AMEMPLOY – Algorithmic Management in Employment

Introduction to Algorithmic Management in Employment 

—

by

Sini Mickelsson
in

Text written by Sini Mickelsson

Algorithmic management refers to the use of algorithms and digital tools to coordinate, control, and evaluate workers’ performance. Algorithmic management systems are increasingly taking over tasks traditionally performed by human managers, ranging from assigning work schedules to monitoring productivity and, in extreme cases, even making hiring and firing decisions. Directing and monitoring employees is an essential part of an employment relationship. Algorithmic management is nowadays deployed in platform work as well as in more traditional settings of white-collar and blue-collar environments. While algorithmic management can bring great benefits, novel technologies can also enable and automate significant interferences with, inter alia, workers’ privacy and autonomy.

What is Algorithmic Management?

Algorithmic management has been given various definitions by researchers from different scholarly backgrounds. In the context of workplace, it has been defined as “the diverse set of technological tools and techniques to manage workforces, relying on data collection and surveillance of workers to enable automated or semi-automated decision-making.” 1 From a human resource management perspective, algorithmic management can be defined as “a system of control where self-learning algorithms are given the responsibility for making and executing decisions affecting labor, thereby limiting human involvement and oversight of the labor process”2 and from information systems as “the large-scale collection and use of data on a platform to develop and improve learning algorithms that carry out coordination and control functions traditionally performed by managers.”3

To conclude, some of the distinguishing features of algorithmic management include the use of algorithms in connection with digital technologies (e.g., mobile apps and smartphone sensors), as well as a reliance on data collection and surveillance through technology. Algorithmic management systems typically involve automated or semi-automated decision-making.4 In a broader sense, organizations utilize algorithms to automate the oversight and optimization of various management functions, such as workforce scheduling, performance evaluation, and task assignment.

Benefits and Harms

Algorithmic management enables work to be “assigned, optimized, and evaluated through algorithms and tracked data.”5 It can enhance knowledge-based decision-making, reduce operational costs, and help organizations to adapt to market changes. Algorithmic management can also be used to benefit workers, for example, to improve occupational safety or to enforce working hours to maintain a healthy work-life balance. However, as a downside, the use of algorithmic management can also raise concerns, inter alia, about privacy, data protection, algorithmic bias, and detrimental work intensification. If used in an intrusive way, these systems can impact employee autonomy, morale, and well-being. Thus, striking a balance between efficiency and employee protection is critical. 

Employees are inherently in a vulnerable position in relation to their employer, and algorithmic management in the context of employment may shift this imbalance even further. While directing and monitoring employees in the workplace is an essential part of the employment relationship, surveillance and control based on new technologies can be more intrusive than traditional forms of physical surveillance. Algorithmic management also allows new forms of work and affects working conditions. Furthermore, the employer may be able to justify the use of algorithmic management due to the nature of employment relationships, and employees can be exposed to the unwanted consequences of using algorithmic management.

Examples of New Forms of Control

Ride-hailing platforms, such as Uber, and food delivery platforms, like Wolt and Foodora, are well-known examples of companies that utilize algorithmic management. These platforms can use sophisticated algorithms through applications to match drivers/couriers with customers, set dynamic pricing using pricing models, optimize routes to reduce wait times and improve efficiency, and use metrics and ratings to monitor driver/courier behavior. These systems could potentially even take disciplinary actions without human intervention, such as suspending access to accounts.6 Platforms can also algorithmically nudge users to modify their behavior, such as driving at specific times or locations.7 While these technologies increase operational efficiency and provide a novel way to perfom work, they have raised questions about driver/courier autonomy, fair compensation, and the form of the work, as the platforms can impose significant indirect control over the workers without providing them with an employee status (and protection).8

In addition to platform work, algorithmic management is also becoming more common in traditional work settings, such as insurance companies, hospitals, and warehouses.9 To increase efficiency and evaluate performance, task management can be handled through algorithms that push a new task to the employee as soon as the previous one has been completed. The employer can measure the time spent on the task and the number of tasks the employee completes during a defined working period. Organizations can also use algorithmic management systems to plan and optimize employee work shifts, taking into consideration, e.g., employee requests and shift information.

Algorithmic management can have significant implications for employees’ well-being in environments such as warehouses. For example, Amazon’s fulfillment centers are known for their extensive use of algorithmic management to optimize logistics and employee productivity. Each worker carries a handheld scanning device that tells them which item to pick next, from which shelf to retrieve it, and where to place it. Algorithms track and direct the movements of warehouse workers, determining their workflow and the pace at which they work, leading to public rankings of performance and layoffs. Workers have reported the pressure of meeting algorithmically set productivity targets, raising concerns about workplace conditions and the impact on physical and mental health.10

As another example, many call centers use algorithmic management systems to monitor employee performance, manage shift scheduling, and optimize call distribution. These systems can improve customer service efficiency, but they may create high-stress environments where employees are closely monitored and pressured to meet targets.11 In addition to well-being and autonomy, algorithmic management can also affect employees’ personal lives and finances. For example, several retail chains utilize algorithmic management to schedule employees based on predictive analytics of customer demand. Automated systems replace managers’ discretion over scheduling, and workers can be required to maintain open availability to accommodate the algorithm’s last-minute adjustments. While this can optimize staffing levels, “just-in-time” scheduling can create unpredictable work hours for employees.12

In a white-collar work context, remote work has given rise to new forms of intrusive monitoring, through which employers seek to ensure attendance and monitor employees’ performance. One example of this is employee monitoring and productivity software, which can track keystrokes, mouse movements, and application usage and even take periodic screenshots of the employee’s screen. Some systems even use webcam photos or record audio to ensure the employee is at their desk. Office workers can experience a form of algorithmic surveillance that measures not only their output (e.g., lines of code written, calls made) but also their time spent and engagement at the computer, with the software effectively instructing employees to stay constantly active to meet the algorithm’s expectations. Algorithmic management may also manifest in more subtle manners, as recommendations through email, suggesting scheduling time for deep work, or notifying about exceeding hours of specific tasks.13

Beyond surveillance, HR departments can use algorithms and AI to assist in managerial decisions about employees. For example, an AI-driven analytics system for internal talent management can infer employees’ skills and proficiency by analyzing various data points such as work and training history, and even social media or internal communications. Algorithms can also use performance reviews, engagement surveys, email patterns, and digital traces to flag “at risk” employees or to recommend promotions and raises to those who score well on performance metrics. While these AI tools promise more objective and efficient decision-making, they essentially turn employees into data points for algorithms to sort and evaluate.14 Organizations can also use similar systems for hiring and promotions (e.g., resume screening algorithms or sales performance algorithms that decide bonuses).

Conclusions 

The above examples illustrate the diverse nature of algorithmic management in contemporary work life. Algorithmic management has expanded from optimizing driver and courier behavior into warehouses, delivery fleets, retail stores, and offices. In each case, algorithms take on managerial functions: they distribute work assignments, set the pace of work, monitor compliance, and evaluate performance. Organizations seek to capture efficiency, consistency, and cost savings, but these deployments also raise important questions about privacy, transparency, fairness, and the impact on employee well-being.

  1. Alexandra Mateescu and Aiha Nguyen, ‘Explainer Algorithmic Management in the Workplace‘ (2019) Data & Society, <https://datasociety.net/wp-content/uploads/2019/02/DS_Algorithmic_Management_Explainer.pdf> accessed 22 April 2025 ↩︎
  2. James Duggan and others, ‘Algorithmic management and app-work in the gig economy: A research agenda for employment relations and HRM‘ (2017) 30 Human Resource Management Journal 114 ↩︎
  3. Alexander Benlian and others, ‘Algorithmic Management: Bright and Dark Sides, Practical Implications, and Research Opportunities‘ (2022) 26 Business & Information Systems Engineering 825 ↩︎
  4. See e.g. Benlian and others (n 3), 825; Mateescu and Nguyen (n 1), 3 ↩︎
  5. Min Kyung Lee and others, ‘Working with Machines: The Impact of Algorithmic and Data-Driven Management on Human Workers’ (2015) Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems ↩︎
  6. See, for example, Heather Steward, ‘Fired by AI? Just Eat UK couriers ‘deactivated for minor overpayments’ (2023) The Guardian <https://www.theguardian.com/business/2023/apr/22/fired-by-ai-just-eat-uk-couriers-deactivated-for-minor-overpayments> accessed 22 April 2025 ↩︎
  7. See, for example, Noam Scheiber, ‘How Uber Uses Psychological Tricks to Push Its Drivers’ Buttons’ (2017) New York Times <https://www.nytimes.com/interactive/2017/04/02/technology/uber-drivers-psychological-tricks.html>
    accessed 22 April 2025 ↩︎
  8. About the legal issues concerning employment status see, for example, a recent case of the Supreme Administrative Court of Finland KHO:2025:41. <https://www.kho.fi/fi/index/paatokset/ennakkopaatokset/1747729406840.html> ↩︎
  9. See for examples of logistics and healthcare Uma Rani, Annarosa Pesole and Ignacio Gonález Vázquez, ‘Algorithmic Management practices in regular workplaces: case studies in logistics and healthcare’ (2024) Publications Office of the European Union <https://op.europa.eu/en/publication-detail/-/publication/bff25994-cfc2-11ee-b9d9-01aa75ed71a1/language-en> accessed 26 April 2025 ↩︎
  10. Steven P. Vallas and others, ‘Prime Suspect: Mechanisms of Labor Control at Amazon’s Warehouse‘ (2022) 49 Work and Occupations 421 ↩︎
  11. See, for example, Virginia Doellgast and Sean O’Brady, ‘Making Call Center Jobs Better: The Relationship between Management Practices and Worker Stress‘ (2020) ILR Research Studies and Reports Faculty Publications – Labor Relations, Law, and History <https://ecommons.cornell.edu/server/api/core/bitstreams/e4ae98ca-9034-4f68-8be9-86885905f403/content> accessed 23 April 2025 ↩︎
  12. See, for example, Kaye Loggins, ‘Here’s What Happens When an Algorithm Determines Your Work Schedule‘ (2020) Vice Digital Publishing <https://www.vice.com/en/article/heres-what-happens-when-an-algorithm-determines-your-work-schedule/> accessed 24 April 2025; Jodi Kantor, ‘Working Anything but 9 to 5 Scheduling Technology Leaves Low-Income Parents With Hours of Chaos‘ (2014) The New York Times <https://www.nytimes.com/interactive/2014/08/13/us/starbucks-workers-scheduling-hours.html?module=inline> accessed 24 April 2025 ↩︎
  13. Tammy Katsabian, ‘The Telework Virus: How the COVID-19 Pandemic Has Affected Telework and Exposed Its Implications for Privacy and Equality‘ (2020) SSRN Electronic Journal. <https://ssrn.com/abstract=3684702> accessed 25 April 2025 ↩︎
  14. See, for example, Armin Granulo and others, ‘The Social Cost of Algorithmic Management‘ (2024) Harvard Business Review <https://hbr.org/2024/02/the-social-cost-of-algorithmic-management> accessed 25 April 2025 ↩︎

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *