The translation of Lean management: Prospects of a relational approach for successful practice

IMPACT

Rather than focusing on a functionalist implementation of Lean in a hierarchical, top-down manner, this article shows how public sector managers can look at Lean in a new way. They should be luring others into a Lean network by creating conditions in which they define problems and solutions, seek to interest others in these problems, enroll and commit them to possible solutions and mobilize them as one network. Rather than a linear, straightforward implementation, in this way Lean becomes embedded in a network that is capable of entertaining the multiple interpretations and uses of Lean, thus adding to its organizational success and viability.

ABSTRACT

This article reconceptualizes Lean (management) from a concept that is instrumental to managers to a concept that develops through its relations with other (human) actors. Lean is not a guided process of implementation; rather, Lean transforms through network dynamics into forms that are not predefined. Using a case study from a public sector organization, the authors show that translation is not a linear implementation process and does not result in one stable Lean. Instead, it results in many different (sometimes temporal) ‘local’ Leans. The article contributes to theorizing about Lean, as well as to success factors in making Lean work in practice.

KEYWORDS:

• Actor–network theory (ANT)
• Lean
• mediators
• relational perspective
• translation

Introduction

Since the 1980s, Lean management principles have been applied and studied in many sectors of industry, as well as for service delivery (Benders & Santbergen, 2007; Van Dun, 2015; Bateman et al., 2018). Developed in the Japanese car industry (Ohno, 1988; Womack & Jones, 2003), Lean principles reached public sector organizations early this century, where initially there was little doubt about their overall applicability to the public sector (Andersson et al., 2020; Radnor & Boaden, 2008).

However, after two decades of research into Lean’s application in the public sector, the evidence is mixed. In a Public Money & Management theme published in 2018 (Vol. 38, No. 1), it was concluded that much of Lean’s potential in the public sector still remains an unfulfilled promise, with Lean’s fit with many public sector activities being questioned (Bateman et al., 2018). More generally, many organizations have struggled with the implementation of Lean principles (Knol et al., 2018; Langstrand, 2012; Van Loenen & Schouteten, 2016; Van Loenen, 2021), and failure rates are very high (Bhasin & Burcher, 2006). This is problematic, because Lean is intended to help public sector organizations become more efficient and effective in service delivery and policy implementation, and much effort and energy on the part of managers and employees has been drained from organizations by unsuccessful Lean applications. Part of the problem may reside in the functionalist perspective on Lean management. This perspective largely views Lean as a linear and hierarchical process. Adopted as a global concept, at the level of a specific organization Lean is tailor-designed and then implemented, where implementation is conceived as a top-down transportation process that spreads the concept across an organization, from the centre to its peripheries (Radaelli & Sitton-Kent, 2016; Van Oss & Van ‘t Hek, 2009). If the concept does not reach its pre-given destinations, it is considered to be a failure. Such a perspective downplays the dynamics that are involved in Lean.

In this article we propose a different perspective. Rather than looking at the development of Lean in an organization as a process of adoption and implementation in a hierarchical context, we perceive of Lean’s development as a relational enactment through which Lean is adapted and transformed. When the concept of Lean enters an organization, multiple actors associate with it—they interpret and translate it. Through a focus on association and translation in networks of relations, we account not only for how and why Lean is being transformed in the network surrounding it, but also how Lean, as an actor itself, transforms its organizational context (see Van Erp et al., 2019, for a similar perspective in the context of a control system). In such a relational perspective, Lean is not a static object, but an actor that is affected by other actors in the network and, in return, it affects the other actors in its network. Thus, the focus is not on individual decision-makers and change agents and their (linear) actions in the organization, but on networks of interacting human and non-human actors.

More specifically, we adopt an actor–network theory (ANT) perspective to explore the dynamics of a Lean translation process in a large Dutch public sector organization that we call ‘InfraOrg’. InfraOrg is responsible for the management and maintenance of the main Dutch highways and waterways infrastructure. Our case study presents a nuanced picture of Lean’s development as it transforms through associations and interactions in a Dutch public sector organization:

• How do multiple actors associate with Lean?

• Can a relatively stable network around Lean be mobilized, so that it more or less acts as one actor?

In the next section, as an alternative to a functionalist perspective, we present a relational ANT perspective on Lean.

A relational perspective as an alternative to a functionalist perspective

The functionalist perspective on Lean management largely views its implementation as a linear top-down process, where Lean is seen as instrumental to managers who wish to design more efficient production or service processes catering to customer demands. In this perspective, humans play a central role, either as managers who act as (top-down) initiators, designers and change agents (Radaelli & Sitton-Kent, 2016; Van Oss & Van ‘t Hek, 2009), or as receivers of the organizational change initiatives (Holmemo & Ingvaldsen, 2018). Adopted as a global concept, at the level of specific organizations, Lean is tailored to their needs and then implemented in a way that spreads the concept across the organization. It may encounter resistance: wilful attempts within the organization to counter implementation (for example Benders & Van Bijsterveld, 2000; Bresnen et al., 2004; Guler et al., 2002), which have to be overcome so that the concept can successfully reach its destination.

Research that fits with this functionalist perspective has particularly addressed success factors for implementing the concept of Lean (Ingvaldsen & Benders, 2016; Knol et al., 2018; Marodin & Saurin, 2015). It has been complemented with research from an institutional perspective that studies the diffusion of global concepts across organizations (Abrahamson, 1996; Czarniawska & Joerges, 1996; Van Grinsven, 2017).

As an alternative, we adopt a relational perspective, conceptualizing the development of Lean management not as a simple process of transportation, but as a process of translation. The notion of translation is anchored in ANT, which was most prominently espoused by Bruno Latour (for example 1987; 1996; 2005), Michel Callon (for example 1984; 1999; 2007) and John Law (for example 1999; 2004). They were particularly concerned with a sociology of translation, accounting for how ideas become materialized and localized in specific situated practices (Latour, 1987, 1996; Callon, 1984). ANT’s focus is on how Lean is made and remade by many actors in a chain, thus explicitly focusing on transformation rather than simply transportation. The (re)making of Lean is done through its network, through which Lean as an actor is ‘made to act by many others’ (Latour, 2005, p. 46). Importantly, ANT holds that non-humans should be treated as similar to humans (the so-called ‘principle of symmetry’), which at that time came as a shock: by putting humans and non-humans on the same plane it removed the human actor from the centre of the stage. However, the principle of symmetry should not be understood as if non-humans, such as the concept of Lean or information technologies, have intentionality which is similar to humans. Essentially, what ANT scholars propose is that the power to act (agency) is located in the ‘relating’ between humans and non-humans: both are made to act through their association (as a verb) with others. Both are not the sources of an action, but ‘the moving target of a vast array of entities swarming toward it’ (Latour, 2005, p. 46).

Non-humans are not simply tools or instruments, subordinate to their human users. Put differently, a non-human, such as the concept of Lean, is not simply an intermediary that transports without transformation. The concept of Lean is a mediator and, in that sense, it is similar to a human, who in ANT is also conceptualized as a mediator (Latour, 2005). The term ‘intermediary’ fits within linear and causal thinking: it stands in between a cause and an effect and does not have any power of its own. The term ‘mediator’, however, points at the enactment of performative power through an actor–network. Mediators play an important role in ANT research (like ours). Such research follows Latour’s advice to ‘follow the actor’, that is to follow the concept of Lean in its development, tracing the links and ‘networking’ through which it is made to act. Through translation many actors purposefully enrol and mobilize Lean’s network (for example Doorewaard & Van Bijsterveld, 2001). Translation relies on the ability of Lean as an actor to bring others in to accept its conception of problems and solutions (Lowe, 2001). Translation always involves a ‘chain of mediators’. That is, it is not strictly top-down directed by a particular human (for example a top manager), but results from the energy directed to the concept of Lean by every mediator in the chain who does something with it. Every mediator has different interests and may modify, deflect, betray, or appropriate the concept accordingly.

Importantly, Callon (1984, p. 196) has noted that translation is ‘a process, never a completed accomplishment, and it may … fail’. In order to become successfully translated, the actor–network of Lean (Callon, 1999) has to pass through four moments of translation:

• Problematization.

• ’Interessement’.

• Enrolment.

• Mobilization.

This means that in moments of problematization a problem is defined and potential solutions are suggested. In moments of interessement certain mediators (for example change agents, consultants) attempt to interest others in Lean’s network, by recognizing and drawing-in reciprocal interests in the translation. However, interessement does not ensure enrolment as the others do not necessarily commit to the mediators’ worldview and expectations. Enrolment is only achieved when actors understand the problematization surrounding Lean as their own and accept the roles attached to them, when translation forms a ‘single composite goal’ (Latour, 1999, p. 88). Without enrolment, there can be no mobilization of the network. Once Lean’s actor–network is not betrayed, and once it ‘comes to speak as one’, it is really mobilized. Lean may then become a black box: a thing that no longer needs to be considered, a thing whose content has become a matter of indifference.

From an ANT perspective, top management’s role is not simply to diffuse the concept of Lean across the organization in a top-down manner. Rather, top management is considered to be one of the mediators in Lean’s network, who may enable or afford the translation of Lean by bringing ‘inscription devices’ into circulation. An inscription device is any apparatus which can transform a material substance (including a concept) into a figure or diagram (Latour & Woolgar, 1986, p. 51). The inscriptions (texts, figures, graphs, diagrams or tables) that are produced are the forms of Lean that can travel; they are the ‘circulating references’ (Latour, 2005) of Lean. They provide the focus for the discussions about and translation of the concept of Lean. As performative mediators, inscriptions do not simply represent the concept of Lean, as they, on the one hand, reduce the concept into text, figures or diagrams and, on the other hand, amplify the concept by enabling and affording (new) discussion, meaning-making and interpretation.

Methods and data

The data for this study were collected through semi-structured interviews, non-participating observations and additional documents (Gibbert & Ruigrok, 2010; Swanborn, 2013). By applying source triangulation, we validated the data by comparing interview data with observation notes, supplemented with collected feedback from internal Lean training courses, intranet stories, and newsletters. Such a confrontation of data offered the opportunity to discover (in)consistencies in the storylines and to test the data against each other. This approach also provided insights into missing and conflicting data, leading to follow-up questions in the subsequent interviews. We communicated all observations to the participants and provided them with an explanation of the purpose of our research activities (Boeije, 2008).

The interviews were conducted from 2013 to 2016, with 11 directors, six heads of departments, 11 employees, four internal Lean experts, two external Lean experts, three Lean implementation team members and one top consultant. We selected directors and heads of departments, because they were required to take a Lean ‘servant leadership’ course and expected to propagate Lean and to actively facilitate the Lean implementation process. In addition, we interviewed six internal and external Lean experts who facilitated Lean implementation within so-called ‘Lean sessions’. We also interviewed three members of the Lean implementation team (hereafter LIT) who coordinated, directed and reported on the progress of the implementation project to top management. Eleven employees who had to apply the Lean philosophy to their daily work were also interviewed. They were expected to receive support from directors, department heads and internal and external Lean experts.

All interviews were recorded with respondents’ permission and, on average, lasted 57 minutes. This resulted in a total of 53 hours of interview material, which was transcribed and presented to the informants for verification. We asked all respondents to add comments and reflections to the reports, and to do so in a different colour, so that a distinction could be made between primary and secondary information. In this way, we reached agreement about what the informants indicated and intended to say in the interviews.

The observations took place at a number of Lean-related meetings, such as Lean summer schools, Lean meetings at management teams, and Lean introduction days. The aim of these observations was to learn how those involved gave meaning to Lean and how they provided substance to the concept based on that meaning. All transcribed interview and observation reports were analysed with the help of the data analysis programme Atlas.ti.

The development of Lean at InfraOrg

InfraOrg had approximately 9,000 employees geographically dispersed across the country when we started our case study. Around 2004, InfraOrg’s board of directors (hereafter the ‘board’) faced major challenges with regard to its core tasks and raison d'être. Critics found InfraOrg too cumbersome, bureaucratic and opaque. They were of the opinion that other, market-based, organizations might be able to provide InfraOrg’s services more effectively and efficiently. Politicians demanded change; the Dutch government required an increase in effectiveness and efficiency, while at the same time imposing budget cuts. InfraOrg’s board responded by adapting the behaviours for 2005 and after in the required directions. Simultaneously, a manager in one of the departments who experienced internal challenges regarding employee commitment and lagging service quality towards customers got inspired by the Lean philosophy. An external business consultant suggested Lean management as a solution to cope with these pressures. In 2008, approved by the board, the departmental manager started a Lean experiment.

In following this experiment, our translation perspective shows that, through its relations with other human and non-human actors, Lean is enacted upon and incites to act performatively in different ways, but ultimately fails ‘to act as one’.

We present our narrative by demarcating four chronological phases in the emergence of Lean as an actor–network. In each phase, various moments of translation, as discussed previously, can be distinguished.

Phase 1: A local experiment (2009–2012)

In a first moment of translation (problematization), the departmental manager defined the problems of a lack of employee commitment, a lack of customer service and increasing external pressures. In consultation with the external business consultant, this manager proposed Lean as a solution. In order to get a better idea of Lean management in practice, the departmental manager started reading Lean literature and visited Scania (a truck manufacturer), a company that had already successfully implemented Lean. The departmental manager made a first translation of the concept of Lean for department staff. Thus, the manager and the business consultant connected Lean (a solution) to the department’s problems. The manager explained:

I am the initiator. I mainly focused on the idea and idea development by listening to others. In my view, the business consultant had a brilliant idea to initiate an organizational change process. But I made others actually do it. I was only the manager and keeper of the idea … It fits my interpretation of the Lean philosophy to engage employees of all levels in the change process. That is what Lean aims for and what I was told. By engaging employees in such change projects, they feel recognized and respected, which helps to realize the change.

In this way, problematization and interessement went hand in hand. As initiating and mediating actors, the departmental manager and the business consultant succeeded in interesting three other departmental employees in associating with Lean’s network. The three potential allies were encouraged to read books on Lean by Womack and Jones (2003) and Mann (2005); the books acted as inscriptions of the Lean management philosophy that afforded the potential allies to translate Lean to their own environment. The three others engaged in conversations and discussions about the departmental challenges, and came to see Lean as a possible solution for the challenges the department was facing. Furthermore, they organized company visits to Scania and Achmea—both private companies with successful Lean programmes.

After a positive evaluation, in order to learn more about Lean and its potential to solve the departmental problems, the three allies participated in a Lean education programme at the Dutch Lean Management Institute. Through this programme, they also learned about convincing others to associate with Lean. The three new actors, as Lean ambassadors, started to further build, strengthen and stabilize the network. The initiators encouraged the new allies to spread the Lean discourse in the department. The five human actors in the Lean network now became the primum movens (or prime movers) that actively looked for new associations with Lean’s network.

In order to further enrol Lean’s network in InfraOrg at large, the five primum movens aimed to ‘pull’ the chief financial officer (CFO) of InfraOrg in Lean’s network with their first translation of Lean for the organization as a whole. They organized a presentation and a visit to Scania. This resulted in the CFO’s approval of a local experiment with Lean. So far, enrolment had formed a single composite goal. Further communicating this goal to other human actors in the department, for example by the use of presentations and an introductory, participative ‘Lean-game’, helped them to begin to understand the problems surrounding Lean (particularly waste in the departmental service processes and long lead times in the licensing processes) as their own and to accept the roles attached to them. Inscriptions that reflected successful examples of Lean in other companies were brought into circulation and became important non-human actors in the enrolment of Lean. Actors in the department came to translate Lean and to negotiate their translations with others. By the end of 2010, the Lean network had 15 human actors. As a non-human actor, Lean allowed human actors to connect and to enact Lean in their day-to-day activities. After three years of experimenting, the Lean actor–network managed to reduce the lead time of an important licensing process from nine weeks to one week. Moreover, the level of employee commitment increased and employees actively participated in the change processes.

Phase 2: Countrywide roll-out (2012–2013)

Having established a motivated Lean actor–network in one department, the Lean team invited InfraOrg’s director general (DG) and the minister of infrastructure and water management to a meeting in which the results of the local experiment were presented. The minister immediately saw Lean’s potential to meet efficiency and effectiveness targets across the organization with a smaller budget. It was helpful that the minister already had some experience with Lean in previous posts.

In 2011 a new business plan (‘business plan 2015’) was created in which Lean was prominently positioned as a way of reducing costs:

We will achieve this along two lines: each employee improves his/her work processes with the use of Lean and we renew our work processes and organizational structure.

The cost reduction target in business plan 2015 was 18% of total costs and the number of staff had to be reduced by 13% by 2015. Although cost reduction was never an intended aim in the local experiment, it now became an important actor in the network.

Implementation rather than translation

The board took a functionalist rather than relational perspective; they talked about implementation rather than translation. The LIT was set up with 10 members, two from the department that experimentally enrolled and mobilized Lean. Taking a top-down implementation approach, LIT developed a national Lean introduction programme and Lean educational programmes for managers (‘servant leadership’ courses) and for other employees (courses to become Lean experts or Lean practitioners). These programmes explained what was expected from employees. The servant leadership programme focused on the organization’s leaders explaining the leadership required to coach and facilitate employees to enact Lean in their daily work processes.

Led by the LIT, the first 37 employees were trained as Lean experts and Lean practitioners. Their motivations to engage in the network varied from personal (‘I was curious about what Lean could mean to me and the organization’) to more professional (‘I was invited to participate in the course because I was supposed to co-ordinate the implementation of Lean in my department’). In the eyes of the LIT, these Lean experts and practitioners were to become important Lean ambassadors and spokespersons to mobilize other actors. Their assigned task was to facilitate the implementation of Lean throughout the entire organization by supporting directors and management teams to implement Lean in their departments. With silver briefcases and coloured markers, they assisted with developing brown-paper sessions and visual management boards. In these brown-paper sessions, employees were invited to engage in detecting waste in their work processes. Visual management whiteboards served to present an overview of the challenges and problems (‘waste’) and action plans for dealing with them.

Although it was not an explicitly stated goal to establish a network surrounding Lean and, although the LIT did not see its role as one of a mediator but rather as an intermediary, moments of interessement emerged. The LIT convinced other actors to associate with the Lean network because they experienced Lean as a solution for the problems they perceived. They organized events such as ‘Lean experience days’, ‘Lean workgroup meetings’, and a ‘Lean exhibition’. Gradually, there was a change in perspective and approach: from a rather hierarchical implementation trajectory toward the emergence of a trajectory of network building. So, the type of interplay between Lean (and its growing number of inscriptions) and the human actors in the network was gradually changing.

Enrolment and (de)mobilization

As the LIT, in a top-down fashion, focused on mobilizing actors to implement and use Lean in the organization, a formal reorganization began. The board, inspired by the Lean philosophy of efficient working processes and a flat organizational structure, repositioned many employees into different jobs and/or different departments. This damaged the Lean network, because many actors in the network lost track of each other. Even the network’s initiating actor, the manager of the department that originally experimented with Lean, was moved to another department; and so many human actors in that department lost (part of) their interest in Lean. Other actors were repositioned into roles in which their Lean expertise was no longer important. As a result, the Lean expertise and knowledge got dispersed and was barely used. The need to associate with Lean faded and the actors lost their interest in building and defending the network. The power of the connections and relations decreased and, as a consequence, the network lost its strength.

Moreover, during the same period, some shortcomings of the functionalist hierarchical approach towards the development of Lean became visible. First, it turned out that many managers participated in the servant leadership programme only because they were required to participate, not because they saw the benefits of it. Many felt that participating in the programme was the best way to survive the cost-cutting reorganization. Hence, their participation was rather opportunistic or symbolic; not sincere. Second, as many other initiatives also were mandatory, some employees complained about the LIT’s top-down and instrumental approach. They found it difficult to translate Lean to their work environment. One trained Lean expert explained:

It is all rather unclear, because several initiatives and developments tumble over each other … As Lean expert, I do what I am asked to do, even when I disagree with the vision of the LIT and how they interpret and execute Lean. Moreover, they view Lean as a goal in itself. In my view, this can never be the real intention. As a result, the actual motivation to enact Lean is deformed.

This resulted in confusion about what Lean was and what it could contribute. Many different interpretations of Lean started to circulate in the organization, complicating its development. Employees found the concept too abstract and difficult to translate. Managers failed to explain Lean to their employees in a clear way. As one manager put it:

We are not even aware of Lean’s basic principles, which [makes it difficult for us to] explain it in a clear fashion to our employees … We do not go beyond attempting to make visual management whiteboards or unfinished brown-paper sessions.

As the network, nevertheless, grew through new associations, new moments of interessement appeared. Humans at all levels of the organization were searching for legitimization and they thought that Lean could be helpful in this respect.

Yet, the success of Lean did not really spread beyond the initial pilot experiment. The relevance of this story faded over time, leaving managers with the difficult task to convince others of the significance of Lean without having clear answers to the questions that were being asked. Even the LIT and Lean experts had difficulties responding to these questions. A Lean expert explained:

I do not really know the story. I know that we need to become more efficient, that society no longer wants to spend a lot of money for our services … Such a story is not appealing for others to identify with Lean.

A final translation dynamic during this period was related to the connection of Lean to reorganization. In business plan 2015, Lean was mentioned in the same paragraph as the goal of cutting costs and jobs. Consequently, many people involved in the network interpreted Lean as a cost-cutting instrument, giving it a negative connotation. As a result, employees did not feel attracted to the network and refused to actively participate. Moreover, newly-hired managers, with prior Lean experience from other organizations, were not sure why they had to work with Lean in this organization. They did not see the benefit of Lean for their new tasks of rebuilding their departments. As a result, confusion about Lean grew and the network stagnated.

Phase 3: Resuscitating Lean (2014)

Despite the problems, the LIT was determined to pursue its top-down Lean implementation trajectory to efficiency enhancement and cost reduction targets. With a new top consultant, experienced in implementing Lean in a public sector organization, the LIT embarked on reviving the Lean implementation process. They included Lean in the annual management contracts between the board and departmental managers. These contracts contained targets regarding participation in Lean training and use of Lean tools, such as visual management whiteboards and A3 problem analyses. Brown-paper sessions in 2012 generated 350 improvement suggestions. Top management and the LIT wanted to see these suggestions materialized and used their power to motivate and convince other actors to realize this. They forced moments of interessement, which resulted in the appearance of a growing network. However, the association of many managers with the network was not related to a real interest in Lean. Rather, their formal accountability to top management forced them to engage with Lean. The same applied to managers at lower levels in the organization; they participated because they felt they had to. One manager said:

In my opinion, all these managers have to show to care for Lean, but deep down they are not really interested … If management is not interested and only shows symbolic leadership, and the department is only assessed regarding production targets, it is not going to work; it is a waste of energy.

The enactment of Lean remained an issue. Immediately after their training, participants in training programmes were enthusiastic. But, after two weeks of unsuccessful experimenting in their own working environment, their motivation faded. Particularly, the support from Lean experts and practitioners was perceived as insufficient. The training of Lean experts and practitioners was focused on the instrumental application of Lean tools, such as A3 or brown-paper sessions. They were less well trained in actually solving practical problems.

The mandatory use of Lean tools resulted in a large diversity of visual management whiteboards throughout the organization. Employees engaged, at least to some extent, in working with these boards as these required active association in order to result in useful outcomes. For example, green and red smileys were used to indicate the progress of production targets. However, these mandatory engagements with Lean also, and to a large extent, resulted in symbolic behaviour. Managers participated in the servant leadership training to check their to-do list. And an observation in 2015 revealed that the visual management whiteboards were only pulled out of their closets when senior management visited the site. Some of these boards had not been updated since May 2014, and management had not noticed this.

In this phase, many actors acted upon Lean and associated, but also dissociated with the network. There was a large diversity in the ways in which Lean was acted upon, varying from intrinsically motivated actors, determined to use Lean for solving actual problems, to actors who associated mainly out of opportunistic and symbolic motivations. Although the network grew, it was diverse and unstable. And as soon as the term Lean disappeared from the next business plan (‘Roadmap2020’), the actor Lean dropped out and the network ceased to exist.

Phase 4: The rise of multiple smaller networks (2015–2016)

Despite the original Lean network falling apart, in 2015–2016 a variety of new and smaller Lean networks started to appear. As former Lean experts, practitioners and managers were dispersed throughout the entire organization, and they began to use their Lean experience and knowledge as ‘regular’ employees in order to improve the production processes in the new (project) teams they were assigned to. For example, visual management whiteboards, rather than bulky reports, were used in meetings to discuss day-to-day progress. In this way, new success stories (for example reducing production time from eight months to one month) emerged at local levels and then spread across the organization. Former Lean experts were convinced of Lean as a solution to certain organizational problems and kept on using and translating their expertise to the local context, resulting in the emergence of new and smaller Lean networks. These networks, however, were not geared towards the overarching organizational goals; instead, they solved local issues in InfraOrg. As a concept, Lean even disappeared from InfraOrg’s intranet in 2017.

Conclusions and discussion

In this article, we have reconceptualized Lean from an instrument to an actor–network. A conception of Lean as an actor–network focuses on relations rather than (human and non-human) entities. Lean is not a tool that is purposefully adopted, implemented and then used by human actors—it is a ‘moving target of a vast array of entities swarming toward it’ (Latour, 2005, p. 46). Our case study demonstrates that, through its relationships, Lean will develop into forms that are not pre-defined (see also Van Erp et al., 2019). Translation did not result in one stable Lean, but in many different temporal ‘local’ Leans.

ANT has previously been used as a lens for exploring the dynamics of the concept of Lean in public organizations (Papadopoulos & Merali, 2008; Papadopoulos et al., 2011; Andersson et al., 2020). These earlier studies view the relational perspective that ANT offers as subordinate and instrumental to a functionalist perspective. For instance, Papadopoulos et al. (2011, p. 167) found that ANT is useful for ‘identifying objects and actions that are effective in engaging individuals in networks which enable transition to a Lean process’. Abdallah et al. (2020) drew upon ANT to describe relevant factors for a successful implementation of Lean. Thus, both studies use insights from ANT to produce functionalist knowledge that may enhance successful implementation by managers.

In this article, we have developed a different perspective by relocating the research focus away from implementation in a hierarchical context to the various emerging and often unpredictable associations of the concept of Lean with other human and non-human actors. Rather than viewing the relational perspective as subordinate to a functionalist perspective, we present it as a distinct perspective that produces enactive knowledge; knowledge about how Lean is enacted in the organization. Such a distinct relational perspective has the potential to study the full dynamics of Lean. It opens up possibilities to study Lean as a journey without a pre-given destination. It relaxes the binary distinction between successful and unsuccessful, as it recognizes and makes visible unexpected adaptations and transformations, both beneficial and harmful to the organization. Thus, from our relational perspective, perhaps the question whether Lean is a panacea or paradox for public service organizations (Bateman et al., 2018) is not the first question to ask, as it downplays the dynamics that are involved. These dynamics can go in every direction.

It is important to distinguish the research perspective on the development of Lean from the managerial perspective in the organization. In InfraOrg, after an initial experiment in one department, top management adopted a functionalist and hierarchical approach to the development of Lean. However, our relational research perspective reveals that the centrally-orchestrated implementation process of Lean did not reach its intended destination, as the process did not result in a stable tool that was functionally useful to managers and employees. Although there was an appearance of participation in the process of implementation, in many cases this was a pseudo-participation: managers were going through the motions, but were not really committed to Lean.

Our relational perspective can not only be a valuable perspective for researchers, but also for public sector managers. We recommend that managers take a relational perspective on the development of Lean in their organizations. Rather than focusing on implementation in a hierarchical, top-down manner, they should focus on processes of translation in Lean’s actor–network. Being mediators themselves in Lean’s actor–network, they can bring others into the network by creating conditions in which moments of translation emerge. That is, they should create conditions for problematization, interessement, enrolment and mobilization, following which human actors willingly attach to Lean and translate it to their own environments. Moreover, in the interests of the organization, they should be open for Lean to become multiple: more than one, but less than many (see Mol, 2002; Van Erp et al., 2019). This, we feel, might avoid too much ‘symbolism’.

In a general sense, a relational perspective on the development of Lean in the organization points to a need for top managers to have ‘an awareness requiring a de-centralization of the selves that presume change agency in order to appreciate the others—human and not-human—that shape as much as these managers [think they] do’ (Lancione & Clegg, 2013, p. 23). By interacting with multiple other actors, managers may enhance processes of learning and mediate in the development of multiple Leans that are somehow still related and overlapping. They should recognize that a successful centrally-orchestrated implementation of Lean is an illusion.

Finally, although not the focus of our study, an ANT perspective on Lean has implications for the way we look at spreading the concept of Lean across organizations. Rather than seeing such spreading as a purposive process of diffusion across organizations, from a relational perspective we can see a process of travelling. Traveling and translation go hand in hand; they move away from central orchestration and locate agency in the relations between Lean and other actors.