Viral Media: Audience Engagement and Editorial Autonomy at BuzzFeed and Vice

As organisations ‘native’ to the digital environment, sites like BuzzFeed, Vice, Vox and The Huffington Post have been well placed to take advantage of new technologies and pioneer new approaches to creating and distributing media. Despite this, they remain conspicuous by their absence in contemporary media scholarship. This article will focus on two North American digital-native media organisations: BuzzFeed and Vice. As two of the largest and most popular digital natives in the world, these organisations merit closer critical attention. Little remains known, for instance, about the types of content these organisations produce, or the routines, cultures, and practices that undergird their sites of content production. Given their expanding role in the contemporary information ecosystem, it is vital that scholarship does more to take these social actors into account. In the interest of advancing our theoretical and empirical understanding of virality in media, this paper examines the extent to which news production at BuzzFeed and Vice is impacted by the ‘quantified’ audience, and the normative implications of these findings with regards to journalistic autonomy. The findings of this research are based on semi-structured qualitative interviews conducted with 22 journalists from BuzzFeed and Vice, in the US and UK.


Introduction
As organisations 'native' to the digital environment, sites like BuzzFeed and Vice have been well placed to take advantage of new technologies and pioneer new approaches to creating and distributing media (Carlson and Usher, 2016). Since their launch in the mid 2000s, several of these companies (' digital natives' henceforth), have grown to occupy a prominent position in the realm of news and entertainment, to the point that they now compete with legacy media for both attention and advertising (Jurkowitz, 2014;Nicholls et al., 2016Nicholls et al., , 2017Newman et al., 2016Newman et al., , 2017Newman et al., , 2018. In the process, digital natives have helped introduce new norms and practices to the broader field of cultural production: participation, openness, subjectivity, and transparency -values that we associate with the logic of digital media and culture (see Lewis, 2012).
In this regard, it is curious that these organisations have been largely absent from the scholarly discourse on digital media. There has been a tendency to stay within the more comfortable bounds of the familiar and explain media change from the perspective of traditional institutions. This parochialism becomes less tenable as the field of cultural production grows ever more heterogeneous, particularly when one attends to subjects that share a strong intellectual lineage with more peripheral actors, or those outside the field's dominant core.
The subject of virality would seem to represent a case in point. As one of the first organisations to seriously experiment with virality, the news and entertainment website, BuzzFeed, was one of the first organisations on the internet to pioneer the approach of using digital intermediaries like Google and Facebook to 'supercharge' the spread and reach of its content (see Nicholls et al., 2016). By combining aspects of data science with media content creation, ' analysing user data to decode how and why content is shared and distributed' (Küng, 2015: 58), the organisation helped spawn a new wave of imitators, some with more legitimate motives than others (e.g. Bakker, 2012).
In the interest of advancing our theoretical and empirical understanding of virality in media, this paper explores the extent to which news production at BuzzFeed and Vice is impacted by the ' quantified' audience, and the normative implications of these findings with regards to journalistic autonomy.
Based on semi-structured interviews with 22 journalists from the US and UK offices of both organisations, this research reveals how the needs and wants of an increasingly ' quantified' audience (Carlson, 2018) are becoming the 'seemingly more democratic … substitute for the internal standards' by which journalists judge news stories (Bourdieu, 1998: 73). This raises challenging questions about the nature and normative qualities of journalistic autonomy and invites further speculation on the future of journalism as a bounded profession (Waisbord, 2013).

Literature review
In research examining the changing relationship between journalists and audiences, scholars have often invoked notions of the journalist as 'gatekeeper': a term first introduced to journalism studies by White (1950) to describe journalists' subjective power in the process of deciding what news and information should be disseminated to the public. The theory of gatekeeping has been revitalised over the past decade or so, as the jurisdictional claim by journalists to decide news for the public has diminished. As Deuze (2005: 451) writes, 'new media technologies challenge one of the most fundamental "truths" in journalism, namely: the professional journalist is the one who determines what publics see, hear and read about the world'. In research on this topic, an important underlying question has been what level of gatekeeping control journalists are willing to relinquish over news content in order to facilitate greater levels of participation from audiences in the news process (Lewis, 2012).
What studies in this area have indicated so far is a general unwillingness on behalf of journalists to give up on established practices of publishing, filtering and selecting information (Robinson, 2007;Domingo et al., 2008;Thurman, 2008;. In this regard, Lewis (2012: 836) argues that journalists remain ' caught in the professional impulse toward one-way publishing control'. This also helps to explain why user-generated content (UGC) has so far only occupied a marginal space in news coverage (Domingo et al., 2008). Such behaviour is thought to be the result of journalists wanting to maintain control over familiar processes of newsgathering, for fear of surrendering some of their public legitimacy and cultural authority (Lowrey, 2011).
Nevertheless, other research has argued that the audience-journalist relationship might be slowly changing, and with it, the gatekeeping function of journalism (Gillmor, 2004;Lowrey, 2006;Singer, 2008;Bruns, 2008). This is partly attributed to the growing importance of social media in journalism, which has become an important channel of news distribution and also a new tool for journalists to engage with the audience day-to-day in their work (Hermida, 2013). Distribution in these social networks relies heavily on users sharing content within their own personal networks, and in this respect, Singer (2013) has argued that internet users have acquired more power in determining what news is circulated in wider society. Singer (ibid) argues that the audience now effectively function as secondarygatekeepers, who choose news 'not only for their own consumption but also for the consumption of others, including those within their personal circle of acquaintances and those who are part of an undifferentiated online public' (Singer, 2013: 22). She argues this shift toward 'user-generated visibility' suggests a 'new way of looking at one of the oldest conceptualisations of the journalist's role in our society', namely, that of the gatekeeper (Singer, 2013: 22).
For Bruns (2008) the gatekeeping function of journalists is shifting to a new practice of 'gatewatching': ' a form of reporting and commenting on the news which does not operate from a position of authority … but works by harnessing the collective intelligence and knowledge of dedicated communities to filter the newsflow and to highlight and debate salient topics of importance to the community' (Bruns, 2008: 176-77). What these debates suggest is that audiences might slowly be reconfiguring the autonomy and authority of journalists, which might eventually lead to a more collaborative relationship between journalists and their publics. At this stage, however, it might be too early to conclude this is occurring. Anderson (2011a), for example, is arguing that the present rhetoric around audience empowerment and participation, at least in newsrooms, often takes place in the context of audience analytics and metrics.
According to Tandoc (2014), it is in the context of economic instability and shrinking audiences, that news organisations are increasingly turning to web analytics to understand the preferences of their audiences, with the aim of increasing traffic to their websites and earning more revenue from advertising. Although audience considerations have played a visible role in news selection since the 1970s (Nadler, 2016), the rise of new audience feedback mechanisms has transformed the previously 'imagined' audience (see Schudson, 2003) into a ' quantified' aggregate, providing real-time access to a range of behaviour-based metrics. Those optimistic about these changes have argued that analytics enable journalists to 'figure out who their audiences are, learn what they want, and in real time, track their behaviours in order to be more responsive to their needs' (Usher, 2010: 1). However, others such as Anderson (2011a), Tandoc and Thomas (2015), and Cohen (2015), have expressed concern that the growing presence of web analytics in newsrooms is influencing news selection in ways that undermines traditional editorial judgement and journalistic values of autonomy. In his study of the Philadelphia news website, Philly.com, Anderson (2011a: 561) observed that 'website traffic often appeared to be the primary ingredient in Philly.com news judgement', resulting in sports, gossip, and human-interest stories being selected, sometimes at the expense of more serious and complex news stories. Similarly, in a study of three online newsrooms, Tandoc (2014) noticed news stories and headlines were often being selected according to a ' consumer-driven logic', with audience metrics playing a key role in influencing editorial decisions.
Some research has also pointed to potential differences in the way traditional and online journalists interact with web analytics, with the latter group accepting metrics -and the growing visibility of the 'quantified' audience -as a normal feature of daily Stringer: Viral Media 8 newswork (Agarwal and Barthel, 2015). Beyond ideological factors, Boczkowski (2004) has drawn attention to the ways in which technological adoption is shaped by organisational influences as well as the professional culture of journalism and normative visions of the audience. Hence, 'the availability of a technology does not mandate its use' (Zamith, 2018: 418), which speaks to the critical need of studying the affordances of new technologies in context.

Method
To explore the notion of virality and the impact of the quantified audience in news production at BuzzFeed and Vice, 22 interviews were conducted with a range of editorial staff based in the UK and US offices of both companies. In addition, I conducted around half a day of non-participant observation at the offices of both organisations. Approximately half of the interviews were conducted face-to-face, while the other half were conducted over the phone. Fourteen staff from BuzzFeed were interviewed in total: ten women and four men. Four of these staff were editors while the remaining ten were a combination of generalist and specialist reporters. Eight staff were interviewed from Vice: seven men and one woman. Half of these staff were editors while the other half were generalist reporters.
Anonymity was offered to all participants in the study. This was to help protect their identities and prevent any potential blowback in case of sensitive information being disclosed during the interview process. Ethical approval was also sought from my university before I carried out my fieldwork.
Interviews were semi-structured and lasted approximately one hour. Respondents were asked a predefined list of questions about topics such as their background, professional dispositions, their organisation's approach to journalism, their working routines and ideas about the role and purpose of journalism. All interviews were recorded and transcribed in their entirety. Transcripts were then uploaded to the research management tool, NVivo, and general themes were identified before formal coding categories were developed. These categories were then further developed and refined during a second-stage of coding.

Rhythms of news production at BuzzFeed and Vice
Perhaps the first and most obvious thing to note about BuzzFeed and Vice is that they mainly produce news content for the web, which has a significant impact on the spatial and temporal dimensions of daily news work. In contrast to the print or television news cycle, where news production tends to follow an 'industrial' logic, the rhythm and organisation of online news is more fluid, decentralised, and flexible, with publishing taking place continuously throughout the day (Gade, 2011;Anderson et al., 2012). The journalists I spoke with tended to view these conditions favourably, sometimes contrasting them with ' older' methods of organising news work, which were often criticised for being slow, inefficient, and overly-bureaucratic; ' a tottering, petering pile of management on top of management', as one editor put it to me (editor, BuzzFeed UK). As a consequence, journalists frequently made out the news process at BuzzFeed and Vice to be more nimble, reactive, and efficient, partly because a 'flat[ter] organisational structure' meant stories that did not have to move through ' as many layers of hierarchy' (editor, BuzzFeed UK). At Vice, one US editor offered a slightly more haphazard description, suggesting that the news process was a 'soupy mess of collaboration … of having various people doing various things and having it all just kind of come together; that's what Vice News was' (editor, Vice US).
This arrangement was advantageous in that it facilitated an emphasis on editorial practice oriented around the coverage of breaking news events and trending topics, both of which promised to deliver high returns in traffic and attention. As one Vice editor explained: We had to get cracking on the site immediately and so [stories] would be assigned, deadlines would be assigned. Much of the turnaround would be day of … We had to be highly selective. Whatever we thought was trending [on social media] the most in the morning that wasn't already spent [or] something that had hit the wires within a reasonably short amount of time; there was still room to carve some space out on that. That was the rationale for much of morning assignments … We [also] did a fair amount of enterprise reporting … but that wasn't going to generate 10 million monthly uniques … The stuff that's however many thousand tweets on Twitter, you know, if you can get something in there, that's in the flow, in the feed, in the continuum where millions of people are already looking for that sort of thing, then you're going to get a lot more aggregated traffic out of it. So, it's just logic (editor, Vice US).
This fits with observations made by other scholars about the newfound importance of social media as a source of breaking and engaging news -an ' ambient', always-on environment that journalists from BuzzFeed and Vice could tap into to boost the virality and impact of their news stories (e.g. Hermida, 2010Papacharissi and Oliveira, 2012). As a consequence of this arrangement, 'locating and sharing "new" news' appeared less important than 'managing multiple fast-moving flows of information already in circulation' (Boyer, 2013: 2).
In this routine, journalists' decisions about exactly what news to cover seemed strongly shaped by their understanding of what would appeal to the audience. As one Vice reporter explained: So, once you've identified the story [published elsewhere], you then have to, sort of, think what you're going to be doing that's different. And, you know, our style was always different to everyone else … because we're making this stuff for maybe sixteen to maybe thirty-year olds (reporter, Vice UK).
One reporter from BuzzFeed shared similar sentiments, describing the process of taking already published information, and repurposing it to make it more useful and relevant to their own readers: You might cover the same issues [as traditional media] but it's whether you can lead on it or not. So, on the doctor's strike: firstly, junior doctors are in the BuzzFeed age group, they read us, they already read us, but our audience are interested too … So it was very easy for us to get traction with the doctors as well as the audience. And so, it's an issue all the papers are covering but we weren't just doing every slip and move … We could sort of decide which bits we were going to do, and what days we were really going to go in on an issue … And so, some things you decide to do but do them in a BuzzFeed way and others you just go 'yeah, this story is good, it's interesting, but it's not for our audience and everywhere else has done it quite well (reporter, BuzzFeed UK).
Decisions about news therefore routinely centred on whether particular issues could be made fresh and relevant for the audience at hand, which affected not only the process of news selection, but also the very structure of news at both organisations.

Beats and 'passions'
At BuzzFeed and Vice, there appeared to be less beat specialisation compared to traditional news outlets, with both organisations favouring generalist reporters who could ' code-switch', and cover a range of news subjects, depending on what was trending and deemed of interest to their particular audience. As one reporter from BuzzFeed explained: We have a lot of general reporters grouped into junior reporters and senior reporters depending on your level of expertise. And we sort of cover everything. We're not beat reporters, we do have particular subject areas of interest and people will gravitate towards different things … In the same day I might write about the EU referendum, and then a local issue in London, a crime story, or something funny … So it's quite a range (reporter, BuzzFeed UK).
Still, as the quote above suggests, there were certain news topics that journalists at BuzzFeed and Vice felt were better predictors of audience interest compared to others. Many of these areas of coverage were based on so-called 'millennial passions' (Vice, 2016) -concerning issues such as civil rights, gender equality, drug liberalisation, lesbian, gay, bisexual and transgender rights, and mental health -which collectively seemed to reflect a progressive political ideology (see Stringer, 2018). This would also seem indicative of a broader trend towards more personal, emotional and identity-driven forms of news gaining currency online (Ip, 2015;Beckett and Deuze, 2016;Wahl-Jorgensen, 2016).
In interviews, journalists were cognisant of this fact too, recognising that covering topics like mental health, LGBT issues, and gender issues, was partly a ' deliberate strategy' to target younger audiences, that had different interests compared to 'standard paper audiences' (reporter, BuzzFeed UK). As one freelance journalist for Vice said: Vice is slightly younger and are quite progressive and tuned into the kinds of subjects that young people are interested in. Things like -you mentioned mental health and LGBT issues -which are being talked about much more generally among young people (reporter, Vice UK).
One reporter from BuzzFeed UK offered a similar summary of the audience: 'We obviously know our audience is millennials, and they're often left-leaning. There's a complete gender mix. But they're quite open, liberal, and young basically' (reporter, BuzzFeed UK). As a consequence, one reporter explained that BuzzFeed had taken quite a ' dedicated position' to issues like LGBT rights and mental health, partly because of their 'general importance', but also because they ' always [did] well with [BuzzFeed's] readers' (reporter, BuzzFeed UK). That some journalist seemed to possess distinct knowledge about what subjects ' did well' with readers pointed to the growing importance of data and metrics in news (Zamith, 2018). One reporter from BuzzFeed UK, for example, spoke about how, in a quantifiable sense, issues like 'mental health, LGBT, housing' performed very well with audiences, which reinforced the need to ' do more' of these stories (reporter, BuzzFeed UK). Another reporter from BuzzFeed, who specialised in social media news, suggested that stories based around identity issues (race, gender, sexuality, etc.) could be particularly effective at driving virality through high engagement, because of their ability to 'provoke emotion … whether it's finding it funny, finding it sad, or making [the audience] feel as though we're making them angry and wanting to support a social justice cause' (reporter, BuzzFeed UK).
At Vice, journalists also recognised the mutually reinforcing relationship between certain subjects and high audience engagement. One senior editor, for instance, spoke about the decision to hire journalists who specialised in covering ' civil rights … gender bathroom rights … queer rights', who had ' a sense of what was shareable [on social media]' (editor, Vice US). In this manner, social media traffic seemed to play a role in determining precisely what content journalists selected for publication. When asked if journalists sometimes selected stories based on what data and analytics had shown to be successful in the past, one editor responded: 'I mean, to a little extent. For example, our audience really like stories about weed, so like, fairly often, we'll say 'hey, what's up on the weed beat'' (editor, Vice US).

The quantified audience
The centrality of digital technologies to the news selection process was a recurring theme, with journalists recognising the range of new tools and techniques for quantifying and measuring audience behaviour (see Carlson, 2018;Zamith, 2018). During interviews with numerous reporters and editors from both organisations, editorial decisions were often discussed in reference to the language of audience metrics, with terms like traffic, hits, clicks, views, and shares regularly deployed in discourse.
At Vice, the metric of web traffic was discussed frequently and appeared to have the greatest impact on newswork. As one reporter from Vice explained: As far as traffic numbers, [there was] not day-to-day [pressure], but there was definite concern … It wasn't necessarily relayed on the news staff but on the editors or supervisors. There were weekly traffic meetings where they looked at the traffic numbers, analysed, figured out where they were deficient and so there was some pressure there, I think. I just don't know how it was projected on day-to-day work (reporter, Vice US).
Editors from Vice were helpful in expounding further how managerial pressures on achieving high traffic manifested in daily work. One interviewee from the US office gave the following description: There were always traffic targets, but then as far as the extent to which they would always be made plain to us, or there would always be a fire under our sheets, that wasn't really the case you know. Sometimes we would be just shy, sometimes we'd hit it better, and we understood that whenever a traffic target was made plain to us, it always seemed very aggressive, if you know what I mean … Ultimately, we were always trying to tell interesting stories and predicating our day to day on the idea that doing what we did well would result in the kind of hits that they were looking for … And we knew what kind of material spiked and we would jump on those kinds of stories and we would want to tell them well (editor, Vice US).
As this description suggests, the presence of data and metrics in the newsroom (and the concomitant pressures of meeting preordained traffic targets), encouraged an emphasis on audience-oriented editorial practices that privileged speed and immediacy. This was a cause for concern among some of the more senior journalists at Vice, who seemed troubled by audiences' newly acquired power in the news production process and the concomitant loss of autonomy in news selection. As one former editor from Vice US reflected: So, your point about the audience … and this is particularly true of digital upstarts, or upstarts in the news media world, that they're keenly aware of the audience and keenly aware of catering to the audience to some degree … And I don't necessarily mean in terms of puppy videos or listicles or things of that sort, but nevertheless, being like, 'look, we've got mouths to feed'. I'm not sure how much the New York Times' editors wake up and be like 'we've got mouths to feed'. They're more like, 'look, we've got stories to cover' and then [the production process] sorts itself out … Whereas at Vice, it was like, 'this is a business', and we have phantom subscribers … that subscribe to this brand of Vice and we need to make them feel at home (editor, Vice US).
From the perspective of this editor, Vice's responsibility to the audience was partly a consequence of being a new entrant to the field, which produced a deficit in certain assets like prestige and reputation, that other, more established news brands could rely upon as capital. Here, it was telling that Vice was contrasted with the New York Times, which was cited favourably as an organisation where news judgement remained under the jurisdiction of journalists. By contrast, Vice was pejoratively referred to as a 'business', implying that commercial pressures (or having 'mouths to feed') might be encroaching upon editors' professional sense of news judgement. Similarly, another reporter from Vice expressed concerns that journalism had become more driven by a 'looming idea of audience engagement', meaning organisations had to work harder to be 'seen and heard more … through titillation, some sort of ideological affinity … or [by establishing] a voice or brand that appeals to a particular segment' (reporter, Vice US).
By contrast, journalists from BuzzFeed appeared far more comfortable integrating audience metrics into news selection decisions. This attitude helped fuel a general curiosity in the use of data and metrics in news production. This was most evident in discussions on the metric of 'sharing', which, as Jonah Peretti (2014), CEO and founder of BuzzFeed, has suggested, underpins much of BuzzFeed's thinking about content creation: [For BuzzFeed] sharing has always been the biggest metric because it shows that someone thinks a piece of content is worth passing on to a friend … Word of mouth is the distribution [of content] … You are getting so much data back about what people like and what people share and that can immediately inform the media you create.
For journalists at BuzzFeed, sharing was held in similar esteem and recognised as distinct from other metrics, which were often reproached for being linked to cruder forms of audience measurement and more culturally debased practices like news aggregation (Anderson, 2011b;Bakker, 2012). As editor-in-chief of BuzzFeed UK, Janine Gibson, once said in an interview: It's a fundamental mistake that people make about BuzzFeed that we are all about getting traffic. If an article is shared by relatively fewer people but really speaks to them, that's incredibly important to us … BuzzFeed journalists, editors, and video-makers study relentlessly what will cause their audience to share their content. We don't deal in big round numbers, we deal in metrics of sharing … The key to the success of news is working out how to marry those learning with more traditional journalistic skills of running down a story, finding out who's doing a bad thing and trying to cover it up (Gibson, 2016).
Reporters I spoke with generally echoed these sentiments, discussing how factors related to the metric of sharing fed into decisions about newsworthiness: Seventy-five per cent of our traffic comes through social media referrals. Not homepage clicks which is extremely rare. Like, the BBC, Guardian; most of their traffic comes through homepage clicks. But BuzzFeed is not a destination homepage. You don't wake up in the morning and be like 'Oh I'm interested in the news, let's see what BuzzFeed is writing about'. You're going on Twitter, on Facebook, and then you go ' oh that story's interesting' and click through that way. So, when we're doing investigations or commissioning one, we're always thinking ' okay, is this something people are going to talk about? Is this something people are going to want to share?' (reporter, BuzzFeed UK).
It would be fair to assert that figuring out what news people felt compelled to share was not an exact science, but rather, an evolving and ever-changing discipline that relied on user data being ' captured, analysed, and manipulated in a perpetual loop of analysis, interpretation, experimentation, feedback, and refinement' (Küng, 2015: 58). As one reporter put it: The huge advantage of writing digitally is that we know exactly what our readers are reacting to because you can see it for each individual story … We can see what people care about and what people want to read and cater to that and that's super important (reporter, BuzzFeed UK).
That catering to audience's needs was seen as 'super important' was partly attributed to a rise in audience autonomy (Napoli, 2012). Reporters were aware that the 'people formerly known as the audience' (Rosen, 2006) had acquired more power in self-selecting the news, and consequentially, felt a stronger need to tailor content to their particular needs and preferences. As the same reporter above argued: 'The big difference with the internet … is that you can't tell people what they want any more … You have to work out what people want and give it to people in the form they think they want it' (reporter, BuzzFeed UK).
Unlike Vice, then, the notion of tailoring news more explicitly around the needs and wants of audiences was viewed by many journalists at BuzzFeed as normatively desirable. One potential reason for this was that journalists, somewhat paradoxically, connected this motive with a greater sense of freedom and independence in work. As one reporter put it: If there's a big breaking news story or we see something that has been developing for a while, my editor might say 'maybe watch this and see if there's something we can do with it'. Less so in my role on story day-to-day and being like 'can you write this up' A lot of the time, I'm very lucky to have my editors believe in my judgement. If I see something that I think will work, or they come to me and say -' do you think this will work on the website?' -they'll trust me. Mainly because being a 20-something female, I kind of represent a demographic of BuzzFeed readers (reporter, BuzzFeed UK).
This corresponded with comments made by an editor at BuzzFeed, who spoke about the 'inversion' of responsibility for selecting news, because BuzzFeed's young journalists could be relied upon to select news relevant to BuzzFeed's young audience: [At BuzzFeed, we think] it's okay to employ talented people in their early 20s and give them a go of it and give them good editing and guidance and they'll have ideas. And so, from those background assumptions that the traditional media maybe does not have … The traditional media likes people to serve their time a bit because all the senior people did, so bloody hell, everyone else is going to have to, and the traditional media likes to say … 'what stories we'll cover what stories are important is' … tends to flow top-down. And … [while] we value experience and our editor-in-chief and our head of news do have final say about what stories we do, there is a slight inversion of that … Reporters should be able to pursue things they think will work. Hire people that are like your target audience, you know. They can … they will have good ideas and will do them well and [we] trust them to do that (editor, BuzzFeed UK).
In other words, reporters were trusted to self-select news because of their ability to integrate audience preferences into their own sense of news judgement. This would seem to stand in direct contrast to much of the earlier literature describing the relationship between reporters and editors, which has typically portrayed reporters as more source-oriented and editors as 'more conscious of what will have audience appeal' (Reese, 2009: 288; also see Gans, 2004: 89-90). Indeed, this seemed to be the case at Vice, with editors appearing to shoulder most of the responsibility for making stories appealing to the audience, rather than reporters.

Conclusion
In an effort to progress our understanding of virality, this article has explored audience engagement in news production from the perspective of journalists at two leading digital native media organisations: BuzzFeed and Vice. The discussion has taken place against a backdrop of larger media upheaval; a time when broader social, cultural, economic, and technological trends seem to be reworking definitions of what journalism 'is', and how it defines itself as a profession (Broersma andPeters, 2012, 2017).
One manifestation of this is the shift from more centralised, hierarchical, bureaucratic, and stable forms of industrial production, towards more fluid, decentralised, flexible, networked, and individualised modes of work (Deuze and Witschge, 2018). At BuzzFeed and Vice, this arrangement spoke of a production system optimised to be responsive to the needs of the market, with a focus on breaking news and trending topics bringing higher returns in traffic and attention compared to other forms of reporting. Journalists found this work appealing because it imbued them with a greater sense of ownership and control over news stories, even if, paradoxically, it resulted in them sacrificing some of their own 'gut feeling' about what was newsworthy (Schultz, 2007).
That the 'informational wants and needs' (Beam, 2001: 467) of the audience were viewed as important is nothing new in journalism. In fact, since at least the 1970s journalism has been in a 'post-professional' period 'marked by greater institutional acceptance of the idea that consumers' preferences should factor into news production' (Zamith, 2018: 420; also see Nadler, 2016). This is particularly true of audience-oriented publications, such as tabloids or many digital natives, where 'it is it is not so much the relation with sources that circumscribes journalistic practice, but the requirements of the market or the community that is catered for' (Van Zoonen, 1998: 136). Still, what distinguishes the current era from earlier periods is the relative abundance and availability of digital technologies that 'give journalism outlets unprecedented ability to track, measure, and quantify audience activity' (Cohen, 2015: 108). Consequentially, older conceptions of the audience as the 'missing link' in news production (Schlesinger, 1978) have been rendered less apposite, as journalists' intuitions about the audience have steadily been replaced by a rhetoric of the ' active' audience, 'laying the groundwork for a vision of the professional reporter that is less autonomous in his or her news decisions and increasingly reliant on audience metrics as a supplement to news judgment' (Anderson, 2011a: 550).
At BuzzFeed and Vice, this rhetoric animated many conversations about the audience, albeit in ways that were specific to both organisations. While at Vice, journalists seemed generally more sceptical about metrics, connecting them with a loss of gatekeeping power and autonomy in work, BuzzFeed journalists seemed to embrace them, viewing metrics like 'sharing' as effective substitutes for audience wants. This finding speaks to the importance of organisational and managerial influences in shaping attitudes and uses of technology, as highlighted in other studies of news production (e.g. Boczkowski, 2004;Usher, 2014;Anderson, 2011a).
More broadly, these developments inevitably raise concern about the nature of editorial autonomy, adding support to the notion that journalist's 'gut feeling' about news is being encroached upon by the ' quantified' audience (Schultz, 2007;Anderson, 2011a;Carlson, 2018). This would seem indicative of a broader trend towards integration, whereby journalists become more comfortable with ' combining established editorial values with the values of collaboration, adaptation and business thinking' (Cornia et al., 2018: 3). While the separation of editorial and business interests has always been a somewhat of a tenuous commitment (Coddington, 2014), these findings would collectively seem to suggest that editorial autonomy might also be waning as a normative aspiration, as organisations trade in some of their cultural capital and autonomy in exchange for the promise of economic sustainability.
From this perspective, the intensive pursuit of virality can be viewed as one offshoot of journalism's deeper economic crisis. Yet whether a news production system increasingly oriented around audience engagement -interpreted through the narrow prism of audience data and metrics -is desirable, remains highly questionable. Rather, virality in news (and the various metrics that propel this ambition) would seem to support a very narrow definition of consumer choice (Tandoc and Thomas, 2015), ultimately driven by an institutional need to extract more (economic) value from the news process (Anderson, 2011a;Cohen, 2015;Carlson, 2018).