The transformative potential of control systems for sustainable innovations in industrial contexts

Innovation is key for industrial firms as they move beyond their linear, production-based pasts, towards increasingly sustainable and/or circular business models (Bocken and Ritala, 2021). Here, organisational “success” can be explained in terms of how the innovation of industrial products, processes and services is managed or controlled for improved sustainability performance (Adams et al., 2006; Geissdoerfer et al., 2017; Nidumolu et al., 2009). Put simply, sustainability performance regards the outcomes of business activities that contribute to an environmentally friendly and socially responsible workplace or wider society, in addition to the traditional financial security of the organisation (see Johnstone and Beusch, 2025). For this, innovations require accurate and timely information-exchange both within (i.e. intra-organisational) and between (i.e. inter-organisational) industrial organisations and other up/downstream actors such as customers and suppliers (Aureli et al., 2023; Johnstone, 2024b). Notwithstanding the importance of management control for industrial innovations, and despite the internal-managerial perspective being the most evaluated in the literature (Cillo et al., 2019), scholarly attention to how management controls shape sustainable innovations in such settings remains nascent (e.g. Beusch et al., 2022; Johnstone, 2024b). This study addresses that gap with new empirical evidence and analysis.

Sustainable innovations are defined as those related to products, processes or services that generate social and environmental value, as well as economic returns (see Bocken and Geradts, 2020; Bos‐Brouwers, 2010). This definition suggests that the innovations are the means for improved sustainability performance to be achieved. Meanwhile, sustainability controls are the organisational tools used to guide, monitor, motivate and coordinate behaviours that support the sustainability performance within or between organisations (see Guenther et al., 2016; Johnstone, 2019; Wijethilake et al., 2017). Unlike traditional management controls, sustainability controls explicitly integrate environmental and social objectives with financial ones. Yet, even though controlling sustainability issues in supply chains or value networks is increasingly important for industrial firms in different institutional contexts, few sustainability control studies address innovation topics (Adams et al., 2016; Cillo et al., 2019).

Of those studies that do exist, Beusch et al. (2022) recognised the importance of interactive controls for focusing operational managers’ attention to sustainability issues and facilitating information exchange at strategic and tactical levels for product innovations (see also Schaltegger and Burritt, 2018). Others suggest the importance of cognitive integration or cultural controls for supporting sustainable product innovations through creativity (ibid., Baird et al., 2019; Lopez-Valeiras et al., 2016). Meanwhile, Johnstone (2024b) highlights service agreements as key for planning and performance-control decisions at the organisation–customer interface, with implications for environmental and social sustainability. For this, key data (e.g. machine performance and product/component tracing in the supply chain) are considered important for actionable insights that support the organisation’s sustainability strategy (ibid.).

While sentiments on the role of interactive controls and creativity for innovation purposes are shared in the wider management control literature (e.g. Ditillo, 2012; Henri, 2006), others find that diagnostic controls are more effective in contexts focused on exploiting markets or capabilities (Bedford, 2015; Bisbe and Otley, 2004). In this respect, there remains a wider debate on whether management controls can hinder or facilitate innovation (for an overview see Lövstål and Jontoft, 2017). Further still, management control studies not focusing on sustainability often emphasise product innovations in closed (i.e. intra-organisational) settings (e.g. Akroyd and Maguire, 2011; Bisbe and Otley, 2004), rather than on how innovation is achieved in open settings, that is, in collaboration with external parties (see Chesbrough and Appleyard, 2007). Others even focus on innovations in the accounting technologies themselves (Ax and Greve, 2017; Modell, 2009). Such findings make for an overall fragmented conceptualisation and understanding of innovation in the management control literature, including studies that focus on sustainability control.

A more limited group of studies explore open innovation strategies in an inter-organisational setting. In this context, innovation is often driven by collaboration, trust and horizontal relationships, with control mechanisms enacted through centralised governance and contracts, as addressed in the wider business literature (e.g. Aagaard and Rezac, 2022). The interaction between different types of control adds another dimension to scholarly understanding. Biswas and Akroyd (2022b) observed that while diagnostic controls initially dominate in open innovation settings (i.e. inter-organisational setting), they gradually become replaced by interactive controls as relationships mature, fostering trust and collaboration. In another study, Biswas and Akroyd (2022a) found that while management controls may be implemented to support open innovation, they can inadvertently reinforce closed innovation strategies by streamlining product development for the focal organisation, thus enhancing customer satisfaction.

Together, the multiple and potentially contradicting ways through which different management controls impact (mainly product) innovation in (often) a traditional intra-organisational management control setting (Bedford, 2015; Davila et al., 2009) indicate an obscured role of management control on innovation in inter-organisational contexts. There are even less sustainability control studies looking into innovation to draw on. We argue that given the importance of innovation for improved sustainability performance (see also Beusch et al., 2022; Johnstone, 2024a), more research needs to be conducted in this area. Therefore, the aim of this study is to explore the role of control in fostering sustainable innovations within inter-organisational contexts by asking:

These questions build upon prior work on management control for innovations, operationalised around the paradoxes or tensions of control for innovation purposes (see Lövstål and Jontoft, 2017), but extend this into newer contexts (i.e. inter-organisational, industrial) and for different types of sustainable innovations as the control problem (i.e. product, process and service innovations). This inter-organisational aspect of sustainable innovation is important in industrial contexts because industrial innovations typically require collaboration (between customers, and other up/down stream partners), whereby inter-organisational information, behaviour and decisions – as functions of control – require alignment (Johnstone, 2024b). The questions also nuance the understanding of the transformative potential of controls in practice through the notion of change in inter-organisational, industrial settings. The transformation here can be described in terms of degrees of innovations that range from compliance-driven innovations to meet regulations and avoid penalties to co-created innovations that generate regenerative social and environmental benefits beyond the business itself (see Schaltegger and Burritt, 2018; Brown and Fraser, 2006).

Through a longitudinal case study of two industrial firms, we use Simons (1995)LOC framework in an inter-organisational context to understand why and how inter-organisational controls are enacted, making the following contributions to research and practice. Simons’ LOC framework is considered particularly useful as it offers a structured lens for examining the tensions and complementarities that shape sustainable innovation in inter-organisational settings (see also Biswas and Akroyd, 2022a, 2022b).

First, we contribute by suggesting the emerging role of control systems for social innovations that support sustainability in the inter-organisational context for industrial firms. We define social innovations and describe their role alongside the other product, process and service innovations already described related to sustainability (Bos‐Brouwers, 2010; Johnstone, 2024b). Second, we empirically contribute to discussions on the role of diagnostic and interactive controls for exploitation and exploration purposes, showing that diagnostic controls are also important for exploration (see also Bisbe and Otley, 2004) that supports sustainability. For practitioners, we emphasise the need to use a combination of controls in the inter-organisational context to align behaviours that support sustainable innovation. Particularly, social innovations require integration into inter-organisational control systems and industrial companies should focus more on customer education and support programmes for far-reaching effects of the sustainability strategy into the value chain.

The paper begins by elaborating on the different types of innovations for sustainability strategies in industrial organisations, connecting this to the theoretical framework that guides this study. Following from this, the qualitative case method is described. Thereafter, the findings of the two cases are presented before being discussed more analytically. Finally, the conclusions, contributions and future research opportunities stemming from this study are offered.

Innovation is key for industrial firms as they transition towards more sustainable or circular business models (Bocken and Ritala, 2021). Often conceptualised in terms of its type as “closed” (i.e. achieved within the organisation) or “open” (i.e. achieved in collaboration with external parties) (Chesbrough and Appleyard, 2007), innovation is important for competitive advantage. Innovation occurs at different paces and levels; from strategic business model innovation (Geissdoerfer et al., 2018) to improvements on efficiency, operational or service optimisations, among others, at the micro-foundational level (Adams et al., 2016; Coffay et al., 2024, p. 5386). While some innovations are more radical, others involve incremental improvements that yield sustainability performance benefits to industrial actors at different scales. All degrees of “change” are important for advancing sustainability within corporate and industrial strategies (Schaltegger and Burritt, 2018). Still, critical sustainability accounting scholars caution that lasting impact is hard to achieve (Dey, 2002; Milne et al., 2006) because it requires moving beyond eco-efficiency to approaches that respect ecological and social limits (Dyllick and Hockerts, 2002).

The relationship between control and innovation for transformations in sustainability performance is discussed in various ways. Brown and Fraser (2006) suggest that the transformative potential of accounting and control systems depends on how they are conceptualised and applied. They suggest that if systems are narrowly tied to compliance or profit, they reinforce business-as-usual. One the other hand, if systems include more critical perspectives and stakeholder values, they can drive transformative sustainability change.

Similarly, Schaltegger and Burritt (2018) put forward four business cases (BCs) that illustrate progression towards sustainable innovation, ranging from reactionary to collaborative. The BCs reflect different levels of innovation and their transformative potential in terms of improved sustainability performance. Reactionary firms (BC1) focus on compliance-driven, minimal innovations to meet regulations and avoid penalties, such as installing mandated pollution controls. Reputational firms (BC2) adopt marketing-orientated innovations to enhance image and attract stakeholders, like launching eco-friendly packaging lines. Responsible firms (BC3) invest in product and process innovations that genuinely reduce harm and create triple-bottom-line value, for instance through circular design and repairability. At the highest level, collaborative firms (BC4) pursue transformative, co-created innovations that generate regenerative benefits for the firm, society and the environment. This is also the only form where the focus is outside-in, making it more altruistic in nature compared to the other three forms.

Typically, the prior innovation focus of management control research has been on “products” and economic returns in the intra-organisational context (e.g. Bisbe and Otley, 2004; Davila et al. (2009); Henri and Wouters, 2020). Such studies are often quantitatively designed and focus on the characteristic type of controls that support innovation, rather than the innovations’ quantities, qualities or adequacies (see Bisbe and Otley, 2004).^[1] However, various challenges to innovation in the inter-organisational context have been noted.

To achieve sustainable innovations, whatever the scale, understanding how the decision-making and behavioural alignment functions of control interact is central (Johnstone, 2024b). In industrial contexts, this relies on accounting information being shared between business actors (ibid.). However, this is complex because product, process or service innovations require coordination across organisational boundaries (see Johnstone, 2024b; Rauter et al., 2023). These conditions make it difficult to define clear metrics or rely on a single control mechanism between firms, as different phases of innovation call for different types of control. Controls therefore play a dual role: they not only constrain risks and align behaviour but also enable learning, adaptation and collaboration necessary for innovation to succeed in inter-organisational settings (Ibid.).

While the need for non-hierarchical relationships, flexibility and creativity have been put forward as important for solving control problems in the inter-organisational context in both mainstream and sustainability control studies (see e.g. Cäker, 2008; Dekker, 2004; Rauter et al., 2023; Waerness et al. (2023), a significant knowledge gap remains in terms of understanding the “decision-making tensions” related to control system design inherent for sustainable innovations in the inter-organisational context. These tensions often involve multiple stakeholders beyond in-house decision-makers and thus require diverse types of information to effectively navigate the complexities involved.

While particular controls can be used to support sustainable innovations in industrial contexts, Bocken and Geradts (2020) suggest that certain controls (e.g. standard procedures, planning, resource allocations, incentive systems, financial performance metrics or organisational culture) can act as barriers to innovation; a claim suggested by others who champion breaking down cognitive barriers for innovation purposes (e.g. Beusch et al., 2022). However, others imply that the very same controls can support behavioural alignment for sustainable performance outcomes (see Adams et al. (2016); Johnstone, 2024b). This leads us to question the very role of control systems for sustainable innovation purposes in an inter-organisational context, as we simply do not know much about how controls are enacted (i.e. put into practice) for supporting sustainability performance in industrial networks, which can be regarded as the main control problem driving this research.

While the connection between product and process innovations has been well documented since the early work of Abernathy and Utterback (1978), which describes the rate of innovations in the product cycle model, service innovations framed in relation to sustainability can be considered as something “new” (Calabrese et al., 2018; Johnstone, 2024a). To further clarify how different innovation types vary in their sustainability impact, it is useful to distinguish between innovations that alter core product architectures and those that only marginally improve efficiency. For example, modular product designs, where components can be repaired, upgraded or remanufactured, are widely shown to generate substantially greater environmental and social impact than isolated efficiency measures such as installing solar panels (Bocken and Geradts, 2020; Geissdoerfer et al., 2017; Bos‐Brouwers (2010). Modularity supports circular strategies by extending product lifetimes, enabling refurbishment or retrofitting and reducing material throughput across value chains (Johnstone, 2024a). By contrast, single-point technical fixes often improve only site-level efficiency without altering underlying resource-use patterns or value-chain relationships. These distinctions matter for assessing the transformative potential of sustainable innovations (Calabrese et al., 2018; Schaltegger and Burritt, 2018) and help explain why some innovations remain incremental while others support deeper sustainability transitions.

In today’s industrial context, product, process and service innovations interlink to support sustainable strategic performance along economic, environmental and social terms (see Johnstone, 2024a). This relates to broader conceptualisations of value and longer-term performance horizons, which distinguishes sustainability control from traditional management control conceptualisations (see Guenther et al., 2016; Johnstone, 2019). Service innovations that support sustainability are particularly crucial in industrial contexts (Bocken and Geradts, 2020), meaning that existing or new services are offered to industrial customers; services that necessitate accounting information and inter-organisational control (Johnstone, 2024b).

Enacting controls means understanding how controls are “used”, in this case, for different types of sustainable innovations in an industrial context. This “use” often centres around their design features, which are regarded as “loose or tight” (e.g. Davila et al., 2009), “enabling or coercive” (e.g. Mundy, 2010) or “interactive or diagnostic” (Artto et al., 2011). For this study, we use the distinction of interactive or diagnostic as we build upon the existing knowledge bases of innovation studies that adopt Simons’ LOC framework to understand how controls interact for innovation purposes (Chenhall and Moers, 2015). We also seek to elaborate not only on the types of controls but also on the types of sustainable innovations that such controls are implemented to support. This is because sustainable innovations embrace a plurality of forms and scopes in the industrial context.

Simons (1995)LOC framework, a practice-informed model (Martyn et al., 2016), serves as a cornerstone for understanding how management control systems enable and constrain organisational behaviours, including innovation. Its strength lies in its holistic approach, which emphasises multiple controls and diverse styles to balance creativity and efficiency (Chenhall and Moers, 2015). The LOC comprises four levers: belief, boundary, diagnostic and interactive control systems that are now discussed (Simons, 1995). This framework is often used in mainstream management control studies dealing with innovation that we draw on here to build our theoretical framework (Baird et al., 2019; Barros and Ferreira, 2022; Martyn et al., 2016).

Belief and boundary systems, formalised as “statements of procedures” (Burns and Scapens, 2000), guide employees’ activities in the following ways. Belief systems articulate an organisation’s core values and inspire innovation by fostering alignment with a shared vision (Simons, 1995). Flexible cultures often emphasise beliefs controls (Heinicke et al., 2016). In contrast, boundary systems define acceptable behaviours through rules or risk frameworks, limiting some innovation but also focusing efforts and mitigating misaligned initiatives (Barros and Ferreira, 2022). Together, belief systems are perceived as nurturing creativity, while boundary systems are perceived as ensuring strategic alignment and ethical compliance (Beusch et al., 2022; Marginson, 2002).

Diagnostic and interactive control systems, described as “procedures in use” (Burns and Scapens, 2000), manage individual actions. Diagnostic systems monitor performance to meet goals but, in some instances, may stifle innovation by prioritising efficiency and hierarchical accountability (Bedford, 2015). Meanwhile, interactive systems promote learning, experimentation and adaptability through dialogue and collaboration; thus, counterbalancing the perceived constraints of diagnostic systems (Bisbe and Otley, 2004). Balancing these levers is essential, as diagnostic and boundary systems maintain focus, while belief and interactive systems foster creativity (van der Kolk et al., 2020).

Notwithstanding its widespread use in mainstream management control studies, critiques of the LOC framework exist. Some suggest its insufficient focus on socio-ideological and informal controls, as well as its emphasis on top-management’s belief systems as inappropriate (see Martyn et al., 2016), especially for sustainability concerns (Johnstone, 2019). A growing stream of research suggests that sustainability-related decisions often require knowledge and perspectives that extend beyond top management, as shown in studies on sustainability accounting and control (Egan and Tweedie, 2018; Frostenson and Johnstone, 2023; Rodrigue and Picard, 2022). At the same time, top managers remain central for steering innovation processes (Pan Fagerlin and Lövstål, 2020). There is also the wider discussion on how the design of particular controls contributes to and/or hinders innovation (see Lövstål and Jontoft, 2017). Others see the levers as dynamic and mutually reinforcing, not purely separate tools (e.g. Bisbe and Otley, 2004), a view that aligns with studies emphasising the interdependence and integration of control systems (e.g. Widener, 2007), particularly in sustainability contexts (e.g. Narayanan and Boyce, 2019; Beusch et al., 2022).

For this study, we suggest that the LOC can simultaneously encourage innovation (exploration) and maintain operational stability (exploitation) in industrial contexts. We also highlight its potential in the inter-organisational setting (i.e. across organisations) as a lens for understanding the types of controls enacted (see also Biswas and Akroyd, 2022a, 2022b) for different types of sustainable innovations that support sustainable strategic “change” in industrial business networks as now explained.

In industrial organisations, beliefs systems are important for fostering organisational commitment to sustainability by embedding sustainable and/or circular values into the company’s mission and culture to drive innovation and ensure alignment with long-term sustainability goals (see e.g. Jollands et al., 2015). In addition, interactive controls can facilitate innovation through active dialogue on sustainability challenges (Beusch et al., 2022). This can be achieved through cross-departmental or inter-organisational collaboration on sustainable and/or circular solutions that encourage experimentation in products, processes and services for industrial firms (Johnstone, 2024a). Meanwhile, boundary systems help set limits on unsustainable practices such as the overuse of resources or non-compliance with environmental or industrial regulations. Diagnostic controls help track progress towards sustainability and/or circularity through key performance indicators on waste reduction, resource efficiency or carbon neutrality in the value chain, which both helps ensure accountability for actions as well as supports (inter)organisational learning from performance data. In addition, the combination of controls appears to be important for supporting innovation in industrial contexts because combining inspirational (belief and interactive) and constraining (diagnostic and boundary) levers could reinforce proactivity without stifling creativity. Finally, prior research shows that belief, interactive and diagnostic controls can strengthen organisational resilience by shaping how sustainability-related goals and priorities are communicated and acted upon (Baird et al., 2023).

In summary, the levers play different, important roles in supporting sustainability performance outcomes in wider business networks. They do this by offering different categories of control tools to balance and align organisational actions with external pressures for change. These changes may not be as far-reaching or radical as the more critical social and environmental scholars would like, but they can be connected to the various business cases already described (e.g. Schaltegger and Burritt, 2018) as summarised in Table 1.

Building on this, we now turn to how institutional forces shape the enactment of controls in inter-organisational contexts and how the four levers interact to support progression across the business cases.

Industrial organisations are driven to adopt increasingly sustainable and circular production systems due to various institutional demands, including regulatory pressures, market expectations, societal norms and competitive drivers (see Arroyo, 2012; Wijethilake et al., 2017). While a detailed exploration of these demands is beyond the scope of this paper, the transformative potential of control systems lies in their ability to guide focal organisations and their broader industrial ecosystems towards long-term performance outcomes and expanded notions of value that incorporate social and environmental concerns. Thus, while Simons (1995)LOC helps explain how controls are enacted for sustainable innovations, it does not explain why certain controls are enacted in the inter-organisational context in the first place. This is where drawing loosely on institutional perspectives is important for understanding transformation.

The transformative potential of controls for sustainable innovation purposes in the inter-organisational context lies in the alignment between external and internal institutional forces for change as already indicated through the above-described business cases. Change here regards the extent to which new sustainability controls are implemented that support innovation purposes whereby the change(s) indicate(s) a trajectory of controls towards increasingly sustainable solutions (see also e.g. Jollands et al., 2015; Battaglia et al., 2016; Wijesinghe et al., 2023; Johnstone and Skoog, 2025). This nuances the discussion on the enactment of controls by outlining potential reasons for such “enactment” (or resistance to) in our case contexts.

For example, the external aspect of institutionalisation helps explain “change” in terms of how controls are enacted to support sustainable innovation in industrial contexts, which may be more reactionary to begin but move increasingly towards more responsible or even collaborative business cases over time. Various coercive (e.g. legislative, regulative demands), normative (e.g. societal or professional expectations of behaviour/practices) and mimetic (e.g. following other industrial actors’ practices) pressures on industrial organisations help explain the formal design of sustainability controls (Arjaliès and Mundy, 2013; Wijethilake et al., 2017). This has been captured as operational sustainable changes (see Arroyo, 2012), for example, by adapting or (re)designing management controls for sustainable innovation purposes. At the same time, Schaltegger and Wagner (2011) emphasised the internal aspects of institutionalisation, referring to different types of entrepreneurships, as crucial for enabling sustainable organisational change and sustainability-orientated innovation. Also, studies on the role of institutional activists or entrepreneurs that promote changes to sustainable working practices or innovations from below are deemed increasingly important for sustainability control within organisations (Ball, 2002; Brown and Fraser, 2006; Johnstone, 2024a, 2024b). We can only assume their importance in the inter-organisational context for sustainability issues, as implied in the wider business literature (Gabler et al., 2023; Velter et al., 2020).

Together, belief, boundary, diagnostic and interactive systems operate jointly (see Bisbe and Otley, 2004) to facilitate change in a wider industrial network (whether incremental or radical) by responding to external institutional pressures and internal organisational needs. The levers are expected to clarify acceptable practices, track performance, support organisational learning and foster a shared commitment to sustainable innovation across products, processes and services in both intra- and inter-organisational settings. Similar to Bisbe and Otley (2004), we suggest that for true transformation in inter-organisational contexts (i.e. a collaborative business case), the levers should be regarded as interdependent; the optimal configuration for driving transformative change.

Figure 1 summarises these connections and, in doing so, extends earlier LOC conceptualisations by embracing inter-organisational aspects into each of the levers for our research context. It also implies the mutually reinforcing effects of external systemic changes in the institutional context and the intra- and/or extra-organisational changes through controls that support innovation as necessary to understand the role of control systems for sustainable innovations in the industrial context and even beyond.

To increase our understanding of the role of control systems for sustainable innovations in an inter-organisational context, this study draws on the enactment of controls in two industrial organisations through a multiple case study design. Case studies are useful for exploring the actual practices of organisations as well as for understanding how or why something came into being (Cooper and Morgan, 2008). Specifically, the multiple case study approach provides a rich, contextual understanding of complex phenomena by revealing both common patterns and unique variations that can generate broader insights (Eisenhardt, 2021). This makes it well suited to examining how controls for sustainable innovations are enacted across organisations, while also considering the contextual factors that shape them (ibid.).

The cases in this study are multinational, private-listed, industrial organisations operating on the B2B market and headquartered in Scandinavia. CONSTRUCT (pseudonym) is an international manufacturer of heavy-duty machines for the mining and construction industries, with production sites and customers in over 150 countries worldwide and around 20,000 employees. Its products are highly customised and produced in low volumes, meaning that most revenue currently comes from service innovations. CONSTRUCT has the vision for complete circularity by 2030 and is subject to various European legislations on sustainability (e.g. Corporate Sustainability Reporting Directive [CSRD 2022/2464/EU] and associated European Sustainability Reporting Standard [ESRS] range), making sustainability high on the corporate agenda. Access to CONSTRUCT was given in 2022, mediated through the process and application manager at the group level, although drawing on data from earlier and continues to this day.

TECH (pseudonym) manufactures components and subassemblies a variety of industrial equipment for more than 40 industries globally, especially rail, automotive and other heavy industries, as well as for the industrial distribution market. TECH mainly sells through dealers and intermediaries, not directly to end-consumers. Operating in over 100 countries and employing over 40,000 people around the globe, TECH is subject to the same European legislation as CONSTRUCT, yet its industrial products are more standardised and the proactive push for sustainability has come mainly from within TECH as part of the business strategy and overall direction. Access to TECH was established in 2013 and continues to this day.

As contrasting cases (Eisenhardt and Graebner, 2007; Eisenhardt, 2021), the two companies operate in distinct industries and entail different business activities to be “innovated”, yet both are headquartered in Sweden and subject to the same legislation. Given our focus on the role of control in sustainable innovation, the findings from these cases can be compared and may also be transferable to other research contexts addressing the same phenomenon.

For this study and its focus on the role of control systems for sustainable innovations in the inter-organisational context, we take the respective firms described above as the focal points of departure similar to other studies on inter-organisational sustainability control e.g. (Spence and Rinaldi, 2014; Johnstone, 2024b; Beusch et al., 2022). This means that our empirics stem directly from the focal companies and, for the purpose of this research, have been limited to just over ten years of investigation (i.e. 2014–2024 inclusive) ^[2]. This allows for a comparative analysis of how the respective companies are “innovating” and “controlling” for sustainable solutions considering similar contextual or institutional factors as Scandinavian-based industrial multinationals. This timeframe is also considered important from an institutional perspective given the vast developments regarding sustainability accounting, reporting and control in the European context.

Both primary (e.g. interviews, observations) and secondary (e.g. internal documents) data were used in the cases as outlined in Table 2 below, which also provides codes for empirical material that we refer to in the findings ^[3]. All formal interviews were recorded and transcribed, whereas notes were taken at meetings (formal and informal), lectures, site visits or from video footage. The interviews were conducted with personnel at group and tactical levels for both organisations, with personnel both directly and indirectly implicated in the sustainability work. This constituted 19.25 h of primary material and 1,380 pages of secondary data for CONSTRUCT and 39 h of primary material and around 1,500 pages of secondary data for TECH. The secondary data were used as reference points ex ante to guide the interviewees towards particular aspects. They were also ex post used to understand particular types of diagnostic or interactive controls, beliefs or boundary systems used and reported (e.g. in the corporate reports). We also drew on national and international legislation and directives (e.g. National Accounts Act, CSRD) for contextualisation in our empirics as to why certain controls were enacted from an institutional perspective.

Interviews for both companies entailed a deliberate sampling of personnel involved in sustainability issues as well as a snowballing approach whereby initial interviewees suggested other employees that we should speak to in terms of our research interest. Importantly, interviewees from other departments or functions (other than sustainability) were included given that sustainable innovations involve personnel with different roles and responsibilities. This was also considered necessary as prior research shows that sustainability accounting and control work often involves actors beyond those formally responsible for sustainability, highlighting that relevant expertise is distributed across the organisation (Egan and Tweedie, 2018; Rodrigue and Picard, 2022; Johnstone, 2024b). The interviews were mainly conducted in person, but online and telephone interviews also occurred. This was not a disadvantage and was, in fact, sometimes necessary due to the vast geographic base of managers for both multinational organisations; thus, providing research opportunities that would have otherwise not been possible (de Villiers et al., 2021).

Interview guides were broadly operationalised around the types of controls designed and used to support sustainability within the industrial organisations in lay terms, i.e. that did not include the theoretical constructs related to Simons (1995)LOC such as diagnostic-interactive controls or belief-boundary systems in the questions. There was also a focus on understanding the specific sustainability strategies of the industrial organisations, particularly in relation to circular solutions or electrification. In this sense, the interview guides were deliberately kept broad to understand how each organisation was working towards increasingly sustainable and/or circular strategies.  Appendix offers examples of the types of questions asked, which were then tailored to specific interviewees and roles.

The theoretical framework guiding this research was used to frame, structure and explain the findings even though the theoretical concepts were more inductively induced from the data. This means that, as a first step, the authors sought fragments from the data (e.g. transcripts, meeting notes, corporate documents) that could be operationalised broadly around:

• the main sustainability strategies of the industrial organisations considering their wider institutional context;

• the types of sustainable innovations (product, process and service) explicitly discussed to meet these strategies; and

• the means of controls used to support these strategies.

A form of content analysis was conducted in the secondary data whereby keyword searches were used to expedite fragments on “sustainability” (e.g. sustainable, sustainability, circular, environment/al, etc.), “innovation” (e.g. R&D) or “strategy” (e.g. vision, mission, goals, objectives, aims) ex post. Influenced by the ideas of institutional theorising, we deduced some reasons as to why our industrial organisations were moving towards (more) sustainable innovations within their particular contexts and the involvement of other up or downstream actors in this – as forces for change or resistance – as an inter-organisational control problem.

As a second step, attention was given to the distinct types of controls designed and/or implemented “within” our case organisations for sustainable innovations. In this sense, control becomes the means to connect strategy with operations (see Guenther et al., 2016). We categorised the types of sustainability controls designed and used for innovation purposes in relation to Simons (1995) levers, but with the added nuance from our theoretical framework that incorporates the “external” aspects of control imperative to industrial organisations, especially those with circular business strategies such as ours.

Following from a narrative presentation of our findings, we then mapped out the main controls used by both cases in a summary theoretical model. This served as a reflection point to structure our following discussion in terms of the role of these controls for encouraging alignment, fostering duality and, ultimately, driving systemic change in our research context.

Note that the third author is a sustainability manager for one of the case companies, who was affiliated with a local university at the time of access. Having an insider is beneficial given that they have intimate knowledge of the case company, its context and control systems (see Beusch et al., 2022). This author did not interview other employees at the company or conduct the initial analysis of the interview transcripts. This author participated in subsequent analysis and verification work and in writing up the findings, alongside the other authors who were also involved in the interpretation to mitigate bias (see Rowe et al., 2012).

As already indicated in the case descriptions, both CONSTRUCT and TECH have been increasingly affected by more coercive institutional aspects in recent years, which require them to make adaptations to their respective strategies and especially when it comes to issues of sustainability control in the value chain. Both companies are subject to European legislation and standards (e.g. CSRD, ESRS, Restriction of Hazardous Substances in Electrical and Electronic Equipment [RoHS, 2011 / 65/EU], Low Voltage [LVD, 2014 / 35/EU], Electromagnetic Compatibility [EMC, 2014 / 30/EU] and Radio Equipment [RED, 2014 / 53/EU] Directives, etc.), as well as are expected to, in various ways, support the home country’s national objective of complete circularity by 2040.

Beyond that, CONSTRUCT explicitly mentions the Battery Directive (2006/66/EC) and New Battery Regulation (2023/1542), part of the European Green Deal, as important for its operations. It has had electrification as a key strategic objective since 2011 and KPIs for selling aftermarket services related to batteries and charging infrastructures. Innovation is also driven by the wider industrial standards or norms of the construction and mining industrial customers it provides solutions for (e.g. ISO19296 on Machine Safety). These can be viewed as customer requirements that CONSTRUCT must consider in terms of how it designs its sustainable innovations.

From a critical perspective, such legislation may be seen as external institutional pressure for change, yet both companies pursue innovations that extend considerably beyond compliance to address broader social, environmental and financial performance goals. CONSTRUCT prides itself as a company built on innovation; particularly in terms of the “speeding up the sustainable innovations” (ASR, 2022) that it provides for its customers. The company sees itself as a business partner that ensures its customers “can work in the safest, most environmentally friendly, and efficient way” (ASR, 2023); thus, meeting most sustainability performance aspects. It also views the mining industry as necessary for supporting energy transitions (e.g. through mining lithium for batteries). This customer focus is also important, especially when it comes to the use of products sold as this constitutes to 80% of its reported (Scope 3) emissions (National Accounts Act). The company actively seeks to engage and educate its customers, for example on automated machinery that reduces workplace accidents (a health and safety and employee well-being issue linked to social sustainability) and on electrification initiatives. Thus, working together with upstream partners to provide customer solutions is part of its increasingly sustainable business model that, while costly in the short-term, can promote long-term improved performance for all aspects of sustainability.

Automation, electrification and digital solutions are put forward as the three main trends that support CONSTRUCT’ innovation efforts in collaboration with partners by “driving productivity and the sustainability transformation of the industry” (ASR, 2022, emphasis added). Particularly, the automation of industrial products relates to CONSTRUCT’s social sustainability objectives of zero work-related injuries and electrification (i.e. replacing diesel machines with batteries) promotes its strategic ambition to half carbon emissions of 2019 levels by 2030 (SP). Both objectives relate to the wider business network as per the incoming legislation but also to CONSTRUCT’s own values, as its servitisation and electrification strategies existed long before mandated. Meanwhile, having better integrated digital solutions (e.g. telematics, connected CRM and ERP systems) provide the means to monitor machines at site (i.e. customer locations) for optimal performance (with financial implications for customers), as well as incorporate feedback from up and downstream partners into other internal sustainable innovations.

Around 10% of CONSTRUCT’s global staff base is dedicated to R&D, innovating the industrial products and services that it provides in different ways and especially in the three strategic areas of digital solutions, automation and electrification. These R&D initiatives have been needed especially for the design of modular products whereby diesel machines can be replaced or retrofitted with battery engines, linking to further service innovations. CONSTRUCT perceives industrial collaboration as especially important for sustainable innovation both because it cannot meet certain innovations by itself due to geographical, technical or resource constraints and because it seeks to learn through its involvement in cross-section industrial knowledge networks. It also relies on customer onboarding, since the product and service innovations described above – supporting social and environmental sustainability, as well as long-term financial performance for industrial customers (e.g. automation, electrification and digital solutions) – could not occur without these customers.

CONSTRUCT also works with its 1,800 suppliers to drive product innovations (e.g. types of materials used, repairability, reusability, functionality, etc.) and offers complementary services such as charging infrastructures for battery machines (ASR2022, R14). The company also participates in joint-industrial projects with its customers and competitors to develop intelligent and sustainable mining practices (SP). Without its customers, CONSTRUCT would not be able to offer sustainable innovations and the institutional pressure experienced by mining companies around the world appears to drive the need for inter-organisational collaboration and innovation. For example, there are projects for technological development such as providing 5G services in remote locations to be able to track machine use for repairs and the development of original equipment manufacturer (OEM) standards in terms of charging systems for electric machines (SR, R6). It also involves continually innovating its traditional products through increasingly modular designs that can be replaced or retrofitted when needed in a more sustainable manner. More recently, CONSTRUCT has been acquiring partner organisations to speed up certain innovations or complementary services, for example, in relation to firms that have competencies in battery conversions for diesel machines or those specialised in charging infrastructures (R12).

These innovation trends are supported by the growing service offerings that help ensure customer lock-in such as remanufacturing, midlife upgrades, productivity-enhancing technology, health and safety training and battery conversion services (SP). As of 2023, aftermarket solutions constituted 68% of all orders received and were the main revenue-generating stream for CONSTRUCT as it transitions from its linear production-based past towards 42% of its products having emission-free versions. In this sense, the service innovations are not only sustainable in terms of improved safety, wellbeing and environment, but also ensure financial sustainability for CONSTRUCT and its customers in the long-term. Therefore, the impact of such innovations reaches beyond the immediate performance goals of CONSTRUCT and its customers.

Subject to many of the same legislative demands as CONSTRUCT, since 2013, TECH has had the goal of fully integrating sustainability into operations; supported by increasingly formal controls, new responsibilities and sales targets related to sustainability (T1–T8) ^[4]. In incorporating sustainability issues more clearly into strategy, TECH has been transitioning from an industrial heavyweight to a more agile cleantech leader, meaning “clean and profitable growth while creating shared value for [TECH], customers, and the environment” (ASR2023). By 2030, TECH aims to grow its business at improved margins by leading sustainable solutions; innovations driven by digitalisation, sustainability, electrification and regionalisation, which are reflected in its strategic balanced scorecard (ibid.). For that, TECH’s focus is on driving R&D and innovation across the value chain to accelerate product development, focusing on high-growth segments, where digital transformation prioritises process re-engineering, data and automation, with the aim of becoming a data-driven company (ibid.). By leveraging advanced analytics, TECH further intends to optimise operations through automation and robotics while ensuring efficiency through streamlined processes (ibid.).

So far, TECH is on track to meet Scope 1 and 2 greenhouse gas emissions reduction by 2030, validated by the SBTi, alongside five other long-term financial goals (T13 and T14). Yet, challenges persist, such as scaling circular economy (CE) solutions globally and balancing “financial” growth with other types of sustainability and TECH acknowledges that “it’s all about scaling up,” aiming to achieve this through a “company-wide circularity movement” (T16).

With the CSRD becoming mandatory, TECH frames its goals as “ESG-driven circularity targets to lead and innovate,” positioning ESG reporting as a baseline (T15). Yet this strategic framing appears equally aimed at satisfying investor, employee and market expectations, raising doubts about whether the commitment reflects genuine transformation or merely strategic compliance. Although TECH invokes circular principles to reinforce its industry leadership and innovation profile, the transition from a linear “take, make, dispose” model demands a fundamental reconfiguration of operations, one that challenges not only financial and environmental trade-offs but also stakeholder inclusion and deeper cultural change (T16). While TECH also presents CE as crucial for its future competitiveness and relevance, questions remain as to whether this reflects genuine integration of sustainability or is primarily framed to advance strategic interests.

Beyond these more strategic aspects of sustainable innovation, stemming from the headquarter context, the enactment of controls regards how different types of controls are used for sustainable innovations in the wider inter-organisational, industrial context. As indicated, innovations that support sustainability in both cases broadly relate to digitalisation, electrification and automation trends ^[5]. These innovations lead to an array of environmental (e.g. battery conversions), social (e.g. customer health and safety, wellbeing) and financial performance outcomes for the case companies and their customers (e.g. reduced downtime, added revenue). For both companies, these innovations are not only the result of external institutional aspects whereby the drive for change is coming from outside, but also because of internal strategic visions that drive change throughout the industrial network from within, as now explained.

The beliefs systems of both cases are clearly positioned at the intersection between intra and inter-organisational systems; that is, to be successful, CONSTRUCT and TECH require their beliefs to permeate into the value chain. This is illustrated in the findings as ways to “future-proof” the products and services offered. Without industrial customers, suppliers and other up/down stream actors, their strategic visions would be nothing.

CONSTRUCT “dares to think differently” by transforming the industries using its machines to more sustainable ones. “Innovation” and “value in people” are its core values (corporate website). CONSTRUCT perceives itself as an “entrepreneurial” company, supported by courage, curiosity and creativity, reinforced by its internal mantra of being “a 150-year-old start up” (R11).

As examples of its innovative focus when it comes to sustainability, the company began to develop its electrification infrastructure back in 2011 when the (now) senior zero emission manager noted the implications of emissions (environmental and health related) from diesel machines in underground mining and construction (e.g. tunnels) “was not going to be a viable option going forward”. Furthermore, the value that “people matter” (relayed through various strategic goals such as no work-related injuries) extends beyond CONSTRUCT’s immediate employees and especially into its customer activities in hazardous industries such as mining: “the sustainability work within the organisation is considered key for pushing new ideas not only internally but for our customers” (R11).

The beliefs systems of CONSTRUCT are implemented through various controls that are enacted to support different sustainability projects for a variety of inter-organisational actors. One example includes its employees being able to donate SEK 5 of their salary a month, which is then matched and doubled by CONSTRUCT, to fund projects that support clean water extraction in some of its key markets (Event R²). Given that CONSTRUCT has 20,000 employees worldwide, the potential is SEK 300,000 per month or SEK 3.6m per year (€330,000). Even if, for a company like CONSTRUCT, this amount could be considered marginal, it reinforces the belief that “people matter” – not only its employees and customers (i.e. operators) – but the wider communities in the value chain, which are affected by mining operations around the world.

Another way of reinforcing CONSTRUCT’s beliefs is to encourage and educate its mining customers to “mine for a wider purpose” (e.g. to provide minerals that support renewable energies rather than for other less sustainable projects) (R11, Event R3). These are customer events planned around the world to educate customers on the broader impacts of mining and the sustainable innovations in products and services offered by CONSTRUCT. There are also dedicated service centres with sales force trained to meet customers and promote more sustainable customer solutions.

Local events further illustrate this, such as the annual “bring your baby to work day” (Event R²) in one of the main Swedish production sites, which illustrate the values of the company’s employees having a healthy work-life balance, and its consideration of its employees as its key “asset”. It also provides a symbolic commitment to gender equality by normalising babies/children in the workplace; again at no immediate cost to the company so a financially sustainable activity.

While not related to innovation in a more traditional sense (i.e. in terms of product and process innovations), such strategies are nevertheless innovative for promoting social cohesion within CONSTRUCT and throughout its wider business network and the communities that it serves. Thus, they offer examples of low-cost social innovations that are not mandated by law but nevertheless, being used by CONSTRUCT to help solve a range of socio-environmental problems in the inter-organisational business network and those affected by it (i.e. in the wider value chain).

From the outset, TECH has had a strong history of the social welfare of its employees, providing robust services for them in terms of healthcare, training initiatives and community care (T3, T6, T7). This was long before such issues were mandated by law. Beyond these more localised, social sustainability elements for employees, in more recent years, TECH has embraced a belief system rooted in “authentic commitment to the CE” (T14) as crucial for its strategic goals. To stay competitive against the continental European firms adopting circular models, TECH aims to set an industrial benchmark by embedding CE principles into its organisational ethos (ibid.). Using its belief systems, TECH frames circularity and Net-Zero ^[6] at the core of its identity, emphasising enhanced product properties and improved customer application performance that focuses on concrete measures. In 2023, it identified four key factors for driving circularity, which in various ways relate to the inter-organisational context:

1. flexibility for sourcing components and servitisation:

2. data traceability and AI integration;

3. employee upskilling; and

4. embedding circularity into product and process design throughout the lifecycle.

For TECH, this requires a fundamental shift in value chain dynamics, demanding deeper collaboration than traditional linear business models. As one manager noted, “We can’t close the loop on our own … circularity only works when our partners move with us” (T20).

To minimise waste, optimise resource use and create closed-loop systems, TECH launched an ambitious strategic partnership development programme in 2024 (T20). The programme aims to foster win-win outcomes by partnering with both existing and potential customers, as well as other industry players. Key elements of this initiative include market analysis of potential partners, the creation of tailored engagement strategies and the development of collaboration frameworks and knowledge-sharing platforms. In this context, TECH uses its belief systems to build trust, inspire collaboration and foster a culture that prioritises circularity in its wider business network. In addition, TECH promotes “success stories” on the intranet and internal meetings, as well as in the annual report, sales presentations and other external communications that highlight tangible benefits, such as waste reduction and improved product quality through remanufacturing. This approach seeks to make the circular economy transformation meaningful and motivating for other organisations, rather than imposing a top-down directive. Notwithstanding the attempts made by TECH to embed its circular beliefs systems throughout the organisation and into the industrial network, various challenges persist. Within TECH and its divisions across the globe, employees must acquire new “circular” mentality skills (T16). Meanwhile, customers need to trust the quality and reliability of remanufactured products (T18).

Due to the perception of inferior quality, as well as legal and market reasons related to boundary systems, TECH must sell its remanufactured products (e.g. steel products) at lower prices/large discounts, which both incites customer doubts in the quality but also entails reduced profit margins for the company, despite the environmental benefits in terms of reducing the use of virgin materials (T17 and T18). As a strategy level manager explained: “Every time we remanufacture, we know we are losing money on the product compared to selling new steel, but we continue because otherwise the circular idea will never take hold” (T18). This illustrates how TECH accepts direct financial sacrifices to promote circularity, underscoring an engagement with sustainability that extends beyond a narrow business-case rationale, a move that is not always appreciated by middle-level management, as their performance is still most strongly measured in terms of financial contribution. Finally, suppliers and distributors must align with TECH’s CE goals, despite potential conflicting incentives or operational constraints, which appear to be the most difficult challenge (T20). But, beyond this, TECH is also increasingly addressing workers’ rights or social sustainability issues in the supply chain via the introduction of ESRS S2. This has led to a stronger internal emphasis on “people in the supply chain” reflecting a growing recognition that social responsibility must extend beyond the company’s own workforce to include those employed by suppliers and partners.

Considering these examples, belief systems alone, although powerful in terms of the visions and narratives that they construct, are insufficient for different types of sustainable innovations. Yet, as Simons (1995) posits, they must be supported by other structural mechanisms and interactive control systems in the industrial context to ensure the products or services of industrial firms have a lasting impact. We now turn to these other levers.

The visions and beliefs of the case companies are supplemented with other boundary systems that define the scope of operations within the industrial ecosystem in which they operate. These boundaries provide safeguards against unsustainable practices, such as excessive resource use or waste generation, as highlighted in the example of remanufactured products above. For clarity between the orientation of boundary systems (i.e. why they come into being and for whom), we broadly capture these systems as “adherence to” (i.e. to external mandates) or “adherence through” (i.e. through inter-organisational adherence to company-developed boundary systems) boundaries in the following.

“Adherence to boundaries” regard adherence to a broader array of institutional mandates, demands or expectations on our industrial organisations such as being signatories to the UN Global Compact or subject to the CSRD, among others. CONSTRUCT is also required to comply with industrial standards and specifications on, for example, tools and attachments, as well as producing mining machinery as already mentioned. The same is true for TECH as it follows specific industry standards (e.g. automotive, railway and aerospace industry) but also “occupational health and safety” and “international labour organisation” standards, to mention just a few. CONSTRUCT is also involved in collaboration with other large industrial actors to develop testbeds for sustainable underground mining which “requires new control systems, new and improved mining equipment, as well as complex and efficient management systems that meet future demands for a sustainable industry” (CONSTRUCT website, 2020). In this sense, CONSTRUCT is attempting to develop boundary systems with others in the industry from the “outside-in”.

Meanwhile, “adherence through boundaries” regard adherence through the construction of company-specific sustainability policies, codes of conduct and supplier codes of conduct, digital infrastructures, among others, which are implemented both internally and throughout the industrial network but take the firm as the focal point of reference. For TECH, the Code of Conduct serves as the primary boundary system, setting ethical standards that guide policies across both group and local levels, superseding all other policies (ASR2023). It is aligned with sustainability regulations and standards (e.g. CSRD/ESRS) and is designed to prevent greenwashing, resource overuse and unethical labour practices. The code includes a strong emphasis on health and safety management, aligning with ISO 45001, supporting a zero-accident programme, which is applied consistently across operations, suppliers (a specific supplier code of conduct), distributors, agents and intermediaries (a specifically designed code of conduct). Through the code, TECH mentions that it provides “non-discriminatory working conditions, promotes diversity and ensures that wages and other related benefits meet at least the legal or industry minimum standard in the country in question”, which are rendered in full compliance with laws and collective agreements (ASR2023).

TECH has also designed and implemented a CE-programme for its circular operations. While the programme was designed in response to external factors, which focus on aligning with regulatory frameworks (e.g. CSRD regulations, ESG reporting requirements and the EU taxonomy) and is responding to market demands by understanding customer needs (T12), it deals with TECH-specific operations for circular solutions (T14). This CE-programme is therefore considered by TECH as important to stimulate sustainable innovations and, as a control, it is supposed to be implemented across the value chain, using “an iterative process” driven by a core (around seven strategic level members) and an extended core team (around 40 members on all levels and in all regions).

These different forms of company-centric boundary systems nevertheless entail various challenges in their enactment related to the inter-organisational context. For instance, adherence to the Code of Conduct for CONSTRUCT relies heavily on self-regulation and periodic training internally (R1). While managers are required to conduct annual training to stay updated on the changes stemming from the wider legislative or institutional context, the practical enforcement and monitoring of such principles, particularly in complex global supply chains, remain opaque or difficult to control. For TECH, establishing and enforcing clear operational boundaries, such as mandating minimum levels of recycled content in designs, introducing design-for-disassembly requirements to ease future remanufacturing or implementing material passports for traceability across the supply chain, remains challenging, with various bottlenecks evidenced (T17, T18 and T19).

This is further complicated by TECH’s vast global reach, which makes the coordination of strategic boundaries or codes of conduct across regions difficult to enforce (T18). Geopolitical and cultural differences add to this complexity: while sites within the US market remain cautious towards sustainability-related investments, neighbouring Mexico and Canada show greater openness, shaping TECH’s future investment opportunities. Although the company recognises the importance of locating remanufacturing centres close to key markets, such uneven regional conditions make this difficult in practice. Reflecting these uncertainties, mandatory targets for waste reduction, material efficiency and circular-ready products are initially kept internal to avoid committing to goals that may later prove unachievable (T20).

Yet, in both cases, a central concern involves the boundaries of connected digital infrastructures in the inter-organisational context. For instance, CONSTRUCT highlights the need for digital connectivity among partners, sometimes even competitors, ^[7] to share information not only about products but also about customers’ mining processes. According to R12, this is all about “knowing how the customer treats the machines” and making appropriate innovations related to sustainability from this. Still, there remain limits to getting accounting information related to machine performance from customer machines in remote parts of the world, where internet connectivity may be limited and real-time data not accurate (R4). There are also limits to the sheer amount of incoming data to make sense of, which can affect the resultant innovations, even with the support of customer support and development operations engineers:

These boundaries of connected digital infrastructures are made even more complex as they require extensive IT solutions that CONSTRUCT is struggling to organise due to resource constraints (R5). Thus, boundary systems appear to both facilitate (such as the adherence to mandates and through internally designed controls) and hinder the development of sustainable innovations in industrial contexts; that is, in terms of their reach and impact.

Both CONSTRUCT and TECH have extensive diagnostic systems with concrete targets related to, for example, emissions/waste reductions or having a certain percentage of service agreements sold on sustainable offerings. Often, these diagnostic performance measures relate directly to the boundary and beliefs systems described above. Beyond the clearer strategic level targets of both companies, which are manifold and often incorporated into company-level balanced scorecards, diagnostic controls within and from the inter-organisational context are becoming increasingly important for strategic sustainability aims.

Yet, machine performance (e.g. in terms of operational hours), customer experience or perceptions (e.g. in terms of safety of machines for CONSTRUCT or the quality of remanufactured steel products for TECH) clearly indicate diagnostic systems that extend beyond the focal organisations and allow them to innovate processes, procedures or service offerings related to sustainability in the inter-organisational context. Accounting data coming in from customers is not only important for inventory management between up/downstream partners for both companies but also for product development as customer information is exchanged between the case companies and their suppliers. This information also informs the sustainability reporting for CONSTRUCT and TECH.

As examples, through service innovations that include telematics on machines or integrated CRM or ERP systems (R1), CONSTRUCT can track the use and disposal of machines and spare parts, which leads to the exchange of inter-organisational accounting information for quality control purposes related to health and safety or product innovations in the wider business network:

These integrated systems for exchanging performance information are also important when the companies cannot provide all products or services themselves. For example, CONSTRUCT currently does not have the capacity or means to recycle its batteries and needs to do that in collaboration with an upstream partner that is informed via diagnostic controls when incoming batteries will be received (R14).

Notwithstanding, for TECH, one of the most pressing challenges regards reconciling financial imperatives with environmental goals. This is where “mixed” diagnostic control systems that include CO₂ emissions and other non-financial metrics (e.g. share of CE-solutions in total sales, CO₂ reductions achieved) become essential (T16). TECH has a dual reporting system that incorporates an “acceptable range for profit margin reductions” in the early stages of CE-solution adoption to ensure that financial pressures do not overshadow its commitment to sustainability (T19). Thus, the challenge here lies in defining KPIs and metrics that are both stringent enough to enforce CE principles and flexible enough to accommodate the uncertainties of market dynamics. These diagnostic controls are typically used to establish and monitor clear performance metrics for the sales force (T18, T19), ensuring alignment with organisational goals in promoting CE-solutions and followed up at the regional level.

Yet the regional differences are striking: “In France the gap is simply too wide – we can’t price CE competitively when logistics and demand don’t support it”, explained one manager (T18). By contrast, sales managers in the Benelux countries noted that “here, customers ask for sustainable solutions, and shorter transport distances make them viable” (T19). Such contrasts reveal how the institutional context and market demand shape the extent to which CE-solutions can realistically be advanced. By leveraging this data, TECH not only evaluates the effectiveness of its sales strategy but also identifies where regional conditions hinder or enable progress (T14, T20). Therefore, empowering the sales team to prioritise CE-solution sales requires a carefully structured approach that integrates performance incentives, specialised expertise and robust training (T20).

A key issue, however, remains for TECH in that selling CE-solutions often involves longer sales cycles (T18), something not mentioned by CONSTRUCT, as it requires educating and persuading customers about the benefits of circularity. To address this challenge, TECH is evaluating options to further incentivise sales representatives for the additional time and effort required, such as offering bonuses tied to time spent on customer education or providing centralised funding for these initiatives across different regional contexts (T19 and T20). Meanwhile, the bonuses tied to selling sustainable service agreements at CONSTRUCT are tied only to the managerial level of customer service centres, rather than individuals in the sales force (R10).

The types of diagnostic controls mentioned in relation to obtaining data along the customer interface are noted as important for innovation in CONSTRUCT, even with the limits of connectivity and having enough customers sold on service agreements as mentioned earlier. The incoming data are used as a reference point for what can be regarded as interactive discussions on sustainable innovations both within CONSTRUCT (e.g. in its R&D team for product innovations and service division on service innovations that support sustainability), as well as for sharing with other upstream partners, such as the suppliers that can modify the design of materials and various parts (R8). Similarly, customer education and long-term relationship building are inferred through TECH’s recent focus on service innovations that support circularity (T15–T20). In this sense, interactive and diagnostic controls work together in various ways as now discussed.

While TECH aims to strike a balance between diagnostic controls with more dynamic and interactive approaches, for example, by fostering a sales force that is both results-driven and empowered to address broader strategic objectives (T15, T16, T20, T21), CONSTRUCT uses diagnostic controls to stimulate interactive discussions on product, component or service development (R1, R8) with suppliers and customers. For both cases, the interactions that facilitate customer learning in relation to various sustainability issues are evident.

For CONSTRUCT, interactive discussions with customers, via the customer service centres and CRM systems, regard promoting the three core service agreements that support sustainability (i.e. remanufacturing, batteries as service and automation services) (R3, R10). While the first two support circular and other environmental sustainability ambitions by introducing service innovations that entail customer lock-in, the latter supports the social sustainability dimension of health and safety or employee wellbeing throughout the value chain. The interactions between customer-facing staff are key interactive controls and CONSTRUCT has developed training programmes on sustainable service offerings for both sales staff and the field technicians who service both CONSTRUCT and its competitors’ machines in remote locations. These interactions allow for knowledge exchange between operational staff and industrial customers, which induce information for actionable insights. However, one product engineer (R15) questions the extent to which his team receives useful diagnostic information, from the vast array of incoming data, from customers to make product improvements.

Meanwhile, TECH relies on interactive control systems to foster ongoing dialogue between central leadership and regional teams. While balancing a unified vision with localised execution is presented as critical for adapting global sustainability goals to regional market demands (T12, T14, T16, T17, T18), the extent to which this dialogue translates into substantive change remains sometimes unclear. Regular stakeholder engagement, training programmes and even discussions about internal sustainability certifications along the value chain, are intended to demonstrate commitment, yet they risk becoming symbolic if not matched by concrete operational shifts (T19). Similarly, collaborative initiatives and frequent workshops (weekly, bi-weekly, monthly) are framed as vehicles for education and promotion of CE solutions (T22), but they may also serve as managerial rituals, reinforcing discourse rather than overcoming structural barriers. While these interactive systems establish feedback loops and enable the sharing of best practices, the unevenness of regional conditions, as mentioned earlier, raises questions about whether such mechanisms genuinely reconcile local challenges with TECH’s unified vision for circular transformation.

For CONSTRUCT, internal dialogue between the different divisional councils (e.g. service, R&D) regarding sustainable innovations is not yet embedded into the formal control system structure (R1). Rather, councils are tied to specific divisions and sometimes contain individuals at the group level from other divisions that are important from an internal control perspective on “quality control”.

Finally, beyond these interactions linked to diagnostic systems, other interactive systems are positioned as supporting sustainable innovations that reach beyond the focal companies and their industrial networks. Yet, while such initiatives are framed as an impetus for sustainable innovation, their actual impact remains rather difficult to gauge. CONSTRUCT, for example, participates in an inter-organisational network, including multinationals from different industries and research institutes, aimed at sharing and developing knowledge on making traditional linear business models more sustainable (R3). Yet the open-ended nature of such collaborations raises questions about whether they lead to tangible transformation or primarily serve as platforms for signalling engagement. Similarly, alongside its CE programme, TECH highlights partnerships with technical universities and value-chain actors on hydrogen use in industrial processes and energy systems to accelerate fossil-free products (ASR2023). While potentially important, these projects also risk being framed more as showcases of ambition than as mechanisms that confront the deeper operational and cultural changes required for systemic transformation.

Figure 2 summarises the findings in terms of the main beliefs, boundary, diagnostic and interactive systems and relations therein between these controls, for the case contexts. While a more detailed analysis comes in the discussion section, some general trends can be noted, not least the impact of external institutions on beliefs and boundary systems. Firstly, social innovations appear to be incorporated into beliefs controls, but not clearly translated into other boundary, diagnostic or interactive systems in any explicit sense. Secondly, boundary systems entail some limits in the inter-organisational context (e.g. remoteness, connectedness, data overload) for the effective use of diagnostic systems for sustainable innovations. Thirdly, diagnostic systems mainly inform interactive ones for sustainable innovations in the inter-organisational context. This means that a meaningful analysis of data is crucial, as many sustainable innovations are difficult to compare in terms of financial performance and therefore require a mixture of KPIs.

Despite these general trends, we also find differences between the cases in terms of the role of controls for sustainable innovations in the industrial context. For example, TECH notes challenges in customers accepting circular service innovations and intends to use more tangible incentive systems linked to the sales force’s KPIs to improve performance in that respect. In addition, remanufactured product quality seems to be compromised by price even though the quality is not lower; something that was not explicitly mentioned by CONSTRUCT. This may be down to the fact that the machines that CONSTRUCT sells are highly customised and therefore expensive in the first place, whereas TECH’s business model and customers appear to be more price sensitive. In the following discussion we build upon these initial findings in relation to our theoretical framework and research aim.

In our theoretical framework at the front end of the paper, we implied that controls are enacted for:

• the alignment of industrial organisations in their institutional contexts with internal sustainable innovations, which moves beyond symbolic actions to substantive improvements in terms of meeting sustainability goals; and

• fostering duality between institutional compliance (exploitation) and long-term sustainability visions (exploration).

Together, the combination of controls is deemed important for driving forward systemic change that supports sustainable innovations, from both the outside-in and the inside-out. For industrial organisations, we emphasised the importance of enacting control systems with various up/downstream actors for success; that is, success in terms of moving forward with sustainable innovations in different forms. In the following, we discuss how different types of controls shape sustainable innovations in our industrial settings and how they are enacted. The structure covers the types of sustainable innovations in our cases and nuances the discussion on key controls and lever interactions.

The examples of sustainable innovations related to different types of inter-organisational controls found in our cases can be categorised around product innovations (e.g. R&D solutions of industrial products stemming from diagnostic and interactive controls in the inter-organisational network), process innovations (e.g. efficiencies in relation to science-based targets, optimising machine performance from boundary and diagnostic systems) and service innovations (e.g. solutions offered to industrial customers such as electrification or remanufacturing of core products) as part of beliefs and interactive systems, informed by boundary and diagnostic controls. In addition, we find another type of innovation that appears to be somewhat important for, at least, the beliefs systems of industrial organisations, namely, social innovations.

Although definitions of social innovations vary in the wider business literature, they are commonly seen as mission-driven efforts aimed at creating societal impact, whether social, environmental or otherwise, whereby organisations balance the profit-driven and mission-driven practices that generate financial returns, alongside intended societal benefits (e.g. in Komatsu Cipriani et al., 2020). Choi and Majumdar (2015) describe social innovations as those that aim to create social value, such as human and environmental wellbeing, through processes of societal change. Considering the other types of sustainable innovations already described, social innovation is distinct in that it relates to the societal level of wellbeing generated through innovation, rather than merely focusing on company-centric innovations (e.g. as product, process or service) or those related to the immediate industrial network.

For CONSTRUCT, social innovations are embedded at the strategic level within its belief systems, reflecting an explicit focus on the value of “people” or “communities” affected by their (directly) or their customers (indirectly) industrial operations. In contrast, for TECH social innovation emerges more as an implicit undertone supporting social sustainability through longstanding initiatives in healthcare, training and community care for employees and surrounding communities. TECH’s attention to social aspects has recently intensified with the introduction of ESRS S2, which specifically addresses workers across the value chain. Yet, in both cases, social innovation remains primarily embedded within belief systems, though also expressed through boundary systems such as Codes of Conduct that promote non-discriminatory working conditions, diversity and inclusion. Thus, the social innovations often relate directlymainly to social sustainability issues but can also support other financial and environmental performance concerns for both the industrial organisations and their wider value chain.

At the same time, it is necessary to remain critical of how far these corporate claims of social and sustainable innovation genuinely reflect transformative sustainability outcomes beyond the business case. Large industrial companies such as those studied here are deeply embedded in global extractive and resource-intensive value chains. As such, while they promote sustainability and circularity narratives, their operations can arguably never be fully sustainable or circular in the truest sense (e.g. as discussed in Dyllick and Hockerts, 2002). The organisations themselves admit that resource constraints limit the follow-through of certain initiatives, meaning that they are not quite there yet.

The beliefs, boundary and diagnostic systems illustrated in the cases clearly attempt to encourage behavioural alignment in the industrial context. However, compared to earlier research in this area (e.g. Adams et al., 2006, 2016; Akroyd and Maguire, 2011), these systems primarily influence process, service and social innovations rather than merely product-related innovations, the latter of which in this setting is mainly tied to the decision-making function of control. For example, there have been clear attempts by both case companies to push intra-organisational beliefs systems regarding sustainability and/or circularity onto industrial customers. Adherence to external mandates and from internal boundary systems also reflect efforts to standardise behaviours that align with the industrial organisations’ sustainability strategy across the entire value chain. For service innovations, the behavioural alignment aspect comes down to incentive systems or KPIs for sales staff (TECH) or customer service centre managers (CONSTRUCT) internally, and through education programmes between customer-facing staff and industrial customers externally. Albeit the role of interactive controls in achieving behavioural alignment across organisations remains less clear.

While TECH attempts to encourage alignment in its beliefs systems through its recently introduced CE programme that embraces partnership and education initiatives and CONSTRUCT educates its customers as described, the degree of “interaction” (i.e. in terms of learning, experimentation and adaptability with partners in the industrial network) remains limited. This is because the attempts to communicate are mainly coming from the focal industrial organisations to their industrial network partners, rather than through any bi- or multi-lateral communication attempts. Of course, the knowledge networks that both companies participate in stand out as exceptions to this general trend (e.g. the international multi-sector industrial knowledge network that CONSTRUCT participates or the Networking through the Sustainability Partner Programme of TECH and the Recycling Partnerships of both companies). However, when it comes to interactions for sustainable innovations, they largely remain within the case organisations, for instance, within cross-divisional working groups at TECH or R&D initiatives focused on products.

This highlights that the interplay among belief, boundary, diagnostic and interactive systems is not simply additive but shapes how sustainable innovation is (or is not) enabled across organisational and inter-organisational boundaries. Our findings reveal a distinctive dynamic: beliefs and boundaries tend to cascade outward to industrial partners, while diagnostics loop back inward to stimulate organisational learning. Together, these movements create specific patterns of control interdependencies that characterise sustainable innovation in industrial contexts.

The role of controls in fostering duality (e.g. as discussed by Arroyo, 2012; Lövstål and Jontoft, 2017) highlights the importance of their decision-making function, particularly in supporting inter-organisational learning. For both cases, product, process and service innovations are stimulated by diagnostic systems that promote learning, whereby incoming data from customers (either via the telematics linked to machines on site or through formalised complaints procedures) are translated into useful information to make sustainable changes.

This example illustrates the duality between diagnostic and interactive controls (Artto et al., 2011) in the inter-organisational context as necessary for supporting more sustainable innovations in our case companies. However, rather than focusing on diagnostic controls in terms of being “compliance” tools related to the overarching boundary and beliefs systems, the diagnostic systems are, in fact, mainly used to stimulate learning. This nuances prior assumptions about the importance of interactive systems for learning (e.g. Bisbe and Otley, 2004) by emphasising that, without diagnostic systems, there would be little learning, and that the main interaction stems from diagnostic controls to interactive ones, reflecting findings in Beusch et al. (2022). Here, the novelty lies in how diagnostic and interactive systems work together in ways that reconfigure their traditionally assumed roles: diagnostics no longer merely enforce, but actively enable exploration, while interactive systems gain direction from diagnostics, making their interplay central to sustainable innovation (see also Bisbe and Otley, 2004). In this sense, diagnostic controls in this industrial setting serve a dual purpose: ensuring compliance with sustainability standards (exploitation) while also fostering learning and innovation aimed at long-term sustainable change (exploration).

Indeed, while there are degrees of support that inter-organisational diagnostic controls may promote closed innovation strategies (i.e. in terms of product innovations for our case companies) (Biswas and Akroyd, 2022a), our findings reveal that they also promote inter-organisational learning initiatives. This is evidenced in CONSTRUCT whereby customer complaints on machine performance are shared with upstream suppliers to improve tools and attachments, and by TECH through its education initiatives with customers on the performance benefits of its circular services.

This finding suggests that interactive and diagnostic controls are not fully integrated as implied by the notion of duality; although interaction occurs, a true balance between the two is not achieved in the cases, contrary to what Simons (1995) proposes. Instead, the “duality” appears in a more symbolic form, within the diagnostic systems themselves, as when TECH uses mixed diagnostics that combine sustainability metrics and financial KPIs to guide decisions on sustainable innovation.

Building on the duality perspective, our findings refine earlier assumptions about the need to balance exploration and exploitation in sustainable innovation. In both case companies, diagnostic controls play a key role in monitoring, measuring and forecasting specific product and service innovations for industrial customers. However, these controls are not primarily orientated towards short-term financial gains but rather towards enhancing efficiency and ensuring regulatory compliance, both internally and across the value chain.

Importantly, diagnostic controls enacted stem directly from the boundary systems already discussed, which entail “short-termism” in terms of adherence to, for example, industrial regulation or company codes of conduct (i.e. on a daily or annual basis) as behavioural controls, rather than controls that are about minimising costs. At the same time, the enactment of these diagnostic controls for sustainable innovations supports long-term learning and planning, extending to 2030 or 2050, respectively, in areas like sustainability targets and carbon neutrality.

We specifically nuance the exploitation-exploration tension for sustainable innovations in the industrial context from our case findings by highlighting the duality in terms of interaction rather than balance, reformulating our initial theoretical assumptions as follows:

• Diagnostic controls (focused on exploitation) emphasise efficiency in terms of product and process innovations, as well as service offerings through the learnings gained from network interactions (exploration means). They also emphasise short-term performance through adherence to and from boundary systems and the potential for long-term sustainable innovation developments by stimulating learning through exploration.

• Interactive controls (focused on exploration) encourage engagement and learning through diagnostic controls (exploitation means mainly via boundary systems), within our industrial case organisations (closed innovations on products, processes and services) and their wider industrial networks (open innovation on products, processes and services). These controls draw on diagnostic mechanisms to explore new ideas and develop capabilities for long-term success. Illustrative examples include battery recycling partnerships for batteries (CONSTRUCT), customer education initiatives (e.g. CE programme, TECH) and product innovations with suppliers (CONSTRUCT).

Thus, treating diagnostic controls solely as instruments of exploitation and interactive controls as mechanisms of exploration is too simplistic for industrial, inter-organisational contexts, considering our empirical findings. Diagnostic systems also play a crucial role in fostering learning for long-term sustainable performance, performance that extends beyond financial outcomes and underpins the system-level transformations required to achieve sustainability.

CONSTRUCT and TECH both exhibit some degree of change in relation to the institutional aspects that require them to innovate for sustainability in the inter-organisational setting. While TECH can be described as more responsible to begin with, due to its long-term commitment to the social sustainability of its employees (e.g. social security, housing), both companies appear to position themselves in the responsible business case (Schaltegger and Burritt, 2018). Here, sustainable innovations are pursued through product, process and service improvements. Diagnostic systems play a central role here, enabling learning from customer data, complaints and telematics, which in turn feed into interactive controls for developing more sustainable offerings. For example, CONSTRUCT uses customer feedback loops to work with suppliers on tool performance improvements, while TECH introduces KPIs linked to circular services despite challenges in customer acceptance. These actions move beyond symbolic or reputational strategies by embedding sustainability into operational decision-making, reflecting genuine progress towards improved sustainability performance.

Nonetheless, traces of the Reactionary (BC1) and Reputational (BC2) business cases persist. Boundary systems tied to regulation and codes of conduct primarily ensure compliance and short-term adherence to external standards, while beliefs systems communicate sustainability visions outwardly, reinforcing reputational benefits. CONSTRUCT frames its operations around community values and TECH draws on its historical emphasis on healthcare and social initiatives. Yet, these social innovations largely remain embedded in belief systems and are not consistently translated into boundary, diagnostic or interactive mechanisms, creating a risk of sustainability narratives serving more as reputation management than transformative practice.

To clarify how these different innovation types align with the logic of the four business cases, we summarise our empirical insights in Table 3. The table consolidates illustrative examples from both case organisations, identifies the primary control systems associated with each innovation type and assesses their transformative potential along the BC1–BC4 continuum. Together, this enables an analytical link between the observed practices and their underlying capacity to support incremental versus transformative sustainability change.

As Table 3 shows, all innovation types remain primarily situated within BC3, with BC4-level collaboration emerging only in isolated areas such as sustainable service offerings and modular product development. Evidence of collaborative innovation (BC4), for instance, through customer engagement and participation in industrial networks, remains limited. CONSTRUCT’s involvement in international industry platforms and battery recycling partnerships and TECH’s circular economy education initiatives, represent early steps towards co-created solutions by incorporating external perspectives into innovation. Yet these examples show only partial alignment with BC4, as constraints such as resource limitations, fragmented data infrastructures and customer price sensitivities hinder deep collaboration.

Overall, the findings indicate that both companies predominantly operate within the Responsible business case (BC3), exhibiting substantive but bounded sustainability-orientated innovations. While there are glimpses of BC4-style collaborative work, these remain embryonic. Table 3 thus illustrates that sustainable innovation in these industrial settings is progressing, but largely within the constraints of organisational capabilities, institutional pressures and inter-organisational dependencies.

The insights from this study deepen the discussion on how the interplay among belief, boundary, diagnostic and interactive systems is reconfigured for sustainable innovation: instead of a static balance, we observe dynamic, asymmetric interdependencies that support both compliance and transformation. This adds specificity to how controls contribute uniquely to sustainability contexts, thereby strengthening the implications of our findings for both practice and policy debates around system-level change.

The aim of this study was to explore the role of control in fostering sustainable innovations within inter-organisational contexts by asking: What roles do different types of sustainability controls play for sustainable innovations in industrial contexts, and how are these controls enacted?

Returning to this question, we find that for our industrial case companies, beliefs, boundary and diagnostic controls interact in different ways to promote behavioural alignment. This alignment reflects an incremental change rather than a radical transformation of the business model (see Schaltegger and Burritt, 2018), when it comes to sustainable innovations. This change further regards the introduction of social innovations as a novel finding in the beliefs and, to an increasing degree, boundary systems of our case companies and their wider value chains. It also regards process and service innovations such as energy optimisation or sustainable service offerings (e.g. remanufacturing or batteries as services), whereby strategic objectives are realised in practice through the boundary (e.g. industrial standards, internal codes of conduct) and diagnostic systems (e.g. machine telematics, integrated CRM or ERP systems) that work together to coordinate sustainable innovation decisions and behaviours in the inter-organisational context. Meanwhile, the degree of interactive systems remains limited for inciting change, in that the interaction mainly comes from our case companies to their customers as unidirectional, rather than interactive.

Secondly, it appears that diagnostic controls form interactive ones in our cases. This implies duality through the interaction of diagnostic and interactive controls, but not duality in terms of balance as asserted in Simons (1995)LOC given that this interaction is also “one-way” or unidirectional. Similarly, the role of diagnostic controls appears to be one of learning (e.g. through data that translates into actionable insights for product, process and service innovations) in addition to compliance. This indicates the role of diagnostic controls as serving both exploitation (compliance to boundary systems and efficiency innovations) and exploration (stimulating learning for long-term sustainable changes) purposes, extending prior conceptualisations of diagnostic controls.

Together, the belief, boundary, diagnostic and interactive systems facilitate incremental change in our industrial contexts in terms of sustainable innovations, which are product, process, service and – to a limited extent – social based. The beliefs and boundary controls particularly respond to various institutional pressures, that are manifested through diagnostics to clarify acceptable practices, track performance, support organisational learning and foster a shared commitment to sustainable innovation across products, processes and services in both intra- and inter- organisational settings. Yet, the role of interactive controls in the inter-organisational context of the case organisations’ industrial ecosystems and value chains remains less clear. This finding partly resonates with Biswas and Akroyd (2022b) who find support for more compliance-driven or diagnostic systems in the inter-organisational context that are replaced over time by increasingly interactive ones as relationships mature (see also Dekker, 2004; Cäker, 2008). At the same time, the study exposes the limits of what current control practices can deliver in such contexts, as overall sustainability performance remains constrained by institutional, organisational and market realities.

This study contributes empirical, conceptual and theoretical insights into the role of control systems in sustainable innovation.

Empirically and descriptively, we address the limited research focus on sustainable innovations from a control perspective (e.g. Beusch et al., 2022) and extend this into the inter-organisational context, given the increasingly important role of control in business networks for innovation (Johnstone, 2024b). We also identify four types of sustainable innovation (e.g. product, process, service and social innovations) and their connections to the different control functions, such as behavioural alignment and decision-making, in the inter-organisational setting.

Expanding the scope, through comparative case studies or large-scale quantitative research in different contexts, could further clarify how control systems support sustainable innovation across diverse inter-organisational networks. Thus, future research should continue to deploy the LOC framework in inter-organisational contexts (cf. Biswas and Akroyd, 2022a, 2022b), focusing on specific sustainability control problems and/or discrete types of sustainable innovations. Both qualitative case studies and large-scale quantitative research could be particularly well suited to address these inquiries.

Conceptually, we emphasise the growing importance of social innovation from both a control system and sustainability perspective, not only to bridge research boundaries (e.g. Bebbington and Thomson, 2013) but also to address sustainability challenges beyond the firm’s boundaries and contribute to systemic change. For example, with the CSDDD coming into force, social innovations will likely need to expand as the value chain impacts of environmental management and human rights issues are required to be accounted for more explicitly. This requires companies to develop sustainability controls that align due diligence with reporting, embedding sustainability into strategy, governance and financial planning (see also Komatsu Cipriani et al., 2020), which ultimately reduces compliance risks and improves value chain resilience. In our case organisations, these social innovations still mainly reside within beliefs and boundary systems. However, future research should further explore how control levers are designed and used to support various types of social innovations across the value chain.

Theoretically, we contribute to Simons (1995)LOC framework and conceptualisations by nuancing the description of the role of diagnostic and interactive controls for exploitation and exploration, respectively. Diagnostic controls appear particularly important for facilitating learning related to long-term sustainable product, process and service innovations. Given that our study is based on two cases of industrial organisations headquartered in Scandinavia, continued research in diverse contexts is needed to enhance the broader applicability and robustness of these insights.

Some questions to consider are: (how) do diagnostic controls simultaneously exploit and explore for sustainable performance in industrial contexts? How do diagnostic and interactive controls interact to influence the success of product, process and service innovations within industrial organisations? What is the balance between diagnostic and interactive controls in promoting both short-term efficiency and long-term innovation in closed and open innovation models? What role do network interactions play in bridging the gap between exploration (new ideas) and exploitation (efficiency in execution) in an industrial setting? What are the mechanisms through which boundary systems influence innovation outcomes in both closed and open innovation contexts? It could also be interesting to elaborate through a conceptual or literature review paper on how the LOC could be applied to inter-organisational accounting.

In addition, as our findings indicate a limited understanding of the role of interactive controls for sustainable innovation in the inter-organisational context, future research should explore how collaborative learning is achieved in network settings. Some questions to ask are: How do interactive controls facilitate learning and capability development in cross-organisational collaborations, such as those seen in industrial networks? How does customer education (e.g. CE programmes, TECH initiatives) interact with diagnostic and interactive controls to foster both innovation and efficiency in industrial organisations? Such questions can help guide further research into the intersection of control systems and sustainable innovation in industrial settings.

Yet, our findings also call for deeper scrutiny of whether these innovations truly advance broader sustainability, or merely reinforce incremental, efficiency-driven improvements. This critique, in turn, invites a more critical engagement with the transformative potential and inherent limits, of control systems in driving genuine sustainability transitions.

This study carries important practical implications, showing that industrial firms must use a combination of control systems to steer and align behaviour across the value chain towards sustainable innovation. Thus, companies should move beyond a narrow focus on environmental or technological advances and give equal priority to social dimensions, such as employee well-being and community engagement, particularly considering new EU legislation. Embedding these dimensions into control systems through responsibility metrics and performance indicators can help cultivate a broader sustainability mindset throughout organisations and their value chains.

Furthermore, the findings highlight the need for stronger customer education and behavioural alignment through interactive controls to foster sustainable practices across value chains. In practice, this requires more deliberate efforts to build bilateral partnerships and co-create solutions with customers, suppliers and other business partners. The joint development of eco-efficient products and services can, in turn, enhance stakeholder engagement and strengthen long-term commitment to sustainability.

The authors are grateful for the comments received at the 28th Nordic Workshop in Management Control at the University of Gothenburg, 6–7 February 2025 and at the Centre for Empirical Research on Organising and Control at Örebro University, Sweden.