descriptive MM. It is intended to support organisations in reflecting on and interpreting their current IIoT integration capabilities. The primary users include practitioners in enterprise architecture, IIoT solution design, digital transformation, and operational improvement. Academic researchers were also identified as an important stakeholder group.

Design Cycle 1 focused on developing the initial IIoT MM artefact and conducting a preliminary evaluation. The model was developed through literature analysis and theory-based structuring, then evaluated via interviews with IIoT practitioners.

The MM development commenced with a structured analysis of literature to identify concepts, capabilities, and progression mechanisms relevant to IIoT integration. Academic literature was retrieved primarily from the Scopus database, which was selected for its broad citation coverage and manageable search results. The search strategy was iteratively refined by combining IIoT and broader IoT keywords with MM-related terms. Synonyms and related terms were identified through preliminary scanning of titles, abstracts, and keywords, as well as discussions within the author team.

An example search string was TITLE-ABS-KEY (“IoT” OR “Internet of Things” OR “Industrial Internet” AND (“maturity model” OR “capability model” OR “maturity framework”)).

The search was limited to English-language publications published between 2015 and 2025, reflecting the emergence of IIoT integration as a significant phenomenon over the past decade. The retrieved records were imported into EndNote 20 software, where duplicate records were removed, resulting in 140 unique publications for screening. Titles were first reviewed to exclude studies outside the research scope. Abstracts and, where necessary, introductions were then examined to assess alignment with the study focus. Publications focused primarily on consumer IoT or broader Industry 4.0 contexts without an IIoT maturity focus were excluded. Additional articles were identified through backward and forward searching and targeted Google Scholar searches using terms such as “IIoT integration” and “IIoT maturity model”. This process resulted in 21 academic papers for further analysis. The inclusion and exclusion criteria are summarised in Appendix A - Part B1.

Given the emerging nature of IIoT and the limited academic literature, grey literature was also reviewed to capture relevant practitioner insights. Sources included white papers, case studies, technical reports, and selected expert-authored articles on IIoT adoption and integration. These were identified through references in academic publications and targeted Google searches of credible industry, consulting, and international organisations (e.g., Deloitte, McKinsey, Harbour Research, CAD IT UK, and the World Economic Forum). Grey literature was used as complementary evidence to triangulate and enrich insights from the academic literature, rather than as equivalent to peer-reviewed evidence.

Grey literature may introduce bias due to its practitioner focus and lack of peer review. To mitigate this risk, sources were screened based on publisher credibility, author expertise, evidential transparency, relevance to IIoT integration and maturity, and consistency with academic or other credible sources. Sources with unclear authorship, limited evidence, or primarily promotional content were excluded.

The final selection comprised 43 sources: 21 academic publications and 22 grey literature sources. This source base was intended to support conceptual model development rather than provide an exhaustive review of IIoT literature.

Following the literature search, the 43 selected sources were analysed through a two-level qualitative coding process. Level 1 involved open coding, in which statements and phrases relating to IIoT integration, maturity, and progression logic were identified and coded. Level 2 involved axial coding, in which conceptually related open codes were grouped into subcategories and subsequently abstracted into higher-order categories. The first author conducted the primary coding, with two other authors participating in coder corroboration sessions at each level to review code definitions, category boundaries, and interpretations. Further details are provided in Appendix A - Part B2 (see Table B.2).

The resulting subcategories and higher-order categories informed the development of the model's capability factors and dimensions. Capability factors represent specific capability areas within broader dimensions, similar to sub-dimensions in existing MMs. To refine and label these components, the derived subcategories were compared with relevant sub-dimensions in existing IIoT MMs, while the higher-order categories were compared with existing MM dimensions. This process informed the refinement and labelling of the capability factors and the four dimensions of object, data, process, and service. Appendix A - Figure B.1 illustrates the refinement process, while Table B.3 provides an example of component labelling. Together, these appendices provide an audit trail from the source material to the initial model components.

Following this coding and refinement process, the Semiotic Ladder (SL) (Stamper, 1993), introduced in Section 2.4.2, was applied as a theoretical lens to structure and interpret the derived dimensions and clarify their relationships. Its layered logic provided a basis for examining IIoT integration across levels ranging from physical entities and data structures to meaning, interpretation, and organisational action. The dimensions were mapped and refined using the SL to support conceptual coherence across these layers. Importantly, the SL was used to structure and interpret, rather than generate, the dimensions. Section 3.2.1.3 further explains the alignment between the SL layers and model dimensions.

The maturity levels were derived separately - from the literature - rather than from the SL. Terminology describing stages of IIoT deployment and scaling was analysed, with particular attention to progression in scope and scale (e.g., pilot, shopfloor, organisation-wide, and ecosystem-level). Based on this analysis, maturity-level titles were developed using commonly recognised industrial terminology to support clarity, comparability, and practical relevance. Further details are provided in Appendix A - Part B3, Table B.4.

Finally, capability mappings were established by analysing how each capability factor manifests across the different maturity levels and aligning these capabilities with stages of progression to support internal consistency and logical coherence. The dimensions, capability factors, maturity levels, and capability mappings were then consolidated into the initial descriptive MM, presented in Appendix A - Table B.5.

This section explains how the layers of the Semiotic Ladder (SL) align with the dimensions of the IIoT MM-also visually summarised in Figure 2. In this study, the SL serves as a classificatory and interpretive framework for organising IIoT-related components across multiple levels of abstraction. It shows that IIoT integration extends beyond physical technology deployment. It provides a structured basis for examining IIoT-related phenomena across physical entities and measurable signals, structured data, operational meaning, and business-oriented action. The SL clarifies how these areas are related within organisational practice.

The physical world layer corresponds to the object dimension. This layer concerns tangible entities that can be connected, monitored, and coordinated through IIoT.

The empirics and syntactics layers inform the data dimension. The empirics layer concerns measurable outputs generated by IIoT objects, which provide the inputs for further analysis. The syntactics layer concerns the formal structures and rules used to represent, organise, and exchange these data.

The semantics layer informs the process dimension. At this layer, structured IIoT data become meaningful through interpretation in specific operational contexts. Organisational actors and systems interpret data using contextual knowledge, process rules, performance expectations, operational routines, and domain-specific thresholds. These actors may include operators, engineers, managers, and service stakeholders, depending on the operational context and level of use. This interpretive process helps determine whether observed patterns indicate normal performance, emerging risks, inefficiencies, or intervention opportunities. For example, a vibration reading becomes meaningful only when interpreted against equipment condition, process thresholds, or quality standards. Accordingly, the semantics layer informs how IIoT data are interpreted within operational processes.

The pragmatics layer informs the service dimension. It concerns how interpreted IIoT data and process-relevant insights are purposefully used to support service delivery, service improvement, service innovation, and business value creation. Illustrative examples associated with each IIoT component are provided in Figure 2.

After developing the initial MM (V1.1), a preliminary evaluation was conducted to obtain early expert feedback on the completeness and relevance of the model dimensions and maturity levels. Semi-structured interviews were conducted with six IIoT practitioners from different domains, recruited through LinkedIn. Ethics approval was obtained from the institutional Human Research Ethics Committee, with all data collected and handled in accordance with the National Statement on Ethical Conduct in Human Research (2023). Participant details are provided in Appendix A - Part B4, Table B.6. All interviews were anonymised, transcribed, and analysed, as described in Section 3.4.

The evaluation informed several refinements to the initial MM (V1.1). The four dimensions, object, data, process, and service, were retained. However, a new capability factor, Data security and governance, was added to the data dimension. In addition, a new initial maturity level, Pre-Pilot, was introduced to represent minimal or no IIoT integration. These new components were selected for further evaluation in the next design cycle. The remaining model components were retained, as no significant concerns were raised. Appendix A - Part B4, Table B.7 summarises the refinements, while Part B5 presents the revised model (V1.2).

Design Cycle 2 focused on further evaluating and refining the MM (V1.2), including both newly introduced and retained components. The cycle was conducted in two stages. Stage 1 focused on evaluating the model dimensions and maturity levels, while Stage 2 focused on evaluating capability factors and capability mappings within individual dimensions. This supported a more detailed evaluation of the model components.

• Stage 1 involved semi-structured interviews with 10 IIoT experts. It evaluated the completeness and relevance of the model dimensions and maturity levels, including the components introduced in Design Cycle 1 (Kırmızı and Kocaoglu, 2022; Sonnenberg and vom Brocke, 2012; Gökalp and Martinez, 2022). Participants were senior IIoT practitioners with managerial and technical experience across several sectors, including manufacturing, oil and gas, transportation, and mining. They were selected through purposive sampling based on direct experience in IIoT implementation, DT, or smart industries. Experience with developing or applying MMs was considered an additional qualification. Participants were recruited mainly through LinkedIn and professional networks, with snowball sampling used when participants recommended other relevant experts. Participant details are provided in Appendix A - Part C1, Table C.1.

• Stage 2 involved semi-structured interviews with 20 IIoT experts. It evaluated individual dimensions, including their capability factors and capability mappings. Participants were recruited based on expertise in areas related to the dimensions - such as data, industrial processes, IIoT infrastructure, devices and systems, and industrial services. Where necessary, their expertise was verified through direct clarification and publicly available professional information. Some practitioners had cross-functional experience, which enabled them to provide feedback on multiple dimensions. A recruitment process similar to that used in Stage 1 was followed. The interviews assessed the relevance, clarity, and completeness of the capability factors and maturity mappings within the dimensions reviewed by each participant. Participant details are provided in Appendix A - Part C1, Table C.2. Across both stages, the aim was to obtain relevant expert insight through purposive selection rather than statistical representativeness.

The analysis was conducted progressively after each interview, as detailed in Section 3.4. Insights from both stages informed further refinement, resulting in the final IIoT MM presented in Section 4 (Table 2).

All interviews were recorded, transcribed, anonymised, and analysed using NVivo 14. As the objective was to evaluate and refine the MM, the predefined model components provided the analytical structure for coding and analysing the interview data.

In the preliminary evaluation of Design Cycle 1 and Stage 1 of Design Cycle 2, the dimensions and maturity levels were used as the initial codes. This enabled participant feedback to be mapped against the structure of the model and assessed in terms of completeness, relevance, and clarity. Additional codes captured comments on capability factors, capability mappings, and other emerging insights. In Stage 2 of Design Cycle 2, the coding structure was expanded to include individual capability factors and their associated capability mappings, reflecting the more detailed evaluation undertaken during this stage. Subsequently, during the Stage 2 evaluation in Design Cycle 2, the coding structure was expanded to include each capability factor and its associated capability mappings.

The first author conducted the primary coding. After each stage, the other authors reviewed the coded outputs, emerging interpretations, ambiguous comments, and proposed refinements. The team then compared feedback across participants to identify agreement, concerns, and potential changes. Differences in interpretation were discussed until consensus was reached. Together, these steps supported coding consistency, interpretive alignment, and traceability between expert feedback and model revisions.

To support qualitative rigour, the analysis used systematic coding, author-team corroboration, consensus-based interpretation, reflexive discussion, and transparent documentation (Barbour, 2001; O'Connor and Joffe, 2020; Miles et al., 2014). Formal inter-coder reliability metrics, such as Cohen's Kappa, were not calculated because the study did not involve independent coders applying the same coding framework to the same interview transcripts. Rather than assessing coding agreement among independent coders, the analysis emphasised collaborative interpretation and refinement of expert feedback through iterative author-team review and consensus-building. This approach is appropriate when the objective is interpretive artefact evaluation rather than measurement of agreement among independent coders (Barbour, 2001; O'Connor and Joffe, 2020).

Following Design Cycle 2, the two components introduced in Design Cycle 1 - the pre-pilot maturity level and the data security and governance capability factor-were retained because participants confirmed their relevance. No components were added or removed. Instead, selected terminology was revised for clarity and relevance. Accordingly, Appendix A - Part C2 records all refinements, while Tables D.1-D.3 in Part D link expert evidence to retained dimensions, capability factors, and maturity levels. Collectively, these tables provide a transparent audit trail linking expert feedback to model refinement decisions. Section 4 presents the final IIoT MM resulting from Design Cycle 2.

This section presents the final conceptual maturity model (MM) for Industrial Internet of Things (IIoT) integration. The model reflects the outcomes of the second design cycle and is intended as a descriptive reference framework. It comprises four main components: dimensions, capability factors, maturity levels, and capability mappings. The following sections describe the dimensions, capability factors, and maturity levels, respectively. The capability mappings, which show how each capability factor is demonstrated at different maturity levels, are presented in Table 2 which presents the complete current MM.

The proposed IIoT MM comprises four dimensions: physical object, data, process, and business service. Informed by the Semiotic Ladder, they capture the technical and operational components of IIoT integration. Table 3 summarises the dimensions and their focus.

The IIoT Integration MM comprises eleven capability factors across the four dimensions. These represent the key organisational capabilities that develop as IIoT integration matures. Table 4 summarises the capability factors and their definitions.

The proposed model comprises five maturity levels: Pre-pilot, Pilot, Operational, Organisation-scale, and Ecosystem-scale. These describe progression from no or minimal IIoT implementation to operational, organisation-wide, and ecosystem-level integration. Rather than representing only a gradual accumulation of capabilities, the maturity levels indicate broader shifts in the scope and coordination of IIoT integration. Table 5 summarises the dominant organisational focus, nature of integration, and key capability transition associated with each level.

The discussion interprets the proposed MM, states its limitations and contextual trade-offs, proposes a future research agenda, and highlights its theoretical and practical contributions.

IIoT integration maturity is an organisation's capability to progressively integrate physical entities, data, processes, and services to support meaningful action. It involves more than accumulating technologies. The Semiotic Ladder (SL) provides an interpretive structure for examining how these components are connected through measurement, structuring, interpretation, intended use, and action. In particular, the transition from data to process depends on actors and systems interpreting structured data through domain knowledge, rules, thresholds, and routines. This interpretation transforms technical signals into process-relevant insights. Decision-making is not confined to the data dimension but emerges through interactions among data, process, and business service capabilities. Together, these capabilities support monitoring, optimisation, service delivery, and value creation.

Within this interpretation of maturity, the model directly represents the SL's physical-world, empirics, syntactics, semantics, and pragmatics layers. The social world, which encompasses culture, governance, institutional norms, and stakeholder expectations, is treated as a contextual influence rather than a separate dimension. This positioning reflects how social-world factors shape the prioritisation and enactment of capabilities.

Viewed through the SL, uneven maturity across dimensions can create a semiotic bottleneck. For example, an organisation may have advanced capabilities related to IIoT assets that generate large amounts of data, while its data and process capabilities remain underdeveloped. Weak connections between data generation, interpretation, and action may then limit the translation of available data into operational understanding and responsive action.

At the same time, the staged presentation should not imply a strictly linear progression. The MM adopts a continuous maturity perspective. Organisations need not fully satisfy every factor at one level before developing more advanced capabilities. However, higher maturity requires substantial development of the foundational capabilities associated with earlier levels. Accordingly, trajectories may vary according to organisational strategy, legacy infrastructure, operational context, and constraints.

In practice, progression may require coordinated investments in technology, data management, process redesign, governance, skills, and change management. However, identifying the specific mechanisms that enable advancement requires further empirical study.

Practitioners can use the model as an initial reflection tool to examine how IIoT capabilities are developed across the dimensions. For example, the model can help them identify whether physical object capabilities provide a sufficient foundation for related data, process, and business service capabilities. This can support focused discussion about next steps, such as improving data integration, standardising data flows, or linking IIoT outputs to process-level decision-making. Nevertheless, the model's practical application remains preliminary and requires further organisational validation.

This study has several limitations. First, the proposed MM remains primarily a conceptual and descriptive artefact. Although a self-assessment grid has been developed to support practical application and reflection, it is not yet a fully validated instrument. Further work is required to refine and validate its assessment indicators, scoring approaches, maturity thresholds, and application procedures across different organisational contexts. Accordingly, the model should currently be regarded as an empirically informed conceptual artefact, rather than a fully operational instrument.

Second, grey literature was included to capture timely practitioner insights into IIoT adoption and integration. However, its use may introduce bias due to variations in evidence quality, organisational interests, and the absence of peer review. This risk was mitigated through source screening, triangulation with academic literature, and expert evaluation, although practitioner-oriented bias cannot be fully excluded.

Third, the empirical evaluation was limited to semi-structured interviews with IIoT experts. Although the interviews supported the evaluation of the completeness, relevance, and clarity of the model, the study did not include longitudinal investigation, large-scale quantitative evaluation, or application in organisational settings. Therefore, the practical utility and assessment outcomes of the model require further organisational deployment and empirical testing.

Fourth, the study focuses on developing and evaluating a descriptive MM rather than identifying the specific mechanisms through which organisations progress between maturity levels. Although the model describes capability progression, it does not empirically explain the managerial interventions, governance arrangements, architectural decisions, resource allocations, or organisational changes that may enable maturity advancement.

Finally, although designed as a generic framework, IIoT integration maturity may vary across organisational, sectoral, regulatory, technological, and ecosystem contexts. Future studies should therefore examine the model's boundaries and where contextual adaptation is required.

Maturity progression is rarely linear or frictionless. Organisations may advance unevenly, experience periods of stagnation, or encounter constraints despite substantial technology investments. Governance requirements, organisational capabilities, contextual conditions, and trade-offs therefore shape feasible progression pathways and appropriate maturity targets.

One set of challenges concerns governance and risk management. As IIoT infrastructures become more interconnected and information flows extend across organisational boundaries, concerns about cybersecurity, data quality, ownership, privacy, and accountability become more pronounced. Consequently, higher maturity requires clearer responsibilities and more formal mechanisms for governance, monitoring, and control.

As one participant explained: “The ecosystem level is even more challenging because not only the data dimension, but also the process and service dimensions become significantly more complex. This is because organisations must manage data and process flows that extend beyond their own boundaries and into those of trusted suppliers and partners, who also need to be brought on board.” - Participant 7.

Skills and capability gaps represent another significant challenge. Organisations may possess the required technological infrastructure but lack the analytical, operational, managerial, or governance expertise needed to interpret data, redesign processes, coordinate across organisational units, and manage external partners. Moreover, interoperability constraints, conflicting priorities, unclear responsibilities, and resistance to change help explain why many IIoT initiatives remain at the pilot stage (Abbatemarco et al., 2022).

The pace and pathway of progression also depend on contextual factors. These include legacy technologies, resource availability, digital capability, organisational size and geographical dispersion, operational complexity, regulatory requirements, strategic priorities, financial capacity, and risk tolerance. At advanced maturity levels, an organisation's position within its ecosystem becomes increasingly influential. Suppliers, customers, regulators, and technology partners may either enable or constrain further integration. Therefore, maturity progression is context-dependent rather than universally prescribed.

In addition to these challenges, organisations must manage tensions as maturity increases. Enterprise-wide standardisation and information sharing can improve visibility and coordination, but they may reduce local autonomy. Similarly, advanced IIoT capabilities can enable innovation and responsiveness while requiring stronger controls for cybersecurity, compliance, data sharing, and operational reliability.

These tensions create practical trade-offs. Expanded connectivity improves monitoring but increases data-management and governance burdens. Process automation improves efficiency and consistency but creates greater dependence on interoperable systems, high-quality data, and standardised processes. Likewise, ecosystem-level service innovation enables collaboration while intensifying concerns about data ownership, privacy, accountability, and control. Organisations must therefore balance agility with governance and coordination with flexibility.

Advanced IIoT integration may also support organisational adaptability and resilience. Greater visibility into assets, processes, and operating conditions can enable earlier disruption detection, faster access to information, and more coordinated responses under uncertainty. Drawing on the notion of information-driven resilience (Wang and Zhang, 2026), advanced IIoT integration may help organisations convert real-time information into adaptive responses. However, resilience also depends on managerial, organisational, and environmental conditions. It is therefore treated as a potential outcome supported by advanced IIoT maturity rather than a maturity dimension.

Overall, maturity progression involves more than accumulating technological capabilities. Organisations must balance competing objectives, address evolving governance and capability requirements, and adapt to contextual constraints. Consequently, higher maturity is neither universally attainable nor necessarily desirable. The appropriate level depends on strategic objectives, operating conditions, resources, risks, and ecosystem relationships. Recognising these challenges, contextual influences, tensions, and trade-offs is therefore essential for understanding the contingent nature of IIoT maturity progression.

Building on the study's limitations, this section presents five themes for advancing IIoT integration maturity research. Although the proposed MM is informed by semiotics, the agenda extends beyond this theoretical lens to support broader empirical, methodological, theoretical, and practical inquiry.

The themes are not equally urgent. The immediate priorities are to validate the model across organisational and industrial settings and to develop reliable indicators, scoring mechanisms, and assessment procedures. Establishing these foundations will enable subsequent research on maturity-progression mechanisms, contextual adaptation, ecosystem-level integration, and decision-support tools.

Empirical validation is essential for testing the relevance, robustness, practical applicability, and explanatory value of the IIoT MM. Future studies could apply the model within a single organisation, across multiple organisations, or across sectors. Comparative case studies could examine whether the dimensions and maturity levels reflect observed integration patterns and identify sector-specific pathways. Longitudinal research could investigate how the dimensions develop over time, whether particular dimensions act as leading indicators, and how capability imbalances affect progression. In addition, future research could test the boundaries of the model as a design science artefact. This should identify where the generic model remains applicable, where adaptation or extension is required, and where its explanatory and practical value may be limited by organisational, sectoral, regulatory, technological, or ecosystem conditions.

Illustrative research questions include:

1. How accurately does the IIoT MM explain observed patterns of integration across the physical object, data, process, and business service dimensions in different organisational contexts?

2. How do interactions among the physical object, data, process, and business service dimensions empirically shape progression across maturity levels over time?

3. Under what organisational, sectoral, regulatory, technological, or ecosystem conditions does the proposed IIoT MM require adaptation, extension, or boundary refinement?

As a near-term priority, future research should further develop and validate assessment and measurement instruments for the proposed IIoT MM. Building on the preliminary assessment grid presented in Appendix B, studies can refine observable capability indicators, maturity thresholds, scoring mechanisms, and systematic assessment procedures. This work may produce enhanced maturity grids, survey instruments, diagnostic tools, and digital assessment platforms that support consistent and actionable organisational assessments. Researchers could also develop quantitative metrics and qualitative evaluation criteria for measuring capabilities across the four dimensions. They should be tested for reliability, validity, and diagnostic usefulness. They should also account for interdependencies and uneven development among the dimensions to provide a more holistic representation of IIoT integration maturity.

Illustrative research questions include:

1. What indicators and metrics can be used to assess capabilities across the physical object, data, process, and business service dimensions at different maturity levels?

2. How can interdependencies among the physical object, data, process, and business service dimensions be incorporated into maturity assessment approaches?

3. To what extent do assessment instruments derived from the proposed IIoT MM demonstrate reliability, validity, and diagnostic usefulness?

Future research should extend the descriptive MM towards a prescriptive MM by identifying actionable pathways for maturity progression. While this study conceptualises the dimensions, capability factors, and maturity levels of IIoT integration, it does not specify the mechanisms through which organisations transition between maturity levels. Such guidance would strengthen the model's theoretical and practical utility.

Studies could investigate the organisational, managerial, technical, and governance mechanisms that enable transitions. These may include managerial interventions, resource allocation, architectural changes, governance arrangements, process standardisation, and cross-functional coordination. For example, organisation-scale integration may require scalable digital infrastructure, enterprise-wide data governance, cross-functional coordination, and alignment between IIoT initiatives and business strategy.

Research could also examine alternative maturity pathways, as organisations may not progress through the levels in identical ways. Different contexts may require different sequences of capability development. Moreover, progress in one dimension may enable, accelerate, or constrain progress in others. Studies could therefore investigate cross-dimensional dependencies, sequencing effects, trade-offs, and path dependence across maturity trajectories. They could also identify conditions that enable or inhibit advancement.

Illustrative research questions include:

1. What organisational, managerial, technical, and governance mechanisms enable successful transitions between maturity levels?

2. What dependencies, trade-offs, and sequencing effects influence progression across maturity levels?

3. How can maturity roadmaps be developed to guide organisations along different IIoT integration pathways while accounting for contextual variation?

Future research should examine how the proposed MM can be adapted and deployed across different organisational, technological, regulatory, and ecosystem contexts. Although the model is intended as a generic framework, organisations operate under diverse conditions that may influence how maturity is interpreted, assessed, and achieved in practice. Comparative studies could examine its application in SMEs and large enterprises, greenfield and legacy environments, and highly regulated and less regulated industries.

Research could also explore how organisations tailor the model to their requirements. This may include adapting maturity criteria, assessment processes, governance arrangements, and improvement priorities. Studies could also examine how the relative importance and progression of the dimensions vary across organisational and industrial contexts. Additionally, research may examine how contextual factors such as organisational size, digital capability, resource availability, ecosystem position, and regulatory requirements influence maturity development and progression pathways.

Finally, research could examine how the MM can be integrated with enterprise-architecture and IIoT reference frameworks, including RAMI 4.0 (Reference Architectural Model Industrie 4.0) and IIRA (Industrial Internet Reference Architecture). Its alignment with process frameworks such as SCOR (Supply Chain Operations Reference) and digital-platform ecosystem approaches should also be explored. Such research could clarify how maturity assessment complements architectural and process design, supporting the model's practical deployment.

Illustrative research questions include:

1. How can the IIoT MM be adapted for different organisational, technological, regulatory, and ecosystem contexts?

2. How do the relative importance and progression of the physical object, data, process, and business service dimensions vary across different organisational and industrial contexts?

3. How can the IIoT MM be integrated with enterprise architecture frameworks and IIoT reference architectures?

Studies could develop tooling and decision-support mechanisms to facilitate IIoT-enabled transformation based on the proposed MM. These could help organisations assess maturity, identify improvement opportunities, and plan progression pathways. Examples include roadmap generation tools, transformation planning tools, and systems that prioritise capability development across the model dimensions.

Research could also examine how maturity assessments can inform IIoT investment decisions and resource allocation. Decision-support tools could help align technological investments, organisational capabilities, and transformation priorities with current and target maturity levels. To provide realistic guidance, these tools should account for contextual conditions, interdependencies among dimensions, and trade-offs between alternative progression pathways.

Further work could explore the development of automated maturity assessment tools, including digital diagnostic tools, AI-supported assessment platforms, and interactive dashboards. These tools could track maturity over time, identify emerging capability gaps, and evaluate transformation initiatives. Such artefacts may improve the model's practical usability and support evidence-based IIoT decision-making. Their development and evaluation are well suited to DSR.

Illustrative research questions include:

1. How can decision-support systems be designed to guide IIoT maturity improvement and transformation planning?

2. How can decision-support tools support the prioritisation of capability development across the physical object, data, process, and business service dimensions?

3. How can maturity assessments inform IIoT investment decisions, resource allocation, and transformation priorities?

The study's contributions are twofold: theoretical contributions to conceptualising IIoT integration maturity and practical contributions to supporting organisational reflection and improvement.

This study makes three theoretical contributions. First, it extends the application of organisational semiotics, particularly the Semiotic Ladder (SL), to IIoT maturity research. The SL provides a structuring, classificatory, and interpretive lens for organising IIoT integration across physical objects, data, processes, and business services. It also provides a conceptual pathway for understanding how signals generated by physical entities become structured data, process-relevant meaning, and purposeful service outcomes. Consequently, the model conceptualises IIoT maturity as more than technological deployment. It highlights cross-layer relationships and provides a basis for interpreting how weak alignment may result in semiotic bottlenecks. However, the SL is not presented as a causal theory of organisational transformation or maturity progression.

Second, the study provides an integrated conceptualisation of IIoT integration that explains maturity across four interconnected capability domains: physical objects, data, processes, and business services. In contrast to models centred on broad digital readiness, this structure focuses specifically on how IIoT capabilities become embedded in business operations, thereby addressing conceptual fragmentation across existing IIoT MMs.

Third, the study contributes an MM artefact that describes capability development across five levels. This progression is descriptive rather than deterministic because organisations may develop unevenly across dimensions and follow different trajectories depending on their context, priorities, and capability interdependencies. The model therefore provides a theoretical foundation for examining cross-dimensional progression, contextual variation, and maturity imbalances. It also supports future empirical validation, refinement, and theory development in IIoT integration maturity.

The model helps practitioners, including managers, consultants, and digital transformation leads, understand how connected objects, data capabilities, operational processes, and service outcomes relate to one another. It can also support early-stage organisational sensemaking. For example, practitioners can use the model to discuss whether IIoT initiatives remain focused on isolated devices and pilots, or are integrated with data management, process improvement, and service delivery.

Appendix B provides a preliminary self-assessment grid that translates the model into observable indicators. The grid guides potential users through the assessment questions and identifies the evidence to consider when determining their current maturity level. It can help reveal uneven maturity and support improvement discussions but requires further empirical validation before use as a robust assessment or benchmarking instrument.

This study develops a descriptive, conceptual MM for IIoT integration into business operations. It addresses limitations in existing models, which provide incomplete accounts of how IIoT becomes integrated within organisations. Using the SL as a structuring and interpretive lens, the model organises IIoT integration across physical object, data, process, and business service dimensions and describes their development across five maturity levels. The model frames maturity as coordinated, potentially uneven capability development rather than technology accumulation or a deterministic progression. Formative expert evaluation refined the artefact, but organisational application and further empirical validation are required. Nevertheless, the MM and preliminary assessment grid support future research and practitioner reflection on IIoT integration maturity.

The authors would like to thank all interview participants who generously contributed their time and insights to the empirical evaluation of the maturity model.

The supplementary material for this article can be found online.