How digital technology is reshaping the art of management

Abstract

This study describes how the digitisation of the workplace may contribute to the emergence of data-driven management, and how this, in turn, may affect work organisation and aspects of job quality, such as occupational health and safety. It summarises and defines the technologies that enable data-driven management and provides a snapshot of their diffusion across Europe. The paper shows how some technologies are fairly widespread, while other are found only in a minority of establishments. The descriptive analysis presented in this research report shows that technologies enabling data-driven management may have both positive and negative impacts on workers and working conditions. Better establishment performance, provision of training, greater job complexity and worker autonomy are among the positive impacts associated with the presence of some digital technologies enabling data-driven management at the establishment level. By contrast, lower workers’ well-being and a higher prevalence of reported psychosocial risks in the workplace are among the negative impacts. The paper suggests that workplaces can introduce a number of measures to mitigate the potentially negative impact of data-driven management on workers’ wellbeing.

Full text

Centre
JRC Technical Report
How digital technology is reshaping
the art of management
JRC Working Papers Series on
Labour, education and Technology
2022/05

This publication is part of a working paper series on Labour, Education and Technology by the Joint
Research Centre (JRC). The JRC is the European Commission’s science and knowledge service.
It aims to provide evidence-based scientific support to the Euro
pean policymaking process. The
contents of this publication do not necessarily reflect the position or opinion of the European
Commission. Neither the European Commission nor any person acting on behalf of the Commission
is responsible for the use that mig
ht be made of this publication. For information on the
methodology and quality underlying the data used in this publication for which the source is neither
Eurostat nor other Commission services, users should contact the referenced source. The
designations employed and the presentation of material on the maps do not imply the expression of
any opinion whatsoever on the part of the European Union concerning the legal status of any
country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or
boundaries.
Contact information
Name: Maria Cesira Urzì Brancati
Address: Joint Research Centre, European Commission (Seville, Spain)
Email: cesira.URZI-BRANCATI@ec.europa.eu
Tel.: +34 9540-38200
EU Science Hub
https://joint-research-centre.ec.europa.eu
JRC130808
Seville: European Commission, 2022
© European Union, 2022
The reuse policy of the European Commission is implemented by the Commission Decision
2011/833/EU of 12 December 2011 on the reuse of Commission documents (OJ L 330, 14.12.2011,
p. 39). Except otherwise noted, the reuse of this document is authorised under the Creative Com-
mons Attribution 4.0 International (CC BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/).
This means that reuse is allowed provided appropriate credit is given and any changes are indicated.
For any use or reproduction of photos or other material that is not owned by the EU, permission
must be sought directly from the copyright holders.
All content © European Union 2022
How to cite this report: Urzì Brancati, M., C., Curtarelli , M., Riso, S., Baiocco, S. How digital technology
is reshaping the art of management, European Commission, Seville, 2022, JRC130808.

1
How digital technology is reshaping
the art of management
Maria Cesira Urzì Brancati, Maurizio Curtarelli, Sara Riso, Sara Baiocco
Abstract
This study describes how the digitisation of the workplace may contribute to the emergence of
data-driven management, and how this, in turn, may affect work organisation and aspects of job
quality, such as occupational health and safety. It summarises and defines the technologies that
enable data-driven management and provides a snapshot of their diffusion across Europe. The
paper shows how some technologies are fairly widespread, while other are found only in a minority
of establishments. The descriptive analysis presented in this research report shows that the use of
technologies enabling data-driven management may have both positive and negative impacts on
workers and working conditions. Better establishment performance, provision of training, greater
job complexity and worker autonomy are among the positive impacts associated with the presence
of some digital technologies for data-driven management. By contrast, lower workers’ well-being
and a higher prevalence of reported psychosocial risks in the workplace are among the negative
impacts at the establishment level. The paper suggests a number of measures to mitigate the
potentially negative impact of data-driven management on workers’ wellbeing.
Authors: Cesira Urzì Brancati (JRC Sevilla), Maurizio Curtarelli (EU-OSHA), Sara Riso (Eurofound),
Sara Baiocco (DG EMPL)
Keywords: data-driven management; algorithmic management; psychosocial risks; digitisation;
work organisation; working conditions.
Joint Research Centre reference number: JRC130808
Related publications and reports:
• Urzí Brancati, C., Curtarelli, M., Digital tools for worker management and psychosocial risks
in the workplace: evidence from the ESENER survey, Seville: European Commission, 2021,
JRC125714.
Available at: https://joint-research-centre.ec.europa.eu/publications/digital-tools-worker-
management-and-psycho-social-risks-workplace-evidence-esener-survey_en
• Baiocco, S., Fernández-Macías, E., Rani, U. and Pesole, A., The Algorithmic Management of
work and its implications in different contexts, Seville: European Commission, 2022,
JRC129749
Available at: https://joint-research-centre.ec.europa.eu/publications/algorithmic-
management-work-and-its-implications-different-contexts_en
Acknowledgements:
The authors would like to acknowledge the contribution of Gjis van Houten for providing the data
upon which Chapter 4 is based and comments to previous versions of the report. Our gratitude is
also extended to the two reviewers, William Cockburn and Enrique Fernández-Macías, who provided
valuable suggestions and insights that greatly helped us improve the report.
The authors are also very grateful to Melisande Cardona for her thoughtful and valuable review,
comments and suggestions.

How digital technology is reshaping the art of management
2
Contents
Executive summary ..................................................................................................................................................................... 3
1 Introduction ........................................................................................................................................................................... 5
1.1 Policy context in the EU .................................................................................................................................... 7
2 Simplified analytical model and key definitions .............................................................................................. 9
2.1 Enabling technologies ..................................................................................................................................... 10
2.2 Data driven management ............................................................................................................................. 12
2.2.1 What does data-driven management look like? Examples from the literature ....... 14
2.3 Data driven management and its potential impacts on work: an overview .................... 15
2.3.1 Work organisation and workplace practices................................................................................. 15
2.3.2 Workplace wellbeing and occupational safety and health .................................................. 16
3 Enabling technologies for data-driven management: a snapshot ...................................................... 18
3.1 Data collection and processing across Europe ................................................................................. 19
3.1.1 Computers and connectivity .................................................................................................................. 19
3.1.2 Internet of Things (IoT) ............................................................................................................................. 19
3.1.3 Artificial Intelligence (AI) ......................................................................................................................... 21
3.1.4 Big data and Data Analytics .................................................................................................................. 22
3.1.5 Wearable devices......................................................................................................................................... 24
3.1.6 Management software applications and other enabling technologies ......................... 26
4 Data analytics and work organisation: insights from the ECS .............................................................. 28
4.1 Job complexity and autonomy ................................................................................................................... 28
4.2 Outsourcing and collaboration ................................................................................................................... 30
4.3 Variable pay schemes ..................................................................................................................................... 31
4.4 Training provision .............................................................................................................................................. 32
4.5 Establishment performance and workplace wellbeing ................................................................ 33
5 Wearables and other digital devices enabling data-driven management: impacts on
Occupational Safety and Health ....................................................................................................................................... 34
5.1 Enterprises adopting data-driven management enabling technologies ............................ 35
5.2 Psychosocial risks .............................................................................................................................................. 37
5.3 Workplace-level measures to prevent risks to worker´s safety and health ................... 39
6 Conclusions ......................................................................................................................................................................... 43
References .................................................................................................................................................................................... 45
List of figures .............................................................................................................................................................................. 49
List of tables ................................................................................................................................................................................ 50

How digital technology is reshaping the art of management
3
Executive summary
This report illustrates how the process of workplace digitisation facilitates the emergence of data-
driven management, and how this, in turn, may affect work organisation and aspects of job quality,
such as occupational health and safety. Workplace digitisation is the process through which
analogue or mechanical elements of the physical world are converted into bytes (and vice versa)
via means of sensors or rendering devices (Fernandez-Macias, 2018); it is enabled by technologies
such as the Internet of Things (IoT), Artificial Intelligence (machine learning or deep learning), or
wearables, which use cameras, sensors, and other devices to collect, transmit, store and process
massive amounts of data, not only on the work environment, but also on workers’ activity in itself.
This process is the basis for data-driven management in that the collected and processed data can
be used to assist or even (partly) automate managerial decision making.
Data-driven management is an umbrella concept that may refer to decision-making in the more
general process of goods production or service delivery (e.g. use of raw material or energy, logistics,
sales, etc.), or to specific work-related aspects (e.g. recruitment, workshift organisation, tasks
assignment, performance evaluation, etc.). An extreme example of how the availability of big data
can revolutionise management practices can be found in the digital labour platform economy,
where data-driven management takes the form of algorithmic management. In digital labour
platforms, most managerial functions are replaced by software algorithms, that is, ‘computer-
programmed procedures for transforming input data into a desired output’. Albeit typical of digital
labour platforms, data-driven forms of management are not uncommon in traditional workplaces,
especially in large firms and in sectors such as logistics (both in warehouse and delivery work) and
to a lesser degree retail, manufacturing, marketing, consultancy, banking, hotels, call centres, and
among journalists, lawyers and the police.
This report is the product of a collaboration between the JRC, EU-OSHA and Eurofound. It draws
from different data sources – The Community survey on ICT usage and e-commerce in enterprises;
The European Company Survey (ECS); and The European Survey of Enterprises on New and
Emerging Risks (ESENER) – to provide a snapshot of the distribution of technologies enabling data-
driven management across Europe. The distribution of enabling technologies suggests that the
potential for data-driven management practices to spread to traditional workplaces is quite large,
especially in some EU member states. In particular, some enabling technologies are fairly
widespread, while others are found only in a minority of establishments; for instance, Enterprise
Resource Planning (ERP) and Customer Relations management (CRM) software are present in
approximately more than a third of European enterprises with 10 or more employees; the Internet
of Things is present in approximately 29% of the EU27 enterprises; AI in 8%; and wearable devices
only in fewer than 6% of European enterprises. Finally, according to the ICT survey, 14% of the
enterprises with 10 or more workers analyse big data internally from any source, whereas
according to the ECS, data analytics is present in just about over half of the establishments. The
discrepancy with the ICT survey reported above probably relates to the more general understanding
of data analytics in the ECS survey where the relevant question makes no reference to use of ‘big
data’. It should also be noted that there is substantial heterogeneity across countries, and that the
presence of each technology is more likely in larger establishments.
In terms of impacts, the report examines the relationship of selected technologies with work
organisation and workplace well-being. In particular, the report investigates the use of data
analytics, that is, the ‘use of digital tools for analysing the data collected within the establishment
or from other sources’ (Eurofound and Cedefop, 2020) and shows its association with several
aspects of work organisation and workplace practices, namely: job complexity and autonomy,
collaboration with other establishments, pay schemes, training opportunities, establishment
performance and workplace well-being. The descriptive analysis shows that establishments using
data analytics (to improve production and monitor employee performance) tend to perform better,
provide higher levels of training, and are characterised by performance-related pay, greater job
complexity and worker autonomy; in addition, these establishments are more likely to engage in the

How digital technology is reshaping the art of management
4
design and development of new products or services. However, the use of data analytics is also
associated with somewhat lower worker wellbeing, especially if used for employee monitoring.
The report also examines other technologies that can be used for data-driven management, such
as wearables, and other devices determining the content or pace of work, or monitoring worker
performance, and their association with occupational health and safety, as measured by the
(reported) presence of psychosocial risks in the workplace. Estimates drawn from the ESENER
survey suggest that the presence of such technologies is associated with a higher prevalence of
reported psychosocial risks in the workplace, and in particular time pressure and long or irregular
working hours.
The report suggests a number of measures that workplaces can introduce to mitigate the
potentially negative impact of data-driven management on workers’ wellbeing. For instance, prior to
adopting a technology that can enable data-driven management, establishments can involve their
workforce and discuss its possible impacts with them. Having an action plan to reduce stress in
place would also help mitigate some of the negative effects on Occupational Health and Safety
(OSH).
At the EU level, it should be recognised that, besides platform workers, workers in traditional
workplaces may also be subject to risks related to algorithmic and data-driven management. For
this reason, all establishments could carry out risk assessments that cover the automated or semi-
automated worker monitoring and decision-making systems, and ensure that such systems do not
in any manner put undue pressure, or otherwise put at risk the physical and mental health of
workers.

How digital technology is reshaping the art of management
5
1 Introduction
Every time we go to work, clock in or scan our id-card, turn on our computer, or use our company’s
digital equipment, we produce an extraordinary amount of data that machines, digital devices,
sensors and so on are able to capture, store, and process, often without our knowledge. The vast
amount of data collected and processed through digital devices can then be used to make the
workplace safer, our performances more efficient, or the companies we work for more productive.
At the same time, there is a real danger that employers use these data to invade employees’
privacy, exert unfettered control, or make potentially unfair decisions affecting their working lives,
for example on wages or promotions. The process of data collection, storage and processing by
means of sensors or digital devices is what is commonly referred to as the “digitisation of the
workplace” (1), a key driver of the digital revolution (Eurofound 2018). Employers’ ability to back-up
their decisions with the information provided by these data is what can be referred to as “data-
driven management”. More formally, data-driven management can be defined as decision-making
backed up by a (perceived) objective representation of the situation provided by the (available)
data.
Like algorithmic management, data-driven management is a term described by Lee, Kusbit, Metsky,
and Dabbish (Lee, et al. 2015) referring to work settings, in which “human jobs are assigned,
optimized, and evaluated through algorithms and tracked data” (Lee, et al. 2015). More generally,
data-driven management may apply to decision-making in the process of goods production or
service delivery (e.g. use of raw material or energy, logistics, sales, etc.), as well as to specific work-
related aspects (e.g. recruitment, workshift organisation, tasks assignment, performance evaluation,
etc.), giving rise to what has been defined as the “datafication of the workplace” (Sánchez-
Monedero and Dencik 2019). In both cases, it affects workers and the workplace; however, when
data-driven management is applied to generic aspects of production or service delivery processes,
its impacts on work are indirect. For example, choices related to the timeline or the size of
production or concerning the overall business model can affect the level of employment,
employment relationships and the nature and quality of work. By contrast, when applied specifically
to work-related aspects, it has direct implications on work organisation, working conditions and job
quality (Kellogg, Valentine and Christin 2020, Wood 2021). This report only focuses on the direct
effects of data-driven worker management.
To a certain extent, worker management has strived to be data driven since the dawn of the
industrial revolution, with Taylor’s scientific management theories and the Gilbreths’ motion studies.
By breaking down jobs into smaller tasks and collecting enough information on how each task was
carried out, it was deemed possible to devise the optimal (most efficient, fastest and cheapest) way
to perform a task and impose it upon workers. In Taylor’s conception of scientific management,
workers were not allowed to make decisions about their work or evaluate actions that might
produce a better result; monitoring was a means to make sure work tasks were executed properly
and to prevent workers from slowing or sabotaging the modes of production; high performers
would be rewarded, whereas low performers would be dismissed. In this sense, a worker in a pin
factory of two centuries ago is not that different from a modern warehouse picker or a driver for a
ride hailing app. In addition, Taylor’s scientific management showed no concern for workers’
satisfaction or motivation and was only interested in increasing productivity (OTA 1987, Ball 2010,
Rosenblat, Kneese and Boyd 2014). Data were already crucial in that they invested employers with
the knowledge and determination of how work was performed, therefore increasing their ability to
control the workplace (OTA 1987).
(1) A formal definition of digitisation can be found in chapter 2.

How digital technology is reshaping the art of management
6
The difference in today’s workplace is that the availability of massive amounts of data coupled with
technologies that collect them in real time and the increased processing power of enabling
technologies has implications that are not only quantitatively, but also qualitatively different both
for worker management and for the concept of management in itself. On the one side, many
managerial functions, and especially the most standardised, can be increasingly automated; on the
other, workers who are constantly monitored and evaluated may be subject to an increase in their
workload, a decrease in their autonomy or creativity, and an invasion of their privacy, and therefore
experience a drop in their well-being, as well as in productivity.
An extreme example of how the availability of big data can revolutionise management practices
can be found in the digital labour platform economy, where data-driven management takes the
form of algorithmic management. In digital labour platforms, most managerial functions are
replaced by software algorithms, that is, ‘computer-programmed procedures for transforming input
data into a desired output’ (Kellogg, Valentine and Christin 2020, 341, Barocas, et al. 2014,
Gillespie 2014). As pointed out by Kellogg, Valentine and Christin (2020), based on extensive
evidence, algorithms – and the data they process – can be more comprehensive than any kind of
system previously used to support the management of workers precisely because they can rely on
large amounts of data collected through a wide range of digital devices.
As suggested by the literature and confirmed by the remainder of this report, albeit typical of
digital labour platforms, data-driven forms of management are not uncommon in traditional
workplaces, especially in large firms and in sectors such logistics (both in warehouse and delivery
work) and to a lesser degree retail, manufacturing, marketing, consultancy, banking, hotels, call
centres, and among journalists, lawyers and the police (Wood 2021)
Theoretical models, such as the Job Demand-Control (JDC) model by US sociologist Robert Karasek
(R. A. Karasek 1979) and some qualitative studies (Wood 2021, Bérastégui 2021) suggest a
potentially negative impact of data-driven management on workers’ well-being, as a result of
combined low control (autonomy) and high demand (as a result of intensification of work and
increased workload). However, not many studies have so far produced robust quantitative evidence,
with some exceptions. Urzì Brancati and Curtarelli (2021) carried out an empirical analysis drawing
on a unique set of establishments survey data (ESENER 3, described in the following chapter) to
investigate the impact of management technologies on occupational safety and health (OSH).
This report contributes to the literature by providing some descriptive evidence and a simplified
analytical framework to illustrate how, as a result of enabling technologies, such as the Internet of
Things (IoT), Artificial Intelligence (machine learning or deep learning), or wearables, the process of
workplace digitisation facilitates the emergence of data-driven management, and how this, in turn,
may affect work organisation and aspects of job quality, such as occupational health and safety.
Secondly, it provides a summary and a definition of the technologies likely to enable data-driven
management and a snapshot of their diffusion across Europe. Thirdly, the report zooms in on
selected technologies and looks at their potential impact on work organisation and workplace well-
being. In particular, chapter 4 investigates data analytics, that is, the ‘use of digital tools for
analysing the data collected within the establishment or from other sources’ (Eurofound and
Cedefop, 2020) and shows its association with several aspects of work organisation and workplace
practices, namely: job complexity and autonomy, collaboration with other establishments, pay
schemes, training opportunities, establishment performance and workplace well-being. Chapter 5
illustrates the relationship between technologies that can be used for data-driven management,
such as wearables, and other devices determining the content or pace of work, or monitoring
worker performance, and occupational health and safety. Chapter 6 concludes.

How digital technology is reshaping the art of management
7
1.1 Policy context in the EU
“A Europe fit for the digital age” is one of the European Commission’s six political priorities for the
years 2019-2024(2). One of its key pillar is the European strategy for data (2020)(3), which aims to
make the EU a leader in a “society empowered by data to make better decisions – in business and
the public sector”. At the same time, the strategy sets out the need to protect citizens from the
potential harm associated with emerging digital technologies by building “on a strong legal
framework – in terms of data protection, fundamental rights, safety and cybersecurity”
(COM(2020/66)). The strategy sets out the plan to create a single market for data; new regulation
for data governance to boost data sharing across sectors and Member States (4); a European Data
Act aiming to ensure fairness by setting up rules regarding the use of data generated by Internet of
Things (IoT) devices (5). In addition, the European Commission has proposed a declaration on digital
rights and principles to the European Parliament and Council (COM(2022)27)(6) in line with
European values.
Currently, the three most relevant legal references for data-driven management at the EU level are:
the (EU) 2016/679 General Data Protection Regulation (henceforth GDPR) (7); the 2021 proposal for
a ‘Directive on improving working conditions in platform work’ (COM(2021) 762) (8); and the 2021
Artificial Intelligence Act (COM(2021/206) (9), which builds on the Communication on Artificial
Intelligence of 2018 (COM (2018) 237) (10).
The GDPR provides a single legal interface to the entire EU digital market; it regulates the
processing of personal data, whereby ‘processing’ means ‘any operation or set of operations which
is performed on personal data or on sets of personal data, whether or not by automated means,
such as collection, recording, organisation, structuring, storage, adaptation or alteration, retrieval,
consultation, use, disclosure by transmission, dissemination or otherwise making available,
alignment or combination, restriction, erasure or destruction’ (GDPR, art. 4, definitions). Even though
the GDPR does not include specific protection for workers or employees, many of its provisions
apply to the employment relationship. According to article 19, processing personal data during a
person’s employment may take place if it is both necessary and proportionate. Consent is not used
as a legal ground for data processing in the workplace due to the power imbalance between
employers and employees: because the employer has authority over the employee and the
employee is financially dependent on the employer, permission from an employee to an employer
to process their personal data is not considered to be freely given (Todolí-Signes 2019). Article 22
prohibits fully automated decisions in terms of dismissing or punishing individual workers (based on
processed data) without meaningful intervention of a human manager (Wood, 2021). However, the
exact meaning of what would constitutes a meaningful intervention is debated.
According to Todolí-Signes (2019), only human intervention that can change the outcome produced
by a computer system by accessing different aspects can be interpreted as such. Article 22(3) also
(2) https://european-union.europa.eu/priorities-and-actions/eu-priorities_en
(3) EU Communication 2020/66: ‘A European strategy for data’
https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52020DC0066&from=EN
(4) Data Governance Act https://digital-strategy.ec.europa.eu/en/policies/data-governance-act
(5) Dta Act https://digital-strategy.ec.europa.eu/en/policies/data-act
(6) https://op.europa.eu/en/publication-detail/-/publication/6a2cfdb8-2101-11ec-bd8e-01aa75ed71a1/language-en
(7) Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with
regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data
Protection Regulation) (Text with EEA relevance). https://eur-lex.europa.eu/legal-
content/EN/TXT/PDF/?uri=CELEX:32016R0679&from=EN
(8) Proposal for a Directive of the European Parliament and of the Council on improving working conditions in platform work,
https://ec.europa.eu/commission/presscorner/detail/en/ip_21_6605
(9) Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence
(Artificial Intelligence Act) and Amending Certain Union Legislative Acts, COM (2021) 206 final (April 22, 2021). https://eur-
lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52021PC0206
(10) Communication from the Commission to the European Parliament, The European Council, The Council, The European Economic and
Social Committee and the Committee of the Regions, Artificial Intelligence for Europe, COM (2018) 237 final (June 26, 2018).

How digital technology is reshaping the art of management
8
sanctions the right to an explanation; when an automated decision is taken, the data controller has
the obligation to inform the data subject of the reasons that led to that decision and provide details
as to what parameters it has used to reach the decision in question.
The 2021 proposal on platform work (COM(2021)762) tries to overcome some of the limitations of
GDPR regarding automated decisions and establishes a number of requirements for digital labour
platforms to strengthen (platform) workers’ rights. Chapter III, article 6, 8 and 9 refer to the use
and key features of automated monitoring systems and automated decision-making systems and
require that platforms inform, explain and consult the workers or their representatives when such
systems affect their working conditions. In addition, article 6 reiterates the limits to data processing
already established in the GDPR, namely, that platforms are not allowed to process data that are
not strictly necessary for the performance of the worker’s contract, such as data on conversations,
on the health or emotional state of the worker and so on. Article 7 recognises the potential time
pressure that automated systems might create for platform workers and establishes that digital
labour platforms must “evaluate the risks of automated monitoring and decision-making systems
to the safety and health of platform workers and ensure that such systems do not in any manner
put undue pressure on platform workers or otherwise put at risk the physical and mental health of
platform workers”. To be noted that a similar disposition is not (yet) envisaged for workers outside
digital labour platforms.
Finally, the 2021 Artificial Intelligence Act (AI ACT) lays down the basis for the regulation of
algorithmic management, and in particular it addresses i) AI systems used for recruitment; and (ii)
AI systems used for promotion and termination of contracts for work, for allocating tasks, and for
monitoring and evaluating workers’ behaviour (p.26, n. 36). The regulation for high risk AI systems
set out in chapter 2 of the Act essentially sets out design criteria that providers must put in place: a
risk management system that identifies, evaluates, and mitigates risks (article 9); criteria for data
governance (article 10) including “examination in view of possible biases” and that “Training,
validation and testing data sets shall be relevant, representative, free of errors and complete”. It
should be noted that, according to the AI Act, most of the obligations fall upon the AI provider, that
is, the person/body that develops or markets the AI, and not on the user (or the employer) who
simply has to make sure the instructions are followed and the data inputted are relevant.
As this report shows in the next section, the potential for data-driven management practices to
spread to traditional workplaces is quite large, especially in some EU member states, and it may
carry some unwanted side-effects, in terms of increased time pressure, work intensity and so on.
For this reason, EU legislation must be comprehensive and cover all potential impacts of data-
driven management on workers. The remainder of this report investigates some impacts of data-
driven management on workers that should be taken into account by the legislator – besides
automation of managerial functions, privacy issues and biases that may lead to discrimination.

How digital technology is reshaping the art of management
9
2 Simplified analytical model and key definitions
The digitisation of processes ―that is, the conversion of analogue or mechanical elements of the
physical world into bytes (and vice versa) via means of sensors or rendering devices― is one of the
main channels through which the digital revolution is changing the nature of work (Eurofound
2018). It is driven by a number of technologies able to collect, store, process and analyse data. Not
all enabling technologies considered in this report – and described in the next section - are able to
perform all data functions: some can only collect and store data, while others can only process and
analyse them; for this reason, firms tend to adopt several technologies at once and exploit
complementarities.
Figure 1 summarises how, as a result of enabling technologies, the process of workplace
digitisation facilitates the emergence of data-driven management, and how this, in turn, may affect
work organisation, workplace practices and well-being, establishment performance and aspects of
job quality, such as occupational health and safety.
The next subsections will describe and define the concepts expressed in Figure 1.
Figure 1. Simplified analytical approach
Source: Authors’ elaboration
Enabling Technologies
Artificial intelligence
Wearable devices
Internet of Things
Management software
Data Analytics
Personal Computers
(desktops and laptops)
Data
Data-driven
Management
Leading and
directing
Staffing
Evaluating
Planning
Organising
Workplace
Digitisation
Direct Impacts
On Work
Establishment
performance
Workplace practices
Workplace well-being
Work organisation
Occupational health
and safety

How digital technology is reshaping the art of management
10
2.1 Enabling technologies
To understand how the digital technologies considered in this report enable data-driven
management, we first group them according to the functions performed on/with data (Klingenberg,
Viana Borges and Valle Antunes 2019). In particular, some technologies allow for data collection
and transmission, while others allow for data processing and analysing; many of them, however,
can do all functions. The former are technologies that collect data from their environment and are
able to transfer it from one place to another; typical examples are the Internet of Things or
wearable devices. The latter are technologies that transform the data they are fed with into an
output, still in the form of data; examples are Artificial Intelligence, or computer software. The large
amount of data involved in these processes tends to require data storage technologies, such as
cloud services (11), which often accompany the other types of technologies mentioned below, but
fall outside the scope of this report.
Digital devices often embed many of these technologies, to allow for several data functions to be
performed simultaneously. Digital devices, such as computers and wearables, can potentially
perform data collection and transmission, as well as data processing if they embed technologies
that allow these functions: often wearables used in warehouses collect and transmit data, but they
also process them and send automated instructions based on the processed data. The same applies
to personal computers, which can collect and transmit data (providing that they are programmed to
do so and connected to the internet/intranet), and process data with the software included in the
device. Digital devices are therefore crucial in the implementation of data-driven forms of
management.
Data-driven management relies on interconnectivity and combination of different types and layers
of technologies that in the end are all interrelated. In the implementation of data-driven
management, these technologies are often adopted together, allowing the use of data within an
organisation to coordinate operations, or work specifically. Digital devices are the ‘visible’ part of
this technology combinations, embedding different layers of technologies and those to perform
different data functions at the same time.
Internet of things
Internet of things (IoT) is a system of interconnected sets of entities encompassing smart devices,
sensors, and products collecting and analysing data and autonomously performing actions; it relies
on ubiquitous and pervasive connectivity, computing capability, and large amounts of data. The
fields of application for IoT technologies are numerous and diverse, including transportation and
logistics; healthcare; smart environments (home, office, plant); and, of course, the personal and
social domain (Atzori, Iera and Morabito 2010, Eurofound 2020, Khodadadi, Dastjerdi and Buyya
2016). The term IoT was coined by British technology pioneer Kevin Ashton in 1999 to illustrate
how connecting Radio-Frequency Identification (RFID) tags to the Internet would allow the
collection, processing and transmission of data without the need for human intervention, thereby
greatly reducing costs (Kramp et al, 2013). Although IoT systems can be used in any setting, when
used in the workplace they collect data on the production or service delivery processes, and
therefore enable data-driven forms of management.
Artificial Intelligence
The European Commission Artificial Intelligence Act (AI Act) defines AI as software that, for a given
set of human-defined objectives, can generate outputs such as content, predictions,
(11) According to the definitions in the Community survey on ICT usage and e-commerce in enterprises 2021 – General outline of the
survey “Cloud computing refers to ICT services that are used over the internet to access software, computing power, storage
capacity etc”.

How digital technology is reshaping the art of management
11
recommendations, or decisions influencing the environments they interact with. AI systems are
developed with one or more of the techniques and approaches listed below:
— machine learning, including supervised, unsupervised and reinforcement learning, using a wide
variety of methods including deep learning
— logic and knowledge-based approaches, including knowledge representation, inductive (logic)
programming, knowledge bases, inference and deductive engines, (symbolic) reasoning and
expert systems;
— statistical approaches, Bayesian estimation, search and optimization methods
The AI technologies surveyed by Eurostat’s ICT Survey include: text mining, computer vision, speech
recognition or image recognition, natural language generation, machine learning or deep learning.
The technologies surveyed also include AI-based software for robotic process automation and
technologies enabling physical movements of machines thanks to autonomous decisions based on
observation of surroundings (e.g. self-driving vehicles, autonomous robots or drones).
Data analytics
In statistics, Data Analytics (DA) is the process of examining datasets in order to draw conclusions
about the information they contain, increasingly with the aid of specialised systems and
software(12). Data analytics can be categorised into descriptive, predictive and prescriptive analytics
(T. Davenport December 2013, Sharda, Delen and Turban 2015, Berndtsson, et al. 2018). While the
descriptive approach uses a dataset for a better understanding of past events, predictive and
prescriptive analytics are more advanced techniques. Predictive analytics are used to predict future
outcomes and trends by uncovering previously unidentified data patterns, and prescriptive analytics
assesses current data to help companies automate decision making. These techniques can be
applied to sets of big data that often contain unstructured and semi-structured data. Both
predictive and prescriptive analytics increasingly exploit AI techniques (machine learning or natural
language processing) for more precise predictions and larger scale business optimisation. The main
difference between artificial intelligence and data analytics is that, while both techniques are able
to generate predictions, an AI system is also capable of observing its environment, learning, and
based on the knowledge and experience gained, take intelligent actions or propose decisions.
Autonomy of decision processes and interaction with other machines and humans are other
dimensions that need to be considered (Annoni, et al. 2018).
Wearable devices
The reflection and orientation paper on smart wearables published in 2016 by the European
Commission defines wearable devices as “body-borne computational and sensory devices which
can sense the person who wears them and/or their environment” (13). Wearable devices can interact
with each other directly either wirelessly or through another device (such as a smartphone). Smart
wearable devices may have control, communication and data storage functions. A wearable may
contain several types of sensors, integrated as a system on a module. The sensors sense
physiological phenomena from the user or from surrounding environment. The wearable processes
this information and acts (e.g. display some visual content, play some audio content, send an alert,
provide force feedbacks) on the user.”
Management software applications
Management software applications, such as software for Customer Relationship Management
(CRM) and Enterprise Resource Planning (ERP), have long been used for data elaboration to inform
(12) https://ec.europa.eu/eurostat/cros/system/files/dawosreport_0.pdf
(13) Smart Wearables: Reflection and Orientation Paper, available at: https://ec.europa.eu/newsroom/document.cfm?doc_id=40542

How digital technology is reshaping the art of management
12
management decision-making, relying on data collected through interconnected computers across
different business functions (Sánchez-Monedero and Dencik 2019).
CRM refers to an integrated management information system that is used to schedule, plan and
control the sales and pre-sales activities in an organization. CRM systems comprise of hardware,
software and networking tools to improve customer tracking and communication. The software
processes data about customers to improve customer service by facilitating communication, for
instance with more user-friendly mechanisms to log a complaint, or pre-empt problems (14).
ERP is a commonly used software application to facilitate the flow of data across several functions
of an enterprise, such as planning, purchases, marketing, sales, customer relationship, finance and
human resource (15).
Other management technologies
This report also presents descriptive statistics on the distribution and potential impacts of other
technologies enabling data-driven management, such as ‘machines, systems or computer
determining the content or pace of work’ or ‘monitoring performance’.
These technologies can be employed to facilitate the management of different aspects of an
organisation, from security to energy consumption, from the supply of raw materials or components
to marketing and sales. When enabling technologies are adopted and used for purposes that
directly involve workers, for example for monitoring purposes or for Human Resources (HR)
practices, the implications of data-driven management are directly affecting workers. Otherwise the
impacts are likely to be indirect. Regardless of the purpose for which a technology is introduced in
an organisation, its deployment should take into account both intended and unintended
consequences, direct and indirect effects for workers (Edwards and Ramirez 2016).
2.2 Data driven management
In the words of F.W. Taylor, management is an art of knowing what to do, when to do and see that
it is done in the best and cheapest way. In the context of a firm, it can be described as the process
by which an organisation uses resources, mainly people, to efficiently achieve a set of goals (Cole
and Kelly 2020). It then follows that under data-driven management, the use of resources is made
efficient as a result of decisions backed up by data. This definition is, however, somewhat
tautological, so to try and understand what data-driven management actually means, the best
option is to break down the concept of management itself into its components, that is, into
managerial functions, and see to what extent each function can be informed or automated as a
result of data availability.
The first attempt to classify managerial activities into specific functions dates back to the French
engineer Henri Fayol at the beginning of the 20th century. According to Fayol, the five essential
functions of management were: forecast & plan, organise, command, coordinate, and control. The
more modern and commonly accepted version of the five functions of management was proposed
by Koontz and O'Donnell (1972) who replaced harsh words such as “command” with more
employee friendly terms such as leading and directing. The basic ideas remained however
unchanged. According to Koontz and O’ Donnell, the five essential functions of management are:
1. Planning: the process of appraising the present position of an organisation (setting the
premises); anticipating and forecasting the future environment; formulating long term and short
term goals; selecting the strategies for their realization, and evaluating alternatives. Planning
involves thinking creatively about issues at hand.
(14) https://ec.europa.eu/eurostat/cros/content/Glossary%3ACustomer_Relationship_Management_%28CRM%29_en .
(15) https://ec.europa.eu/eurostat/statistics-explained/index.php?title=E-business_integration#Enterprise_resource_planning_.28ERP.29 .

How digital technology is reshaping the art of management
13
2. Organising: the process of identifying and classifying activities; assigning duties; and providing
the necessary authority to carry out such duties – in other words, it involves the design of
individual jobs within an organisation. The organising function is vital because it provides an
organisation with a structure.
3. Staffing is the process of filling positions within the organisation through proper and effective
selection. The staffing process has the following steps: personnel acquisition (which may
include forecasting staffing needs, recruitment, and selection) and personnel retention (which
may include transfer and promotion, training, and remuneration).
4. Directing and leading is the process in which the manager instructs, guides and oversees
workers’ performance to achieve the planned goals. It involves influencing and motivating and
can be considered the executive function of an organisation (whereas the other managerial
functions can be seen as preparatory).
5. Controlling is the process of ensuring the objectives of the organisation are met. Controlling
ensures that all other functions are executed correctly and everything occurs in conformity with
standards. An efficient system of control helps to predict deviations before they actually occur.
It includes the establishment of standard performance, measurement and evaluation of actual
performance, and corrective action.
These activities are common to each and every manger irrespective of their level or status.
While, to the best of our knowledge, not much has been written on the managerial functions
affected by data-driven management, the literature on algorithmic management can provide some
useful insights.
Data-driven management is an umbrella concept that includes algorithmic management.
Algorithmic management is defined as the use of computer-programmed procedures (i.e. the
algorithms), which can be both AI or non-AI powered, to coordinate labour input in an organisation
(i.e. the management) (Baiocco, et al. 2022). As such, algorithmic management is a particular form
of data-driven management that is applied specifically to worker management and it involves
some degree of data processing through algorithms, in addition to data collection. In general terms,
an algorithm is ‘a set of rules that must be followed when solving a particular problem’. For the
scope of this report, however, the term is more specific and refers to software algorithms, that is,
‘computer-programmed procedures for transforming input data into a desired output’ (Kellogg,
Valentine and Christin 2020). Algorithmic management is most developed in digital labour
platforms, where it leads to higher degree of automation of management functions in comparison
to its applications in other work settings (Wood, 2021). However, even in digital labour platforms,
algorithmic management does not reach a full degree of automation, but rather requires a
combination of human activity and algorithms (Wood, 2021).
Whether automating or assisting, algorithms can deeply transform several management functions.
Kellogg and co-authors refer to the “6 Rs” mechanisms through which algorithms coordinate work
(i.e. restricting and recommending, recording and rating, replacing and rewarding), while Wood
(2021) mentions three main functions, namely direction, evaluation and discipline of workers
through algorithmic management. Finally, Baiocco, et al. (2022) claim that the introduction of
algorithms to perform the different functions of management implies a fundamental departure
from traditional human management practices, with new forms of control, monitoring and
disciplining of workers emerging on digital labour platforms.
The main difference between algorithmic management and generic forms of data-driven
management is that the former implies more or less a substitution of human (managerial) activities
with algorithm based automated procedures, whereas under the latter, management can be not
only complemented, but also and especially enhanced and improved due to the use of data. In other
words, under algorithmic management, especially if AI-powered, algorithms can be programmed to
make autonomous decisions regarding workers (within the limits set by the law, as illustrated in
section 1.1), whereas the concept of data-driven management is somewhat more nuanced and can

How digital technology is reshaping the art of management
14
be applied also to less sophisticated technologies that provide managers with information based on
the collected and processed data, and help them make better decisions.
2.2.1 What does data-driven management look like? Examples from the literature
Logistics and warehousing is arguably one of the most automated and digitised economic sector
and offers a plethora of examples of data-driven management to draw from. Typically, it is the use
of the Internet of Things or wearable devices - worn on the head, wrist, chest, fingernail and feet -
that provide the organisation with a variety of performance, environment and body data (Ball
2021).
Alessandro Delfanti’s study of Italian Amazon warehouses reports how data collected by handheld
and wearable devices are used to direct, control and evaluate workers (Delfanti 2019). Sensors
track workers' every movement around the floor, and the collected data are used by algorithmic
systems to allocate tasks, re-route workers (based on their location and the proximity to item’s
position on the shelf), and set a maximum amount of time a worker has to complete their task
(referred to as ‘pick rate’). At the same time, warehouse managers can see real-time data on
workers’ pick rate, and use such data to evaluate their performance, set standards, discuss firm
productivity during meetings, and so on (Wood, 2021). Incidentally, managers can also see how
many times a worker went to the bathroom, giving rise to what is known as ‘function creep” (16).
Another example of data-driven management is illustrated in the study of a port terminal in the
South of Spain (TTA Algeciras) conducted by Grande, Vallejo-Peña and Urzì Brancati - also featured
in Eurofound’s latest report on digitisation (Eurofound 2021). The terminal uses the Internet of
Things and other ancillary digital technologies to track and maintain containers and other
equipment. It is endowed with a tailor made Terminal Operating System (TOS) - an advanced
computer system managed remotely - which generates automated orders by processing all the
data received through IoT technology. In addition, the TOS can process and analyse data at high
speed, providing useful information (e.g. a longitudinal analysis of the energy consumption of
machines) to engineers and managerial staff. In this way, further improvements can be introduced
by management. Finally, data collected via IoT allows engineers and other technical staff to apply
predictive maintenance, as they can observe the operation of machinery in detail. The system
provides warnings of any deficiencies that could develop into breakdowns, or on parts that need to
be replaced in advance, based not only on the time elapsed but also on their condition. Eurofound’s
report also describes other examples of IoT based data-driven management, notably for the
Estonian gelling agent producer Est-Agar and the Italian Centro Seia plant nursery. In both cases,
the IoT technology was introduced specifically to monitor the quality and quantity of the work done
by employees. In Est-Agar, the IoT-based monitoring influenced to some extent employees’
behaviours at work, for example reducing informal exchanges and interactions during working time.
In Centro Seia, the collected data resulting from the monitoring was used to inform important
employment decisions, for instance about promotions and financial incentives (Eurofound 2021).
In the logistics industry, transport firms use real-time fleet management systems and GPS tracking
in vehicles. Whilst this is partly to comply with working time regulations and to manage liability in
the event of an accident, the performance of the driver and their vehicles are also monitored by
these means (Levy 2015). Amazon installed AI powered cameras on its vans to, ostensibly, improve
drivers’ (or ‘delivery service partners’) safety (17). The cameras, named Driver(i), are able to capture
the road ahead, the driver, and both sides of the vehicle; when detecting risky behaviour by drivers
they will bark verbal warnings like “Distracted driving,” “No stop detected” and “Please slow down”
(16) Function creep” occurs when information is used for a purpose that is not the original specified purpose.
(17) https://vimeo.com/504570835/e80ee265bc .

How digital technology is reshaping the art of management
15
(Wood, 2021). The data gathered by Driver(i) are also used to evaluate workers' performance
scores and can affect their chances of getting bonuses, extra pay, and prizes (18).
Among digital labour platforms, arguably the best-known example of data-driven management is
provided by the taxi hailing app Uber. Through its smartphone app, Uber collects GPS, gyroscope
and accelerometer data from drivers and store them long term for a number of purposes, such as
directing, evaluating and disciplining workers, as well as informing its research in self-driving cars
(Ball 2021, Wood 2021). Wood (2021) also explains that the Uber app directs the route that divers
are supposed to take and fine them (at least in the US) if they follow a less efficient route or how it
directs drivers to charge customers a maximum fare based on distance, city and labour supply.
2.3 Data driven management and its potential impacts on work: an overview
As mentioned in section 2.1, the impacts of data-driven management on work take place through
changes in managerial functions – in other words, when based on the extensive data collected by
digital tools, managerial functions such as directing, organising, evaluating workers and so on, may
be intrinsically transformed and that has an impact on work organisation, workplace practices,
establishment performance, and occupational safety and health. For example, with the data-driven
management, rewarding and penalising can happen in real-time, underperformance can be
instantaneously identified and can result in dismissal and immediate reallocation of job tasks to
other workers (Kellogg, Valentine and Christin 2020). Data-driven management, but especially
algorithmic management, is reported to reduce workers’ autonomy and control over how a job is
done. Incentivising and influencing workers’ behaviours towards desired goals through a system of
nudges and penalties (Kellogg, Valentine and Christin 2020) can result in a number of issues for
workers’ mental health. Data-driven forms of management could also affect firms’ business
models, as a result of a more centralised control over processes while at the same time
fragmenting production or service delivery, in time and space or across several actors (Adams
Prassl 2019). This can lead to new ways of organising work and cooperation within and among
enterprises. These ways could include outsourcing and subcontracting (Loebbecke and Picot, 2015).
2.3.1 Work organisation and workplace practices
Work organisation, and how jobs are designed to support it, is part of the management of the
employment relationship (ECS 2019) (19). In its broadest sense, work organisation is about the
division of labour, the coordination and control of work. It encompasses the tasks performed, who
performs them and how they are performed in the process of making a product or a service. It
refers to ‘how work is planned, organised and managed within companies via production processes,
job design, task allocation, rules, procedures, communication, responsibilities, management and
supervisory styles, work scheduling, work pace, career development, decision-making processes,
interpersonal and interdepartmental relationships (Eurofound 2022).
Within the framework provided by Eurofound and Cedefop’s European Company Survey, aspects of
work organisation potentially affected by data-driven management are collaboration and
outsourcing and job complexity and autonomy. In addition, the impact of data-driven forms of
management can lead to new ways of organising work and cooperation within and among
enterprises, such as outsourcing and subcontracting (Loebbecke and Picot 2015).
Job complexity and autonomy are linked to the managerial function of control. Employers (or
management) can exert direct control by designing jobs which give individual workers or teams
more or less autonomy or challenges (problem solving). They can also exert indirect control through
incentives. Managerial control is also reflected in the way tasks and responsibilities are allocated
and in the way teamwork is used (Eurofound and Cedefop 2020).
(18) https://www.vice.com/en/article/88npjv/amazons-ai-cameras-are-punishing-drivers-for-mistakes-they-didnt-make
(19) European Company Survey 2019: Workplace practices unlocking employee potential.

How digital technology is reshaping the art of management
16
Often, monitoring is linked to incentive and penalty mechanisms. Traditional forms of management
rely on systems of incentives and penalties in which workers with desired behaviours are rewarded
with promotions, increased pay, progressively higher responsibility, more benefits, preferred tasks,
etc. while those who do not behave as desired can be dismissed according to existing regulation,
policies, etc. (Ball 2021). The reward system and its quick responsiveness can lead to workers’
frustration and stress when the system is intentionally not transparent; in addition, it can result
more generally in a significant increased (both subjective and objective) job insecurity and work
intensification (Bérastégui 2021, Eurofound 2022).
2.3.2 Workplace wellbeing and occupational safety and health
One of the functions of management that is more likely to be data-driven consists in the control
(surveillance and monitoring) of workers. Henderson, Stanford and Swann (2018) explain how
employers use tools to directly control the pace and intensity of work in combination with
monitoring and supervising systems, to extract maximum work effort from employees.
Data-driven management can result in higher isolation and lack of social support of workers, for
example in cases where human supervisors are replaced by algorithms and tasks are allocated
automatically on a screen of a handheld device, with little to no time for interaction. When workers
are deprived of organisational support (20), they can experience increased stress, anxiety and even
burnout (Bérastégui 2021). Co-workers and supervisors’ support are both positively related to
organisational support, though supervisor support has a significantly stronger effect because
supervisors are seen as acting on behalf of the organisation through their responsibilities of
directing and evaluating workers’ performance (Kurtessis, et al. 2017). A supportive work
environment is characterised by positive social interactions helping workers to cope with uncertainty
or stressful circumstances.
The literature has found that management associated with technologies that determine the content
or pace of work is linked to intensification of work, and therefore an increase in workers’ stress and
mental health issues. For example, in working contexts that rely on algorithmic management,
evidence has shown that the speed of computing processes and the data-driven directions and
instructions can result, for workers, in a hectic pace of work to meet deadlines and achieve targets,
hence resulting in quantitative overload (21) (Bérastégui 2021). Working under time pressure to
meet tight deadlines has been linked to high levels of strain, stress, depression and anxiety (Cooper
and Roden 1985, Kushnir and Melamed 1991, Narayanan, Menon and Spector 1999) that may
ultimately result also in a lower level of worker’s performance (Westman and Eden 1992). To make
things worse, workers subject to algorithmic management are more likely to be exposed to
information overload, and are therefore more likely to experience feelings of guilt and anxiety
about their inability to meet the demands placed upon them. This can result in “relentless self-
exploitation, often justified by both workers and employers as ‘flexible working’” (Bérastégui 2021,
36). But if overload is an issue, underload is nonetheless also of concern as it is associated with
frustration, stress and anxiety. The breakdown of jobs into very simple or standardised tasks, and
the overall simplification of work favouring the development of repetitive tasks, are related to
qualitative underload (that is, performing tasks and assignments that are well below worker
abilities), and therefore are associated with higher levels of psychological distress and job
dissatisfaction (Bérastégui 2021). Urzí Brancati and Curtarelli (2022) find that the presence of
management technologies in an establishment is associated with a higher number of psychosocial
risks and that this relationship is only partly mediated by OSH preventive measures. In particular,
only allowing workers to take more decisions on how to do their job and having an action plan to
(20) Social support refers to the degree to which individuals perceive that they are valued and supported in the workplace ( (Eisenberger,
et al. 2002, Kossek, et al. 2011, Sias and Gallagher 2009)
(21) Quantitative overload can be defined as performing a high amount of work in a given timeframe.

How digital technology is reshaping the art of management
17
reduce work-related stress are associated with lower reported psychosocial risks. In other words,
being proactive and having a plan in place to pre-empt problems can be the most effective way to
reduce stress in the workplace.
Working under data-driven forms of management, and especially algorithmic management,
appears to be particularly demanding and straining for workers; it can therefore have a significant
impact on workers’ mental health, which is detrimental also for the organization in which they are
employed. Workers can feel frustrated when recommendations are unintelligible to them, or feel
powerless, especially when the algorithmic direction is perceived as unfair. When algorithms allow
specific behaviours while preventing others (Kellogg, Valentine and Christin 2020) workers can
experience alienation, losing control over their work and being “deprived of the right to conceive of
themselves as directors of their own actions” (Kellogg, Valentine and Christin 2020, 375).
Applying the Job Demand-Control (JDC) model by Karasek (1979) on occupational stress, the
combined low control (as explained above) and high demand (as a result of intensification of work
and increased workload), which feature jobs under data-driven management practices and in
particular algorithmic management, would result in “high-strain” jobs, associated to increased
stress and health issues for the worker.
In the developed Job-Demand-Control-Support (JDCS) model (Karasek and Theorell 1990, Johnson
and Hall 1988), support from the organization can be seen as a mitigating factor to limit the impact
of “high strain” jobs on workers’ health. Yet, it should be taken into account that data-driven forms
of management frequently result in isolation and lack of support from supervisors and managers.
Therefore, managers’ and supervisors’ support cannot be considered a mitigating factor of the
impact of “high strain” jobs on occupational health in this context. Yet, initiatives or measures can
be adopted by the employer to mitigate the impact of these forms of management on occupational
health.
Figure 2 Job Demand-control model
Source: Adapted from Karasek (1979)
Research dating back to the 1990s already demonstrated a clear link between performance
monitoring – carried out with or without the use of digital devices – and employee stress (Aiello
1993, Aiello and Svec, Computer monitoring of work performance: Extending the social facilitation
framework to electronic presence 1993). The burden of feeling or being under constant monitoring
can be detrimental for the worker’s mental health. Invasive employee control and monitoring may
also lower employee morale and result in higher employee turnover (Ajunwa, Crawford and Schultz
2017). On the other hand, performance monitoring may have a positive impact if workers can use
feedback to advance their careers, get better pay, and have their health and safety protected (EU-
OSHA 2018). However, intensive monitoring is also reported to increase concerns of job insecurity,
pushing workers to work long hours out of fear of not achieving the targets set by the employer. If
workers are concerned that decisions are made based on data that they have neither access to nor
power over, and that such data may be used for appraisal and performance evaluation and can
impact negatively on career development, or may be used for workplace restructuring, job

How digital technology is reshaping the art of management
18
description changes or even firing, their concerns can result in stress and be detrimental for mental
health (EU-OSHA 2019).
If job security can be defined as an employee’s perception that their job (or an important aspect of
their job) is secure (Burchell 2014), job insecurity refers to the potential loss of the job itself (or of
one or more of its key components e.g. pay or role/responsibilities). This definition underlies the risk
of loss of something of value, anticipation of the consequences of losing their job and
powerlessness (Rogelberg 2017). Job insecurity is one of the most important psychosocial risk
factors and a major work-related stressor linked to several negative health outcomes, such as
poorer mental health (Sverke, Hellgren and Näswall 2002, Bonde 2008, Stansfeld and Candy 2006).
Perceived job insecurity may result in burnout, depression and anxiety and can also be related to
poorer self-rated physical health with issues including fatigue or pain, and cardiac problems such as
high blood pressure or ischemic heart disease (Hellgren and Sverke 2003, Burgard, Brand and
House 2009).
According to a recent systematic review of electronic monitoring and surveillance in the workplace,
wearable devices have two applications in the workplace: as part of corporate wellness programs
and as part of performance management in highly automated settings such as logistics
warehouses (Ball 2021). Employees who take part in corporate wellness programmes are
encouraged to use output (data) from their wearables to compete or collaborate with colleagues;
though the data collected by corporate wellness wearables are not seen by the employee only, but
by management too, and can used to evaluate workers’ performance.
Wearables used to monitor the physiological responses of employees, such as heart-rate variability
or electrodermal activity (perspiration), may serve as a proxy for employees’ stress responses or
fatigue. Activity wristbands can thus alert employees to take breaks from work when their
performance degrades in order to optimize productivity (Maltseva 2020).
3 Enabling technologies for data-driven management: a snapshot
As pointed out in chapter two, data-driven management is enabled by the use of digital
technologies that collect, store, process and analyse data. The distribution of such enabling
technology can therefore provide a rough idea of the potential diffusion of the phenomenon itself.
Three main sources are used to show the distribution of digital devices and technologies enabling
data-driven management across Europe: Eurostat’s Community Survey on ICT usage and e-
commerce in enterprises (henceforth the ICT survey); the European Company Survey (ECS), and
European Survey of Enterprises on New and Emerging Risks (ESENER).
Box 1 : The Community survey on ICT usage and e-commerce in enterprises
The Community survey on ICT usage and e-commerce in enterprises is an annual survey conducted
since 2002, collecting data on the use of information and communication technology, the internet,
e-government, e-business and e-commerce in enterprises.
Data provided in this domain are collected on a yearly basis by the National Statistical Institutes
and are based on the annual Eurostat model questionnaires on ICT (Information and
Communication Technologies) usage and e-commerce in enterprises. It supports measuring the
implementation of one of the six priorities for the period 2019-2024 of the von der Leyen
European Commission – A Europe fit for the digital age.
The survey population consists of enterprises with 10 or more employees and self-employed
persons.

How digital technology is reshaping the art of management
19
3.1 Data collection and processing across Europe
3.1.1 Computers and connectivity
Together with mobile phones, computers connected to the internet are the most common devices
that enable data-driven forms of management in traditional workplaces. When employees use
interconnected computers, the employer can collect data on their activity, work processes and
beyond. Between 2010 and 2021, the share of employees using a computer with an internet
connection has risen steadily from 42% to 58% (Figure 3).
Figure 3 Use of computers and the internet by employees - EU27, 2010-2021
Source: Authors’ elaboration based on Eurostat, 2021 – Table [isoc_ci_cm_pn2].
Note: Data for all enterprises with 10 employees or more, excluding the financial sector.
The widespread use of connected computers reflects the digitisation trend while providing a
baseline for data-driven management. Conditional on the adoption of technology for data
processing, which sometimes can be embedded in computers themselves, the data collected from
connected computers can be used for management purposes at any time.
The acceleration of the digital transformation in the workplace is reflected in the broad diffusion of
more complex technologies, among which the IoT, (big) data analytics and artificial intelligence are
primary examples. To inform management decisions, the data collected need to be processed. For
example, the IoT can be complemented by technologies to process the large amount of data
collected, including some forms of AI, like machine or deep learning. However, even before the rise
of AI, other technologies for data elaboration have enabled data-driven forms of management.
3.1.2 Internet of Things (IoT)
As shown in Figure 4, IoT is present in approximately 29% of the enterprises with 10 or more
employees in the EU27, albeit with significant difference among member states, from
approximately 51% of the enterprises in Austria to a little more than 10% in Romania. Predictably,
it is also more widespread among large enterprises (48% in EU27) than medium or small sized
firms, with 37% and 26% of them using IoT respectively.
42
58
40
45
50
55
60
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
% tot empl

How digital technology is reshaping the art of management
20
Figure 4 Distribution of Internet of Things, by country and firm size (2021)
Source: Authors’ elaboration based on Eurostat data 2021 [isoc_eb_iot]
The most common use of IoT is for security purposes at the enterprise’s premises, as, according to
the ICT survey, 21% of enterprises in the EU use IoT for this purpose. Energy consumption and
production monitoring and/or automation are the following two second-most common purposes,
employed by 9% of EU enterprises each. 7% and 6% of the EU enterprises use IoT, respectively, for
tracking movements to offer vehicle maintenance and for logistics management. Finally, 4% of the
EU enterprises use it to improve customer services, monitor customer activities and offer
personalised shopping experience (Figure 5).
All these different purposes indicate that data are largely available to inform management
decisions in these enterprises. Even when IoT is not used to specifically track workers’ movement or
monitor their activity, the processed data can still be used by management to change some aspects
of work organisation or workplace practices. For instance, the use IoT for customer monitoring or
products tracking, can lead to automated task or route assignment; whereas IoT systems, including
interconnected cameras, for security reasons can eventually lead to new forms of pervasive
monitoring of employees (Muñoz Ruiz 2022).

How digital technology is reshaping the art of management
21
Figure 5 Use of IoT by its purposes in enterprises - EU27, 2021
Source: Authors’ elaboration based on Eurostat, 2021 – Table [isoc_eb_iot]
Note: Data for all enterprises with 10 employees or more, excluding the financial sector.
3.1.3 Artificial Intelligence (AI)
Figure 6 shows that the greatest majority of firms in Europe still do not use any AI technology. As it
is present on average in 8% of EU27 enterprises with at least 10 employees. Denmark and Portugal
are the countries where AI is most widespread (25% and 17% respectively), while it is present in
only 1% of the enterprises in Romania. Even in this case, the technology is substantially more
widespread among large firms (28% in EU27) than in medium (13%) or small (6%).
Figure 6 Distribution of Artificial Intelligence technologies, by country and firm size (2021)
Source: Authors’ elaboration based on Eurostat data 2021 [isoc_eb_ai]
AI technologies can be used for several purposes. Figure 7 shows how the majority of enterprises
using AI (24%) use it for ICT security purposes, followed by similar shares that use it for business
administration (23%), for marketing or sales purposes (22%) or for production process (20%).
29
21
99764
0
5
10
15
20
25
30
All uses

How digital technology is reshaping the art of management
22
Lower proportions use it for enterprise management (16%) and for logistics (10%). 9% of
enterprises with AI use it for HR or recruiting purposes. This corresponds to the smallest share of
enterprises with AI, and to approximately 1% of all EU enterprises. The data collected by the ICT
survey specify when the technology is used for work coordination, namely, when enterprises use it
for HR or recruiting, hence directly affecting workers.
Figure 7 Use of AI technology and its purposes)
Source: Authors’ elaboration based on Eurostat, 2021 – Table [isoc_eb_ai]
Note: Data for all enterprises with 10 employees or more, excluding the financial sector.
3.1.4 Big data and Data Analytics
Figure 8 shows that on average in EU27 14% of the enterprises with 10 or more workers analyse
big data internally from any source. Malta and the Netherlands are the countries in which more
enterprises engage in data analytics (30% and 27% respectively), whereas Romania and Bulgaria
are the countries with fewer enterprises adopting this technique (5% and 6% respectively). It should
also be noted that nearly 60% of the large enterprises in Denmark or Belgium use data analytics.
8
2222
111
24 23 22 20 16 10 9
0
20
40
60
80
100
0
2
4
6
8
10
% ent with AI
% tot ent
% tot ent % ent with AI

How digital technology is reshaping the art of management
23
Figure 8 Distribution of Big data analysis, by country and firm size
Source: Authors’ elaboration based on Eurostat data 2020 [isoc_eb_bd].
To show some of the purposes data analytics can be used for in enterprises, we draw on another
source of data, namely the European Company Survey (ECS) described in box 2.
The ECS 2019 provides data on the presence of data analytics in establishments in the EU27 and
the United Kingdom (UK), considering the specific purpose for which this technology is employed.
The ECS 2019 survey distinguished between uses of data analytics for the improvement of
production and service delivery processes and/or for monitoring employee performance (Eurofound
and Cedefop, 2020).
Box 2 : The European Company Survey (ECS)
The ECS has been carried out regularly since its inception in 2004–2005 as the European
Establishment Survey on Working Time and Work–Life Balance (ESWT). The second survey, under
the new title European Company Survey, was completed in 2009, the third in 2013 and the fourth
in 2019 in collaboration with Eurofound’s sister agency Cedefop.
The ECS is a questionnaire-based representative sample survey of business establishments with at
least 10 employees. Interviews take place with the manager responsible for human resources in the
establishment and when possible with an employee representative.
The fourth edition of the ECS was carried out in 21,869 establishments across the 27 EU Member
States and the United Kingdom. The target population was establishments with 10 or more
employees in economic sectors engaged in ‘market activities’, according to the statistical
classification of economic activities in the European Community (NACE Rev. 2). The survey however
does not cover establishments operating in agriculture, forestry and fishing (A), activities of the
household (T) and activities of extraterritorial organisations and bodies (U).

How digital technology is reshaping the art of management
24
First of all, it should be noted that the ECS survey finds that just over half of establishments use
data analytics (51%). The discrepancy with the ICT survey reported above probably relates to the
more general understanding of data analytics in the ECS survey where the relevant question makes
no reference to use of ‘big data’. Of the larger share of establishments identified as using data
analytics in the ECS sample, a relatively low proportion uses it only for employee monitoring
purposes (5%), while 24% use it only for process improvement, and 22% for both purposes (Figure
9).
According to an analysis of the ECS data carried out by Eurofound and Cedefop, the use of data
analytics is more likely in establishments that have adopted to a large extent customised robots
and software, but not necessarily computers. By contrast, establishments that use computers
extensively, but not robots or customised software, are less likely to use data analytics (Eurofound
and Cedefop, 2020: 26).
Summing the percentages reported in Figure 9, in total 27% of the establishments in the EU27 and
the UK adopt data analytics for workers monitoring purposes, either exclusively or in combination
with process improvement purposes.
Figure 9 Use and purpose of data analytics in establishments - EU27 and the UK, 2019
Source: Authors elaboration based on ECS 2019 management questionnaire
3.1.5 Wearable devices
To show the distribution of wearable devices by country and firm size, we draw from a different
data source, the ESENER 3, described in box 3.
Box 3 : The European Survey of Enterprises on New and Emerging Risks (ESENER)
ESENER is a survey of establishments carried out every five years, since 2009, by EU-OSHA. The
third ESENER wave was in 2019. It surveyed a representative sample of 45,420 establishments
with at least 5 employees in all economic sectors from 33 European countries (i.e. EU27, United
Kingdom, Iceland, North Macedonia, Norway, Serbia and Switzerland). The survey is answered by
the person “who knows best about health and safety in the establishment”, that can be the owner,
managing director or manager, health and safety officer, an employee or a representative, and is
characterised by a holistic view on health and safety risks and management in the workplace. It
addresses topics concerning the physical and psychosocial risks in the workplace, how these risks
are managed, drivers and barriers for their management and workers’ participation in OSH
management.

How digital technology is reshaping the art of management
25
The questionnaire underpinning the third ESENER wave included also some questions on the use of
digital technologies and systems that would enable data-driven monitoring. In particular, ESENER
contains information on whether the enterprise uses any of the following: 1) Personal computers at
fix workplaces; 2) Laptops, tablets, smartphones or other mobile computer devices; 3) robots that
interact with workers; 4) Machines, systems or computer determining the content or pace of work;
5) Machines, systems or computer monitoring workers’ performance; and 6) wearable devices such
as smart watches, data glasses or other embedded sensors.
Wearable devices are still fairly uncommon in the EU27, with on average less than 5% of the
enterprises with 5 or more employee using them for work (5.7% if we consider enterprises with 10
or more employees, like in the ICT survey). Estimates in Figure 10 also show that, predictably, the
technology is more frequently adopted in larger establishments and in countries such as Austria,
which also ranked first for the use of IoT. Interestingly, Romania is also one of the top adopters,
even though it always ranked last for the adoption of technologies such as the IoT, AI or Data
analytics. One possible reason is that, in the ESENER dataset, 50% of the enterprises with more
than 10 workers belong to sectors such as manufacturing, wholesale and repair trade, and
construction, which have an above average adoption of wearables, even though they are not the
top sectors (see Figure 11).
Figure 10 Distribution of wearable devices, by country and firm size
Source: Authors’ elaboration based on ESENER 3 data

How digital technology is reshaping the art of management
26
Figure 11 Distribution of wearable devices, by industrial sector
Source: Authors’ elaboration based on ESENER 3 data
3.1.6 Management software applications and other enabling technologies
In 2021, 38% of EU enterprises used Enterprise Resource Planning (ERP) software, whereas
Customer Relationship Management (CRM) software is found in 35% of enterprises across the EU.
Belgium leads the ranking for both technologies, with over 57% and 53% of its enterprises using
ERP and CRM respectively, while Hungary and Romania have the lowest levels of adoption.
Figure 12 Distribution of management software applications, by country
Source: Authors’ elaboration based on Eurostat 2021 data; table [isoc_eb_iip].

How digital technology is reshaping the art of management
27
With the rise of the IoT and AI, CRM and ERP software, and management software applications in
general, are increasingly sophisticated. On the one hand, these software applications are enriched
by the large amount of data collected from a wider range of interconnected devices that work often
automatically. On the other hand, they are empowered by advanced techniques for data
elaboration, including AI.
The ESENER dataset contains information on whether an establishment uses machines, systems or
computers either determining the content or pace of work, or for monitoring workers’ performance
(or both). The questionnaire does not specify what these machines or systems may be, but only that
they can be used to carry out or assist with the performance of some managerial functions, such as
controlling (monitoring) or directing (determining content or pace of work).
Estimates in Figure 13 show that the proportion of enterprises that use devices to monitor
performance ranges between 5% in Austria and 24% in Bulgaria, with a much higher proportion of
large enterprises doing so; the proportion of enterprises using digital devices to set the content or
pace of work ranges from approximately 10% in Austria to about a quarter of the Hungarian
enterprises surveyed in the ESENER 2019. Again, the use of these devices is substantially higher in
large enterprises (more than 50% in Bulgaria), suggesting that a potentially large number of
employees may be affected by it.
Figure 13 Distribution of other data-driven management enabling technologies, by country and firm
size
Source: Authors’ elaboration based on ESENER 3 data

How digital technology is reshaping the art of management
28
4 Data analytics and work organisation: insights from the ECS
The adoption and use of digitisation technologies in companies is mostly driven by the need to
increase productivity (and profitability) and reduce costs (Eurofound 2021). Data analytics, a
technology enabling data-driven management, is no exception to this. Modern theories on the value
of information depart from information theory (Blackwell 1953) and information-processing view
of organisations (Galbraith 1974) and suggest that the availability of more accurate information
can enable more informed decision making and lead to higher firm performance (Brynjolfsson, Hitt
and Kim 2011). Much of the available research has found a positive relationship between the use
of data analytics and firm performance resulting from better informed decisions (Davenport and
Harris 2006, Loveman 2003, Brynjolfsson, Hitt and Kim 2011, LaValle, et al. 2011, Ferraris, et al.
2019, Brynjolfsson, Jin and McElheran 2021). Empirical research drawing from German firm-level
data also found that big data analytics is a determinant of innovative performance in terms of
product innovation (Niebel, Rasel and Viete 2019).
Recent research - using survey data of over 30,000 American manufacturing establishments – has
however found that productivity gains from the use of predictive analytics occur only in the
presence of a significant accumulation of IT capital, educated workers, or so-called high flow-
efficiency production systems (Brynjolfsson, Jin and McElheran 2021). Other empirical research
exploring the impact of data analytics on firm performance found high level of complementarities
between IT implementation, performance pay schemes, and HR analytics practices used to monitor
employee performance and suggested that firm productivity is higher when these practices are
implemented together, rather than separately (Aral, Brynjolfsson and Wu 2020).
As already mentioned in chapter 2, work organisation refers to ‘how work is planned, organised and
managed within companies via production processes, job design, task allocation, rules, procedures,
communication, responsibilities, management and supervisory styles, work scheduling, work pace,
career development, decision-making processes, interpersonal and interdepartmental relationships’
(Eurofound, 2017).
This chapter illustrates the relationship between one of the enabling technologies of data-driven
management, namely data analytics, and aspects of work organisation, such as: autonomy and job
complexity, variable pay schemes, or training provision, well as aspect of the business model
(outsourcing and collaboration), establishment performance and workplace wellbeing. The reported
estimates are merely descriptive and do not represent causal relationships. The data source used
for this chapter in the 2019 European Company Survey (ECS), described in box 2.
4.1 Job complexity and autonomy
To assess the level of autonomy that workers have within the workplace, the survey asked
managers about the proportion of the employees in their establishment who organise their tasks
and their time autonomously or whose pace of work is determined by machines or computers.
Figure 14 summarises the proportion of establishments in which the pace of work is determined by
machines or computers by the use of data analytics. The estimates show that the use of data
analytics is more pervasive in establishments where the pace of work for a large share of
employees (at least 40%) is determined by machines or computers. This suggests that the use of
data analytics is more likely in establishments where work processes are significantly digitised or
automated, for example in advanced manufacturing. The positive correlation is more pronounced
for use of data analytics for both employee monitoring and process improvement, which suggests
that employee monitoring is as much a by-product of technological processes of production as of
performance monitoring.

How digital technology is reshaping the art of management
29
Figure 14 Pace of work significantly determined by machines or computers and use of data
analytics, EU27 and the UK
Source: ECS 2019 management questionnaire
Note: *For at least 40% of employees.
Latent class analysis from previous analysis of the ECS data (Eurofound and Cedefop, 2020) also
identified three types of establishments based on the complexity of employees’ jobs and the
degree of autonomy they can exercise. These three types were: ‘high complexity and autonomy’,
‘selective complexity and autonomy’, and lastly ‘command and control’.
The ‘high complexity and autonomy’ type is the least common, accounting for 6% of
establishments in the EU27 and the UK. This type has many of the features of so-called high-
performance workplaces, which are characterised for example by high degree of autonomy and
complex problem solving on the part of individual employees as well as the presence of self-
directed teams. Establishments using data analytics only for process improvement tend to be more
represented in this group compared to establishments using data analytics for employee
monitoring only and those not using data analytics at all.
The ‘selective complexity and autonomy’ type is more prevalent, representing 57% of
establishments. This type exhibits some of the characteristics of the ‘high complexity and
autonomy’ type, for example managers create the conditions for employees to work independently,
but workers have typically lower degree of autonomy and exposure to complex problem-solving
compared to the first type. This is the type of establishment most frequently associated with use of
data analytics particularly for process improvement only (63% vs. 53% with no data analytics in
place).
The ‘command and control’ type, which accounts for 37% of establishments, is characterised by
more Tayloristic work arrangements, for example limited autonomy and less complex problem
solving, lower incidence of teamwork, and greater management oversight on tasks performed. Use
of data analytics is relatively less observed in this group of establishments (Figure 15). This lower
prevalence of data analytics use in the ‘control and command’ category may be due to the fact that
this establishment type is characterised by limited levels of computer use and digitalisation more
generally, which can be considered a requirement for application of data analytic techniques
(Eurofound and Cedefop, 2020).
16%
21% 22%
28%
0%
5%
10%
15%
20%
25%
30%
No data analytics Data analytics for process
improvement only
data analytics for employee
monitoring only
Data analytics for both
purposes

How digital technology is reshaping the art of management
30
Figure 15 Use of data analytics by establishment type – job complexity and autonomy, EU27 and
the UK (%)
Source: ECS 2019 management questionnaire
4.2 Outsourcing and collaboration
Data-driven forms of management – including those enabled by data analytics – could affect firms’
business models, for instance by boosting outsourcing or contracting out, as well as collaboration
among firms. No clear correlation was however found between data analytics usage and
outsourcing or contracting out of key activities in the surveyed establishments. Estimates shows
that innovative establishments – which use data analytics – are more likely to engage in the design
and development of new products or services and carry out such activities internally. This may have
to do with the size of these establishments, with large ones more likely to have the capacity to do
so. This correlation is more prominent for establishments using data analytics for process
improvement, either exclusively or together with employee performance monitoring.
Furthermore, the survey identifies whether establishments engage in relationships with other
establishments, either for production or design and development of goods and services. In this
respect, the data shows that establishments with no data analytics are less likely to engage in
design and development of new products or services. Estimate in Figure 16 summarise the share of
establishments where design and development of new products is carried out and those in which is
isn’t by the use of data analytics and show that engagement in such activities carried out internally
is more prevalent in establishments with data analytics.
5% 8% 5% 6%
53%
63%
57% 58%
42%
29%
38% 36%
0%
10%
20%
30%
40%
50%
60%
70%
No data analytics Data analytics for
process improvement
only
Data analytics for
employee monitoring
only
Data analytics for both
purposes
% establ
High complexity and autonomy Selective complexity and autonomy Command and control

How digital technology is reshaping the art of management
31
Figure 16 Share of establishments engaging in design and development of new products or services
by use of data analytics, EU27 and the UK (%)
Source: ECS 2019 management questionnaire
The use of data analytics is also more observed in establishments that collaborate with other
establishments (Figure 17).
Figure 17 Collaboration with other establishments by use of data analytics, EU27 and the UK (%)
Source: ECS 2019 management questionnaire
4.3 Variable pay schemes
‘Variable pay’ refers to additional compensation beyond the basic pay and may vary in amount. If
properly designed, variable pay schemes can serve as effective tools to stimulate and reward extra
effort, attract talent and enhance staff engagement and commitment to the organisation.
The ECS 2019 survey investigates practices that affect the performance of employees, as well
remuneration and monetary incentives, including use of four forms of variable pay schemes. These
were ‘payment by results’, ‘individual performance-related pay’, group performance-related pay’
and ‘profit-sharing schemes’. On the basis of these four forms of variable pay, latent class analysis
23% 27%
34% 34%
68%
62%
50% 49%
0%
10%
20%
30%
40%
50%
60%
70%
80%
No data analytics Used for employee
monitoring only
Used for process
improvement only
Used for both purposes
% establ
Yes, mainly carried out internally Not engaged in design and development
17%
20%
25% 25%
0%
5%
10%
15%
20%
25%
30%
No data analytics Used for employee
monitoring only
Used for process
improvement only
Used for both purposes
% establ

How digital technology is reshaping the art of management
32
identified three types of establishments (Eurofound and Cedefop, 2020). In establishments with a
comprehensive approach to variable pay, all four forms of variable pay are likely to be used and
applied to a large proportion of employees. This type represents only 13% of establishments. The
comprehensive approach to variable pay is relatively more common in the case of establishments
using data analytics for employee monitoring (19%) and data analytics for both employee
monitoring and process improvement (18%).
The most pronounced correlation is however found in the establishment type with a selective
approach to variable pay, which accounts for 55% of establishments. These establishments are
likely to use at least some forms of variable pay, which apply to a minority of employees.
Establishments using data analytics are more likely to fall into this group, particularly if the
technology is used for both employee monitoring and process improvement. By contrast,
establishments with no use of data analytics are more likely to belong to the establishment type
with no variable pay - found in 32% of establishments.
Figure 18 Use of data analytics by establishment type – variable pay, EU27 and the UK (%)
Source: ECS 2019 management questionnaire
4.4 Training provision
Previous analysis of the ECS 2019 data suggests that the higher the level of digitalisation of an
establishment, the greater the use of data analytics for both process improvement and employee
monitoring (Eurofound and Cedefop, 2020). As AI and IoT technologies generate vast amounts of
data and data analytics techniques become more sophisticated, establishments will require new
skills for their data operations and analysis. This skills gaps may be filled through training and skill
development. With regard to skills development activities, the ECS 2019 asked managers about two
types of training: training sessions delivered to employees during paid working time and on-the-job
learning. Previous analysis of the ECS data shows that training during working time and on-the-job
learning was provided to a limited proportion of workers (less than 20%) respectively in 34% and
28% of establishments (Eurofound and Cedefop, 2020).
Looking at the provision of such training in establishments by use of data analytics, estimates
shows that the lack of provision of paid training and on the on-the-job learning is associated with
no use of data analytics in the establishment (Figure 19). Of all establishments, those using data
analytics for process improvement, either exclusively or together with employee monitoring
purposes, are more likely to provide either form of training to their employees. Where data
51%
58% 55%
61%
38%
30%
26%
21%
12% 12%
19% 18%
0%
10%
20%
30%
40%
50%
60%
70%
No data analytics Data analytics for
process improvement
only
Data analytics for
employee monitoring
only
Data analytics for both
purposes
% establ
Selective No variable pay Comprehensive

How digital technology is reshaping the art of management
33
analytics is used for employee monitoring only, the share of establishments not offering these
training possibilities is higher, though not as high as in the firms not using data analytics at all.
Figure 19 Establishments with no employees availing of paid training or on the job learning, EU27
and the UK (%)
Source: ECS 2019 management questionnaire
4.5 Establishment performance and workplace wellbeing
Possible changes on work organisations and workplace practices can reflect on business
performance and workers’ well-being. Prior analysis of the ECS data used two composite indicators
scaled 0–100 to monitor outcomes at establishment level in relation to workplace well-being, on
the one hand, and establishment performance, on the other (Eurofound and Cedefop, 2020). Four
questions were used to indirectly measure workplace well-being: one captured the quality of the
relationship between management and employees and the other three questions concerned
challenges in terms of human resources (such as low motivation, absenteeism and staff retention).
Establishment performance was measured in relation to the following four variables: the
profitability of the establishments, the profit expectation, the change in production volume and the
expected change in employment.
Using these two indicators, the analysis shows that establishments using data analytics for both
employee monitoring and process improvement score higher on establishment performance. By
differentiating between the two types of data analytics, the score of establishment performance is
higher in establishments using such technologies for process improvement only and marginally
higher in the case of those using the technology for employee monitoring only.
The use of data analytics exclusively for employee monitoring is associated on the other hand with
somewhat lower workplace wellbeing, in comparison to not using data analytics at all. The use of
data analytics for process improvement seem to make almost no difference on the score on
workplace well-being, in comparison to the establishments with no data analytics.
16%
8%
12%
7%
13%
5%
8%
5%
0%
5%
10%
15%
20%
No data analytics data analytics for
process
improvement only
Data analytics for
employee
monitoring only
Data analytics for
both purposes
% establ
No paid training No on the job learning

How digital technology is reshaping the art of management
34
Figure 20 Use of data analytics and workplace outcomes, EU27 and the UK (%)
Source: ECS 2019 management questionnaire
5 Wearables and other digital devices enabling data-driven
management: impacts on Occupational Safety and Health
Data-driven forms of management, and in particular algorithmic management as defined in
Chapter 2, can affect the mental (but also physical) health of workers. This chapter explores the
association between digital technologies that enable data-driven forms of management and
Occupational Safety and Health (OSH) as measured by the presence of psychosocial risks in
European workplaces. In addition, it assesses the potential mitigating role of OSH preventive
measures in managing risks related to digitisation and reducing the potentially negative effects of
data-driven management technologies.
Technological advances have been accompanied by an increased prevalence of psychosocial risks in
most economic sectors (EU-OSHA, 2010; Leka et al., 2011). Psychosocial risks are among the most
challenging issues in OSH, nowadays. The EU-OSHA defines psychosocial risks as “those aspects of
the design, organisation and management of work, and its social and environmental context, which
can cause psychological, social or physical harm” (Brun and Milczarek 2007). In practical terms,
workers experience stress when the demands of their job exceed their capacity to cope with them.
In addition to mental health problems (e.g. depression), prolonged stress can lead to physical health
problems, such as heart disease and musculoskeletal disorders. From an employer’s perspective,
they are an issue as they lead to absenteeism, lower productivity and financial losses (Leka and
Cox 2008).
The empirical analysis presented in this chapter draws on a unique set of data, the EU-OSHA
ESENER 3 (described in box 3). Such data allow to identify a set of digital technologies and systems
enabling data-driven management of workers, as well as the presence of psychosocial risks and
mitigating measures at the workplace level in a representative sample of European establishments.
In line with the literature summarized in chapter 2, the underlying hypothesis is that the use of
selected technologies may be associated with increased psychosocial risks by boosting job
demands and decreasing workers’ autonomy. However, OSH measures in the workplace can be of
help to manage and mitigate such increased risks, so they are also considered in the empirical
analysis, in the following section.
72
71
69
71
63
66
64
68
58 60 62 64 66 68 70 72 74
No data analytics
data analytics for process improvement only
Data analytics for employee monitoring only
Data analytics for both purposes
Establishment performance Workplace well-being

How digital technology is reshaping the art of management
35
5.1 Enterprises adopting data-driven management enabling technologies
The digital technologies and systems that can enable data-driven management included in the
ESENER dataset are:
• Machines, systems or computer monitoring workers’ performance – enabling the management
function of control. In order to make sure the objectives of the organisation are met, the
management function of control relies on performance measurement, evaluation and
corrective action (discipline).
• Machines, systems or computer determining the content or pace of work; these enable the
management function of organisation and direction, by assigning duties and establishing how
they should be performed.
• Wearable devices, such as smart watches, data glasses or other (embedded) sensors (allowing
for data gathering and feed into the other systems).
As illustrated in Chapter 3 and summarized in Figure 21, such managing technologies are relatively
little widespread in European establishments overall, although substantial differences exist across
countries, sectors and firms size.
Figure 21 Selected management technologies by country, EU 27]
Source: Authors’ elaborations using ESENER 3
The ESENER 3 dataset allows for a more granular description of the distribution of technologies, by
establishment characteristics. In particular, the survey provides information on whether the
enterprise employs workers with atypical work contracts (subcontractors, temporary agency workers
or volunteers), whether the enterprise employs people who have difficulties understanding the local

How digital technology is reshaping the art of management
36
language; and whether any of the employees work somewhere else outside the premises of the
establishment (other than working from home). (Figure 22).
Estimates in Figure 22 show that the prevalence of all three digital technologies differs across
different establishments. For instance, devices for monitoring performance are more likely to be
found in establishments that use atypical work contracts (10.6% vs. 8.9%), have employees who
can’t understand the local language (11.7% vs. 9.3%), or that have employees who work from
home (10.9% vs 9.3%). By contrast, having employees working outside the premises seems to
make little difference.
Similarly, devices that determine content or pace of work are more likely to be found in
establishments that use atypical work contracts (14.5%% vs. 11.7%), have employees who can’t
understand the local language (17.7% vs. 12.3%) or work from home (13.9% vs. 12.5%).
The prevalence of wearables follows a similar pattern, however in this case they are also more
likely to be found in establishments where people work outside the premises (5.9% vs. 4.1%).
Figure 22 Distribution of digital technologies by selected establishment characteristics, EU 27]
Source: Authors’ elaborations using ESENER 3; data are weighted using establishment weights.
What these findings suggests is that some data-driven management enabling technologies, for
instance those for performance monitoring, may be related to lower trust, for instance to exert
more control over employees working from home, or lower cohesion, as in the case of enterprises
using management technologies to monitor the performance of subcontractors or agency workers.
In other cases, the adoption of management technologies, and especially those enabling the setting
of content or pace of work, may be sector-related: for instance, Urzí Brancati and Curtarelli (2021)
show that management technologies are more prevalent in sectors such as transportation and
0%
2%
4%
6%
8%
10%
12%
14%
16%
Monitoring
performance
Determining
content/pace
Wearables
Enterprise uses atypical contracts
NO
YES
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
Monitoring
performance
Determining
content/pace
Wearables
Employees have difficulties understanding the
language spoken on premises
NO
YES
0%
2%
4%
6%
8%
10%
12%
14%
16%
Monitoring
performance
Determining
content/pace
Wearables
Employees working from home
NO
YES
0%
2%
4%
6%
8%
10%
12%
14%
Monitoring
performance
Determining
content/pace
Wearables
Employees working outside premises
NO
YES

How digital technology is reshaping the art of management
37
storage, manufacturing, or mining and quarrying, which are also, arguably, more likely to employ
foreign workers (so workers who have difficulties understanding the language spoken on premises)
or workers on atypical contracts.
5.2 Psychosocial risks
Psychosocial risks refer to those aspects of the work environment that are likely to affect mental
health. They are called “psycho-social” because they represent the influence of social factors on
workers' psychological response. Recording the presence of psychosocial risks in the workplace is
important because they result in work-related stress, and possibly depression, anxiety or sleeping
disorders (22).
The ESENER dataset contains five variables related to the reported presence of psychosocial risks in
the workplace, namely time pressure, long or irregular working hours, having to deal with difficult,
customers, patients, pupils, etc., fear of job loss, and poor communication or cooperation within the
organisation.. Note that the presence of a psychosocial risk is reported at the establishment level by
the person interviewed, and therefore it does not reveal how many workers are exposed to it and it
can be influenced by the level of awareness the respondent has of the existing risks. Indeed, it
should be mentioned that since risk reporting may vary according to who answers the survey,
different respondents can be sources of bias.
As shown in Figure 23, having to deal with difficult customers, patients or clients and time pressure
are the most widespread psychosocial risks, as they are present in approximately 60% and 45% of
the EU27 establishments, respectively. They are followed by the risk of long/irregular working hours
(present in 21.5% of the establishments in the EU), poor communication of cooperation (17.8%)
and fear of job loss (11%).
Figure 23 – Establishments by type of psychosocial risk in the EU27 (%, 2019)
Source: Authors’ elaboration on EU-OSHA ESENER-3 data.
The distribution of psychosocial risks by country presented in Figure 24 shows that generally when
one risk is widespread in a country, the others tend to be reported more frequently than in other
countries. For instance, 74% of the enterprises in Sweden report the presence of time pressure and
66% report having to deal with difficult patients, customers, etc., whereas in Italy only 19% of the
enterprises report time pressure and about 36% report having to deal with difficult patients,
customers, etc. While it is possible that this is due to the existence of ‘healthier’ enterprises, with
lower psychosocial risks overall in certain countries, we should not discard the possibility that
higher incidence might depend on reporting culture, or on having an OSH responsible on the
(22) https://www.etui.org/news/gig-economy-and-psychosocial-risk-interview-pierre-berastegui
44.9
17.8
11.0
59.5
21.5
Time pressure
Poor communication or cooperation
Job insecurity
Difficult customers
Long or irregular working hours

How digital technology is reshaping the art of management
38
premises, and therefore countries in which a higher proportion of establishments report the
presence of psycho-social risks may be the ones in which such risks are simply more visible.
Another possibility is that the higher incidence of psychosocial risks actually depends on more
intensive use of technology, which is what the rest of this chapter is going to investigate.
Figure 24 Distribution of psychosocial risk by country, EU27
Source: Authors’ elaborations using ESENER 3
The association between the technologies enabling data-driven workers’ management and the
psychosocial risks in the workplace is described in Table 1. The first row of table summarises the
psychosocial risks composite index (calculated as sum of the five psychosocial risks shown in Figure
24) (23), whereas the following rows summarise the incidence of individual psychosocial risks and
compare it in establishment which have introduced one of the management technologies under
study, with those who didn’t. Estimates are carries out on the sample of EU27 countries.
The psychosocial risks index is nearly half a point lower in establishments that haven’t adopted
monitoring technologies, compared to those that have (i.e. 1.51 vs. 1.98). Similarly, the score on the
psychosocial risks index is approximately a third of a point higher in establishments that have
adopted technologies determining content/pace of work compared to those which did not (1.80 vs.
1.51); the same can be said for the last technology, since establishments using wearables report an
average psychosocial risks of 1.91 vs. 1.53 of those that do not use such technology.
As per the prevalence of individual risks, it is consistently higher in establishments which use one of
the enabling technologies: while approximately 44% of the establishments which do not use
technology to monitor workers’ performance report time pressure, the proportion rises to 57%
(23) Psychosocial risks indicators are dichotomous variables equal to 1 if they were reported in the establishment and 0 otherwise. The
cumulative index for workplace psychosocial risks (PR) is equal to the sum of all dichotomous psychosocial risk indicators at the
establishment level (0-5).

How digital technology is reshaping the art of management
39
among the establishments using such technology. The difference in incidence of time pressure is
very similar for the other two technologies. Looking at the indicator “fear of job loss”, its prevalence
is nearly double in establishments that adopted technology to monitor performance compared to
those that didn’t (18.9% vs. 10.4%). The difference is slightly smaller with respect to the
technology determining content or pace of work (15.6% vs. 10.5%) and substantially smaller, but
still positive, considering wearables (12.4% vs. 11.0%). Similarly, poor cooperation/communication
is always relatively more frequent in establishments that use one of the technologies, but the
higher prevalence is found in establishments that use technology to monitor performance (26.4%).
By contrast, working long or irregular hours is reported more frequently in establishments that use
wearable technologies (31.2%). Finally, focussing on the most widespread psychosocial risk (that is
dealing with difficult customers, patients, pupils etc.), its prevalence is highest in establishments
monitoring performance (67.4%) or using wearables (66.2%), while using technology to determine
content or pace of work does not appear to be associated with this psychosocial risk.
Table 1 Prevalence of psychosocial risks in establishments by presence of different types of
technology – EU27, 2019
Machines, systems, or
computers monitoring
workers’ performance
Machines, systems, or
computers determining
pace/content of work
Wearables
Not
Present
Present
Not
Present
Present
Not
present
Present
Psychosocial risk
Index
1.51 1.98 1.51 1.80 1.53 1.90
% of establishments reporting a psychosocial risk
Time pressure
44.0%
57.3%
43.8%
54.7%
44.4%
58.0%
Fear of job loss 10.4% 19.3% 10.5% 15.6% 11.0% 12.4%
Poor cooperation/
communication 17.2% 26.4% 17.1% 24.2% 17.7% 22.6%
Working
long/irregular hours 20.9% 28.2% 20.9% 26.1% 21.0% 31.2%
Difficult clients
59.0%
67.4%
59.7%
60.1%
59.4%
66.2%
Source: Authors’ elaborations using ESENER 3
Estimates in Table 1 are purely descriptive and do not account for potential confounding factors;
however, they are in line with results from a regression analysis, not limited to EU27 but including
all 33 countries in the ESENER dataset, which was carried out by Urzì Brancati and Curtarelli (2021).
So we can conclude that the positive correlation between management technologies and
psychosocial risks is present even ceteris paribus, that is, controlling for all factors such as
establishment size and characteristics (such as those summarised in Figure 22), and sector fixed
effects.
5.3 Workplace-level measures to prevent risks to worker´s safety and health
A number of workplace measures are aimed at preventing psychosocial risks as reported in ESENER
data. The OSH measures that are likely to be introduced to prevent psychosocial risks in the
workplace and are considered in the analysis are: reorganisation of work in order to reduce job
demands and work pressure; confidential counselling for employees; training on conflict resolution;
intervention if excessively long or irregular hours are worked; allowing employees to take more
decisions on how to do their job. In addition to those, the analysis considers whether the

How digital technology is reshaping the art of management
40
establishment had in place an action plan to prevent work-related stress. (24) This OSH preventive
measure is likely to consist of a document or other hard evidence detailing the actions specifically
foreseen by the management to prevent work-related stress.
According to the ESENER data, a majority of establishments have put in place measures to prevent
psychosocial risks. The OSH measure introduced in most establishments, regardless of whether
they have introduced digital technologies enabling data-driven workers’ management, consists in
allowing employees to take more decisions on how to do their job (67.7%), therefore increasing
their autonomy and counteracting the potentially negative effect of digital technologies. Around
43% of the establishments have introduced measures allowing the reorganisation of work to
reduce job demands and work pressure. Only 29.1% of the establishments have adopted measures
to mitigate the risk of long or irregular hours, whereas confidential counselling for employees has
been introduced by 42% and conflict resolution training by 34% of the establishments. Finally,
approximately 33.5% of the establishments (with 20 or more employees) report having an action
plan to prevent work related stress.
Measures to prevent psychosocial risks are relatively less commonly introduced in establishments
using technologies enabling data-driven management of workers, as displayed in the table below.
Among the establishments using digital technologies, the most frequently introduced measure is
the action plan to prevent stress, followed by the intervention to modify working hours.
Unsurprisingly much less common – while it is the most common option for the European
establishments as a whole – is allowing employees to make more decisions about how to do their
job, as data-driven workers’ management practices restrict workers’ autonomy and room for
manoeuvre.
(24) It should be noted though that the latter question is only asked to establishments with 20 or more employees, therefore
substantially reducing the final sample size on which the empirical analysis has been carried out.

How digital technology is reshaping the art of management
41
Table 2 Measures adopted by EU27 establishments to prevent psychosocial risks (% - 2019)
Establishments using digital technologies
All
establishments
Machines,
systems or
computer
determining the
content or pace
of work
Machines,
systems or
computer
monitoring
workers'
performance
Wearable
devices
Reorganisation of work 43.3% 13.6% 9.6% 5.7%
Confidential counselling for employees 42.0% 13.2% 10.2% 5.6%
Training on conflict resolution 34.0% 13.8% 10.6% 6.2%
Intervention if long or irregular hours
are worked 29.1% 15.0% 10.7% 6.7%
Allowing to take decisions on how to
do the job 67.7% 12.0% 8.5% 5.2%
Action plan to prevent work-related
stress 33.4% 17.3% 12.3% 7.5%
Source: Authors’ elaboration on EU-OSHA ESENER-3 data.
In addition, a proportion of the establishments that adopted management technologies report
having discussed the effects of such technologies on OSH with their workforce as a specific
measure to raise awareness and prevent risks associated to the use of management technologies.
Around 24% of establishments that report using any of the digital technologies have internally
discussed the effects of such technologies on OSH (see Figure 25). Such discussions are reported
by about 11% of establishments that use wearable devices, such as smart watches and glasses,
approximately by 14% of establishments using machine, systems, or computers monitoring worker
performance, and by 17% of establishments using machine, systems, or computers to monitor
content or pace of work.
Figure 25 Establishments in the EU27 which have discussed the possible health and safety impacts
of impact of digital technologies by type of digital technologies used (%, 2019)
Source: Authors’ elaboration on EU-OSHA ESENER-3 data.
* Including personal computers, laptops and other portable devices and robots interacting with workers, in addition to the
technologies enabling data-driven management practices (Q310 of survey questionnaire).
Overall, the topic most often discussed in establishments that use digital technologies is the need
for continuous training to keep skills updated (by 77% of establishments that use any
24.1%
17.3% 13.7% 10.8%
Establishments using any digital
device*
Machines, systems or computer
determining the content or pace
of work
Machines, systems or computer
monitoring workers' performance
Wearable devices

How digital technology is reshaping the art of management
42
technologies). Discussions on prolonged sitting (65.5%) and increased flexibility for employees
(63%) are reported by around two thirds of the establishments. Increased work intensity or time
pressure and information overload, commonly reported in association with the use of data-driven
workers’ management practices, are discussed in about respectively 58% and 52% of the
establishments using digital technologies. The issue of blurring boundaries between work and
private life is discussed in 47% of establishments using digital technologies. Finally, the fear of job
loss is the issue discussed by only 21.1% of establishments that use digital technologies (Figure
26).
Figure 26 Establishments in the EU27 which have discussed the possible health and safety impacts
of impact of digital technologies by type of issue discussed (%, 2019)
Source: Authors’ elaboration on EU-OSHA ESENER-3 data.
* Including personal computers, laptops and other portable devices and robots interacting with workers, in addition to the
technologies enabling data-driven management practices (Q310 of survey questionnaire).
Discussions on the impact of digital technologies on health and safety are more likely in
establishments operating in economic sectors with higher digital intensity, such as information and
communication, education, administrative and support service activities, and finance and real
estate. Establishments in lower digital intensity sectors such as accommodation and food services,
art, entertainment and recreation, and manufacturing are less likely to discuss the impact on health
and safety of digital technologies.
On a final note, it is worth mentioning that workplace risk assessments are carried out by 75.3% of
EU27 establishments, but establishments using technologies enabling data-driven forms of
workers’ management seem more keen in carrying out risk assessments: 80.3% of establishments
using technologies to monitor content or pace of work, 80.5% of those using technologies to
monitor workers’ performance and 78.9% of establishments using wearables report regularly
carrying out risk assessments.
The evidence presented above, complemented by the regression analyses on ESENER-3 data in Urzì
Brancati and Curtarelli (2021), suggests that the presence of technologies enabling data-driven
management of workers is associated with an increase in reported psychosocial risks in the
workplace, especially when it comes to technologies monitoring performance. This is consistent with
the Job Demand-Control (JDC) model on occupational stress, the combined low workers’ control on
their job (due to the presence of monitoring technologies, devices setting the content and pace of
work and wearables) and high demand (as a result of intensification of work and increased
workload) result in “high-strain” jobs, associated to increased psychosocial risks and potentially
work-related stress and other health issues for the worker.
21.1
47.2
52.3
57.5
57.9
63.4
65.5
76.6
Fear of job loss
Blurring boundaries between work and private life
Information overload
Repetitive movements
Increased work intensity or time pressure
More flexibility for employees in terms of place of work…
Prolonged sitting
Need for continuous training to keep skills updated

How digital technology is reshaping the art of management
43
However, the descriptive analysis presented in this chapter suggests that only a minority (24%) of
the establishment that have adopted any of the data-driven management enabling technology are
aware of their potential negative impact on occupational health and safety (as measured by the
presence of psychosocial risks). Indeed, this is in line with previous findings by Urzì Brancati and
Curtarelli (2021) whose empirical analysis cannot confirm a mitigating role of OSH policies, except
for having an action plan to reduce stress. In addition, the relatively small proportion of
establishments that do discuss the risks of digitalisation seem to focus on the need for continuous
training to keep skills updated. This suggests that the impact of such technologies on workers’
wellbeing may be overlooked.
6 Conclusions
‘The processing of personal data should be designed to serve mankind’ (GDPR, pg. 2, (4)).
The EU digital strategy, together with the EU regulation on digital labour platforms, AI and data
privacy (GDPR), aims, first and foremost, to expand the digital economy and make the EU a leader
in a “society empowered by data to make better decisions – in business and the public sector”. At
the same time, the strategy recognises the importance of protecting citizens from the potential
harm associated with emerging digital technologies. This report contributes to the policy debate by
focusing on the potential for digital technologies to enable data-driven management of workers
and the implications for occupational safety and health.
In particular, this report shows how data-driven and algorithmic management are not a marginal
phenomenon limited to digital labour platforms, but are progressively spreading to traditional
workplaces. Several technologies enabling data-driven management are widespread in the EU,
especially in some member states and in large firms. Eurostat data show that the most widespread
technologies enabling data-driven management are Enterprise Resource Planning (ERP) and
Customer Relations management (CRM) software, present in more than a third of European
enterprises with 10 or more employees. The Internet of Things is also fairly widespread, as it is
present in approximately 29% of the EU27 enterprises; other technologies are relatively less
common: for instance, data analytics is present in approximately 14% of the EU enterprises, AI in
8% while wearable devices in less than 6% of the enterprises with 10 or more employee. Such
technologies can affect primarily the managerial functions of controlling, leading and directing, by
facilitating worker monitoring, enabling the automatic allocation of tasks, setting the content or
pace of work and so on. The number of workers subject to data-driven management is uncertain;
however, the fact that large enterprises tend to adopt enabling technologies more often than the
smaller ones implies a large number of workers may be affected. More detailed data at individual
level, could help to assess the size of the phenomenon and to improve our understanding.
The descriptive analysis in this report showed that the use of technologies enabling data-driven
management can have both positive and negative impacts for the employers and the workers.
The analysis presented in chapter 4 illustrated how establishments using data analytics for
production improvement and employee performance monitoring have better outcomes in terms of
establishment performance than those with no use of data analytics. In addition, the use of data
analytics (particularly for process improvement) tends to be associated with features of so called
‘high-performance work practices’, for example, higher levels of training provision, performance-
related pay, greater job complexity and worker autonomy. Possibly due to their large size,
establishments with data analytics are also found to be more likely to engage in the design and
development of new products or services than their counterparts with no use of data analytics.
However, the technology is also associated with somewhat lower worker wellbeing, especially if
used for employee monitoring.
Similarly, as displayed in the analysis presented in chapter 5, the use of digital technologies to
monitor workers’ performance or determine the content or pace of work is associated with a higher

How digital technology is reshaping the art of management
44
prevalence of reported psychosocial risks in the workplace, and in particular time pressure and long
or irregular working hours.
The analysis highlighted that workplaces can introduce a number of measures to tackle the issue
and prevent risks related to data-driven management on workers’ wellbeing; for instance, prior to
adopting a technology that can enable data-driven management, establishments can involve their
workforce and discuss its possible impacts with them. Having an action plan to reduce stress in
place would also help mitigate some of the negative effects on OSH. Nevertheless, the analysis
displayed as well that establishments using the worker management enabling technologies are also
those comparatively less inclined to have psychosocial risk preventive measures in place, which
could indicate that they tend to have a limited awareness about the risks that the digital managing
technologies can bring along. Awareness-raising campaigns and other similar measures could
certainly help in increasing awareness among establishments about the implications of the use of
digital technologies for psychosocial risks and worker’s health and get the information and
knowledge needed to design and introduce specific measures to tackle the issue.
At the EU level, it should be recognised that workers in traditional workplaces may be subject to the
same risks related to algorithmic and data-driven management as platform workers, and all
establishments should carry out risk assessments that cover also the automated or semi-
automated worker monitoring and decision-making systems and ensure that such systems do not
in any manner put undue pressure, or otherwise put at risk the physical and mental health of
workers.

45
References
Adams Prassl, J. 2019. "What if Your Boss Was an Algorithm? The Rise of Artificial Intelligence at
Work." Cpmparative Labour law & Policy journal 41 (1).
Aiello, John R. 1993. "Computer-based work monitoring: Electronic surveillance and its effects."
Journal of Applied Social Psychology 499-507.
Aiello, John R., and Carol M. Svec. 1993. "Computer monitoring of work performance: Extending the
social facilitation framework to electronic presence." Journal of Applied Social Psychology
537-548.
Ajunwa, Ifeoma , Kate Crawford, and Jason Schultz. 2017. "Limitless worker surveillance." California
Law Review 735-776.
Annoni, A., P. Benczur, P. Bertoldi, P. Delipetrev, G. De Prato, C. Feijoo, E. Fernàndez-Macìas, et al.
2018. Artificial Intelligence - A European Perspective, EUR 29425 EN, JRC113826. Edited by
M. Craglia. Publication Office Luxembourg.
Aral, Sinan, Eric Brynjolfsson, and D. J. Wu. 2020. Which Came First, It or Productivity? The Virtuous
Cycle of Investment and Use in Enterprise Systems. Available at SSRN:
https://ssrn.com/abstract=942291 or http://dx.doi.org/10.2139/ssrn.942291.
Atzori, Luigi, Antonio Iera, and Giacomo Morabito. 2010. "The Internet of Things: A survey." Computer
Networks 2787-2805.
Baiocco, Sara, Enrique Fernández-Macías, Uma Rani, and Annarosa Pesole. 2022. The algorithmic
management of work and its implications in different contexts. JRC Working Papers Series
on Labour, Education and Technology, Seville: European Commission.
Ball, Kirstie. 2021. Electronic Monitoring and Surveillance in the Workplace. Literature Review and
Policy Recommendations. JRC125716, Luxembourg: Publications Office of the European
Union. doi:10.2760/5137; ISBN 978-92-76-43340-8.
Ball, Kirstie. 2010. "Workplace surveillance: an overview." Labor History 87-106.
Barocas, Solon, Alex Rosenblat, Danah Boyd, Seeta Pena Gangadharan, and Corrine Yu. 2014. "Data
& civil rights: Technology primer." Paper presented at the Data & Civil Rights Conference.
Bérastégui, Pierre. 2021. Exposure to psychosocial risk factors in the gig economy: a systematic
review. ETUI report.
Berndtsson, M, D Forsberg, D Stein, and T. Svahn. 2018. "Becoming a data-driven organisation."
European conference on Information System (ECIS).
Blackwell, D. 1953. "Equivalent Comparison of Experiments." In Ann. Math. Statist., 265-272.
Bonde, J.P. 2008. "Psychosocial factors at work and risk of depression: a systematic review of the
epidemiological evidence." Occupational and Environmental Medicine 438-445.
Brun, E., and M. Milczarek. 2007. "xpert forecast on emerging psychosocial risks related to
occupational safety and health (OSH)." Available at
https://osha.europa.eu/en/publications/report-expert-forecast-emerging-psychosocial-risks-
related-occupational-safety-and-health-osh.
Brynjolfsson, Eric, Wang Jin, and Kristina McElheran. 2021. "The power of prediction: predictive
analytics, workplace complements, and business performance." Business Economics
pages217–239.
Brynjolfsson, Erik, Lorin M. Hitt, and Heekyung Hellen Kim. 2011. "Strength in Numbers: How Does
Data-Driven Decisionmaking Affect Firm Performance?" Available at SSRN:
https://ssrn.com/abstract=1819486.
Burchell, B. 2014. "Job insecurity." In Encyclopedia of quality of life and well-being research, edited
by A., C. Michalos. Springer.

How digital technology is reshaping the art of management
46
Burgard, S., J. Brand, and J. House. 2009. "Perceived job insecurity and worker health in the United
States." Social Science and Medicine 69 (5): 777-785.
Cole, G., and P. Kelly. 2020. Management Theory and Practice, 9th Edition. Cengage.
Cooper, Cary L., and Jim Roden. 1985. "Mental health and satisfaction among tax officers." Social
Science & Medicine 747-751.
Davenport, T.H. December 2013. Analytucs 3.0. Harvard Business Review.
Davenport, Thomas H., and Jeanne G. Harris. 2006. Competing on Analytics: The New Science of
Winning. Harvard Business Press.
Delfanti, Alessandro. 2019. "Machinic dispossession and augmented despotism: Digital work in an
Amazon." (New Media and Society) 39-55.
ECS . 2019. "(European Working Conditions Survey)."
Edwards, Paul, and Paulina Ramirez. 2016. "When should workers embrace or resist new
technology?,." New Technology, Work and Employment 99-113.
Eisenberger, R., F. Stinglhamber, C. Vandenberghe, I., L. Sucharski, and L. Rhodes. 2002. "Perceived
supervisor support: contributions to perceived organizational support and employee
retention,." Journal of Applied Psychology 87 (3): 565-573.
EU-OSHA. 2018. "Foresight on new and emerging occupational safety and health risks associated
with digitalisation by 2025." Accessed 2022. Available at:
https://osha.europa.eu/en/publications/foresight-new-and-emerging-occupational-safety-
and-health-risks-associated/view.
EU-OSHA. 2019. "OSH and the Future of Work: benefits and risks of artificial intelligence tools in
workplaces." Available at: https://osha.europa.eu/en/publications/osh-and-future-work-
benefits-and-risks-artificial-intelligence-tools-workplaces/view.
Eurofound and Cedefop. 2020. European Company Survey 2019: Workplace practices unlocking
employee potential. European Company Survey 2019 series, Luxembourg: Publications
Office of the European Union.
Eurofound. 2018. "Automation, digitisation and platforms: Implications for work and employment."
By Enrique Fernández-Macías. Luxembourg: Publication Office of the European Union.
Eurofound. 2021. Digitisation in the Workplace. Luxembourg: Publication Office of the European
Union.
Eurofound. 2022. Ethics in the digital workplace. Publications Office of the European Union
Luxembourg. https://www.eurofound.europa.eu/publications/report/2022/ethics-in-the-
digital-workplace.
Eurofound. 2020. "Game-changing technologies: Transforming production and employment in
Europe." By Eleonora Peruffo, Ricardo Rodriguez Contreras, Irene Mandl and Martina Bisello.
Luxembourg: Publications Office of the European Union.
—. 2022. "Work Organisation." Accessed 06 03, 2022. Available at:
https://www.eurofound.europa.eu/topic/work-organisation.
Ferraris, Alberto, Alberto Mazzoleni, Alain Devalle, and Jerome Couturier. 2019. "Big data analytics
capabilities and knowledge management: impact on firm performance." Management
Decision 1923-1936.
Galbraith, Jay R. 1974. "Organization Design: An Information Processing View." Interfaces 28-36.
Gillespie, Tarleton. 2014. "The Relevance of Algorithms." In Media Technologies: Essays on
communication , materiality, and society, by Tarleton Gillespie, Pablo Boczkowski and Kirsten
Foot, 167.
Grande, Rafael, Alberto Vallejo-Peña, and Maria Cesira Urzì Brancati. 2021. The impact of IoT and
3D printing on job quality and work organisation: a snapshot from Spain. JRC Working
Papers Series on Labour, Education and Technology, JRC125612, Seville: European
Commission.

How digital technology is reshaping the art of management
47
Hellgren, J., and M. Sverke. 2003. "Does job insecurity lead to impaired well-being or vice versa?
Estimation of cross-lagged effects using latent variable modelling." Journal of
Organizational Behaviour 215-236.
Henderson, Troy, Jim Stanford, and Tom Swann. 2018. Under The Employer’s Eye: Electronic
Monitoring & Surveillance in Australian Workplaces. Center for Future Work.
Johnson, Jeffrey V., and E. M. Hall. 1988. "Job strain, work place social support, and cardiovascular
disease: a cross-sectional study of a random sample of the Swedish working population."
American Journal of Public Health 1336 1342.
Karasek, Robert A. 1979. "Job Demands, Job Decision Latitude, and Mental Strain: Implications for
Job Redesign." Administrative Science Quarterly 285-308.
Karasek, Robert, and Tores Theorell. 1990. Healthy Work: Stress, Productivity, and the Reconstruction
of Working Life. New York: Basic Books.
Kellogg, Katherine C, Melissa A Valentine, and Angèle Christin. 2020. "Algorithms at work: The new
contested terrain of control." Academy of Management Annals 366–410.
Khodadadi, F., A.V. Dastjerdi, and R. Buyya. 2016. "Chapter 1 - Internet of Things: an overview." In
Internet of things: Principles and Paradigms, edited by Amir Vahid Dastjerdi Rajkumar Buyya,
3-27. Morgan Kaufmann.
Klingenberg, Cristina Orsolin, Marco Antônio Viana Borges, and José Antonio Valle Antunes. 2019.
"Industry 4.0 as a data-driven paradigm: a systematic literature review on technologies."
Journal of Manufacturing Technology Management 570-592.
Koontz, Harold, and Cyril O'Donnel. 1972. Principles of Management: An Analysis of Managerial
Functions. New York: McGraw-Hill.
Kossek, E., S. Pichler, T. Bodner, and L. Hammer. 2011. "Workplace social support and work-family
conflict: a meta-analysis clarifying the influence of general and workfamily-specific
supervisor and organizational support." Personnel Psychology 64 (2).
Kurtessis, J., R. Eisenberger, L., C. Buffardi, K., A. Stewart, and C., S. Adis. 2017. "Perceived
organizational support: a meta-analytic evaluation of organizational support theory."
Journal of Management 6: 1854-1884.
Kushnir, T., and S. Melamed. 1991. "Work-load, perceived control and psychological distress in Type."
Journal of Organizational Behavior 155-168.
LaValle, Steve, Eric Lesser, Rebecca Shockley, Michael S. Hopkins, and Nina Kruschwitz. 2011. "Big
data, analytics and the path from insights to value." MIT Sloan Management Review 21-32.
Lee, Min Kyung, Evan Metsky, Laura Dabbish, and Daniel Kusbit. 2015. "Working with Machines: The
Impact of Algorithmic and Data-Driven Management on Human Workers." 33rd Annual
ACMSIGCHI Conference on Human Factors in Computing Systems. Seoul, South Korea: New
York: ACM Press. 1603-1612.
Leka, S., and T. Cox. 2008. The European Framework for Psychosocial Risk Management. WHO
publications.
Levy, Karen E. C. . 2015. "The contexts of control: Information, power, and truck-driving work." The
Information Society 160-174.
Loebbecke, Claudia, and Arnold Picot. 2015. "Reflections on societal and business model
transformation arising from digitization and big data analytics: A research agenda." The
journal of Strategic Information Systems 149-157.
Loveman, Gary W. 2003. "Diamonds in the Data Mine." Harvard Business Review. Accessed 06 03,
2022. Available at: https://hbr.org/2003/05/diamonds-in-the-data-mine.
Maltseva, Kateryna. 2020. "Wearables in the workplace: The brave new world of employee
engagement." Business Horizons 493-505.

How digital technology is reshaping the art of management
48
Muñoz Ruiz, A. 2022. "Could safety in the workplace be a pretext to control workers?" Accessed 06
03, 2022. Available at: https://www.eusocialcit.eu/could-safety-in-the-workplace-be-a-
pretext-to-control-workers-blog/.
Narayanan, Lakshmi, Shanker Menon, and Paul Spector. 1999. "Stress in the workplace: a
comparison of gender and occupations." Journal of Organizational Behavior 63-73.
Niebel, Thomas, Fabienne Rasel, and Steffen Viete. 2019. "BIG data–BIG gains? Understanding the
link between big data analytics and innovation." Economics of Innovation and New
Technology 296-316.
OTA. 1987. The Electronic Supervisor: New Technology, New Tensions. Washington DC: Office of
Technology Assessment; US Government Printing Office.
Rogelberg, S. G. 2017. "The Sage encyclopedia of industrial and organizational psychology, Second
Edition." Thousand Oakes: Sage.
Rosenblat, Alex, Tamara Kneese, and Danah Boyd. 2014. Workplace Surveillance. Data & Society
Research Institute.
Sánchez-Monedero, Javier, and Lina Dencik. 2019. The datafication of the workplace. Accessed May
2022. https://datajusticeproject.net/wp-content/uploads/sites/30/2019/05/Report-The-
datafication-of-the-workplace.pdf.
Sharda, R, D. Delen, and E Turban. 2015. Business intelligence and analytics: systems for decision
support (10th edition). Pearson Education.
Sias, P. M., and E. Gallagher. 2009. "Developing and maintaining workplace relationships." In Friends
and enemies in organisations: a work psychology perspective, edited by Morrison R. and
Wright S. (eds.), 78-100. Basingstoke: Palgrave-Macmillan.
Stansfeld, S., and B. Candy. 2006. "Psychosocial work environment and mental health: a meta-
analytic review." Scandinavian Journal of Work, Environment and Health 32 (6): 443-462.
Sverke, M., J. Hellgren, and K. Näswall. 2002. "No security: a meta-analysis and review of job
insecurity and its consequences." journal of Occupational health Psychology 7 (3): 242-264.
Todolí-Signes, Adrián. 2019. "Algorithms, artificial intelligence and automated decisions concerning
workers and the risks of discrimination: The necessary collective governance of data
protection." Transfer: European review of labour and research (Transfer) 465-481.
Urzì Brancati, Maria Cesira, and Maurizio Curtarelli. 2021. Digital tools for worker management and
psychosocial risks in the workplace: evidence from the ESENER survey. JRC125714, Seville:
European Commission.
Westman, Mina, and Dov Eden. 1992. "Excessive role demand and subsequent performance."
Journal of Organizational Behavior 519-529.
Wood, Alex J. 2021. Algorithmic Management: Consequences for Work Organisation and Working
Conditions. JRC124874, Seville: European Commission.

How digital technology is reshaping the art of management
49
List of figures
Figure 1. Simplified analytical approach .................................................................................................................................. 9
Figure 2 Job Demand-control model ....................................................................................................................................... 17
Figure 3 Use of computers and the internet by employees - EU27, 2010-2021 .......................................... 19
Figure 4 Distribution of Internet of Things, by country and firm size (2021) .................................................. 20
Figure 5 Use of IoT by its purposes in enterprises - EU27, 2021 ........................................................................... 21
Figure 6 Distribution of Artificial Intelligence technologies, by country and firm size (2021) ............... 21
Figure 7 Use of AI technology and its purposes) .............................................................................................................. 22
Figure 8 Distribution of Big data analysis, by country and firm size .................................................................... 23
Figure 9 Use and purpose of data analytics in establishments - EU27 and the UK, 2019 ..................... 24
Figure 10 Distribution of wearable devices, by country and firm size ................................................................. 25
Figure 11 Distribution of wearable devices, by industrial sector ............................................................................ 26
Figure 12 Distribution of management software applications, by country ....................................................... 26
Figure 13 Distribution of other data-driven management enabling technologies, by country and firm
size .................................................................................................................................................................................................... 27
Figure 14 Pace of work significantly determined by machines or computers and use of data
analytics, EU27 and the UK ................................................................................................................................................. 29
Figure 15 Use of data analytics by establishment type – job complexity and autonomy, EU27 and
the UK (%) ..................................................................................................................................................................................... 30
Figure 16 Share of establishments engaging in design and development of new products or services
by use of data analytics, EU27 and the UK (%) ...................................................................................................... 31
Figure 17 Collaboration with other establishments by use of data analytics, EU27 and the UK (%) 31
Figure 18 Use of data analytics by establishment type – variable pay, EU27 and the UK (%) ............ 32
Figure 19 Establishments with no employees availing of paid training or on the job learning, EU27
and the UK (%) .......................................................................................................................................................................... 33
Figure 20 Use of data analytics and workplace outcomes, EU27 and the UK (%) ....................................... 34
Figure 21 Selected management technologies by country, EU 27] ....................................................................... 35
Figure 22 Distribution of digital technologies by selected establishment characteristics, EU 27] ...... 36
Figure 23 – Establishments by type of psychosocial risk in the EU27 (%, 2019) ......................................... 37
Figure 24 Distribution of psychosocial risk by country, EU27 ................................................................................... 38
Figure 25 Establishments in the EU27 which have discussed the possible health and safety impacts
of impact of digital technologies by type of digital technologies used (%, 2019).............................. 41
Figure 26 Establishments in the EU27 which have discussed the possible health and safety impacts
of impact of digital technologies by type of issue discussed (%, 2019) .................................................. 42

How digital technology is reshaping the art of management
50
List of tables
Table 1 Prevalence of psychosocial risks in establishments by presence of different types of
technology – EU27, 2019 ..................................................................................................................................................... 39
Table 2 Measures adopted by EU27 establishments to prevent psychosocial risks (% - 2019) .......... 41

GETTING IN TOUCH WITH THE EU
In person
All over the European Union there are hundreds of Europe Direct centres. You can find the address of the centre nearest you online
(european-union.europa.eu/contact-eu/meet-us_en).
On the phone or in writing
Europe Direct is a service that answers your questions about the European Union. You can contact this service:
— by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),
— at the following standard number: +32 22999696,
— via the following form: european-union.europa.eu/contact-eu/write-us_en.
FINDING INFORMATION ABOUT THE EU
Online
Information about the European Union in all the official languages of the EU is available on the Europa website (european-
union.europa.eu).
EU publications
You can view or order EU publications at op.europa.eu/en/publications. Multiple copies of free publications can be obtained by
contacting Europe Direct or your local documentation centre (european-union.europa.eu/contact-eu/meet-us_en).
EU law and related documents
For access to legal information from the EU, including all EU law since 1951 in all the official language versions, go to EUR-Lex
(eur-lex.europa.eu).
Open data from the EU
The portal data.europa.eu provides access to open datasets from the EU institutions, bodies and agencies. These can be
downloaded and reused for free, for both comme rcial and non-commercial purposes. The portal also provides access to a wealth
of datasets from European countries.

How digital technology is reshaping the art of management
2