Analysis of Labour Market Expectations in the Digital World Based on Job Advertisements

Abstract

Job advertisements play a key role in human resource management as they are the first contact between employers and potential employees. A well-written job advertisement communicates not only the requirements and expectations of the position but also the culture, values, and goals of the organisation. Transparent and attractive advertisements increase the number of applicants and help to select the right candidates, leading to more efficient recruitment and selection processes in the long run. From a human resource management perspective, effective job advertising can give organisations a competitive advantage. Continuous changes in the labour market and technological developments require new competencies. Digitalisation, automation, and data-driven decision-making have brought IT, analytical, and communication skills to the fore. There is a growing emphasis on soft skills such as problem solving, flexibility, and teamwork, which are essential in a fast-changing work environment. Job advertisements should reflect these expectations so that candidates are aware of the competencies and skills required for the position. The aim of the study is to carry out a cross-country comparative analysis for a few pre-selected jobs based on data extracted from the CEDEFOP database as it is assumed that there are differences between countries in the European Union in terms of the expectations of workers for the same jobs.

1. Introduction

Since the economic crisis of 2008, one of the main objectives of the European Union’s economic policy has been to increase employment in order to achieve both economic stability and growth. Thanks to successful economic policy reforms over the past decade, the number of people in employment in the European Union now exceeds 200.6 million. This means an employment rate of 75.6% in May 2025. For each country, it is important to examine and pay attention to the potential labour supply for policymakers. The potential labour reserve can be divided into four groups: the economically inactive who would like to work but have not looked for work, or the economically inactive who have looked for work but for some reason have not been able to find a job, or the unemployed or underemployed (Tóth et al., 2023). The available labour reserve is low and concentrated in Central and Eastern European countries, with more than three-quarters of new entrants to the labour force being either young women with low-to-medium skills or women over 55 with upper secondary education (Götz & Jankowska, 2020). The ageing of society is also a serious problem for European countries. In the coming decades, the number of people of working age is expected to fall by several hundred thousand. This will also make the situation for employers very difficult in the future (Piątkowski, 2020; Csehné Papp & Hajós, 2011).

The current and future evolution of the labour market is influenced by several factors, including technological developments, digitalisation, economic trends, demographic changes, and social changes. Poland is one of the most dynamically developing economies in the European Union; however, in terms of technology, it lags behind the Czech Republic and Germany. This is due to the structure of the economy and the position of Polish companies within value chains. In Poland, very small enterprises dominate, often operating in local markets, and for them, technological investments pose a significant challenge. Poland builds its competitive advantage on low unit labour costs rather than on technological potential (Hetmańczyk, 2024).

Technological developments are increasingly dominating the labour market today and have caused several changes in job roles and job tasks. The spread of automation and artificial intelligence is leading to the transformation of some jobs and the emergence of new ones in others. Among the positive effects of digitalisation, the most cited is the increase in efficiency and productivity as software allows tasks to be completed much faster and reduces the administrative burden (Adams, 2018). Atypical forms of work (e.g., home office and flexible working hours) have spread faster than before thanks to digitalisation processes. The rapid development of digital technologies means that workers need to continuously improve their skills to remain competitive in the labour market (Mosteanu, 2020).

2. Literature Review

2.1. The Relationship Between Digitalisation and the Labour Market

One of the defining social and economic processes of the 21st century is digitalisation, a phenomenon that is fundamentally transforming the world of work, the structure of the economy, and the expectations placed upon the workforce (Javaid et al., 2023). The globalisation of labour markets, combined with the rapid advancement of technological innovation, demands new skill sets and a heightened level of adaptability from employees. Digitalisation not only gives rise to new occupations but also contributes to the transformation or obsolescence of many traditional professions (Rainnie & Dean, 2020).

Digitalisation refers to the conversion of analogue information into digital formats, as well as the proliferation of technologies, processes, and business models built upon this transformation (Omol, 2024). It constitutes not merely a technological shift but also a structural and functional transformation within both economic and societal systems. The integration of artificial intelligence, machine learning, robotics, automation, and big data analytics is rendering production and service processes more efficient, faster, and more flexible (Rashid & Kausik, 2024; Sjödin et al., 2023).

The impact of digitalisation is most pronounced in the restructuring of labour demand and supply (Feng et al., 2024). The declining labour needs in traditional industries are being counterbalanced by the growth of sectors that require information technology, engineering, and other forms of digital expertise. Simultaneously, there is a growing emphasis on so-called soft skills—such as problem-solving, critical thinking, collaboration, and adaptability—which are becoming increasingly valuable alongside technical knowledge (Li et al., 2024).

Because of the digitalisation, labour market expectations are undergoing a profound transformation (Richiardi et al., 2025). Employers are placing increasing importance on digital competencies, which have become fundamental not only in IT-related professions but across virtually all areas of employment. They expect workers to quickly acquire new technologies, engage in autonomous learning, and adapt flexibly to changing demands. These evolving expectations necessitate a corresponding response from educational and vocational training systems, which must provide knowledge and skills that are relevant and aligned with contemporary labour market needs.

Nevertheless, the benefits of digitalisation are not equally distributed across society. The so-called digital divide—which refers to disparities in access to and ability to use digital technologies—can exacerbate labour market inequalities (Heeks, 2022). Individuals and groups with limited access to digital tools or with lower levels of digital literacy are particularly vulnerable to the disruptive effects of technological change.

At the same time, digitalisation presents opportunities to foster inclusive economic growth. Remote work, online education, and digital entrepreneurship can enable the participation of social groups that have previously faced barriers to labour market integration due to geographical location, health issues, or family responsibilities. While technological change generates new opportunities, it also brings challenges—especially in terms of workforce adaptability and ensuring that individuals possess the appropriate level of education and qualifications.

2.2. The Importance of Competencies for Intellectual Jobs

To excel in intellectual jobs in the labour market, a variety of competencies are essential. These competencies can be broadly categorised into cognitive, non-cognitive, and technical skills, each playing a crucial role in different contexts. The cognitive skills are as follows: problem-solving (this is a critical skill across various sectors, especially in technology-rich environments; it involves the ability to address complex issues and find effective solutions), creative thinking (the ability to think outside the box and generate innovative ideas is increasingly valued), and planning and organisation (effective planning and organisational skills are necessary for managing tasks and projects efficiently) (Grigorescu et al., 2022; Korshunov et al., 2023; Lekashvili & Jamagidze, 2023).

The non-cognitive skills are the following: communication (strong language and communication skills are frequently demanded by employers as they are essential for effective collaboration and information exchange), adaptability (the ability to adapt to changing environments and new challenges is highly valued, especially in dynamic and innovative sectors), interpersonal skills (these include the ability to work well with others, which is crucial for teamwork and maintaining professional relationships) (Heijke et al., 2003; Pater et al., 2019; Poszytek et al., 2023).

In the literature, the concept of Industry 4.0 Competencies is becoming more and more common (Müller et al., 2018). The concept of competencies 4.0 highlights the need for skills that align with the demands of Industry 4.0, including advanced technical and cognitive skills. Industry 4.0 Competencies refers to the knowledge, skills, and abilities required to effectively operate, manage, and innovate within the context of Industry 4.0—the fourth industrial revolution characterised by the integration of cyber–physical systems, IoT, AI, and advanced data analytics in manufacturing and industrial processes. These competencies are generally grouped into three main categories: Technical, Methodological, and Social/Personal skills (Vrchota et al., 2020; Pokrovskaia et al., 2021; Poszytek et al., 2023).

Intellectual jobs in the labour market require a blend of cognitive, non-cognitive, and technical skills. Continuous learning, practical experience, and flexible education systems are key to developing these competencies and ensuring that individuals are well prepared to meet the demands of modern and future job markets.

The digital transformation of the economy is reshaping the labour market in several significant ways. There is a growing demand for workers with advanced qualifications and digital competencies. High-tech companies particularly need employees with ICT skills to manage digital technologies and knowledge systems (Ligonenko et al., 2022).

The rise of automation and AI is leading to structural technological unemployment, where certain jobs are replaced by machines (Dzobelova et al., 2023).

Digital labour platforms are becoming more prevalent, offering flexible work opportunities but also presenting challenges such as job instability, exploitation, and reduced social security (Graham & Anwar, 2019). These platforms simplify interactions between workers and clients, boosting productivity but also necessitating new regulatory frameworks to protect workers.

Developing digital skills is crucial for the future labour market. Higher levels of digital competencies improve the quality of life and job satisfaction for workers. Educational systems must focus on both technical and soft skills to prepare workers for the digital economy (Dieguez, 2024; Ojan et al., 2025).

2.3. Differences in Job Advertising Across the EU

Job vacancies for intellectual job managers in the European Union (EU) exhibit significant differences influenced by various factors such as cultural differences, national legislation, and market characteristics. These factors impact the requirements, recognition of qualifications, and career progression for managers across different EU member states (Stek et al., 2022).

Research on managers’ career factors in the UK, Germany, Malta, Spain, and Lithuania revealed both similarities and divergences. For instance, knowledge of languages is crucial in Spain, Lithuania, and Malta but less so in the UK and Germany. Geographical mobility is more important in the UK and Germany compared to Malta (Zakarevičius & Žukauskas, 2008).

Directive 2005/36/EC and Directive 2013/55/EU aim to extend the rights of employees and self-employed persons to practice their profession across different EU member states. However, differences between regulated and unregulated professions and national legislation pose challenges to labour mobility (Reci & Kokaj, 2023).

Establishing transparency and comparability of qualifications across member states is vital for the free movement of labour. The European Qualifications Framework aims to address these issues, but differences in how qualifications are understood nationally and trans-nationally persist (Brockmann et al., 2011).

The European Working Conditions Survey highlights diversity in job quality indicators across EU countries. Factors such as skills utilisation, task discretion, and employee participation vary significantly, impacting job vacancies and career progression for intellectual job managers (Kornelakis & Veliziotis, 2018).

Public spending on research and development positively impacts employment in creative industries. This sector employs a significant portion of the EU’s economically active population, highlighting the importance of public support for science and research (Baculakova & Harakalova, 2017).

Job vacancies for intellectual job managers in the EU are shaped by a complex interplay of cultural differences, legislative frameworks, and market characteristics. These factors result in varied job requirements, recognition of qualifications, and career progression opportunities across different member states. Understanding these differences is crucial for effective job search and career development in the EU.

2.4. Research Questions

Our research group focuses on how employer expectations have changed in recent years. We primarily analyse managerial, decision-support, and administrative job roles. Our investigation centres on identifying where and how today’s major trends and significant events (such as digitalisation and Generation Z) are reflected. In our study, we seek to answer three research questions:

• RQ1: What similarities and differences can be observed between managerial and administrative business occupations across EU countries? If we did not know the exact position, could we determine whether a job advertisement refers to a managerial or a subordinate role based solely on the competencies listed in the ad?
• RQ2: Among the competencies listed in job advertisements, which can be considered general, and which are occupation-specific across EU countries?
• RQ3: Are there country-specific characteristics that can be identified in the examined occupations?

3. Materials and Methods

3.1. Description of the Database

To answer these research questions, we collected data from the website of CEDEFOP. CEDEFOP is the European Union’s decentralised agency specialised in vocational education and training. It supports EU policymakers in developing and implementing vocational education policies. It monitors labour market trends and serves as a bridge between the world of learning and the world of work. A joint project of CEDEFOP and Eurostat is Skills-OVATE. Skills-OVATE provides information on job vacancies and skills demanded by employers, based on online job advertisements (OJAs) in 27 EU countries and five other European countries. The data are collected from thousands of sources, including private job portals, public employment service portals, recruitment agencies, and company websites. The database covers millions of OJAs across four quarters and is updated four times a year.

This study uses data from the period Q2 2024 to Q1 2025. Skills-OVATE classifies skills based on ESCO version 1.2.0 and occupations based on ISCO-08. In ESCO v1.2.0, the classification of competencies follows a hierarchical structure. The framework includes the following four main categories: Knowledge, Language skills and knowledge, Skills, and Transversal skills. These categories can be further subdivided. The structure and content of the categories were accessed from https://esco.ec.europa.eu/en/classification/skill_main (accessed on 10 June 2025). The classification and content are under continuous development. For our research, we used the categorisation that was current as of 10 June 2025.

To help understand our analysis, we find it important to present the content of the transversal skills category. “Learned and proven abilities which are commonly seen as necessary or valuable for effective action in virtually any kind of work, learning or life activity.” (https://www.cedefop.europa.eu/en/tools/vet-glossary/glossary/transversale-faehigkeiten-und-kompetenzen (accessed on 10 June 2025))

ESCO identifies six main categories of transversal skills and competencies:

• Core skills and competencies;
• Thinking skills and competencies;
• Self-management skills and competencies;
• Social and communication skills and competencies;
• Physical and manual skills and competencies;
• Life skills and competencies.

The database categorises occupations based on the current version of the International Classification of Occupations, i.e., ISCO-08. In our work, we, therefore, adopt the ISCO definition of an occupation. According to this, “The concept of occupation is defined as a set of jobs whose main tasks and duties are characterised by a high degree of similarity.”

The occupational groups examined are the following:

• 121. Business services and administration managers (e.g., Finance Managers, Human Resource Managers, Policy and Planning Managers, and Business Services and Administration Managers Not Elsewhere Classified).
• 241. Finance professionals (e.g., Accountants, Financial and Investment Advisers, and Financial Analysts).
• 242. Administration professionals (e.g., Management and Organisation Analysts, Policy Administration Professionals, Personnel and Careers Professionals, and Training and Staff Development Professionals).
• 331. Financial and mathematical associate professionals (e.g., Securities and Finance Dealers and Brokers, Credit and Loans Officers, Accounting Associate Professionals, Statistical, Mathematical and Related Associate Professionals, and Valuers and Loss Assessors).
• 333. Business services agents (e.g., Clearing and Forwarding Agents, Conference and Event Planners, Employment Agents and Contractors, Real Estate Agents, and Property Managers).
• 334. Administrative and specialised secretaries (e.g., Office Supervisors, Legal Secretaries, Administrative and Executive Secretaries, and Medical Secretaries).

The most recent data download was carried out on 11 June 2025. We downloaded data for the EU27 countries, based on the ESCO skill classification, at the 3-digit ISCO and level 3 ESCO skill depth. The data were downloaded in two phases. For answering RQ1 and RQ2, we downloaded data for the six occupational groups across all EU member states combined. Competencies were retrieved at both level 0 and level 3. The dataset includes, at the 3-digit ISCO and level 3 ESCO skill depth, the number of online job advertisements (OJAs) and the number of OJAs within each occupational group that mentioned a given skill. The second dataset was used for addressing RQ3. In this query, we repeated the first data request with the exception that the data were collected separately for each EU member state, rather than for the EU as a whole. For both datasets, we divided the number of mentions by the number of OJAs, thus obtaining the mention ratio. The available period for download was Q2 2024 to Q1 2025. The number of available OJAs during the examined period is shown in

Table 1. Number of online job advertisements surveyed by country and occupation.

Country Code/Occupation Code	OC121	OC241	OC242	OC331	OC333	OC334	Total
AT	2681	2971	5170	5091	2709	5773	24,395
BE	6870	12,297	18,029	23,163	10,886	19,275	90,520
BG	277	337	957	988	726	174	3459
CY	7	11	13	15	13	11	70
CZ	1536	4973	1415	3551	1571	1795	14,841
DE	44,499	59,150	101,502	63,267	40,159	141,584	450,161
DK	4144	1321	2764	3568	1583	3006	16,386
EE	54	33	286	225	83	148	829
EL	516	1038	695	869	1848	1101	6067
ES	1921	7327	3726	3885	22,390	10,830	50,079
FI	12	22	30	13	12	12	101
FR	50,253	57,862	46,547	146,223	112,139	92,803	505,827
HR	386	92	202	237	272	365	1554
HU	2890	2627	1645	6610	1669	3831	19,272
IE	748	443	798	255	262	893	3399
IT	5773	6223	14,922	8812	14,901	32,746	83,377
LT	2018	1937	1189	2227	1635	1907	10,913
LU	7	3	5	25	5	11	56
LV	182	509	428	570	189	201	2079
MT	352	528	213	254	213	326	1886
NL	1930	1418	4849	3215	5066	5229	21,707
PL	9895	13,411	7297	9170	6112	2158	48,043
PT	668	730	1691	1065	1780	884	6818
RO	19	43	54	62	124	80	382
SE	12,576	2443	10,092	11,408	2652	6059	45,230
SI	199	83	311	337	81	105	1116
SK	2141	1665	2389	4373	2378	2012	14,958
Total	152,554	179,497	227,219	299,478	231,458	333,319	1,423,525

.

3.2. Limitation of the Research

Occupations can be extracted from the database based on at most the first three digits of the ISCO code. As a result, we are only able to examine groups of occupations, not specific individual occupations. This naturally represents a limitation of our research. In our study, we compared six occupational groups from the first three ISCO major groups (1. Managers, 2. Professionals, 3. Technicians and Associate Professionals).

We analysed data from over 1.4 million job advertisements (

Table 2. Number of competencies tested by country and occupation.

Country Code/Occupation Code	OC121	OC241	OC242	OC331	OC333	OC334
AT	69	75	77	79	76	86
BE	99	95	112	122	119	137
BG	36	47	44	38	55	28
CY	7	48	12	28	26	29
CZ	53	69	54	55	53	48
DE	157	146	175	153	161	169
DK	82	66	83	85	78	99
EE	10	10	15	4	4	6
EL	40	49	41	34	64	41
ES	68	86	78	71	122	89
FI	24	26	35	19	13	21
FR	171	161	175	196	200	215
HR	28	29	8	26	30	27
HU	60	74	62	76	73	59
IE	52	59	54	42	69	68
IT	83	89	94	87	113	94
LT	52	73	47	57	47	53
LU	36	18	16	24	19	24
LV	22	35	30	37	32	25
MT	43	52	40	40	51	47
NL	77	86	104	104	129	119
PL	98	98	95	98	108	71
PT	51	64	64	58	71	57
RO	23	43	42	34	54	35
SE	113	94	110	123	100	116
SI	27	15	24	32	17	15
SK	58	77	70	76	69	62

). Of these, 32% were from Germany and 36% from France. The remaining 25 EU countries accounted for 32% of the advertisements. In total, 10.7% of the ads targeted managerial positions, 28.6% were for professional roles, and 60.7% for technicians and associate professionals. A limitation of our research is that the distribution of available OJAs by country does not reflect the actual employment structure of the countries studied. The most significant discrepancies can be observed in the cases of Romania and Finland. In 2023, the number of employed persons aged 15–64 was 2.532 million in Finland and 7.614 million in Romania. However, the number of Finnish OJAs was only 101, while the number of Romanian advertisements was 382. Our results, therefore, must be interpreted with this limitation in mind.

The number of competencies covered per occupation is as follows: OC121: 224, OC241: 218, OC242: 242, OC331: 246, OC333: 257, and OC334: 264. An additional research limitation is that, in the case of several countries, the job advertisements included only a few of the competencies listed in the ESCO version 1.2.0. In the case of Estonia, six occupations, and in the case of Luxembourg, five occupations featured fewer competencies than 10% of those associated with the respective occupation in ESCO. For Finland and Slovenia, there are three such occupations each. Finally, our research possibilities are further limited by the fact that we were only able to download data aggregated over four quarters from the website https://www.cedefop.europa.eu/en/tools/skills-online-vacancies/occupations/skills (accessed on 10 June 2025), which means we are unable to conduct any time-based comparisons.

3.3. Description of the Methods Used

To answer RQ1, we conducted multidirectional association strength analyses. We examined the correlation values between occupations based on Pearson’s correlation coefficient, using the average values across the EU-27. We performed cross-tabulation analysis between ESCO-1 levels (1. Managers, 2. Professionals, 3. Technicians and Associate Professionals) and the main competency groups (knowledge, language skills and knowledge, skills, transversal skills), as well as between the main competency groups and the six occupational groups included in the study. During the cross-tabulation analysis, we conducted Chi-square tests and calculated Cramér’s V as a measure of association strength.

Using ANOVA tests, we calculated the F-test and Eta coefficient to measure the strength of association between the proportion of mentions and (1) ISCO-1 levels, (2) main competence groups, and (3) occupational groups. Analysis of variance is applicable in the case of mixed relationships. A mixed relationship is when we examine the connection between qualitative/categorical and quantitative variables. ISCO-1 levels, main competence groups, and occupational groups were considered as qualitative variables. In the case of mixed relationships, the measure of association is the Eta (η) coefficient. This coefficient ranges between 0 and 1, where 0 indicates no relationship and 1 indicates a functional relationship. In our calculations, we considered the strength of association as weak up to 0.3, moderate between 0.3 and 0.7, and strong above 0.7. The relationship analyses were carried out using SPSS 30.0 and Microsoft Excel 365.

In our cluster analysis, we included the examined occupational groups as variables and aimed to cluster the competencies. For this, we chose the method of hierarchical cluster analysis, specifically Ward’s method. In hierarchical clustering, it is not necessary to define the number of clusters in advance. The number of clusters was determined using a dendrogram. Since we worked with more than two variables in all cases, we used descriptive statistics to characterise the clusters (Székelyi & Barna, 2002; Sajtos & Mitev, 2007; Jánosa, 2015).

To answer RQ3, we examined whether there was any correlation between the main competence groups for each of the six occupational groups. Using cross-tabulation analysis, we examined how competencies are distributed across countries by competence groups. Independence tests were conducted in SPSS for these. The strength of association was assessed using Cramer’s V.

As a first step in reducing the number of variables, we chose the method of factor analysis. During the runs, we checked the following threshold criteria:

• The value of Pearson correlation should be at least 0.3;
• The Kaiser–Meyer–Olkin Measure of Sampling Adequacy (KMO) should be above 0.5;
• The diagonal values of the Anti-Image Correlation Matrix should be above 0.5;
• The cumulative explained variance should be at least 60%.

In the Bartlett’s test, we considered a significance level of 0.05 as the critical threshold. The limited number of countries, as a finite and small sample size, restricted the number of variables that could be included in the principal component analysis. As a result, we were unable to group the competencies into factors while meeting the above conditions.

In the second step, we analysed each main competence group separately for each occupational group. For the knowledge, skills, and transversal skills categories, we determined the frequency of mention for each competence within each main group (see

Table 3. Summary of the most frequently mentioned competencies by occupational group.

Occupational Group\Competence	Personal Skills and Development (Knowledge)	Accounting and Taxation (Knowledge)	Accessing and Analysing Digital Data (Skills)	Demonstrating Willingness to Learn (Transversal Skills)
OC121	1380.25%		1387.40%	1632.63%
OC241	1378.33%		1301.49%	1589.05%
OC242	1240.91%		1019.02%	1597.74%
OC331		1183.39%	849.91%	1230.89%
OC333	1121.36%		814.27%	1270.96%
OC334	821.83%		1053.97%	1431.43%

).

Personal skills are defined by the ESCO classification based on the effects on an individual’s abilities. On this basis, it includes, among others, the following: Argumentation and presentation, Assertiveness training, Communication skills, Co-operation, Development of behavioural capacities, Development of mental skills, Job-seeking programmes, Parenting courses, Public speaking, Self-esteem skills, Social competence, and Time management.

By ESCO, accounting and taxation is the study of maintaining, auditing, and recording financial transactions. Programmes and qualifications with the following main content are classified here: Accounting, Auditing, Bookkeeping, Tax Accounting, and Tax Management. Accessing and analysing digital data means using digital tools to browse, search, filter, organise, store, retrieve, and analyse data, information, and digital content, to collaborate and communicate with others, and create and edit new content. Finally, by demonstrating willingness to learn, ESCO means showing a positive attitude towards new and challenging demands and taking steps to learn from difficulties. A hierarchical cluster analysis was conducted using Ward’s method, selecting the competencies with the highest overall value as cluster variables from each competence cluster. We used the tools of descriptive statistics to look at the mean, standard deviation, and number of elements for the resulting clusters.

4. Results

The CEDEFOP database contains 90 types of knowledge, 263 skills, and 40 different transversal skills and competencies for the period 2024Q2–2025Q1. Only one instance of language skills and knowledge appears. The data for the occupational groups that we examined are presented in

Table 4. Number of competencies by occupational group and competence category.

ESCO_Hier_Level_0	OC121	OC241	OC242	OC331	OC333	OC334
knowledge	70	65	70	75	74	78
language skills and knowledge	1	1	1	1	1	1
skills	130	125	140	142	155	157
transversal skills and competencies	23	27	31	28	27	28

. In the analysis of the six major groups, we examined a total of 84 types of knowledge, 196 types of skills, and 34 different transversal skills and competencies. Thus, in each major competence category, at least 75% of the competencies are represented in our research as well.

In those major competence categories where the number of competencies is significantly higher, we can observe very low average values and very high relative standard deviations. In the case of language, since only one competence is assigned to each job, the average occurrence value is much higher, and the standard deviation is minimal. Based on

Table 5. Average value, standard deviation, and relative standard deviation of mention rates by competence category.

Category	Mean	N	Std. Deviation	Relative Standard Deviation (RSD)
knowledge	0.0359	432	0.1001	279.0%
language skills and knowledge	0.4677	6	0.0315	6.7%
skills	0.0284	849	0.0664	234.1%
transversal skills and competencies	0.1113	164	0.1700	152.8%

, it can be concluded that the more competencies a competence category contains, the lower its average mention value. Larger competence categories are associated with higher relative standard deviations.

There is a moderate (H = 0.071) but non-significant (p = 0.203) relationship between competencies and occupational groups. In this case, the average mention values per occupational group are very similar. For OC334, the average mention rate is below 3%, while for OC121, it is above 5%. For the other occupational groups, the average values range between 4 and 5%. The standard deviations are of a similar magnitude, typically around 0.1. In all cases, the relative standard deviation exceeds 200%.

As previously mentioned, our first research question was as follows: What similarities and differences can be observed between managerial and administrative/business support occupations across EU countries? If we were not aware of the specific job title, could we determine based on the competencies listed in a job advertisement whether the position is a managerial or non-managerial role?

To answer this question, we examined the strength of the relationship between competence categories and occupational groups. The relationship between competence categories and job hierarchy levels is very weak (Cramer’s V = 0.022) and non-significant (p = 0.965). Due to the low number of language-related skills, the assumptions for independence testing are not met. Therefore, our computational result is not suitable for drawing general conclusions—it is only interpretable within the context of this specific dataset. The frequency of occurrence of individual competencies within competence categories, and across job levels (from manager to administrator), shows no meaningful relationship. Consequently, it is not possible to determine the job level being advertised based on the competence categories used in the examined postings.

Similarly, the relationship between competence categories and occupational groups is also very weak (Cramer’s V = 0.036) and non-significant (p = 1.000). Here, too, the assumptions for independence testing are not fulfilled due to the low number of language-related skills.

In the mixed-relationship analysis, our quantitative variable was the frequency of competence mentions. The qualitative variables were job level and occupational group. The relationship between competence mention frequency (quantitative) and job level (qualitative) is very weak (H = 0.071) and non-significant (p = 0.123). Likewise, the relationship between competence mention frequency and occupational groups is also very weak (H = 0.054) and non-significant (p = 0.203). Although we can only make statements about the given dataset, we can conclude the following:

• Competencies cannot be assigned to different job levels.
• Competencies cannot be assigned to specific occupational groups.

Therefore, it is not possible to determine either the job level or the occupational group based on the competencies listed. Based on the 1.4 million job advertisements analysed, there is no difference in the competencies expected for managerial versus non-managerial positions.

Which competencies are general, and which are occupation-specific? In the examination of RQ2, we attempted to reduce the occupational groups using principal component analysis (PCA). The result was a single component with an explanatory power of 84%. However, cluster analysis cannot be performed using only one quantitative variable. Therefore, in our cluster analysis, we treated all six occupational groups as separate variables. Based on this, we were able to classify the competencies into four clusters. Naturally, this was only possible for those competencies that appeared in all occupational groups. Out of the 84 types of knowledge, 54 were included in the clusters; out of the 196 types of skills, 90; and out of the 34 different transversal skills and competencies, 21 were included. The average values of each cluster by occupational group are shown in

Table 6. Average values of clusters by occupational group.

Clusters	Number/Case	Average/OC121	Average/OC241	Average/OC242	Average/OC331	Average/OC333	Average/OC334
1	9	0.265	0.267	0.210	0.399	0.199	0.172
2	122	0.010	0.012	0.012	0.008	0.010	0.011
3	29	0.138	0.102	0.110	0.069	0.139	0.067
4	6	0.576	0.492	0.543	0.416	0.416	0.392
Total	166	0.066	0.059	0.059	0.054	0.058	0.043

.

Cluster 1 includes competencies with an average mention frequency three to seven times higher than the overall average. These are accounting and taxation, computer use, work skills, conducting gaming activities, managing, gathering and storing digital data, performing general clerical and administrative tasks, using digital tools for collaboration and productivity, and showing initiative and working efficiently. Due to the competencies included, we propose naming this cluster High-Level Efficiency and Digital Competencies. Cluster 4 contains those competencies whose occurrence frequency is at least seven times the average. The competencies included in this cluster and their average mention rates are shown in

Table 7. Cluster competencies and average mention rates.

Esco_Hier_Level_0	Esco_Hier_Level_3	OC121	OC241	OC242	OC331	OC333	OC334
knowledge	management and administration	0.650	0.443	0.514	0.324	0.398	0.088
knowledge	personal skills and development	0.570	0.448	0.545	0.139	0.391	0.273
language skills and knowledge	languages	0.478	0.477	0.512	0.453	0.417	0.470
skills	accessing and analysing digital data	0.557	0.454	0.451	0.414	0.372	0.374
transversal skills and competencies	collaborating in teams and networks	0.538	0.492	0.526	0.571	0.406	0.492
transversal skills and competencies	demonstrating willingness to learn	0.663	0.638	0.706	0.594	0.511	0.657

. It can be observed that, as ESCO levels decrease, the mention frequency of a given competence typically also decreases. We have named the fourth cluster Core Cross-cutting Competencies. These are the competencies that employers expect regardless of occupational group.

Cluster 2 is characterised by a dominance of skills. Of the 122 competencies included in the cluster, 57% are skills. The proportion of knowledge is also above average. The most frequently mentioned knowledge items are database and network design and administration, food processing, and law. The most frequently mentioned skills include accompanying and welcoming people, allocating and controlling physical resources, installing interior or exterior infrastructure, monitoring and evaluating the performance of individuals, performing artistic or cultural activities, purchasing goods or services, teaching and training, and weighing. Among these, law, accompanying and welcoming people, allocating and controlling physical resources, and teaching and training have mentioned rates exceeding 2% across all occupational groups. Based on its characteristics, this cluster is proposed to be named “A Little Knowledge and Skill Go a Long Way”.

Cluster 3 features a prominent role of non-knowledge-based competencies. This cluster is dominated by competencies related to communication and creativity. In all occupational groups, the following competencies have a mention rate above 5%: economics (as knowledge) and communicating with colleagues and clients; communication, collaboration, and creativity; developing solutions; managing budgets or finances; planning events and programmes (as skills). Additionally, the following transversal skills and competencies appear to cope with stress, leading others, taking a proactive approach, and thinking creatively and innovatively.

Are the excluded competencies occupation-specific? In the case of managerial job vacancies, six competencies only appear in this occupational group. Three of these competencies have a mention rate of more than 2%. Within the skills category are the following:

• Complying with health and safety procedures (17.0%);
• Developing recipes or menus (2.6%);
• Promoting products, services, or programmes (2.8%);
• Records, reports, or budgets (2.2%).

Among non-managerial occupations, there is only one competence that is exclusive to a single occupational group and has a mention rate above 2%. This is found in group 242. Administration professionals: transversal skill and competence—instruct others, with a mention rate of 2.7%. Some competencies appear in multiple occupational groups but only exceed the 2% threshold in one of them. For group 242, administration professionals, such as competencies, include the following:

• Advising on educational or vocational matters (skills) (2.6%);
• Operating machinery for the manufacture and treatment of textiles, fur, and leather products (skills) (2.2%);
• Adapt to change (transversal skill and competence) (2.6%);
• Manage quality (transversal skill and competence) (7.6%).

In group 331, financial and mathematical associate professionals, there is one exclusive skill: marking materials or objects for identification (14.9%). In group 333, there is also a single skill for business services: providing information and support to the public and clients (2.9%). The remaining two occupational groups have no competencies that are exclusively characteristic of them. Competencies are specifically only to 2. Professionals or only group 3. Technicians and associate professionals could not be identified. However, three competencies are typical for at least four out of the five non-managerial groups, with mention rates above 2%: documenting technical designs, procedures, problems, or activities—mention rates: 8.8%, 3.8%, 7.2%, 1.6%, and 2.8%. Two transversal skills and competencies occur: advise others: 40.7%, 1.5%, 50.3%, 57.4%, and 9.4%; attention to detail: 4.5%, 1.8%, 4.1%, 2.9%, and 4.4%.

Interestingly, the competence ensuring compliance with legislation appears not only in group 1 (Table 6) (Managers) but also in several non-managerial occupations (121. Business services and administration managers 5.6%, 242. Administration professionals 5.7%, and 331. Financial and mathematical associate professionals 6.5%).

RQ3 is as follows: Are there country-specific characteristics observable in the case of the examined occupations? Except for Estonia, in all EU member states, skills represent the largest share within the major competence groups. However, the situation differs by occupational group. For occupational group 121. Business services and administration managers, the OJAs (Occupation and Job Analyses) contain at least as many knowledge elements as skills in Cyprus, Estonia, Lithuania, and Romania (see Figure 1). In the occupational groups 241. Finance professionals and 333. Business services agents, the number of listed knowledge items also reaches or exceeds that of skills in Estonia, Luxembourg, and Slovenia. In the case of 242. Administration professionals, all types of competences (behind the major competence groups) appear in equal numbers in Cyprus—except for language. For 331. Financial and mathematical associate professionals, Estonia is again the exception, where transversal skills and competences are dominant. For 334. Administrative and specialised secretaries, transversal skills are the dominant competence group in all countries.

We conducted a cross-tabulation analysis for each occupational group. The results by occupational group are as follows:

• 121. Business services and administration managers—Cramer’s V = 0.119, p = 0.739.
• 241. Finance professionals—Cramer’s V = 0.097, p = 0.998.
• 242. Administration professionals—Cramer’s V = 0.124, p = 0.371.
• 331. Financial and mathematical associate professionals—Cramer’s V = 0.122, p = 0.418.
• 333. Business services agents—Cramer’s V = 0.108, p = 0.789.
• 334. Administrative and specialised secretaries—Cramer’s V = 0.129, p = 0.137.

Based on the cross-tabulation analysis, in all cases, there is a weak, non-significant relationship between the countries and the major competence groups. The conditions for a test of independence are not met in any of the cases.

As previously presented, for each occupational group, we selected the highest total value competences from each major competence group as cluster variables. The number of clusters and the countries that could not be assigned to any cluster are shown in

Table 8. Number of clusters and countries not classified into any cluster by occupational group.

Code	OC121	OC241	OC242	OC331	OC333	OC334
Number of clusters	4	3	4	5	3	4
Countries not in the cluster	Cyprus Estonia	-	Cyprus Estonia	Estonia Ireland	Cyprus Estonia	Estonia

.

In all six cases, we examined the relationship between the variables and the clusters as qualitative characteristics (Figure 2). In five occupational groups, there is a significant and strong relationship between the clusters and the variables. The only exception is 333. Business services agents, where, for the language competence, the Eta (H) coefficient is 0.467 at a 6.7% significance level.

For 121. Business services and administration managers, the countries in Cluster 1 are characterised by all competences being mentioned at above-average levels. Cluster 2 includes countries slightly above average, while Cluster 3 includes countries slightly below average. The countries in Cluster 4 fall at least 60% below the EU average. A similar pattern can be observed for OC241 and OC333: countries in Cluster 1 are above average, those in Cluster 2 are near average, and Cluster 3 contains countries that fall significantly below average (by at least 65% in OC241 and at least 33% in OC333). For 242. Administration professionals, the first three clusters behave similarly to those in the OC241 group. The exception is Cluster 4, which contains only one country—Luxembourg.

A somewhat different pattern emerges in groups OC331 and OC334. In 331. Financial and mathematical associate professionals, clusters similar to those seen in OC241 appear. Additionally, in Cluster 3, one competence—accounting and taxation—is mentioned more than 50% above average, while the frequency of the remaining competences is 15–40% below average. In Cluster 5, the countries are characterised by two competences—demonstrating willingness to learn and languages—being mentioned significantly more frequently than average (+56%, +31%), while the other two competences occur at around average levels (86%, 94%).

The most diverse pattern is observed in 334. Administrative and specialised secretaries. In Cluster 1, two competences—demonstrating willingness to learn and languages—are mentioned at least 25% more frequently than average, while accessing and analysing digital data and personal skills and development appear at around average levels. Cluster 2 presents the opposite: accessing and analysing digital data and personal skills and development are mentioned at least 25% above average, while the other two competences occur at average levels. Cluster 3 includes countries where the frequency of all competences falls at least 25% below average. In Cluster 4, which includes Ireland and Malta, the language competence is at 75% of the average value, while the other competences are mentioned more frequently than in any other cluster.

5. Conclusions

Leaders typically provide direction, coordinate activities, motivate others, make decisions, and assume responsibility. They are primarily expected to possess soft skills. In contrast, subordinates are tasked with the precise and reliable execution of assigned duties, where professional expertise plays a more central role. Thus, different expectations apply to leaders and subordinate employees. Different occupations and job categories are associated with distinct tasks and responsibilities. However, in terms of competencies, only minimal differences can be observed among the groups examined. The frequency of occurrence of specific competencies in online job advertisements (OJAs), both within main competency categories and across hierarchical job levels, shows no clear relationship. Based on the competency categories used in the advertisements, it is not possible to determine the hierarchical level of the position being advertised. Job level cannot be reliably identified based solely on OJAs, and even the occupational group is difficult to ascertain.

Based on our analysis, we find it important to draw the attention of human resource managers to the following implication: if the statement “a leader is not the same as a subordinate” is to be regarded as valid, then this distinction should be reflected in OJAs as well. This is particularly relevant in light of Eurostat and Cedefop surveys, which indicate that approximately 70–80% of job seekers search for employment online. In most EU countries, the vast majority of job advertisements—around 80–90%—are published online.

We were able to identify competences that are independent of occupational groups. These can be considered constant, “epic epithet-like” competences: management and administration; personal skills and development; languages; accessing and analysing digital data; collaborating in teams and networks; demonstrating willingness to learn.

The results of the cluster analysis by occupational group are summarised in Appendix A 

Table A1. Clustering of EU27 countries by occupational group.

Country/Occupational Group	OC121	OC241	OC242	OC331	OC333	OC334
AT	1	1	1	1	1	1
BE	2	2	2	2	2	1
BG	3	3	3	2	2	1
CY		1		3		2
CZ	2	2	2	4	3	2
DE	2	1	1	5	2	1
DK	2	1	1	5	1	1
EE	4	3
EL		2	3	3	3	2
ES	3	3	2	3	3	3
FI	2	3	2	5	3	1
FR	2	2	2	1	2	1
HR	3	2	3	2	3	1
HU	3	2	2	2	3	3
IE	1	1	1		1	4
IT	1	1	1	2	2	1
LT	3	2	2	2	3	3
LU	2	2	4	3	3	3
LV	2	2	2	2	3	2
MT	1	1	1	5	1	4
NL	2	1	1	5	1	1
PL	4	3	3	4	3	3
PT	2	2	2	2	3	2
RO	3	2	2	3	2	2
SE	2	1	1	5	1	1
SI	4	3	3	4	3	3
SK	3	2	2	2	3	2

. We examined which countries fall into the same cluster at least five times across the six occupational groups.

Based on the classification, we were able to form seven groups. Belgium and France are characterised by consistently providing above-average numbers of competences in their OJAs. For all occupations, they show a slightly below-EU-average demand for demonstrating willingness to learn, while expectations related to computer use and accessing and analysing digital data appear at around or slightly above the EU average.

Czechia, Latvia, Portugal, Romania, and Slovakia show below-average demand for language skills and knowledge as well as for transversal skills and competences.

Greece, Hungary, Lithuania, and Luxembourg display below-average values across all examined cluster variables. The shortfall for language skills and knowledge, and transversal skills, and competences exceeds 30%. A common feature is that at Level 3—Technicians and associate professionals, the shortfall in collaborating in teams and networks and demonstrating willingness to learn exceeds 50%.

In Bulgaria, Finland, and Croatia, the number of competences listed per occupational group in the OJAs is between one-third and two-thirds of the EU average. The shortfall is around 40% for personal skills and development and accessing and analysing digital data. For computer use and collaborating in teams and networks, the shortfall reaches 60% compared to the EU average.

Germany, Denmark, the Netherlands, and Sweden perform at or above average across all variables. They place particular emphasis on personal skills and development and demonstrating willingness to learn, exceeding the EU27 average by 46% and 25%, respectively.

Ireland and Malta form a distinct group. For all frequently occurring competences—except one—they exceed the EU average by at least 25%. In the case of personal skills and development, the difference exceeds 100%. Unsurprisingly, the only competence where they score below average is language.

Poland and Slovenia represent the opposite of the previous group. For all frequently occurring competences, the frequency of mentions is only about 25–35% of the EU27 average.

Human Resource Management (HRM) is no longer merely an administrative function. It has evolved from basic payroll processing to a form of strategic partnership, supporting organisations in adapting rapidly to changing environments. HR professionals are now required to simultaneously address a variety of challenges: managing digital transformation, adapting to new forms of work, navigating intergenerational conflicts among the X, Y, and Z generations, and identifying and developing new competencies necessitated by automation and artificial intelligence. In this context, online job advertisements (OJAs) increasingly emphasise competencies that are difficult to develop but are essential for enabling workers to respond to rapid change and remain resilient during digital transitions. These competencies have gained a dominant presence in OJAs.

Despite global challenges, human resource issues must be addressed locally as the economic structure, level of development, and cultural context differ significantly across individual countries and country groups. For example, Western European education systems tend to be more practice-oriented and place greater emphasis on developing digital competencies than those of countries that joined the EU after 2004. These structural differences are reflected in the characteristics of local labour markets. HR professionals operating in specific regions cannot overlook the competencies already present in the local workforce, even amid overarching global trends. Although certain limitations of our research must be acknowledged—such as the reliance on the first three digits of ISCO codes, the differing ratios of OJAs to employed individuals across countries, and the limited number of competencies analysed in four countries—a clear divergence among Member States can still be identified. Founding EU members and those that joined before 2004 typically emphasise transversal skills and competences in job advertisements. They also place high importance on personal skills and development as well as accessing and analysing digital data. In contrast, most of the countries that joined the EU after 2004 mention these competencies at significantly lower rates compared to the EU average. Thus, regional disparities are evident not only in economic and developmental terms but also in labour market competencies.