Can Work Be Meaningful Under Algorithmic Management? A MacIntyrean Perspective

Abstract

Algorithmic management is deeply changing the way work is performed and the interaction between managers and workers in organizations. It also heavily affects the conditions for meaningful work highlighted by existing literature. Therefore, organizations need an appropriate framework to enable meaningful work when adopting algorithmic management systems. This article presents a normative study of the conditions for work to be meaningful in this new scenario. To fulfil this purpose, it adopts a MacIntyrean approach, according to which work is meaningful when it embodies practice-like characteristics. The article identifies the main threats of algorithmic management and characterizes the normative conditions organizations should meet to enable meaningful work. In addition, the article explores the strategies of resistance that workers use to live up to the standards of meaningful work when organizations are not capable or willing to provide those conditions.

“Algorithmic management” is a specific form of managing workers and labor through algorithm-based processes (Noponen et al. 2024; Baiocco et al. 2022). It is widely used in organizations to coordinate workflows, select personnel, assess performance, and even impose sanctions (Curchod et al. 2020; Möhlmann et al. 2021). Algorithmic management is changing the way managers and workers interact, how results are tracked and rewarded, and how workers are evaluated (Bucher, Schou, and Waldkirch 2021; Stark and Pais 2020). Ultimately, algorithmic management challenges workers’ opportunities for interpersonal relationships, autonomy, and self-development, thus heavily affecting the conditions for meaningful work highlighted by existing literature.

Business ethics scholars have examined whether there are some objective characteristics of meaningful work to which workers are entitled (Lysova et al. 2019; Michaelson et al. 2014) and, therefore, whether organizations have a moral responsibility to provide such conditions (Michaelson 2021). These normative accounts of meaningful work have been built primarily around job content (Hackman and Oldham 1975), autonomy (Bowie 1998), and the job capacity for workers’ self-development (Yeoman 2014). More recently, the advent of new technologies, especially artificial intelligence (AI)-based technologies such as algorithmic management systems, has fueled a new wave of academic interest in meaningful work (Lysova et al. 2023; Mejia 2023; Sison 2024).

However, a normative study of the conditions that need to be guaranteed for work to be meaningful in this new scenario is lacking. There are studies on the consequences of using robots (Smids, Nyholm, and Berkers 2020) and artificial intelligence (Bankins and Formosa 2023) for meaningful work, and it has been affirmed that the effects of these technological changes on the meaning of work are uncertain and deserve further study (Bykov 2024). Nonetheless, there is no systematic normative study of whether meaningful work is still possible under algorithmic management. Beyond a mere gap-spotting approach to the present research question (Sandberg and Alvesson 2011), there is an urgent need to provide organizations with the appropriate framework to facilitate meaningful work when adopting algorithmic management systems. This urgency is dictated by three reasons: 1) without an ethical framework, the dangers connected to the growing deployment of algorithmic management in organizations (which are going to be extensively described in the article) are likely to take over the advantages workers can experience in this technology-driven context, especially in terms of personal and professional development; 2) algorithmic management interferes with the traditional manager-employee relationship creating situations that were not present previously (e.g., the automated and not filtered systems of review and rewards); 3) the job market is now populated by a generation whose expectations of meaning in their work are much higher than previous generations had (Michaelson and Tosti-Kharas 2024). Without guidance on how to preserve meaningful work under algorithmic management, organizations risk losing an important part of the workforce because they are not able to provide for what is now a primary need for the current generation: GenZ members do not want just an employment, they want a work worth doing (Twenge 2023).

This article aims to present the normative conditions that make work meaningful for workers and organizations under algorithmic management. It examines how organizations should meet the conditions for meaningful work, and what undermines the possibility to realize these conditions. To do so, we turn to Alasdair MacIntyre’s goods-virtues-practice-institution scheme and his critique of professional management as a manipulative relationship (MacIntyre 2007), and to relevant MacIntyrean scholarship on meaningful work (Beadle and Knight 2012). MacIntyre’s moral proposal helps us identify some critical threats that algorithmic management poses to meaningful work and how such threats may be overcome under certain conditions, which this article explores in detail. This article contributes to current literature on meaningful work by arguing how organizations should provide workers with the possibility of experiencing meaningful work under algorithmic management, and how workers could realize this potential in their daily work, even when organizations are not capable or willing to support meaningful work.

The article is structured as follows: Section 1 introduces algorithmic management and its core features, reporting examples from organizations using this system. Section 2 describes the essential traits of meaningful work in light of MacIntyre’s theory of practice, and the main threats posed by algorithmic management to meaningful work. Section 3 presents the normative conditions for developing meaningful work in organizations using algorithmic management, and explores strategies that workers put in action when organizations ignore those conditions. Section 4 concludes and provides avenues for further research.

1. WHAT IS ALGORITHMIC MANAGEMENT?

A joint study from the International Labor Organization and the European Commission states, “Algorithmic management can be defined as the use of computer-programmed procedures for the coordination of labour input in an organisation” (Baiocco et al. 2022, 5). Algorithmic management constitutes a combination of digital technologies typical of the Fourth Industrial Revolution (such as AI-based technologies, big data analytics, machine learning, geolocalization, internet of things), in order “to automate or at least support some of the functions previously carried out by human management for the coordination of work” (Baiocco et al. 2022, 8) such as task allocation, scheduling, and performance appraisal (Stark and Pais 2020). It emerged in the context of online platforms and is also being implemented in traditional workplaces.

The main characteristics of algorithmic management applied to organizations are: 1) it helps match consumers and service providers; 2) it helps companies monitor workers from different perspectives (presence, punctuality, performance, etc.); 3) it provides mechanisms for feedback on workers’ performance through rating systems, which then score workers depending on their internal and external reviews. The following paragraphs offer a description of these characteristics.

Matching consumers and service providers. Algorithmic management is present in the management of online platforms such as Uber, TripAdvisor, Upwork, or Freelancer, where it facilitates the encounter between consumers and service providers (Vallas and Schor 2020; Wood et al. 2019). Consumers pay for a service, which needs to be carried out through a worker, and the platform assigns the worker performing the service. Thanks to digital technologies (for example, geolocalization), apps can identify which workers might be in a better geographical position to take the job. Some platforms allow workers to see which jobs they are accepting. In contrast, other platforms suggest options that cannot be explored (e.g., Uber drivers do not know the passenger’s destination before the pickup). Unlike conventional taxis, Uber, Cabify, or Lyft drivers do not need to go around looking for customers: it is the platform that matches them with the passengers (Möhlmann et al. 2021). Similarly, furniture assemblers on TaskRabbit, or software developers on Freelancer, find clients thanks to the platform’s ability to analyze their information and put them in touch (Wood et al. 2019). Algorithms give workers access to consumers through the virtual marketplace.

Monitoring workers. Through algorithmic management, companies easily monitor workers without direct reporting. For example, Amazon uses AI-powered cameras to supervise delivery drivers so that management may know how many deliveries each employee makes daily (Bankins and Formosa 2023). Some authors describe this feature as “algorithmic control” (Wood et al. 2019). Workers frequently experience information asymmetries and dependency relationships in this form of management (Cutolo, Hargadon, and Kenney 2021). Moreover, worker-platform communication is usually automated. The worker with doubts, complaints, or questions often only receives preset answers. There is no room for authentic dialogue (Curchod et al. 2020). Many workers experience frustration and anxiety for being unable to communicate and account for their actions. The algorithm has the power to assess, classify, reward, or sanction, to the point of firing, even without prior notice. Platforms are not required to provide reasons for those actions. Moreover, in their case, the reasons given are invariably cryptic, and the criteria might be externally unclear. Appeals are hardly accepted, and even in the case of a successful appeal, workers or suppliers who were sanctioned, downgraded, or dismissed, do not return to the previous status quo since other workers/competitors will have replaced them (Luca and Zervas 2016).

Some authors have pointed out that there is a new version of panoptic power between platform and worker (Woodcock 2020), a continuous supervision of the individual, both in the form of control and sanction, as well as in the form of shaping and transformation, according to patterns developed by the algorithmic programming of the platform. Algorithms thus become “power technologies” (Galière 2020), which “materialize the needs, wants and desires of the platform owner; they coordinate the collection, analysis and interpretation of data; and they enable, as well as constrain, multiple forms of agency on the online platform” (Curchod et al. 2020, 668).

Rating system. Platforms provide systems of rating, which allow consumers to assign a score to the tasks performed by the workers, so workers receive an overall score of their performance over time. Consumers can decide whether to hire a software developer, or choose accommodation for their holidays, based on the score they see the developer, or the hotel, have accumulated over time (Bucher, Schou, and Waldkirch 2021; Luca and Zervas 2016). It is a fact that ratings work: consumers pay attention to them and decide based on that information.

Positive evaluations are necessary to climb and stay in the rankings. Negative reviews are less common but have a more significant impact than positive or neutral reviews and can ruin a supplier’s or worker’s reputation almost immediately (Curchod et al., 2020). Ratings and rankings are important because the platform economy is an attention economy (Stark and Pais 2020). For those who provide services through platforms, it is necessary to “be seen” and, thus, to be able to be chosen.

Thus, in the platform economy, workers find themselves in need of “satisfying” the algorithm (Bucher, Schou, and Waldkirch 2021). However, the criteria for calculating the rating are not stable or entirely understandable. It is unclear what the minimum score for not being “deactivated” is, nor to what extent the reviews received influence the position in the ranking, nor what parameters are used to establish who is recognized as an “elite worker,” that is, a worker who has high ratings and therefore a high position in the system’s ranking. “It is the algorithms within the system that decide what is valuable and what is not” (Baskerville, Myers, and Yoo 2020, 517). Algorithms act as “black boxes” whose inner operations are not really grasped by users, workers, or managers.

Following existing literature (Noponen et al. 2024), we contend that algorithmic management represents a new model of management as it creates new power structures that differ from traditional ones not just in degree but in kind. Algorithmic management is neither a new Taylorism, nor an automation of bureaucracy, nor a further step in the digitalization of work. Algorithmic management does not tell workers how best to work, but guides them in a distinctive way characterized by a specific combination of autonomy and dependence (Curchod et al. 2020; Möhlmann et al. 2021). Algorithmic management has been originally associated with the platform economy; however, more traditional workplaces and organizations are now using it (Jarrahi et al. 2021). Wood (2021) details the different sectors that are currently using algorithmic management. These sectors range from hospitality to retail, and, as Wood explains, logistics is increasingly using algorithmic management to increase efficiency and requesting the intervention of human managers in “critical moments” (2021, 13). In addition, algorithmic management is seen both in white-collar and blue-collar jobs. Consulting and financial firms use algorithmic management for its potential to organize teamwork and support project managers, while in the manufacturing sector, algorithmic management is used for “work allocation and instruction” (2021, 4).

As we have just detailed, algorithmic management is increasingly present in all kinds of workplaces (Cameron et al. 2023; Acemoglu and Johnson 2023). It touches on essential nerves of the relationship between managers and workers, ranging from autonomy to control, from creativity to trust. Those are also some essential aspects of meaningful work reviewed in the next section: this review of the relevant literature prepares the groundwork for the characterization of the normative conditions for work to be meaningful under algorithmic management.

2. WHAT MAKES WORK MEANINGFUL?

2.1 Characteristics of Meaningful Work

Research based on social psychology and management sciences has identified relevant elements or factors related to meaningful work (Bailey et al. 2019; Hackman and Oldham 1975; Lips-Wiersma and Morris 2009; Rosso, Dekas, and Wrzesniewski 2010). On the one hand, factors related to the content of work contribute to its meaningfulness (Martela 2023; Michaelson 2021; Veltman 2016). These factors entail:

• • Skill variety: the number and diversity of different skills used by workers.

• • Task integrity: the extent to which workers perform a job from start to finish.

• • Task significance: the extent of the achievement or the contribution to a worthy goal; its point and purpose.

On the other hand, factors related to the subjective experience of work influence whether workers perceive it as meaningful or not (Breen 2019; Lips-Wiersma and Morris 2009; Weeks and Schaffert 2019; Yeoman 2014). These factors include:

• • Autonomy: the degree to which the job provides substantial freedom, independence, and discretion to the individual in determining the procedures to carry it out; this trait includes participation in decision-making regarding work and job crafting and design.

• • Self-development: this trait describes how work facilitates skill cultivation and use, and its impact on the realization of a worker’s professional and personal flourishing.

• • Self-esteem: the extent to which work allows integration with other components of one’s life and leads to a meaningful life.

• • Social relationships and social recognition: meaningful work provides a sense of belongingness and relatedness, of being accepted by colleagues, being mentored, and supported; and a shared sense of agency and common purpose, a sense of connection and “unity with others.”

The subjective experience and the objective content of work are connected. Work is often experienced as meaningful when its content has a worthy purpose, in the achievement of which the worker participates from start to finish using a varied set of skills. However, this connection does not always hold (Michaelson 2021). Sometimes, dull and fragmented work, which objectively does not involve an actual achievement or require a set of professional skills, is experienced as meaningful. Conversely, sometimes work that does involve achievement, allows participation in a complete process, and helps to develop one’s skills, is subjectively experienced as meaningless. This possible dissociation between subjective experience and objective content shows that the worker’s motives for engaging in work depend not only on its content but also on other external motives (such as salary and status). In this regard, some authors have pointed out ethical problems concerning workers’ motivation and its potential manipulation by managers, that is, how managers might promote their own interests against those of the workers by implementing certain types of work (Breen 2019) and why workers should desire a job with certain objective characteristics over one that reflects other kinds of preferences (Ciulla 2000; Mortimer 2023).

Understanding the connection between work’s subjective and objective aspects and avoiding possible manipulation requires a normative approach (Michaelson 2021; Sison 2024). Objectively, work is meaningful to the extent that it contributes to a meaningful life (Ciulla 2000; Mejia 2023). Subjectively, work is meaningful to the extent that workers have convincing reasons to engage in it (Mortimer 2023).

Alasdair MacIntyre’s moral sociology (2007; 2016) offers an approach that considers both dimensions from a normative perspective. To show that something has “meaning” or is “meaningful” is to clarify its place in a broader context that makes it intelligible (MacIntyre 1999b). Work, like other human activities, has a teleological structure. Subjective experiences are meaningful not only because they are experienced as such but because “there are good non-circular reasons for that experience” (Michaelson 2021, 421). Meaningful life is not the kind of life one considers good and worth living, whatever that may be, but the kind of life one has reason to consider good and worth living (MacIntyre 2016, 17ff). Accordingly, work is meaningful when the one who works “has good reasons to experience it as meaningful, that others have good reasons to perceive as socially worthwhile, and that is independently (intersubjectively or objectively) meaningful” (Michaelson 2021, 420). MacIntyre’s Neo-Aristotelian ethics offers such reasons in three stages.

First, MacIntyre (2007) presents “practices” (socially established complex activities realizing internal goods and whose excellence is defined mainly by expert practitioners) in contrast to “institutions” (activities or organizations pursuing external goods, such as wealth, status, and power, whose effectiveness can be judged by anyone). Second, he examines how individual practitioners render compatible various roles and practices by ordering different types of goods in their quest for a good life. Third, he argues how excellence in practices contributes not only to practitioners’ good lives but also to the advancement of their communities. It is this normative perspective that justifies and substantiates the legitimacy of the subjective experience of meaningful work (Beadle and Knight 2012). In this article, we adopt this normative perspective.

In the following sections, we present in more detail the MacIntyrean approach to: 1) meaningful work, 2) the tension between meaningful work and professional management, and 3) the conditions that management must meet to sustain and promote meaningful work. We then contrast this approach with the characteristics of algorithmic management and, finally, elucidate whether and how organizations using algorithmic management leave room for meaningful work.

2.2 A MacIntyrean Approach to Meaningful Work

According to MacIntyre (2007), there are goods internal to certain activities that he calls “practices,” in which individuals can work and be educated in a way that informs their sense of meaning (Beadle and Knight 2012, 438). To clarify this claim, it is necessary to explain, first, what kind of activities practices are; second, what kind of goods are internal to practices; and third, why people would prefer those internal goods over other external goods such as money, prestige, or power.

First, a practice, according to an oft-quoted definition by MacIntyre, is “any coherent and complex form of socially established cooperative human activity through which the goods internal to that form of activity are realized in the course of attempting to achieve those standards of excellence which are appropriate to, and partially definitive of, that form of activity, with the result that human powers to achieve excellence, and human conceptions of the ends and goods involved, are systematically extended” (MacIntyre 2007, 187). Examples of practices are found in sciences, arts, and sports. Many professions can also qualify as practices (Higgins, 2010; MacIntyre, 2007, 187–189; Moore, 2017), including interpersonal services like nursing (Sellman 2000), as well as productive activities like farming (Sinnicks 2019).

Second, MacIntyre follows Aristotle in identifying two kinds of goods internal to practices (2007, 189–90). On the one hand, there is the excellence of the product: this encompasses both the excellence of practitioners’ performance (e.g., the skill of the artist who makes portraits or the teacher who lectures) and the excellence of the result (the portrait or the lecture). A productive practice such as painting has standards of excellence, the actualization of which by individual painters produces excellent portraits. The same can be said for teachers (Dunne 2003), nurses who care for people (Sellman 2000), and farmers who cultivate fields (Sinnicks 2019). On the other hand, MacIntyre argues that the actualization of these standards also actualizes something else: “the good of a certain kind of life…. it is the painter’s living out of a greater or lesser part of his or her life as a painter that is the second kind of good internal to painting” (2007, 190). Something similar can be affirmed of the lifestyle of the teacher, the nurse, and the farmer as kinds of life that realize different modes of human excellence, only accessible from within each profession.

Third, practitioners can prefer internal goods over other goods, such as a better salary or better working conditions, because internal goods are valued according to the standards of excellence proper to the practice. Goods such as money, power, or prestige can be attained in many different ways, while internal goods can be obtained only in each specific practice: having more money does not make one a better painter, a better teacher, or a better nurse. Certainly, external goods are necessary to sustain practices and practitioners: good painters, teachers, and nurses need adequate resources and conditions to carry out their work in an excellent way, as do art schools, universities and hospitals. However, for practitioners committed to their profession, it is the internal goods that are essential in the evaluation of their work. The meaning of their work lies in achieving excellence in their profession, both in the results produced and the skills developed, and in the kind of life, specific to their profession, that they aspire to live.

To identify and obtain goods of excellence, workers need some skills and virtues that can only be acquired by becoming expert practitioners. As they become better in the practice, workers also acquire higher levels of skills and insights (better reasons) that were not previously available, that beginners do not know, do not understand well or do not value enough. “Not only skills, maxims and rules but also practical judgment and moral character are learned through work that actualizes the goods of a certain kind of life” (Beadle and Knight 2012, 438). Work is perceived as meaningful when workers take the achievement of its internal goods to be constitutive of their own good.

Meaningful work is thus preferred over other possible goods for moral reasons: it is valued as part of the kind of life one has good reasons to consider good and worth living. “There is a close connection between being a good human being and doing good work” (MacIntyre 2011, 323). This criterion is not purely subjective, nor does it depend on individual preferences. Rather, it depends on work endowing practice-like features (Beadle and Knight 2012). Thus, a work may be meaningful insofar as it is: a) complex and coherent enough to require skill variety and task integrity; b) socially established and cooperative enough to promote social relationships within which a sense of belongingness, unity, and common purpose can be developed; c) able to actualize internal goods, providing purpose, as well as a sense of achievement and self-development; d) able to hold standards of excellence according to which products are judged, and practitioners obtain self-esteem and social recognition; e) with the result that extension of human powers and human conceptions of the ends and goods involved can be pursued with autonomy and discretion in such a way that that work becomes morally educative; and, f) its contribution to society is realized in the advancement of practitioners’ communities.

Table 1 summarizes what has just been described, associating the different characteristics of a “practice” according to MacIntyre (2007) with the characteristics of meaningful work highlighted above.

Table 1: Characteristics of MacIntyrean Practices and Meaningful Work

These categories help us argue why some kinds of work are meaningful and others are not. Some types of work have such limited content that they do not allow the development of professional competencies in quantity and quality. Other types of work are organized in such a way that they do not allow for the development of social and moral competencies conducive to a fulfilling life. Other types of work do not contribute to the good of society. MacIntyre’s moral sociology connects objective and subjective dimensions of meaningful work as it captures the elements that the literature associates with meaningful work and explains the reasons that lead workers to engage in it. Besides, this approach also identifies some important threats to the very existence of meaningful work, which are discussed in the next sections.

2.3 Institutions and Meaningful Work

For practices to thrive they need to be institutionalized (MacIntyre 2007, 193). Practitioners must organize themselves to establish standards, obtain resources, and advance other common objectives. For this purpose, academies of arts and sciences, schools, hospitals, sport federations, professional bodies, and business enterprises emerge. These organizations are a combination of practice and institution (Moore 2012). While practices are concerned with internal goods (also called goods of excellence), institutions focus on the acquisition of external goods (or goods of effectiveness), such as money, power, and prestige necessary for the sustaining and flourishing of practices (MacIntyre 1988). Institutions are supposed to protect practices and promote their healthy development.

The simultaneous pursuit of excellence and effectiveness typically causes tensions in organizations. The acquisition of goods of effectiveness is a necessary and valuable function of an organization, as long as it is subordinate to the sustenance and development of the practices it hosts (Moore and Beadle 2006). However, there is a constant risk that the order of priorities will be reversed and effectiveness considerations will prevail: “The ideals and the creativity of the practice are always vulnerable to the acquisitiveness of the institution, in which the cooperative care for common goods of the practice is always vulnerable to the competitiveness of the institution” (MacIntyre 2007, 194). For MacIntyre, the custodians of the delicate relationship between practices and institutions are the virtues. The integrity of a practice causally requires the exercise of the virtues by at least some of the members of the organization, both managers and employees; conversely, the corruption of institutions is always in part at least an effect of their vices (MacIntyre 2007, 195).

In After Virtue, MacIntyre (2007) warns about the manipulative nature of professional management that arises from rationalization processes in bureaucratic organizations, as described by Max Weber. Professional managers leading all kinds of private and public bureaucratic organizations claim the authority to manage workers on the pretense that they have superior knowledge and experience on how to be effective. Such expertise and authority are assumed to be universally applicable, and transferable from one organization to another regardless of the particular type of activity carried out there (2007, 26–27). However, according to MacIntyre (2007, 90–108), social and organizational studies do not provide a basis for such knowledge and, therefore, do not truly justify the belief in managerial expertise that has become institutionalized in business.

In business organizations, the harshness of competition and the immoderate drive for profit trigger managers to prioritize the achievement of effectiveness objectives over the pursuit of goods of excellence (MacIntyre 2015). Professional management tends to manipulate human resources by treating them as mere means to the fulfillment of the manager’s ends. Managers characteristically insist on dividing the work in such a way that they reserve to themselves the power to decide and expect their subordinates to execute their decisions. As a result, “practices are often distorted by their modes of institutionalization, when irrelevant considerations relating to money, power and status are allowed to invade the practice” (MacIntyre 1994, 289).

Nevertheless, MacIntyre admits that institutions are necessary to sustain practices, formalize and enforce rules, and obtain and distribute resources. To avoid manipulation, management must aim to defend the goods internal to the practice, the standards of excellence and the common goods of the organization’s members (Beadle and Moore 2006). These goods and standards give meaning and purpose to the acquisition of external goods. A management that “defends” internal goods is only possible if managers are also practitioners who have learned to value internal goods through participation in the practice (West 2024). In organizations with this kind of management, managers and practitioners engage in rational critical dialogue regarding how to adequately order the goods of excellence and the goods of effectiveness (MacIntyre 2016). This is the institutional context that promotes meaningful work, based on trust and cooperation relationships between managers and employees (Moore 2012). However, meaningful work is compromised when power structures deny workers control over their activity and, consequently, prevent them from cultivating goods of personal excellence (Beadle and Knight 2012).

As explained in Section 1, the current use of algorithmic management systems introduces new forms of control and power relations in organizations (Noponen et al. 2024). Its increasing diffusion is generating significant consequences for the relationship between practices and institutions: on the one hand, it generates new ways of sustaining practices but, on the other hand, it introduces new threats to the proper order between goods of excellence and of effectiveness. As we have seen above, this order is closely linked to the meaning of work. Therefore, in the following section, we will analyze the threats that algorithmic management specifically poses to meaningful work.

2.4 How Algorithmic Management Threatens Meaningful Work

In the reality of algorithmic management, the three components identified in Section 1 (matching users and service providers; helping companies monitor workers; providing mechanisms for feedback on workers’ performance) are dynamically interrelated. In addition to the many advantages that algorithmic management brings to organizations in terms of improved efficiency, there are specific aspects which threaten exactly those practice-like characteristics identified in Table 1.

The use of algorithmic management tends to heavily affect the aspects related to skill variety and task significance: indeed, algorithmic management divides tasks and assigns them automatically to different workers (Baiocco et al. 2022; Wood 2021), causing workers to lose the context of their work, being assigned a specific task without necessarily knowing how it contributes to the final product or service. This directly affects variety, integrity, and significance, as described above, and it causes deskilling and fragmentation.

Algorithmic-managed platforms use reviews to measure workers’ performance and give or deny them access to the critical resource of visibility. In doing so, the use of algorithmic management indirectly induces workers to care more about ratings and rankings than the achievement of the standards of excellence (Bucher, Schou, and Waldkirch 2021; Roberts and Zietsma 2018). Workers maintain, therefore, a double dependency: they depend on users, who evaluate them (or not), and the platform; this gives them visibility (or not), according to the calculation that the algorithm makes based on the evaluations received, following frequently opaque rules. Thus, work no longer contributes to the moral education and excellence of workers, since it disregards critical thinking and personal growth.

Platform algorithms sustain workers’ as practitioners, as they provide them access to reputation, prestige, and income, but they do so by establishing a power structure that exerts control over workers’ activity, and thus undermines the achievement of internal goods, making the workers prioritize their reputation and standing (Möhlmann et al. 2021; Vallas and Schor 2020). Moreover, algorithmic management is usually implemented through software or apps, which move the conversation between manager and worker from a personal social relationship to a system of automated communication (Cameron et al. 2023; Cutolo, Hargadon, and Kenney 2021).

Platform control is also deployed in other forms of power, in particular, the discourse of the worker as an independent contractor (Galière 2020). Like the repressive power of algorithmic control, the discourse of worker autonomy and entrepreneurship aims to achieve the same compliance, only by other means. The platform presents itself as a tool, an aid for workers to better organize their work. In this discourse, the role of the algorithm is always legitimized by its effectiveness. However, this discourse is not free from contradictions. A paradigmatic example is that of Uber (Roberts and Zietsma 2018; Rosenblat and Stark 2016). On the one hand, the company’s official discourse refers to drivers as partners who join Uber to set up a transportation business together: “Be your own boss.” On the other hand, the company itself implies that drivers are at its disposal as cogs in an automated service: “Uber drops people off everywhere in the city at the press of a button” (Roberts and Zietsma 2018, 208). The high availability of drivers has allowed Uber to relate to drivers as mere executors of orders under the threat of being replaced by other drivers eager to take their position in the ranking (Möhlmann et al. 2021). “The conflicts and contradictions between … a driver-bot role … and driver-partner role … closed off … the positive aspects of the driver-partner identity by eliminating the autonomy and freedom that constructed the work as entrepreneurial, and thus meaningful” (Roberts and Zietsma 2018, 218). The discourse of the independent contractor thus becomes, in this case and many others, a form of manipulative relationship between management and workers that happens through the implementation of algorithmic management. This clearly contrasts the practice-like characteristic related to the extension of human powers and, in terms of meaningful work, the expression of autonomy through work, substituting this trait with surveillance and control.

Indeed, algorithmic management manifests the characteristics that MacIntyre attributes to professional management (2007, 26–32). ¹ It establishes a division of labor in which the algorithm holds the power of decision and assigns the execution to the workers. This authority is exercised under the claim of superior knowledge on how to achieve the greatest effectiveness (algorithmic calculation), which is assumed to be universally applicable, regardless of the type of activity being managed. This approach results in the treatment of workers as mere means in the service of pre-established objectives, clearly moving work from the domain of an intentional positive contribution to society, to a new form of alienation, where ends cannot be put into question.

Table 2 summarizes the threats to meaningful work posed by algorithmic management, following the scheme emerged from the analysis synthetized in Table 1. The first two columns report the findings of Table 1, and the third column pairs them with the threats to meaningful work described in this section.

Table 2: How Algorithmic Management Threatens Meaningful Work

As we have just explored, algorithmic management undoubtedly poses a threat to meaningful work, affecting exactly those practice-like characteristics which constitute the nucleus of meaningful work. Consequently, and given its pervasive diffusion, it is compelling to ask what the conditions for work to be meaningful under algorithmic management are, and who has the responsibility to enable meaningful work under algorithmic management. The next section addresses exactly those questions.

3. MEANINGFUL WORK IN AN ALGORITHMICALLY MANAGED WORKPLACE

As we argued, from a MacIntyrean perspective, the basic condition for work to be meaningful is to be ordered towards goods of excellence within activities endowing practice-like features (Beadle and Knight 2012), that is, activities that order workers towards a meaningful life. For MacIntyre, such meaning is compromised by the manipulative nature of management, which typically treats workers as mere means to the managers’ ends (MacIntyre 2007). Certainly, MacIntyre accepts that management is necessary to sustain practices by effectively providing external resources. In doing so, management actually promotes meaningful work. However, when managers strive for such external goods regardless or even against the internal goods of excellence, practices get distorted and, thus, unable to provide meaning for those who engage in them (West 2024).

Algorithmic calculations transform the power of management. Algorithm-based management systems help owners and managers to achieve pre-established goals (although often in ways that they do not fully understand) by guiding workers as a relational interface, that is, a mediation that shapes the content and the experience of their work (Capone, Rocchi, and Bertolaso 2024). In MacIntyrean terms, algorithmic management becomes an interface between practices and institutions, shaping management’s ability to promote meaningful work and workers’ ability to find meaning in their work. In this section we explore the role of algorithmic management in this scenario, and then we proceed to the characterization of the normative conditions for work to be meaningful under algorithmic management.

3.1 The Practice-Institution Scheme Revisited in the Age of Algorithmic Management

As we noted above, for MacIntyre, managers can promote meaningful work by enabling workers to engage in rational critical dialogues about how to appropriately advance the goods of the practice and the organization (MacIntyre 2016). These shared deliberations should be based on relationships of trust and cooperation among members and managers. To do this, managers need virtues such as courage, justice, constancy, and, above all, practical wisdom (Moore 2017; Moore and Beadle 2006; Sison, Beabout, and Ferrero 2017).

However, in organizations managed by algorithms, these conditions do not seem to be met: the relationship between management and workers is based on automated communication, monitoring and control, rather than trust and cooperation (see Table 2). Thus, decision processes are not based on shared deliberations among members but on opaque algorithmic calculations. Moreover, algorithmic management cannot find a virtuous relation between goods of excellence and goods of effectiveness since it does not have virtues in itself. ²

What algorithms do is to act as an interface between workers and the organization. On the one hand, algorithms improve managerial activities’ effectiveness. On the other hand, they constrain workers’ activities and limit their autonomy by enforcing predetermined patterns of action. However, the outcomes of the deployment of this kind of technologies in organizations are often unintended. As Sandberg and Tsoukas affirm, referencing Orlikowski (2000):

Technology does not have, nor can it be developed to have, certain inherent features that determine how it will be used or appropriated in organizations. Technology, rather, constitutes, along with human agency, a ‘technology structure’ that emerges from the repeated, recursive, and situated interaction between people and particular technologies (Sandberg and Tsoukas 2011, 352).

In Dependent Rational Animals, MacIntyre (1999a) argues that people’s reasons for acting are biologically and socially molded, but they can still regain some autonomy by using their practical reasoning capacity to become aware of and evaluate those reasons. Similarly, we may argue that human work, in addition to being biologically and socially shaped, is technologically mediated (Waelbers 2011). ³ This technological mediation also allows for the exercise of practical reasoning by workers. In one of his few references to the use of technology at work, MacIntyre states that, even when work seems to be purely mechanical, human workers (he offers the example of ticket sellers) always face choices that machines never confront:

How to play her or his role, cheerfully or sullenly, carelessly or conscientiously, efficiently or inefficiently. And for all roles, the way in which the role is enacted presupposes not only an answer to a question posed to and by the role-player: “How is it best for me to play this role?,” but also to such further questions: “By what standards am I to judge what is best?” and “Should I continue to play this role in this way?” It is the inescapability on occasion of such questions that suggests that practical reasoning that is adequate for doing what a particular role requires will itself generate reasons for acting beyond those requirements and even sometimes against those requirements (MacIntyre 1999b, 327, emphasis added).

MacIntyre refers this comment to the social structures in which people find themselves as moral agents, but we can also apply it to algorithmic management, that is, the technological structure in which workers find themselves in organizations managed by algorithms. This technological structure sets specific constraints and affordances for human work, and workers usually conform more or less willingly to its requirements, even deflecting the responsibility for morally questionable courses of action to technology (Köbis, Bonnefon, and Rahwan 2021; Krügel, Ostermaier, and Uhl 2023). Algorithmic management guides the way work is performed and, as a result, how workers perceive their work as meaningful at different levels or according to different aspects (as detailed above).

This relationship between workers and algorithmic management happens within organizations. For this reason, two aspects must be investigated: 1) what conditions must algorithmic management meet for an organization to enable and even promote meaningful work; 2) how workers can act beyond and even sometimes against the requirements of algorithmic management when an organization is not capable or willing to meeting those conditions. The following sections are dedicated to the development of these normative conditions (Section 3.2) and to the exploration of the scenarios where organizations are not willing or capable to enable them (Section 3.3).

3.2 The Normative Conditions for Meaningful Work Under Algorithmic Management

MacIntyre’s scheme has helped us to reframe what meaningful work is and why people engage in it (Section 2.2). Work becomes meaningful when workers take the achievement of specific goods to be part of the kind of life they consider meaningful and worth living. This criterion is not purely subjective, nor does it depend on individual preferences. Rather, it depends on work endowing practice-like features (see Table 1), that is, work that allows for professional and moral excellence and contributes to the advancement of workers’ communities. In addition, we have explored how algorithmic management threatens meaningful work exactly along those characteristics that make work meaningful (Table 2).

The implication of this analysis is, according to our argument, that work can be meaningful under algorithmic management when organizations meet some normative conditions, which are now presented and explained. Thus, to sustain and promote meaningful work, organizations must design and manage jobs in such a way that their employees can actually meet these same conditions. The normative conditions are described in this section, and they are summarized in Table 3, which gives evidence of how each condition enables a specific trait of meaningful work, corresponding to a practice-like characteristic.

Table 3: Normative Conditions for Meaningful Work Under Algorithmic Management

Job crafting: Organizations using algorithmic management should allow workers enough autonomy to craft their activities with some degree of skill variety, task integrity, and task significance. For example, telemedicine increasingly uses algorithms to monitor patients’ health, replacing some of the tasks typically performed by nurses. Nurses must now acquire new skills, such as expertise in reading screens with information about patients and integrating these tasks into a new style of caring (Nicolini 2012). Similarly, teachers must know how to use specialized software for their students’ education (Sandberg and Tsoukas 2011).

Alternative use of technology: Organizations should use their improved capacity for monitoring and control to help workers extend their powers and put their talents to their highest and best uses. For example, automated patient monitoring introduces the risk that nurses find themselves “treating the numbers instead of the patient” (Jones 2014, 907), that is, paying more attention to the data on the screen than to the actual condition of the person, but this automated monitoring allows them to devote their time to more careful attention to each patient. Similarly, teachers can easily access many sophisticated technological devices to conduct their lessons, but they must learn how to use them wisely for the good of their students (Higgins 2010).

Improving results: Organizations should reward workers who take advantage of the system’s affordances to improve results in terms of excellence and not only in terms of effectiveness, reputation, and standing. For example, organizations should use rating and ranking systems not only to help workers access resources and share good practices, but also to further standards of excellence together with colleagues.

Help assess achievements: Organizations should help workers obtain information about their achievements and benchmark their performance. For example, teachers whose performance is evaluated quantitatively may seek to improve their rating to secure rewards, but they may also find out the most valued teaching qualities and try to acquire them, thus improving their teaching. In this way, they can build professional prestige and strengthen their self-esteem.

Connecting people: Organizations should sustain human work by facilitating forms of communication that promote rather than preclude the development of significant social relationships in the workplace. For instance, in large companies, algorithms can be used to identify and connect employees with similar profiles, interests, or abilities, and sustain possible collective initiatives.

Sharing goods: Organizations using algorithmic management should strengthen cooperation and workers’ deliberation about common goods. For example, professional groups of nurses must deliberate on what it means to be a good nurse in a technological context where traditional tasks and standards of excellence have changed. The same applies to teachers and other professions. Participation and shared deliberation about how to reframe one’s work can occur not only in service professions but also in jobs such as drivers or other blue-collar occupations (Acemoglu and Johnson 2023; Wrzesniewski, Dutton, and Debebe 2003; Schwartz 2015). Those workers’ communities may develop guidance and direction not only for colleagues but also for human managers and other stakeholders, such as designers, marketers, and regulators.

Empowerment: Organizations should ensure a working environment that favors professional and moral development. Due to the adoption of algorithmic management systems, human managers also undergo task automation and risk of redundancy. However, those human managers who keep their jobs have a new task of undoubted moral relevance: to ensure that algorithmic systems not only increase the effectiveness in achieving the organization’s objectives but also help obtain its members’ excellence, that is, empower employees to further their professional and moral development. Managing the relationship between algorithms and workers in this way is the ultimate necessary condition for sustaining and providing meaningful work in their organizations.

Table 3 summarizes the essence of what has been explained so far, connecting the normative conditions that organizations using algorithmic management systems need to guarantee to sustain and promote meaningful work with the traits of meaningful work and the corresponding practice-like characteristics. An organization that wants to enable meaningful work for its employees under algorithmic management should enforce the conditions presented in the left column. In this way, the organization ensures that traits typical of meaningful work are realized (second column) in the course of an activity which has practice-like characteristics and is so ordered towards a meaningful life (third column).

The formulation of the normative conditions for work to be meaningful under algorithmic management takes the perspective of the organization: in this sense, the realization of the normative conditions represents the ideal situation, when organizations are willing to make meaningful work possible under algorithmic management and assume the responsibility to ensure the realization of these conditions. However, the responsibility to make work meaningful relies not only on organizations but also on workers. Organizations design and manage workers’ activities in ways that help or hinder their orientation toward goods of excellence and their integration in the workers’ quest for a good life. Organizations influence, but do not determine, the experience that workers have of their own work. Some organizations provide good working conditions, but some individuals do not take advantage of them for their professional and personal development. Other organizations threaten workers’ development by promoting deskilling, dependency, and even alienation, but workers may react in various ways to that influence. Some workers conform to the distressing situation and blame the organization for it; other workers resist the power of algorithmic management and put in place strategies to make their work meaningful. These “strategies of resistance” are described in the next section.

3.3 Strategies of Resistance: How Workers Can Make Meaningful Work a Reality

The normative conditions presented in this article provide a perspective on how to sustain meaningful work under algorithmic management from the point of view of the organization. When organizations are not responsive or do not engage with these conditions, workers do not receive adequate support to make their work meaningful. However, workers are still responsible for their own reaction to the lack of organizational effort to systematically enable the conditions for meaningful work under algorithmic management. If, in the case of an empowering organization, it is anyway the worker who needs to decide whether to engage in their work in a meaningful way or not, the worker’s determination to engage in meaningful work plays an even more determinant role when the organization does not actively pursue or even prevents the realization of the normative conditions. In this latter case, workers are in the position of having to resist such organizational threat and try to create the conditions for meaningful work themselves. This last section explores some strategies of resistance that workers can put in place in non-collaborative organizations.

In a traditional workplace, workers can defend their rights and improve their conditions through various strategies, such as collective action, individual resistance, or collaboration with management (Crozier and Friedberg 1980). In organizations managed by algorithms, in an attempt to balance power relations, in addition to the demands for legislative regulation, some collective action initiatives have arisen. These include the Online Merchants Guild, which brings together entrepreneurs who sell their products on Amazon, and the YouTubers Union, which unites some content creators (Cutolo, Hargadon, and Kenney 2021). ⁴ Cooperatives of platform workers have also emerged (Scholz 2016). However, these attempts are proving insufficient (Schor 2021).

Some platform workers, based on the practical knowledge they are gaining about how the system works, are developing individual strategies to reduce their dependence on the algorithm and achieve some autonomy (Bucher, Schou, and Waldkirch 2021; Möhlmann et al. 2021). For example, on professional services platforms such as Freelancer, Upwork, and Fiverr, some workers, who know how the platform works better than their occasional clients, know how to circumvent the system’s supervision, establish direct communication with the beneficiaries of their services, and secure assignments and positive evaluations (Cameron and Rahman 2022).

An illustrative ethnographic study of platform drivers (Cameron 2022) explores how these workers develop individual strategies that help them find meaning in their work and engage with it. Working in a technologically structured and often socially isolated environment, instead of experiencing alienation, these drivers set their own specific goals in their work through two different games, the “relational game” and the “efficiency game.” In the “relational game,” drivers look for meaning by “crafting their job as a positive encounter service” (2022, 231). To do this, drivers personalize their interactions with clients, trying to generate a connection and, where appropriate, offer physical or emotional support. For example, some drivers offer water bottles or tissues (without being obliged to do so by the system); others listen empathetically to customers’ outbursts; others carefully choose the music they play on each ride. These initiatives allow them to establish rich social exchanges. The purpose of all this work is “to connect with customers and provide good customer service” (2022, 237). The app’s rating system allows them to closely monitor their achievement, based on the customers’ assessment of their efforts. A good rating is “a source of professional pride” (2022, 238). What drivers consider “succeeding” in this game is also what the company wants. Therefore, the company supports this type of behavior with recognition and awards. However, the “relational game” goes beyond incentives and rewards, and incorporates a certain sense of excellence. For drivers, there is no concrete benefit in improving their rating from 4.8 to 4.9, nor in decorating their car with care, and yet they still do it. These drivers see the app as a facilitator of their work, a tool that gives them valuable information to do their job better (2022, 247).

The “efficiency game,” on the other hand, aims to complete rides as quickly as possible, ideally at the highest fare (Cameron 2022, 244). For drivers who play this game, the goal of their work is to optimize the time spent on a trip and the revenue earned. These drivers choose to depersonalize their encounters with customers by emphasizing physical and social boundaries. For example, they do not comfort crying passengers so that their help is not misinterpreted, nor do they offer extra services to protect themselves from potential liability (2022, 242). Some have themselves installed cameras in the car, not as instruments of surveillance and control, but as an effective means of defense against unfounded accusations by clients. This game is about limiting interactions with customers, which can be time-consuming, emotionally draining and financially unprofitable. The purpose of work performed in this way is simply to get passengers to their destination quickly (2022, 242). The measure of success is different in this game than in the relational game. The app does not provide information on the yield of each ride. Drivers distrust the algorithm. They think that it does not always assign them profitable trips and, therefore, they keep their own financial records to monitor their activity and try to secure their income. In this game, drivers see themselves repeatedly trying to thrive in a system in which they are at a disadvantage, and often “feel trapped in a game they cannot win” (2022, 245). Thus, they view the algorithm as an adversary that they must “cheat” to obtain more lucrative results.

In both games, workers relate to the system in such a way that they gain some autonomy to set their own goals and craft their jobs. Algorithmic management indicates to both types of drivers which rides to take and which route to follow, but it does not tell them the purpose of their work, beyond the task of driving. Some drivers set for themselves the goal of providing good service. These drivers find in the algorithm an ally to monitor their achievements, which helps them excel in their job providing valuable information. On the contrary, drivers who mainly aim to optimize revenue experience the algorithm as an obstacle to their work. The system imposes work patterns on them, withholds crucial information from them, and thus stands in their way of achieving their goals. Their only chance of getting away with it is the unlikely strategy of cheating the system.

These examples give us insights into meaningful work under algorithmic management in a non-collaborative organizational environment: they are illustrative of the behaviors and strategies that workers can put in place in such a setting. These strategies reaffirm the central role of workers, as free human beings, in characterizing the ultimate purpose and content of their work, thus highlighting once more the dynamic interrelation between the objective and subjective dimensions of meaningful work. In addition, these examples demonstrate that, ultimately, the threats posed by algorithmic management to meaningful work constitute an additional and novel stimulus for workers who, even in an adverse organizational environment, can creatively use the affordances of the technological structure to find and nurture meaning in their work.

4. CONCLUSION

This article contributes to the characterization of the normative conditions for meaningful work in organizations that adopt algorithmic management systems. Building on MacIntyre’s moral architecture, we argue that there are normative conditions that organizations can realize for work to be meaningful under algorithmic management: these conditions are built on the theoretical basis that for work to be meaningful it needs to be endowed with practice-like features (Table 1). We support this claim by extending the normative theory of meaningful work (Michaelson 2021; Mejia 2023; Beadle and Knight 2012) to algorithmic management, which we depict as setting specific constraints and affordances for human action.

We describe how algorithmic management threatens essential features of meaningful work, often causing deskilling and task fragmentation, putting workers under constant surveillance and control through rating systems that enforce specific relations of power and dependency. We show how these threats undermine exactly those characteristics which make work meaningful (Table 2). We then characterize seven normative conditions, which portray the ideal scenario in which an organization is willing to restore the conditions for meaningful work under algorithmic management (Table 3). Doing so, we develop a theory that enacts the critical conditions (Sandberg and Alvesson 2021) for producing and reproducing meaningful work in such a technological context. However, it is possible that organizations are not willing or capable to realize the normative conditions. In this case, workers need to resist such organizational constraints and try to defend the conditions for meaningful work. Interacting with technology, some workers can develop strategies of resistance to craft their activities so that they not only improve results effectively but also identify and share new and enhanced standards of excellence.

The formulation of the normative conditions for meaningful work under algorithmic management opens avenues for further research. Ethnographic studies are needed to understand better how people suffer from algorithmic management threats and how workers can achieve and sustain excellence both in a collaborative and in a non-collaborative organizational environment. The normative conditions have practical implications for organizations and managers striving to create meaningful work environments, and each condition can be detailed in terms of courses of action, decision-making processes, and design of empowering work relationships under algorithmic management.

Among the current studies on the consequences of algorithmic management in organizations, only a few deal with meaningful work in platform jobs (Möhlmann et al. 2021; Cameron 2022) or more traditional professions (Olsen 2023). Many other professions, as well as many jobs in the platform economy and traditional workplaces, would also provide significant evidence to learn what unforeseen uses of technology, what strategies of resistance, what new roles and professional relationships, and what “games” (forms of practical wisdom) played by workers help them give meaning to their work. It is also interesting to explore potential pedagogies to bring this topic to the business school classroom at different levels: the elaboration of case studies or the use of novels or movie narratives can be an inspiration to design teaching strategies to make students aware of their present and future work scenario, so they can reflect on meaningful work as professionals and future managers.