“Being there”: detailed ethnography, detective work and a little imagination Open Access

Like many other technologies mobilized for informing and communicating, visualization boards are increasingly digitalized in organizations (Haartman et al., 2023). The whiteboards used for sharing information, planning work, visualizing schedules and so on are in many workplaces being replaced by touchscreen LCD (liquid crystal display) monitors. The obvious benefits of these are flexibility and multi-view practices where, for example, users can switch between applications and worksheets during meetings. Digital visual management, in being part of the Lean production philosophy, aims at increasing self-management capacities and transparency across and within work groups (Haartman et al., 2023; Pedo et al., 2023), with benefits of the digital being for instance automatic updating of data (Meissner et al., 2018), non-dependence on a particular place (Hultin and Mähring, 2014) and digital archiving (Nilsson et al., 2019).

During our ongoing ethnographic study at a local plant of a large industrial manufacturing firm we, like van den Ende et al. (2015), have been challenged by the ways in which the material is intrinsic to everyday practices, and how material and social, in entanglement, performatively construct reality (Latour, 2005; Orlikowski, 2007). Dealing with the interweaving of the material and the social is often framed as providing analytical challenges, but it also comes with other methodological challenges. With the Lean production-inspired whiteboards and digital monitors becoming increasingly central artefacts in the daily management in the plant where we performed our study, we were intrigued by the question of how to properly account for the transformational role of information technologies (digital and non-digital) and their effects on work and organizing. This methodological paper is not concerned with whether or not material objects play a critical role in relations between human actors at work, this is well established in numerous fields of research (see for instance Hultin and Mähring, 2014; Sergi, 2016; Hugosson et al., 2019; Lindblad, 2019; Bell and Vachhani, 2020; Fox and Alldred, 2022; Faludi, 2023). Instead, our intention here is to address this issue from the perspective of work and technology research methods. Taking our ontological cues from critical, actor-network and socio-material modes of analysis, we discuss a methodology for researching (digital) information technologies from the perspective of technologies at work.

We know that “being there” is the very foundation for ethnographic research (see for instance Geertz, 1988; Burawoy, 1991) and that following objects or non-human actors in ethnographic research contributes to our understanding of work and technology at work (see for instance Mol, 2002; Bruni, 2005; Hultin, 2019). We do, however, know less about how such research is done in practice; what the actual work on the ground looks like, when we are interested in producing knowledge on technologies and their effects on organizing.

To attend to this, we depart from a view on technologies as entering into, and becoming part of, the organization and regulation of the complex social relations required to make the work of organizing in industry possible. We build on the argument that the material tools for the daily management of work groups – such as visualization boards – need to be understood and researched as actors and influencers within emergent socio-material and network relations (Latour, 1992). With our example from our still on-going ethnographic study in a manufacturing facility, we show that detailed ethnography, detective work and even a little imagination, all have important roles to play.

The paper begins with a selective review of Bruno Latour’s thinking around a sociology of translation and an analysis of how this framework might be used to frame and conceptualize “tools” (in our case boards), as material objects in socio-material relations. We then draw on the socio-material work of Paul Leonardi, as it purposefully frames the detailed detective work that is required and which we then illustrate with an empirical example from our ethnographic study at a manufacturing site. In the final section, we discuss how this approach can be used as a template for how we might better approach researching (digital) information technologies at work in a structured and methodical way. In addition, such an approach also invites to an extended understanding of the “how” of the work on the ground and the work of accounting for what goes on there.

For Latour, talking about the social world without also talking about the material of which it is also composed is non-sensical. Sociology, he observes, in focusing its attention on purely “social interaction” cannot explain how the social is constructed and maintained (Latour, 1992, 2005). As he argues:

Latour (1993) uses the term hybridity to describe more balanced and complete accounts of reality; accounts that mix human and non-human entities. Non-humans, argues Latour (1992, 2005), bring essential force to social relations. This is precisely how we should understand visual tools in a production setting – as the material interventions in human relations that help humans order those relations in particular ways. The visual information board stands-in for the human manager in the tasks of informing, updating, enthusing, motivating, moralizing and directing. Latour (1992) illustrates the point with the example of the “moral injunction” to close doors behind oneself. The moral injunction to close doors, Latour notes, is weak, it carries no real force. A doorman might solve the problem, he suggests, but the doorman may also prove unreliable; fall asleep or wander off. Consequently, the moral injunction to close doors in public buildings has typically been invested in a non-human - in this case a technology, a mechanical door closer or “groom”. The link between the moral injunction to act in a certain way (close the door after you) is thus given material force – acting morally is pre-scribed into a mechanism. Early versions, he notes, lacked spring damping, resulting in them slamming shut in the faces of the unfamiliar user. Later versions, closed doors firmly but slowly – mimicking, Latour observes, the firm but polite morality of a good butler. The great advantage of non-humans is the efficiency, reliability and unwavering morality with which they perform their tasks since non-humans “don’t sleep” and they form associations that “don’t break down” (Latour, 2005). Whereas Latour’s door closer performs a physical action in bringing steel to social relations, the visual board “brings” the manager permanently to the shop floor.

In this paper, we mobilize Latour’s account of the automatic door-closer as an instructive methodological example. Latour pieces together the story through direct observations over time, some deductive reasoning but also a little imagination. Living amongst mechanical door-closers means that he is deeply familiar with the world they inhabit and with the technology itself. He was able to experience, first hand, how the cold air from the outside of public buildings enters and circulates, freezing everyone, including Latour himself, when the door is left open. He also imagines the role of the human doorman, thinking through the implications to make his point, based on what he knows about humans – they can be lazy, ill and they need days off. Latour also explores the detail of the technology itself – he learns how the return-damping piston solves the “rudeness” of the un-aided spring. In short, to do rigorous research into non-human actors, research that escapes the limited ambition of simply highlighting the role that actor-networks play in mobilizing or substituting human action, we need methodologies that will allow us to unpack this level of detail. We need, in other words, to be in a position where we can immerse ourselves in the mundane every day of the contexts inhabited by these machines. To come closer to how we, in practice, can accomplish this, we turn to the work of Paul Leonardi and his 2017 attempt to focus research questions directly on the complex and nuanced relationships between humans and technologies as a means of achieving this.

Leonardi, while not an actor-network theorists per se, offers a pragmatic approach to unpacking the relations between humans and non-humans. Usefully, whereas actor-network theorists tend to view material objects as “inscribed” with human intent (an approach inspired by semiotics (Latour, 2005)), Leonardi draws instead on a “realist” ontology that calls for paying attention to the actual material nature of objects. A chair can be used to break windows, for example, but only if it is made of substantial and weighty enough materials (Leonardi, 2017). Without sufficient weight it does not offer that “affordance” and that affordance only comes into being if the chair is actually used in that way (Leonardi, 2011, 2017). Thus, whether and how a technology matters in a particular context, rests on the unfolding adaptive interactions between the technology (its material properties and related co-produced affordances), the actual practices surrounding its use, and the broader network of relations of which they are a part and through which the results of these interactions; changed actors, new practices, new roles. These effects can reverberate through an adopting organization, creating unanticipated change in power, roles and practices well beyond the original adopting department (Barley, 1990). It follows that the researcher must shift their attention back and forth between the (real) material and the emerging and distributing affordances of the material. Leonardi (2017) proposes a valuable three-step methodology for doing precisely that.

In the first step, Leonardi (2017) suggests three main questions we should ask when accounting for materials. First, we should ask out of what materials is the object made. Second, we should attend to how those materials are arranged into particular features. Third, and last, we should consider what those features do or do not do.

The argument is that the materiality of artefacts creates and restricts opportunities for action. This is the intention of objects like visualization boards, taking a Latourian perspective – to interact with human action in such a way as to alter it in particular ways (encouraging, directing, nudging and so on). In other words, material objects create different social relations, different modes of organizing and different identities and subject positions for the user. Though, at the same time, this depends on how their material properties are exploited – particular ways of organizing do not stem automatically from particular materiality.

As a second step in his methodology, Leonardi (2017) suggests we account for materiality. He proposes asking first what social or cultural institutions have shaped a person’s goals? Second, how those goals mediate the interpretations people make about what a technological artefact’s material can or cannot do? Third, and last, he proposes we ask why people did or did not perceive alternative affordances or constraints based on the actual possibilities offered by materiality itself (uncovered through answering the questions in step 1 above).

This step takes account of how people perceive the artefacts or tools with which they must deal. Technologies have “affordances” (Leonardi, 2011) which are activated only through actual use - how they are used. If particular features are ignored, those affordances do not materialize, they do not come into being. When organizational members interact with tools, such as visualization boards, relations become socio-material; both the user and the tool are defined relationally by the relations that emerge between them. Elsewhere in socio-material study, this is referred to as the “mutual constitution” of the social and the material (e.g. Orlikowski, 2007).

In practice, tools and technologies will have stability of use across different contexts. But this “truth” should come out of the research; it is a question to be answered, it should not be taken as given or self-evident. Culture, including local norms and practices, can also of course shape what a group of individuals “wants to do” with the technology. The affordances of tools or technologies such as visualization boards thus also interact with local norms and practices through individuals whose practices are conditioned by those norms. These norms then stabilize, perhaps not permanently, as a particular way of using a tool.

By answering the second-step questions, Leonardi (2017) argues, the researcher understands how a material object (like a visualization board) becomes a social object. The researcher will be able, by attending to step 2, to provide properly researched and robust accounts of how local norms and practices prompt some features, and not others, to be called forward into use.

The third step in the analysis is to account for how certain technologies and certain features come to matter in organizational terms – how and why they do, or do not, become part of broader organizing processes. In this step, Leonardi (2017) encourages us to first consider in what ways existing patterns of organizing are dependent upon the materiality of certain technological artefacts. Second, we may ask why certain forms of organizing produce a social context in which a technological artefact’s materiality can materialize in ongoing streams of action and interaction. Finally, we ask how the affordances produced by a new technology sustained, altered, or transformed the way that people act and interact with one another.

Leonardi (2017) describes how computer support technicians, when compiling reports, found that the new systems they were using gave them access to previous reports by junior colleagues. They came to rely on these for guidance far more than they did on guidance from senior colleagues. By no longer relying on input from their senior colleagues, but input from more junior colleagues, they had in effect upended the processes by which the organization produced reports. Critically, these new affordances took time to emerge. It follows that a snapshot of how artefacts are used in practice gives only a limited picture and this implies the need for more longitudinal studies of the use of different types of tools in organizing.

In the following section, we tell the story of our own experiences in the field of accounting for material (step 1), for materiality (step 2), and understanding which materials came to matter and why (step 3). We start by introducing the setting, after which we share the story of the “life and death” (c.f. Law and Callon, 1992) of a strategy board.

The Company is a global company with manufacturing facilities across the globe. Our study, which is part of a larger research project stretching over several manufacturing sites across the European Union and the UK, is conducted at The Company’s production plan in Middletown, Sweden. In this particular study, the first author over a period of ten months spent one morning per week at The Company (usually three to five hours), observing the daily work in the manufacturing cells as well as doing interviews with workers and managers. To-date we have 150 h of observations that has provided 120,000 words of ethnographic observation notes from this particular case. In addition, we have 16 transcribed interviews (130,000 words), recordings and notes from meetings and workshops we participated in, and multiple documents and photos gathered throughout the study.

The Company is a global manufacturer, producing advanced, bespoke, high-quality products aimed at enhancing the productivity of customers from a wide range of processing industries. In the facility in Middletown, numerous versions of a highly advanced product are manufactured in seven manufacturing cells, each responsible for producing one or more components contributing to the final product, which is assembled in the assembly cell located within the factory. The factory produces around 300 full units per year, and additionally manufactures spare parts and combinations of replacement kits. There are around 200 blue-collar workers employed at the facility and some 50 white-collar. In Middletown, the facility has been located in the same location for over 100 years, with parts of the original building still intact and in use.

Visualization boards are placed across the facility, often as many as five to six per cell, and in addition there are numerous common factory boards and monitors spread over the plant. Each cell has a particular location within their cell area where all the boards (both analogue and digital) are organized into a stand-up meeting space. Some boards have more or less the same set-up and appearance across the cells, for instance the Lean methodology inspired analogue Status boards look the same across all of the company’s facilities worldwide. These boards have standardized headings in English and are accompanied by green, yellow and red magnets that can be moved across the boards. A different type of board, the Situation board, visualizes ongoing orders in each cell. On these boards, the rows represent unique orders, and the columns represent the machines in the cell. Small paper notes show the details of the orders and magnets indicate where in the flow the order currently is. The digital boards provide digitalized versions of the analogue boards and can also be used to visualize the site’s current order status and other KPI’s to the operators working in the cells. According to a couple of the team managers we talk to, the plan is to replace all analogue whiteboards with digital boards. During our one-year study we noticed, however, a static hybrid set-up in all cells where digital and analogue exist in parallel with rather few changes made during the year.

The visualization boards are central in the daily practices on the site. Staff gather around these boards once every shift to get updates and to share ongoing issues, which are noted on the boards and then tracked daily until solved. When talking to team managers they often mentioned the importance of these boards, and there are numerous projects mentioned during our time there that focus on improving them, the information visualized on them and the daily meeting practice itself. Visualization boards in this plant are, at least in some of the narratives we encounter, central and crucial for the organizing of work. Other narratives provide a more cynical view on the boards, pointing to how organizing is in reality done elsewhere and that the boards in fact do not matter at all.

In this paper, we draw on the parts of our research material that relate to what in this plant is referred to as The Strategy Board. The Strategy Board is not connected to any particular manufacturing cell, but rather to the overall site. It is placed close to the assembly cell, right next to one of the main entrances into the manufacturing facility. In the next section, the first author describes how the strategy board was brought to life and how she experienced how it came to matter, and the instances that seem to have been crucial for understanding the eventual demise of the board.

The strategy board does not catch my attention at all during my first few visits. There is so much going on in this plant: CNC machines run in three-shifts cutting and shaping huge metal blocks, welding operations, fast-moving robots moving the smaller metal parts between automatic welding stations, and not least the people, and the forklifts and trucks racing through the factory that you need to keep an eye out for when moving through the facility. Seeing and acknowledging a simple whiteboard on a wall just doesn’t happen during my first visits – my senses guide me elsewhere in this noisy, lubricant-smelling, varied facility. Also, the board is not highlighted by any of the employees or managers I spend time with the first few times I’m there, and it does not gather groups of people around it at any time I pass. It just hangs there – blended into the wall.

The first time my attention is drawn to the strategy board is the first time I am invited to attend a weekly stand-up Goal fulfilment meeting. This is a meeting that normally takes place on Tuesdays, when I’m not there, but this particular week in March it was moved to the Wednesday and I was invited by Anette, one of the managers, to take part. The Goal fulfilment meeting is a recurring meeting in which most of the top-level managers at the site take part. To me it appears to be a meeting that is expected to be prioritized because, although literally running through the factory is how moving in this site generally is done by managers, being on time to this meeting seems particularly crucial:

The group, a dozen or so managers at the site, transfer to a space in the middle of the factory called the War Room, which is a semi-room with three solid walls and one end open towards the factory. This is where the daily factory meeting is held. The War Room in part is framed by the original white brick walls of the building, which are remnants from when The Company initially opened for business over 100 years ago. These walls are now covered by analogue whiteboards and the newer digital boards. After having “run-walked” from the War Room, the management team from the daily factory meeting plus some newcomers, now instead gather around the Strategy board located next to one of the main entrances into the factory building. Whereas the walls and the boards in the War Room construct a room-like structure with three walls, this board is placed on a wall at one of the end of a long aisle running through the factory. The board holds a number of A3-and A4 printouts attached onto the board with magnets. They reveal current status for the selected goals as columns, with space in each column for writing notes or “to-dos” under each print-out. There are also magnets in red, yellow, and green that can be placed next to the notes made on the current status for a particular goal. During the meeting, goals such as delivery precision and productivity are discussed, and the discussion sometimes gets intense:

Later that day, Anette shares her vision for the space of the strategy board:

A month and a half pass, and I spend time on site with different members of the management team and with the operators in the cell that processes castings. I don’t reflect particularly on the simple whiteboard making up The Strategy Board, although I pass it on my way into the manufacturing facility every Wednesday. It simply isn’t part of the study that I think I’m doing, and it does not really catch my attention when I enter the building. One Wednesday morning I, however, for some reason note that the strategy board has changed:

Noting the material changes on the physical board (the new paper printouts, the new figures and magnets), makes me reflect upon the effects they might have in terms of how the board now is perceived - which is crowded and messy. I remember thinking that this is an example of “you can’t see the forest for the trees”, and how striving to capture more, or even “everything” as expressed by Anette, sometimes leads to the opposite. I also remember reflecting upon the power relations in the management group, and how the plain whiteboard seems to carry and visualize such relations rather clearly.

The study continues into the autumn, and I spend my time in the automatic welding cell and the large components manufacturing cell. In retrospect, I regret not engaging with the strategy board even more, particularly as it would have been easy to stop for a few minutes and make some notes, but during this time I did not fully appreciate the potential in what it could tell us. It is easy to fail to appreciate the ordinary details of everyday organizing.

A month or so into the autumn term, I pass the board and I once more, for some reason, note it. I note how it is filled with more printouts and hand-written notes with lists of things to be done. For instance, one of the printouts is a large excel sheet, taped together from numerous A4’s into one large sheet, with details in rectangular cells. I have to step really close to make out what the cells represent. The traffic light magnets are used for visualizing status on some of the activities (from observation notes September 13th).

Yet another month later, I note:

I find this intriguing: what is happening to this space where strategies are supposed to materialize and become actionable? The week after, the lack of updates of the data on the board is noted in more detail:

The vitality of the strategy board is fading: no written updates for several weeks. I wonder if the recent change of site manager has anything to do with this. The former manager, who was recruited internally many years ago and who now will continue to work in the organization but in a different role, was recently replaced via an external recruitment. For the first time, a woman has been appointed site manager.

Two months later, in December, as I pass the board on my way into the CNC-machine where I currently spend my time, the strategy board is suddenly empty:

What started with the intention of being a space for engaging with the strategy and for visualizing progress, has now become an empty space on the wall in the large assembly hall. I have, almost without knowing it, witnessed the life and death of a strategy board. In the next section, reflecting on the short and, to be fair, not particularly intense life of the strategy board, we discuss what we learn from our engagement with the strategy board, and the contributions related to methodologies and methods for dealing with human and non-human actors in ethnographic studies.

Technology does not always matter (Sage et al., 2010). In the case of an information board, if its viewers are “disciplined” or can be shown to be “better informed”, then this is an affordance of the technology. Conversely, if it inspires conflict, irritation or even indifference, it does not have this disciplining affordance. To be able to talk about how the material and the materiality of the visualization boards come to be part of organizing processes (or not), we must look at our empirical material over time because, as argued by Leonardi (2017), new affordances take time to emerge. We need to observe how the boards are used, not only at one point in time but over longer periods – do organizational members study it, use it, reflect upon it, and discuss it with others, for example?

In this paper, we show why it is of critical importance to follow these “non-human” actors over time. But “following” or “being there” is not always easy, or enough – neither when it comes to humans nor to non-humans. In this case we do see that the board has a life-cycle, and we see how it is part of materializing and dematerializing initiatives, and how it carries with it the desires of management, very explicit in this case, for improved productivity – as such it is both informational and a “call to arms”. Or at least that was its intended role at birth. It also creates, at least in its early incarnation, a place to gather and discuss the future productivity of the firm. A little deconstruction is needed to suggest, tentatively, that it is also, not particularly subtlety, an expression of managerial authority in how it reminds those passing it and noting it that it is management who control the board and have the power to place it so prominently, and to determine the future direction of the company. The board, in being placed where it is and in being mobilized as a place where direction is constructed, also assigns different roles – viewers, discussants and those responsible for updating it. We find that Latour’s (2005) actor-network theory is useful here: the board has viewers, but also servants, meaning those tasked with updating it – the board disciplines up as well as down the hierarchy.

However, the board soon also begins to reflect fractures in managerial power and foresight, as disagreements over the level of detail the board should carry become evident. At this point perhaps, the board already starts losing its power as a sign (of managers’ control over the future, action and, indeed, reality itself) as it becomes crowded with detail and drifts away from its original intent for clarity and inspiration. The board continues to be updated over the next three months or so after which it seems to be abandoned by those responsible for updating it, and the board loses its power to assign even that role. Eventually – having atrophied – its content is removed entirely, and we may ask questions such as when and by whom? It de-materialises – it ceases to offer information or insight and is merely a blank, white space on the wall in a prominent place in the factory reminding those passing of … well, we will have to guess at that!

Two things emerge from this. First, our study shows that it is essential to observe objects over time – in this case to see this object’s materialization, dematerialization, decline and death. Leonardi’s (2017) methodology needs to be “applied” over time as we need to carefully account for materials, materiality, and how materials matter (or not) over time. Secondly, it is essential to recognize the role of multiple human input in supporting (or not) the object and the engagement with it. The humans caught in its influence bring different perspectives and practices, and that too has an effect on what happens to the board. In this case the board fails to stabilize these relations. We also, by observing the board’s emergence and decline, must acknowledge the crucial role played by conversations we could not see. It is clear, from our empirical observations, that what happens with and around the board cannot be understood in isolation – moreover, the action or inaction we observe may have nothing, or everything, to do with the board. Nevertheless, it was also clear from what we saw that there was disagreement about the level of detail and that there must also have been disagreement (amongst managers) about the effects the board was having on its viewers - the (human) actors in our story are also theorizing and filling in the gaps as they go. The cancellation of the board may have been due to other activities taking precedence, or due to the idea that it was failing to have the desired effect – we were not there to observe that, it is not in the data. In other words, even with detailed, on-the-ground, longtime ethnography we must surmise and guess, as to the human intentionality behind certain manifestations of the material, the materiality, and how certain materials come to matter whereas others do not. Just as Latour (1992) assumed that the doorman gets sick or takes days off, we must make some assumptions for which we do not have evidence. We find, in other words, it simply is not possible to follow “the action” as Latour entreats us to do, right into the necessary detail. Some of the doors we come to in following the action are simply locked. However, by doing ethnography and do so longitudinally, we perhaps can learn enough to feel confident in filling in some of the gaps – as Latour felt he was able to do with regards to the habits and frailties of doormen.

It follows that, to explain why and how visualization boards matter as well as the effects they have and do not have on organizing, we need to go beyond the methods, ontologies and assumptions of even relational accounts, to examine the complex relationships of the social and the material in organizing. We need to be meticulous in our methods of inquiry if we are to avoid “rushing to explanation”, as Latour (2005) has put it, but we should not be constrained in where we look for explanation. We should not rush, but we should draw on the deep knowledge gained of the organization elsewhere to “join the dots” and make observations that go beyond the data. We need, in other words, to make sense of events by making assumptions, and by using our imagination. In drawing inspiration from Latour’s earlier work and drawing on Leonardi’s (2017) thinking on methodology, we find that in order to develop methodologies and approaches that will allow researchers to really take artefacts such as visualization boards seriously, we need to push at the boundaries of ethnography.

Consequently, while characterizing visualization boards as actors in the coordination and stabilization of networks is a useful starting point – one that is explored further using Leonardi’s meticulous approach (he would perhaps have spoken to Latour’s doorman – but may still not have got “the truth”) – it is useful to accept also that the ability of the researcher to fulfil this role will vary between organizational contexts and the research time available to them. Tools, such as visualization boards, are performative, but their performativity, their power, relies on the co-production of that power in each of their contexts of use (c.f. Clegg et al., 2006). We should therefore also seek, through our exploration of how tools or information artefacts such as our boards are used and applied, to also explore the cultural and normative constraints within which practitioners operate. For the most part, the users of these tools are neither passively enrolled into the networks being created by those tools, nor are they necessarily intent being powerful actors that shape and enrol others. The operators and the managers we meet in our study are to some extent just trying to get through the day – balancing the Taylorist dogma of the tools they encounter against their own established practices and need to perform the work they have agreed to do.

To do ethnography we must accept, at some point, the need to summarize and bracket the complexity we find into drivers of the action. It is worth noting that Latour and Leonardi both draw on visually accessible, relatively simple and culturally familiar technologies to make their point. Their close understanding of the technology and its context of use allows detailed discussions of the practices and even moralities implicated in and by them. To deal with a complex and large organization like The Company and the rather extensive organizing practices entailing visualization boards, it can be argued, is somewhat different. We note how time and commitment is clearly necessary to genuinely get to understand what the technology does, what its limitations are and perhaps even, why and how it comes to take the form(s) that it does. As noted also by for instance Czarniawska (2007, 2008), Smets et al. (2014), and Akemu and Abdelnour (2020), this is no easy task. It is not enough, for example, to talk only to users about their experience of the technology. We may learn details about their practices, but we will only access those affordances that those practices reveal; we learn nothing about untapped functionality or the dogmas (van der Hoorn and Whitty, 2015) and moralities (Latour, 1992) that shape the tool and its’ intended use. Similarly, precisely how operators and managers are influenced by the implicit assumptions carried by the visualization boards in their day-to-day practice will depend on a range of factors – including (though not limited to) organizational context, the career experience of the individual and the other actors they are dealing with at the time. To access this level of understanding we need ourselves to become users, we may need to talk to designers and disassemble and re-assemble how the technology got to become the actor it is. It follows also that ethnographic research, direct observation of use and wide access to the actors implicated in the technologies production and use is essential, even if it is also always partial. The technologies and the contexts in which they are deployed, need, consequently, to be as familiar to the researcher as chairs and door closers. Thick descriptions of the empirical scene and the network relations beyond the scene at hand, will be essential.

In this article, we explore the challenges of doing ethnographic research where technology is the focus. We follow the tenets of Latour (1992, 2005) and Leonardi (2017) by, respectively, undertaking to “follow the actors” to show in detail how and when technology matters and when it does not. However, we eventually reach the limits of pure ethnography and find ourselves reaching for other resources. Long emersion in an organization can help, as we need to become deeply familiar with the practices and habits of those we are studying and fully understand the technologies we are interested in. There will however always come points in the research where we cannot follow the actors, as those actors disappear behind doors into rooms and times that we simply do not have access to or miss. Or multiple actors engage with the technology at different times and places, while we, as individual researchers, are limited by our human singularity when it comes to “being there”. At this point, we must use our knowledge, developed locally and over time, to re-construct the likely scene, read backwards from what we can see into what we can’t.

The authors would like to thank Viviane Sergi for her valuable feedback to an earlier version of this paper.

Funding: This research was carried out within the Up-skilling for Industry 5.0 roll-out project, funded by EU Horizon (grant no: 101070666).