Opting out of analytics: The tacit compliance of journalists in datafied newsrooms

Abstract

Presenting audiences and journalistic work in quantitative terms has become an established practice in news organisations. The use of audience-generated data as a measurement for journalistic performance is not without problems as questions arise about the extent to which datafied logics generate accurate representations of qualitative, complex, and pluriform journalistic work. In this article, we examine how journalists in a public service media organisation approach analytics by looking at the extent to which they engage in what we call data-related job crafting. We also study how their level of engagement with job crafting relates to their perceptions of their work environment and occupational well-being. The study is based on a survey with 121 journalists working for a Nordic public-funded media organisation. We show that in today’s datafied news media environment, opting out of the use of analytics manifests itself as a kind of private, tacit compliance; it is connected to work-related malaise but not related to the avoidance of analytics extensively at an organisational level. We argue that this reflects the institutionalisation of audience analytics in news organisations, which has made opting out more of a gesture than a real possibility. We conclude the article with a reflection on how the institutionalisation of analytics is challenging the central social justification of public service media.

Keywords

Audience metrics datafication institutionalisation job crafting journalism media work public service media

Introduction

Over the past decade, increased datafication of social action and work life has facilitated new opportunities for journalism and impacted its economic models (Hanusch, 2017). Datafication is associated with collecting, databasing, quantifying, analysing, and using data as resources for knowledge production, service optimisation, and value creation (Flensburg and Lomborg, 2023; Mayer-Schönberger and Cukier, 2013). Audience analytics, that is, the systems and software that enable the measurement of data related to the consumption and interaction of news media content, produce metrics, that is, numerical measures of the attention that the content receives (Tandoc, 2019; Zamith, 2018). By tracking metrics, news media organisations can keep track of their audience’s online behaviour: which news links they click on, which news they share and what kind of retention news articles have (Tandoc, 2019). The use of analytics provides news organisations with a better understanding of their audience and, as a result, a stronger sense of control over predicting audience behaviour (Ahva and Ovaska, 2023). Hence, the collection of audience data is often justified by its economic potential and audience-minded approach (Costera Meijer, 2020; Ferrer-Conill and Tandoc, 2018; Zamith, 2018).

The use of audience metrics has, however, also sparked controversy within and outside news media organisations. There is an ongoing debate about both the impact of analytics on the media economy and the insight that metrics provide into the audiences of journalism (Heikkilä, 2022). Whereas in the past, decisions on journalistic content have been made more explicitly in the editorial offices, in the era of datafication, news media increasingly share decision-making power with audience-generated data and algorithm-based platforms (Salonen et al., 2023).

From the perspective of individual journalists, the main concern about the use of analytics is that it may call into question their professional journalistic autonomy. Editorial autonomy can be defined as journalists’ control over their professional activities and is considered to include the freedom of journalists to choose the stories they consider newsworthy and to emphasise certain aspects of their stories (Hamada et al., 2019). As long as journalists perceive the use of analytics as a threat to their editorial autonomy, they are likely to resist the use of analytics in their work (Bunce, 2019; Petre, 2018). Previous studies (e.g., Ferrer-Conill and Tandoc, 2018; Nelson and Tandoc Jr., 2019; Salonen et al., 2023) paint a picture of how journalists engage in internal negotiations on the use of analytics and seek to align them with the values of professionalism in journalism. In this negotiation, which takes place in the context of everyday news production, journalists reconcile the standards of independent journalism with the standards of providing the public with relevant information on current events (Ekström et al., 2022). Central to this process is the consideration of the extent to which the public should be provided with the content they want, giving priority to the industry market, or the content they need, giving priority to the democratic work of journalism (Barger and Barney, 2004).

Research also shows that there are different views within newsrooms on the importance of data as part of journalistic work (Hendrickx et al., 2021; Lamot and Paulussen, 2020; Omidi et al., 2024; Salonen et al., 2023) and that managers are more positive than their subordinates about the use of metrics to evaluate journalistic content (Ahva et al., 2024; Belair-Gagnon, 2019; Hanusch, 2017). The reasons for this are considered to be both the economic opportunities offered by analytics and the possibility of monitoring both audience behaviour and the performance of journalists through metrics (Bunce, 2019; Zamith et al., 2020). Indeed, previous research (e.g., Ahva et al., 2024; Bunce, 2019; Lamot and Paulussen, 2020; Lee and Tandoc, 2017) shows that newsroom managers use audience metrics to evaluate their employees and move them in a favoured direction. Tensions have been found to arise when managers organise the work processes of their subordinates in such a way that journalists feel that their professional values, such as delivering independent critical content, are threatened (Lischka, 2020). In particular, situations where metrics are used to encourage the profit potential of content produced by journalists rather than in any way justifying the quality of journalistic content, have been found to alienate journalists from their work and the media they represent (Christin and Petre, 2020).

Recent research has also demonstrated a range of intra- and interpersonal negotiation strategies that journalists use to “make peace with metrics” (Christin and Petre, 2020). These include drawing a moral line between “good” and “bad” metrics and focusing attention on moments when metrics indicate that content that serves journalistic values is aligned with audience traffic (Christin and Petre, 2020). Analytics may then take on a less intimidating “informative” role in the eyes of journalists (Ahva et al., 2024; Hendrickx et al., 2021). When journalists accept the presence of analytics in their work and do not perceive it as a threat to their professional self-regard, they also facilitate the institutionalisation of analytics in the news media environment (see Ahva and Ovaska, 2023; Christin and Petre, 2020).

Previous research has largely focused on explaining what drives the adoption of audience analytics in newsrooms and how analytics shape the journalistic profession (e.g., Ferrer-Conill and Tandoc, 2018; Hanusch, 2017) or journalists’ relationship with their audience (e.g., Bodó, 2021; Ovaska, 2024; Zamith, 2020). Less attention has been paid to the deeper institutional change brought about by datafication and its impact on the journalists working in news media. In this study, we aim to find out how journalists relate to the institutionalisation of analytics in newsrooms.

We focus on journalists working for a Nordic public-funded media organisation. The core values of Nordic public service media are generally considered to be quality and universality, the latter of which includes the idea of accessibility and diversity of content (Hokka, 2019). These values are also seen to be in line with the premise of the welfare state, where state-supported media are seen as playing an important role in promoting citizen participation and well-being alongside other public institutions (Ahva et al., 2024). To highlight the specificities shared by Nordic media systems in relation to the welfare state ideal, the term ‘media welfare state’ is often used in literature (Syvertsen et al., 2014). Although the suitability of the term in today’s fragmented media landscape has been questioned (Ala-Fossi 2020), it still retains basic elements shared by the Nordic media, such as the idea of strong editorial freedom and media services as a public good for citizens (Syvertsen et al., 2014). Moreover, research into the professional identities of Nordic journalists shows that their professional views reflect well the characteristics of the media system in which they work (Ahva et al., 2017).

To date, most of the studies on datafication of journalistic work have covered commercial media operators, with only a few studies (e.g., Hanusch, 2017) covering public service media. What makes public service media an interesting and important subject for datafication research is that tax-funded media cannot justify the use of analytics to maintain their market position and financial situation, but they must legitimise the use of analytics in other ways, which inevitably have implications for the journalists working in the organisations. In this study, we examine these consequences and show how they reflect the institutionalisation of analytics in journalism.

Theoretical framework: Institutionalisation of analytics in newsrooms

The term ‘institution’ has two related meanings. On the one hand, it refers to the large organisations in society that exercise power and control through formal procedures and permanence and, on the other hand, to the various routines, norms and rules that circulate within and around diverse social practices and guide people’s thinking and behaviour (Jepperson, 1991; Kosterich, 2019; Moe and Syvertsen, 2007). In other words, institutions are forms of organising and power that are generally accepted and supported in society, upon which different organisations rely and which they reform through their actions. In this paper, we approach public service media as such an institution.

Institutional theory looks at how organisational practices become routinised and taken for granted, that is legitimised (Kosterich et al., 2019; Meyer and Rowan, 1977). Institutional theory therefore looks at how a profession (such as journalism) adopts a social change (such as datafication) and justifies it for itself. The concept of the organisational field is central to institutional theory. The organisational field consists of different organisations and their stakeholders and is located at the level between the organisation and society, influencing how socially constructed expectations and practices are diffused and reproduced within organisations (Greenwood et al., 2002). The meaning systems created in the organisational field define the norms, values and practices of organisations, which over time, are reinforced by regulatory processes (Greenwood et al., 2002).

In the existing literature, the institutionalisation process is divided into different stages, which, broadly speaking, can often be described as three main stages. First, we can talk about a de-institutional phase, where social, technological or regulatory changes occur in the organisational field that force organisations to act innovatively in response to the changes (Cools et al., 2024; Greenwood et al., 2002). An illustrative example of this is the adoption of analytics in news media: Originally analytics served the needs of digital marketing and only became of interest to news organisations when the emergence of competing content providers started to challenge old practices in media consumption (Petre, 2018; Tandoc, 2019).

In the second pre-institutional phase members of the organisations perceive the new innovation and react and build trust in it with different intensity and in different ways, and step by step the use of the innovation in the organisation becomes more widespread and routine (Cools et al., 2024). Early adopters of new tools and technologies play an important role in demonstrating and justifying the added value of the innovation and building trust around it (Cools et al., 2024). Adaptation to innovation is, therefore, a social process, based on negotiations between people at different levels of the organisation (Lamot and Paulussen, 2020). Once an innovation has gained widespread acceptance and trust, it becomes a normative part of the organisations’ activities, and different types of guidelines, policies and regulations are built around it (Cools et al., 2024).

In the final re-institutional stage, the innovation becomes embedded and taken for granted in the everyday life of organisations (Cools et al., 2024; Greenwood et al., 2002). When fully institutionalised, innovations can survive for generations and be uncritically accepted as the ultimate way of operating (Greenwood et al., 2002). It is also possible that even if an innovation has reached a pre-institutional stage in an organisation, it will eventually be abandoned and thus never become fully institutionalised (Greenwood et al., 2002). That the use of analytics seems to have taken the first position among these options has been facilitated by the fact that the introduction of analytics has been a seemingly natural extension of long-standing audience measurement practices and the work processes surrounding them (Lamot and Paulussen, 2020). At the same time, analytics can be seen as a disruptive innovation because it shapes the field in a fundamental way (Ahva and Ovaska, 2023).

Previous empirical research shows that the institutionalisation of analytics in news organisations has largely proceeded as theorised above. The attitudes of news organisations and individual journalists towards analytics have changed radically over the last 20 years as analytics tools have developed and become established in newsrooms (Ahva et al., 2024; Lamot and Paulussen, 2020). Until the early 2000s, journalists had little interest in tracking metrics or using them to make decisions about the structure or content of news sites (MacGregor, 2007). Information about audience behaviour and interests was collected through sales and subscription figures and readership surveys and was mainly considered by the marketing or advertising departments of news organisations (Ferrer-Conill and Tandoc, 2018; Petre, 2018). In the early 2010s, the general attitude towards the use of analytics changed from disinterest or resistance to curiosity and interest (Ahva and Ovaska, 2023; Cherubini and Nielsen, 2016), and by the end of the decade, the general attitude of journalists towards analytics has become more embracing (Cherubini and Nielsen, 2016; MacGregor, 2007; Petre, 2018).

Overall, previous research (e.g., Ahva et al., 2024; Lamot and Paulussen, 2020) suggests that many of the larger Western media organisations are in the final phase of institutionalisation, where the use of analytics is widely embedded in the work of journalists and newsrooms, where data-related tensions have been found to have been defused and where there is a high level of trust in the effectiveness of analytics. For example, in a study by Ahva and Ovaska (2023), individual journalists felt that the era of the most blatant click baiting, which has been at odds with journalistic ethics, is over, as the metrics associated with analytics have become more refined and routine. Journalists and their managers have thus reached a consensus that analytics are needed to effectively navigate the changing media environment (Ahva and Ovaska, 2023; Christin and Petre, 2020). The institutionalisation of analytics is also reflected in new roles and positions in news media organisations, such as social media journalists, demonstrating that an audience-focused approach is well established in today’s news production environment (Ferrer-Conill and Tandoc, 2018). From a broader perspective, serving and engaging audiences through analytics can thus be seen as an illustration of a larger shift where innovation has become a prevailing discourse in journalism (Costera Meijer, 2020).

At the same time, the mundanity of data practices has made it increasingly difficult for journalists to question the everyday use of analytics in news organisations (Ahva and Ovaska, 2023). Studies have found that the adoption of analytics in newsrooms has caused individual journalists stress, anxiety and frustration (Ahva and Ovaska, 2023; Cohen, 2018). These negative feelings are caused by journalists experiencing that they are being evaluated based on purely quantitative goals or that they have diminished workplace autonomy in terms of choosing news topics or headlines (Ahva and Ovaska, 2023).

To (re)gain control and to find meaningfulness in work, journalists may engage in what is called job crafting (Lichtenthaler and Fischbach, 2018; Lippert et al., 2023). Job crafting can be defined as “the self-initiated changes that employees make in their own job demands and job resources to attain and/or optimize their personal (work) goals” (Tims et al., 2012: 173). Data-related job crafting can thus be understood as active choices and actions of employees concerning data and analytics, which in some way support their professional goals and ethics. Conventionally, job crafting is thought to consist of employees taking proactive steps to align their skills and interests with those of the organisation so that they can deliver better results with the resources available (Lippert et al., 2023; Wrzesniewski and Dutton, 2001). More recently, it has also been observed that major institutional and organisational changes (such as datafication of work) may also activate reactive motives for job crafting, that is, employees trying to cope with their changing work conditions (Lazazzara et al., 2020; Lippert et al., 2023).

Job crafting activities can be task-related (redefining the quantity, scope or nature of work duties), cognitive (redefining the meaning and purpose of work) or behavioural (redefining the quantity or quality of work-related social interactions), and can be oriented in two different directions: expansion crafting (sometimes also called approach or promotion-oriented crafting) involves taking on new tasks or skills whereas retraction crafting (sometimes also called avoidance or prevention-oriented crafting) involves avoiding certain tasks or negatively perceived work-related demands (Lazazzara et al., 2020; Lichtenthaler and Fischbach, 2018; Wrzesniewski and Dutton, 2001). Regular engagement in expansion crafting activities has been found to be associated with perceived job autonomy and work meaningfulness, job satisfaction and commitment and, through these, increased occupational well-being (Lazazzara et al., 2020; Lippert et al., 2023). In contrast, retraction crafting activities have been found to be more likely associated with negative outcomes, such as poorer job performance and health problems (Lazazzara et al., 2020; Lichtenthaler and Fischbach, 2018).

In this study, we explore the extent to which journalists engage in different types of data-related job crafting and examine how their job crafting activity relates to their perceptions of their work environment and occupational well-being. By looking at the differences between journalists who opt in to accumulate data-related resources and those who opt out of it, we aim to outline what kind of agency journalists have in the final stage of the institutionalisation of analytics in newsrooms. To this end, we have defined the following four research questions:

RQ1 To what extent are journalists engaged in data-related job crafting?

RQ2 Do journalists differ in their engagement with analytics in their day-to-day work?

RQ3 How is data-related job crafting associated with perceptions of the work environment?

RQ4 Are the effects of data-related job crafting related to journalists’ occupational well-being?

Materials and methods

Sample

To answer our research questions, we conducted a quantitative survey with journalists employed by a Nordic public-funded media organisation. The company produces news and cultural content on its own digital platforms and operates several television and radio channels. Around 3000 people work in the company on a permanent basis. The survey was targeted at journalists working both in the organisation’s centralised news and current affairs unit and its regional units.

More than 800 journalists were invited to answer the survey by email through the media organisation. The online survey ran between 13 May 2024 and 1 June 2024 – resulting in 121 completed questionnaires. This represents a response rate of 14%. Even though the response rate is low, the sample does not systematically differ from the target population of journalists working for the media company: 50% of the respondents identified themselves as female (50% in the media company), and the average age of the respondents was 46 (48 in the media company). Grouped by age, 29% of the respondents belonged to the age Group 24–39, 45% to the age Group 40–54 and 25% to the age Group 55-70. Respondents were mainly university-educated, with 60% of them reporting that their highest level of education was equivalent to a master’s degree. In terms of years of company service, 20% of the respondents had worked for the company for less than 5 years, 20% for 5–10 years, 29% for 11–20 years, 22% for 21–30 years and 8% for more than 30 years.

Measures

Data-related job crafting

In the survey, data-related job crafting was measured with eight items on a five-point scale (1 = not at all, 5 = a great deal). Journalists were asked to assess the extent to which they had engaged in different job crafting strategies in the past week. Four of the items measured expansion crafting (α = 0.80), i.e., how journalists sought to accumulate data-related resources (e.g., “I sought out opportunities to improve how my work is ranked and evaluated”), and four items measured retraction crafting (α = 0.86), i.e., how journalists sought to avoid analytics in their work (e.g., “I actively minimised working on tasks that were measured in data and rankings”).

Opting out/in

A median split of 1,75 was used as a cut-off point for the expansion crafting scale to divide the journalists into two equal-size groups. This led to the recoding of a new binary variable. The purpose of introducing a new binary variable was to examine more closely the differences between those journalists who sought to accumulate data-related resources in their work and those who did so only to a limited extent or not at all. Those journalists who scored 1,75 or below in the expansion crafting scale were grouped as the ones “opting out” and those who scored above 1,75 were grouped as the ones “opting in”.

Perceptions of analytics

In the survey, journalists’ perceptions of analytics were measured using a 7-point scale (1 = strongly disagree, 7 = strongly agree) with nine items of which five measured data optimism (α = 0.86; e.g., “analytics make me more efficient in my work”) and four measured data scepticism (α = 0.90; e.g., “I question whether the datafication of work advances the quality of journalistic work”). The measures were selected and adapted from Walczuch et al. (2007).

Perceptions of work environment

The survey measured journalists’ perceptions of work environment with a five-point-scale (1 = strongly disagree, 5 = strongly agree) with four items measuring competition (α = 0.83; e.g., “everyone at work wants to win by outperforming their coworkers”) and five items measuring collaboration (α = 0.73; e.g., “colleagues frequently discuss common problems in our organisation”). The measures were selected and adapted from Fletcher and Nussbaum (2010).

Occupational well-being

Journalists were also asked to assess their occupational well-being using a 7-point scale (1 = never, 6 = every day) with eight items of which three measured work engagement (α = 0.83; e.g., “I am enthusiastic about my job”) and five work-related exhaustion (α = 0.87; e.g., “I feel emotionally drained from work”).

Findings

Data-related job crafting

Table 1 summarises the extent to which journalists at the time of the survey had engaged in data-related job crafting in the past week. As the table shows, around half of the journalists had engaged in data-related expansion crafting at some level. In particular, the attempt to influence the quantitative metrics of their work was something that many journalists did to varying degrees. In other words, journalists were striving to produce content that could gain popularity measurable in metrics. In addition, many journalists tried to influence to some extent how their work looked in the light of analytics.

Table 1.

Extent of data-related job crafting in %.

		Not at all	A little bit	Moderately	Quite a bit	A great deal
Expansion crafting	I actively tried to influence the quantified metrics of my work	40	19	26	10	6
	I tried to change how my work is reflected in analytics	44	26	19	8	3
	I sought out opportunities to improve how my work is ranked and evaluated	52	17	18	11	3
	I actively took on tasks that would improve data-driven evaluations of my work	58	14	17	11	1
Retraction crafting	I actively minimised working on tasks that were measured in data and rankings	84	8	4	3	0
	I tried to avoid working on projects that rely on data-driven evaluations	86	6	5	3	1
	I sought to make my work less intense by avoiding exposure to data about work performance	85	6	7	2	1
	I changed my work so that I only work on projects that are not quantified through data	93	4	3	0	0

Compared to data-related expansion crafting, journalists did not engage in data-related retraction crafting to the same extent. This means that while about half of the journalists surveyed did not actively seek to influence analytics to their advantage in their daily work, they also did not try to avoid the use of analytics at an organisational level, for example by avoiding projects or tasks that relied on data-driven evaluation. An independent sample t test (Table 2) confirmed that with regard to retraction crafting there were no statistically significant differences between those who engaged in data-related expansion crafting only to a limited extent (i.e., opted out) and those who did so to some extent (i.e., opted in). This illustrates that the resistance to analytics among journalists that is observable in the survey data was therefore rather passive and personal.

Table 2.

Differences in retraction crafting between those opting in and out.

		M retraction crafting^a	SD
All		1.22	0.55
Expansion crafting median split	Opting out (M ≤1.75)	1.18	0.56
Expansion crafting median split	Opting in (M >1.75)	1.27	0.53

^aBased on a five-point Likert scale (1 = not at all, 5 = a great deal).

When looking at which journalists have chosen to opt out of the analytics game, it is difficult to give any unambiguous picture based on the survey data. Opting out is not related to any of the background variables examined: between 46% and 52% of participants in all age and educational groups opt out, and 50% of all opt-outs are female. No statistically significant differences were found in the job crafting mean values by gender or age (table 3).

Table 3.

Engagement with data-related job crafting by gender and age.

		Mean expansion crafting^a	SD	Mean retraction crafting^a	SD
All		2.01	0.92	1.22	0.55
Gender	Female	2.06	0.91	1.18	0.43
Gender	Male	1.97	0.94	1.22	0.55
Age	24–39	1.96	0.84	1.18	0.48
	40–54	2.10	1.03	1.15	0.46
	55-70	1.92	0.83	1.33	0.62

^aBased on a five-point Likert scale (1 = not at all, 5 = a great deal).

Perceptions of analytics and work environment

Survey data on journalists’ perceptions of analytics showed that the general attitude of journalists toward the use of analytics in news media was more filled with optimism than scepticism (Table 4). The result echoes the earlier studies indicating that after a decade of agonising to find a balance between journalistic and market-driven values, journalists have ‘made peace with metrics’ (Christin and Petre, 2020; see also Ahva and Ovaska, 2023). Only a third of the journalists surveyed had doubts regarding whether the use of analytics fulfils its purpose and advances the quality of journalistic work. As can be expected, those journalists who opted in were more data-optimistic than those who opted out, whereas those who opted out questioned more often the purpose of using analytics as a part of journalistic work (Table 6).

Table 4.

Perceptions of analytics in %.

		Disagree	Neither agree nor disagree	Agree
Data optimism	Analytics give people more control over their daily lives	28	26	46
	I prefer to rely on the most advanced analytics	11	31	58
	Analytics make me more efficient in my work	31	27	41
	I find analytics to be mentally stimulating	22	26	52
	I am quite confident that analytics in journalistic work will have the desired effect	24	22	54
Data scepticism	I have doubts that the analytics in journalistic work will achieve its objective	52	21	27
	I need to be convinced that the analytics in journalistic work will achieve its objectives	48	19	33
	I have doubts that the analytics in journalistic work will improve journalistic work	64	22	14
	I question whether the analytics in journalistic work advances the quality of journalistic work	46	22	33

The surveyed journalists largely felt that their work environment was more collaborative than competitive (Table 5). Also, other studies have witnessed a data-led “collaboration culture” emerging in newsrooms that is “breaking down the walls” between editorial and marketing departments (Chua and Westlund, 2019: 162; see also Nelson and Tandoc Jr., 2019). With regards to collaboration, there were no statistically significant differences between those journalists who opted in and those who opted out. This suggests that journalists were rather pragmatic about the use of analytics in the organisation, as they did not perceive their attitude towards analytics as interfering with their collaboration with colleagues and managers. However, those journalists who opted in were more competition-oriented compared to those who opted out (Table 6). Previous studies show similar results: for example, in the study by Ahva and colleagues (2024), managers who were generally more interested in analytics than their subordinates were also more competition-oriented, or as Cohen (2018: 579) noted in their study, “analytics inspire competitiveness”.

Table 5.

Perceptions of work environment in %.

		Disagree	Neither agree nor disagree	Agree
Competition	My coworkers are very competitive individuals	43	31	26
	My coworkers work hard to outperform each other	31	40	29
	My coworkers are constantly competing with one another	65	26	9
	Everyone at work wants to win by outperforming their coworkers	78	17	5
Collaboration	Colleagues share information frequently in our organisation	3	9	88
	Colleagues frequently discuss common problems in our organisation	4	7	89
	Journalists share close ties with people across editorial disciplines	21	30	50
	Overall, colleagues have mutually gratifying and highly cohesive relationships	4	11	85
	I expect that the social relationships in this organisation will exist far into the future	7	29	64

Table 6.

Differences in perceptions of analytics and work environment between those opting in and out.

		M data optimism^a	SD	M data scepticism^a	SD	M competition^b	SD	M collaboration^b	SD
All		4.37	1.08	3.48	1.41	2.42	0.77	3.92	0.57
Expansion crafting median split	Opting out (M ≤ 1.75)	3.95**	1.01	3.72*	1.38	2.27*	0.71	3.94	0.57
Expansion crafting median split	Opting in (M > 1.75)	4.81**	0.99	3.22*	1.41	2.58*	0.80	3.90	0.57

^a Based on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree).

^b Based on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree).

*Independent sample t test results indicated a statistical significance between those who opt out and opt it, p (2–tailed) < 0.05; **p (2-tailed) < 0.01.

Occupational well-being

Questions on occupational well-being showed us that the surveyed journalists were fairly engaged in their work (Table 7). The majority of respondents felt enthusiasm and immersion in their work at least a few times a week. At the same time, the majority of respondents regularly experienced work-related fatigue and exhaustion. Over 80% of the surveyed journalists were emotionally drained and almost 50% felt burned out from work a few times a month or more. When looking at the differences between those journalists who opted in and out, we can observe that those journalists who opted out experienced more work-related exhaustion than those opting in (Table 8). With regard to work engagement, there were not statistically significant differences between the two groups.

Table 7.

Occupational well-being in %.

		Once a month or less	A few times a month	A few times a week or more
Engagement	At my work, I feel bursting with energy	5	34	61
	I am enthusiastic about my job	2	22	77
	I am immersed in my work	6	23	71
Exhaustion	I feel emotionally drained from work	19	49	32
	I feel used up at the end of the workday	33	42	25
	I feel fatigued when I get up in the morning and have to face another day on the job	33	36	31
	Working with people all day is really a strain for me	61	26	13
	I feel burned out from my work	52	36	12

Table 8.

Differences in occupational well-being between those opting in and out.

		M engagement^a	SD	M exhaustion^a	SD
All		5.74	0.98	3.91	1.38
Expansion crafting median split	Opting out (M ≤ 1.75)	5.65	1.02	4.19*	1.52
Expansion crafting median split	Opting in (M > 1.75)	5.84	0.93	3.63*	1.16

Based on a 7-point Likert scale (1 = never, 7 = every day).

*Independent sample t test results indicated a statistical significance between those who opt out and opt it, p(2–sided) < 0.05.

In their international review study, MacDonald and colleagues (2016) explain why journalists experience work-related malaise. They found that factors associated with journalists’ burnout included increased workload and work pressure, lower levels of job satisfaction and organisational support and reduced capacity for innovation and autonomy in the workplace. Conversely, in terms of occupational well-being, high levels of autonomy have been shown to promote creativity and productivity, have a positive impact on job satisfaction and protect workers’ health from the adverse effects of high workloads and work stress (Hamada et al., 2019). One reason why the journalists who opted out experienced more work-related exhaustion can therefore be attributed, in the light of previous research (Hamada et al., 2019; MacDonald et al., 2016 see also Bunce, 2019; Petre, 2018), to their perceived lack of autonomy in their work resulting from their inability to avoid participating in the data-driven culture of the organisation.

Summary

To answer the research questions, based on our survey data we can conclude that journalists engage in data-related expansion crafting (i.e., seek to accumulate data-related resources) to a varying degree, but they largely do not engage in data-related retraction crafting (i.e., seek to avoid analytics in their work; RQ1). Journalists differ in their engagement with analytics in their day-to-day work. Some journalists actively try to influence how the content they produce, and their work performance is seen in the light of analytics, while others do not (RQ2). Engagement in data-related expansion crafting is connected to an optimistic attitude towards the use of analytics in general and a higher sense of competition among colleagues (RQ3). Not actively seeking to use analytics to one’s benefit is connected to work-related exhaustion (RQ4), but not connected to data-related retraction crafting, that is, the avoidance of the use of analytics at an organisational level.

Discussion

The results of this study support the view drawn by previous studies emphasising institutionalisation of analytics in news organisations (see Ahva et al., 2024; Lamot and Paulussen, 2020). In our study, institutionalisation is reflected in the fact that opting out of analytics is more of a gesture than a real possibility in today’s data-driven news media environment; it manifests itself as a kind of private, tacit compliance. Journalists may continue to passively opt out on a personal level, but as our research data shows, such subtle personal resistance is by no means empowering but is associated with higher levels of work-related malaise. At the organisational level, almost none of the journalists surveyed avoided using analytics, suggesting that in order to perform their jobs, a genuine refusal to use analytics was not seen as a realistic option. In other words, independent from being either “data optimists” or “data sceptics”, the respondents tended to conform with what editorial policies expected from them. This reflects not only the institutionalisation of analytics in newsrooms but also the pragmatic agency of journalists in relation to the use of analytics in news production.

The results of this study can be further explained in the light of previous studies showing that that the context of the work environment may influence which forms of job crafting are activated in an organisation: a supportive work context is more often linked to expansion crafting activities, whereas a constraining work context is more often linked to retraction crafting activities (Lazazzara et al., 2020). In a data-driven news media environment, newsroom managers are highly supportive of increasing journalists’ interest in metrics (Ahva et al., 2024; Petre 2018). This leads journalists not only to engage in data-related expansion crafting activities at different levels, but they may also do so for different reasons: some journalists’ motives may be more proactive, that is they focus on achieving goals, while other journalists’ motives may be more reactive, that is they focus on coping with the changing work conditions (see Lazazzara et al., 2020; Lippert et al., 2023). Accordingly, coping with the data culture of an organisation, but not actively avoiding it, is precisely what we consider to be the essence of tacit compliance. In this sense, tacit compliance resonates with the idea of digital resignation, which refers to the rational response of consumers to corporate surveillance practices and to corporate practices that cultivate these responses (Draper and Turow 2019).

Previous literature additionally emphasises that the use and the perceptions of audience analytics must be understood in the light of the media context in which analytics are applied (Omidi et al., 2024; Zamith et al., 2020). Public service media can currently be seen as living under cross-pressure, as they have to compete for audiences in the market driven largely by commercial interests while also proving their distinct value compared to commercial media outlets (Hanusch, 2017; Moe and Syvertsen, 2007). Earlier research shows that public-funded media organisations in particular justify the use of audience analytics on the grounds that by gaining more accurate information about their audience’s interests, they can bridge the gap between what the audience wants to know and what they should know (Boczkowski and Mitchelstein, 2013; Hanusch, 2017). With analytics, public-funded media organisations can arguably better engage their audiences by producing news content that meets the needs of different audiences in the right format and at the right time, thereby strengthening their reach and impact, and in accordance, the institutional legitimacy of being publicly funded. Moreover, by reaching audiences in line with the ideal of universal accessibility, public media can create space for public dialogue, which is generally considered a prerequisite for democracy (Niemi et al., 2021).

At the same time, it is important to consider to what extent the institutionalisation of analytics risks altering the key societal justification of the public service media that is not only based on its role in promoting democratic values and ensuring access to impartial, diverse and high-quality information content for all citizens, but also on proactively combating false information (see Sehl, 2024). Maintaining a diverse information environment and recognising mechanisms of misinformation is arguably increasingly difficult when the “logic of the click” (Hanusch, 2017: 1580) impacts the production practices and the evaluation of journalistic performance in public-funded media organisations. As a result of datafication, journalism has become more efficiency-oriented (Bunce, 2019; Cohen, 2018) but, as this study shows, only a part of journalists feel that analytics make their work more efficient. In addition, the inability of journalists to influence the datafied culture of their organisation, as revealed in this study, has consequences for their well-being at work. In the light of this study, we thus argue that the institutionalisation of analytics in newsrooms can be seen as a threat to editorial autonomy and especially internal autonomy, that is journalists’ sense of professional identity and their freedom to carry out their daily work (see Bunce, 2019; Hamada et al., 2019; Petre, 2018). The loss of internal autonomy may in turn have far-reaching institutional consequences: in line with the ethos of journalism, journalists must be independent in their work if journalism is to retain its legitimacy as a social institution and continue its democratic work (see Hamada et al., 2019).

Building on this argument, we want to challenge the public service media to ask once more “why?” beyond the convenience of using analytics. Why must public-funded media organisations rely so heavily on data as a basis for journalistic decision-making and performance evaluation that individual journalists cannot opt out if they wish? Can the use of analytics still be justified in terms of audience engagement, when studies (e.g., Ahva and Ovaska, 2023; Ekström et al., 2022) show that extensive use of analytics undermines editorial integrity? What are the long-term consequences of a data-driven work culture for journalists’ occupational well-being? Would it be possible for media organisations to support the development of another kind of agency than the one in which journalists, for pragmatic reasons, tacitly comply with the data culture of the newsroom? These are questions that public service media will have to consider when maintaining their institutional role in society in a way that it will also be sustainable for journalists working for the media organisations.

Footnotes

Acknowledgements

The authors wish to thank the team led by professor Ward van Zoonen, who led the project The datafication of media work: Implications for job demands and resources, well-being, and professional identities, under which this study was carried out. The authors would also like to thank the anonymous reviewers for their thoughtful and constructive comments.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research leading up to this article was funded by the Helsingin Sanomat Foundation.

Ethical statements

Veera Ehrlén

Mikko Villi

Data Availability Statement

The dataset supporting these findings is not available due to the fact that it was obtained in collaboration with a public media organisation.*

Author biographies

Veera Ehrlén is a postdoctoral researcher in the Department of Language and Communication Studies at the University of Jyväskylä, Finland. Her work broadly investigates news media, digital platforms, datafication and media consumption. Her particular expertise lies in sports media and journalism. She is currently the president of the Finnish Association for Media and Communication Studies.

Mikko Villi is Professor of Journalism Studies in the Department of Language and Communication Studies at the University of Jyväskylä, Finland. His research interests encompass media management, media consumption, media platforms, news media business models, and the digital transformation of the news media.

References

Ahva

Ovaska

(2023) Audience metrics as disruptive innovation: analysing emotional work of Finnish journalism professionals. Nordicom Review 44(2): 152–171.

Ahva

Van Dalen

Hovden

, et al. (2017) A welfare state of mind? Nordic journalists’ conception of their role and autonomy in international context. Journalism Studies 18(5): 595–613.

Ahva

Salonen

Ovaska

, et al. (2024) The managerial and future-oriented role of audience data in data-informed news organisations. Journalism 0(0): 1–19.

Ala-Fossi

(2020) Finland: media welfare state in the digital era? Journal of Digital Media & Policy 11(2): 133–150.

Barger

Barney

(2004) Media-citizen reciprocity as a moral mandate. Journal of Mass Media Ethics 19(3-4): 191–206.

Belair-Gagnon

(2019) News on the fly: journalist-audience online engagement success as a cultural matching process. Media, Culture & Society 41(6): 757–773.

Boczkowski

Mitchelstein

(2013) The News Gap: When the Information Preferences of the Media and the Public Diverge. Cambridge, MA: The MIT Press.

Bodó

(2021) Selling news to audiences – a qualitative inquiry into the emerging logics of algorithmic news personalization in European quality news media. Digital Journalism 7(8): 1054–1075.

Bunce

(2019) Management and resistance in the digital newsroom. Journalism 20(7): 890–905.

10.

Cherubini

Nielsen

(2016) Editorial Analytics: How News Media Are Developing and Using Audience Data and Metrics. Oxford: Reuters Institute for the Study of Journalism.

11.

Christin

Petre

(2020) Making peace with metrics: relational work in online news production. Sociologica 14(2): 133–156.

12.

Chua

Westlund

(2019) Audience-centric engagement, collaboration culture and platform counterbalancing: a longitudinal study of ongoing sensemaking of emerging technologies. Media and Communication 7(1): 153–165.

13.

Cohen

(2018) At work in the digital newsroom. Digital Journalism 7(5): 571–591.

14.

Cools

Van Gorp

Opgenhaffen

(2024) Newsroom engineering teams as “survival entities” for journalism? Mapping the process of institutionalization at The Washington Post. Digital Journalism 12(2): 171–190.

15.

Costera Meijer

(2020) Understanding the audience turn in journalism: from quality discourse to innovation discourse as anchoring practices 1995–2020. Journalism Studies 21(16): 2326–2342.

16.

Draper

Turow

(2019) The corporate cultivation of digital resignation. New Media & Society 21(8): 1824–1839.

17.

Ekström

Ramsälv

Westlund

(2022) Data‐driven news work culture: reconciling tensions in epistemic values and practices of news journalism. Journalism 23(4): 755–772.

18.

Ferrer-Conill

Tandoc

JEC

(2018) The audience-oriented editor: making sense of the audience in the newsroom. Digital Journalism 6(4): 436–453.

19.

Flensburg

Lomborg

(2023) Datafication research: mapping the field for a future agenda. New Media & Society 25(6): 1451–1469.

20.

Fletcher

Nusbaum

(2010) Development of the competitive work environment scale: a multidimensional climate construct. Educational and Psychological Measurement 70(1): 105–124.

21.

Greenwood

Suddaby

Hinings

(2002) Theorizing change: the role of professional associations in the transformation of institutionalized fields. Academy of Management Journal 45(1): 58–80.

22.

Hamada

Hughes

Hanitzsch

, et al. (2019) Editorial autonomy: journalists’ perceptions of their freedoms. In: Hanitzsch

Hanusch

Ramaprasad

, et al. (eds) Worlds of Journalism: Journalistic Cultures Around the Globe. New York: Columbia University Press, 133–159.

23.

Hanusch

(2017) Web analytics and the functional differentiation of journalism cultures: individual, organizational and platform‐specific influences on newswork. Information. Information, Communication & Society 20(10): 1571–1586.

24.

Heikkilä

(2022) Ethical pitfalls of audience metrics. In: Manninen

VJE

Niemi

Ridge-Newman

(eds) Futures of Journalism. Berlin: Springer, 113–125.

25.

Hendrickx

Montero

Ranaivoson

, et al. (2021) Becoming the data-informed newsroom? The promotion of audience metrics in the newsroom and journalists’ interactions with them. Digital Journalism 9(4): 427–442.

26.

Hokka

(2019) Towards nuanced universality: developing a concept bible for public service online news production. European Journal of Communication 34(1): 74–87.

27.

Jepperson

(1991) Institutions, institutional effects, and institutionalism. In: Powell

DiMaggio

(eds) The New Institutionalism in Organizational Analysis. Chicago, IL: University of Chicago Press, 143–163.

28.

Kosterich

Matthew

Weber

(2019) Transformation of a modern newsroom workforce. Journalism Practice 13(4): 431–457.

29.

Lamot

Paulussen

(2020) Six uses of analytics: digital editors’ perceptions of audience analytics in the newsroom. Journalism Practice 14(3): 358–373.

30.

Lazazzara

Tims

De Gennaro

(2020) The process of reinventing a job: a meta–synthesis of qualitative job crafting research. Journal of Vocational Behavior 116: 103267.

31.

Lee

Tandoc

JEC

(2017) When news meets the audience: how audience feedback online affects news production and consumption. Human Communication Research 43(4): 436–449.

32.

Lichtenthaler

Fischbach

(2018) Leadership, job crafting, and employee health and performance. The Leadership & Organization Development Journal 39(5): 620–632.

33.

Lippert

Kirchner

Wiener

, et al. (2023) Job crafting in the era of algorithmic management. In: Proceedings of Forty-Fourth International Conference on Information Systems (ICIS), Nashville, Tennessee, 14-17, 2025.

34.

Lischka

(2020) Fluid institutional logics in digital journalism. Journal of Media Business Studies 17(2): 113–131.

35.

MacDonald

Saliba

Hodgins

, et al. (2016) Burnout in journalists: a systematic literature review. Burnout Research 3(2): 34–44.

36.

MacGregor

(2007) Tracking the online audience: metric data start a subtle revolution. Journalism Studies 8(2): 280–298.

37.

Mayer-Schönberger

Cukier

(2013) Big Data: A Revolution that Will Transform How We Live, Work, and Think. Boston: Houghton Mifflin Harcourt.

38.

Meyer

Rowan

(1977) Institutionalized organizations: formal structure as myth and ceremony. American Journal of Sociology 83(2): 340–363.

39.

Moe

Syvertsen

(2007) Media institutions as a research field: three phases of Norwegian broadcasting research. Nordicom Review 28: 149–167.

40.

Nelson

Tandoc

JEC

(2019) Doing “well” or doing “good”: what audience analytics reveal about journalism’s competing goals. Journalism Studies 20(13): 1960–1976.

41.

Niemi

Koivula

Laaksonen

S-M

, et al. (2021) Vaakakupissa vaarat ja vastuu: julkisen palvelun median arvot sosiaalisen median alustojen paineessa. Media & viestintä 44(2): 23–44.

42.

Omidi

Dal Zotto

Picard

(2024) Audience analytics and tensions in digital news work: evidence from Swiss news media. Information Technology & People 38: 1991–2014.

43.

Ovaska

(2024) From permissive to resistive tactics: how audience members engage with and make sense of datafied journalism. Convergence: The International Journal of Research Into New Media Technologies 30: 2008–2024.

44.

Petre

(2018) Engineering consent: how the design and marketing of newsroom analytics tools rationalize journalists’ labor. Digital Journalism 6(4): 509–527.

45.

Salonen

Ehrlén

Koivula

, et al. (2023) Post-publication gatekeeping factors and practices: data, platforms, and regulations in news work. Media and Communication 11(2): 367–378.

46.

Sehl

(2024) Public service media as pivotal in combating misinformation and disinformation: prerequisites and approaches. European Journal of Communication 39: 582–594.

47.

Syvertsen

Mjøs

Moe

, et al. (2014) The Media Welfare State: Nordic Media in the Digital Era. Ann Arbo: University of Michigan Press.

48.

Tandoc

(2019) Analyzing Analytics: Disrupting Journalism One Click at a Time. London: Routledge.

49.

Tims

Bakker

Derks

(2012) Development and validation of the job crafting scale. Journal of Vocational Behavior 80(1): 173–186.

50.

Walczuch

Lemmink

Streukens

(2007) The effect of service employees’ technology readiness on technology acceptance. Information & Management 44(2): 206–215.

51.

Wrzesniewski

Dutton

(2001) Crafting a job: revisioning employees as active crafters of their work. Academy of Management Review 26(2): 179–201.

52.

Zamith

(2018) On metrics-driven homepages: assessing the relationship between popularity and prominence. Journalism Studies 19(8): 1116–1137.

53.

Zamith

(2020) Quantified audiences in news production: a synthesis and research agenda. Digital Journalism 6(4): 13–30.

54.

Zamith

Belair-Gagnon

Lewis

(2020) Constructing audience quantification: social influences and the development of norms about audience analytics and metrics. New Media & Society 22(10): 1763–1784.