Abstract
Social media editors were initially appointed by media organizations which aimed to extend their reach and to establish their presence in social media. These editors were hired to create content on behalf of mass media organizations, but with social media in mind. As such, they need constantly to reconcile mass media logic and social media logic. Relying on the frameworks of ‘social media logic’ and ‘network media logic’, this study focuses on the role of social media editors as the arena in which both logics fuse to create a hybrid media system. Twenty in-depth interviews with social media editors, representing a large proportion of the Israeli media, were carried out. Our findings underscore how editors’ every day routines are a result of negotiation between logics, creating hybrid media organizations.
Keywords
‘The influence of social media platforms and technology companies is having a greater effect on American journalism than even the shift from print to digital’ (Bell et al., 2017: 1). The media industry has indeed become dependent on social media in many ways, chief among them being content dissemination. Media organizations therefore created the position of social media editor (SME) as a response to the need to reach out to their audiences via various different channels. The role is similar to that of an editor, yet devoted solely to the world of social media: deciding which content to post on social media and aiding other members of the newsroom in this decision (Sacco and Bossio, 2016).
Given that it is a relatively novel position, little research has hitherto focused on SMEs (Sacco and Bossio, 2016; Tandoc and Vos, 2015). However, the argument advanced here is that the emergence of such a position encapsulates the changes in journalistic norms and work practices: that is to say, SMEs serve as ‘negotiators’ between the two worlds of mass media and of social media, each with its own set of rules and logic (Poell and van Dijck, 2014). As such, SMEs face a constant need to reconcile the two worlds. In that sense, by delineating their organizational position and practices we can exemplify how hybrid media systems (Chadwick, 2013) – systems which operate on various logics, rather than on a single coherent one – come into being.
As our point of departure, we adopt models developed by van Dijck and Poell (2013) and by Klinger and Svensson (2014), which offer an analytical framework presenting the differences between mass media and social media logic. Relying on this framework, we ask “How does a novel media logic (i.e. social media) interact with co-existing logics (Couldry, 2008) and what form does this interaction take?”
From mass media logic to social media logic
The concept of media logic was introduced by Altheide and Snow (1979). According to their original definition, media logic refers to a communication process in which the media institutionalize their guidelines, to the point that they are taken for granted and serve as interpretive schema. This definition assumes one cohesive media logic. While each medium has its own code of interpretation, style, and grammar, there is still a basic underlying conceptual logic which applies to every mediated communication.
This idea of one cohesive logic was contested for decades, and even more so with the rise of digital media (Klinger and Svensson, 2014). The massive changes in the media environment, so the argument goes, constitute a paradigm shift which requires a critical revision of the idea of media logic (Dahlgren, 2014; Thimm et al., 2018). Indeed, recent years have seen attempts to decipher the media logic of a new and highly dominant form of mediated communication: social media communication. This form of communication bears massive differences from key elements of mass media logics, which have been described as the ‘ability to frame reality and partly pertaining to media’s claim towards neutrality or independence’ (van Dijck and Poell, 2013: 3).
Two prominent models were offered in this connection. The first, termed ‘social media logic’, was developed by van Dijck and Poell (2013); the second, from Klinger and Svensson (2014), was termed ‘network media logic’. Both models describe this new logic against the background of mass media logic, a comparative approach that is also embraced in the current work (see Table 1). 1
Comparison between mass media logic and social media logic (van Dijck and Poell, 2013).
Van Dijck and Poell (2013) argue that social media logic (in contrast with mass media logic) is characterized by four dimensions: connectivity, datafication, programmability, and popularity. Connectivity is an ‘advanced strategy of algorithmically connecting users to content, users to users, platforms to users, users to advertisers, and platforms to platforms’ (p. 9). Under mass media logic, media also connect content to citizens or advertisers to consumers. While apparently similar, connectivity manifests differently within each logic. First, as evident from the definitions provided, connectivity in social media can take more forms than in mass media. Second, ‘unlike mass media, social media platforms seldom deal with “natural” geographically or demographically delineated audiences’ (p. 9). Instead, the ‘platform apparatus’ in social media defines and shapes connections and does so in a personalized way. Thus, connections are made on the basis of interest. Third, social media personalization is made possible due to the duality of its connectivity: it is part human made (e.g. users), part automatic (algorithmic). While human connectivity is part of mass media logic, automation is a stranger to it, as will be further discussed in the context of programmability.
Datafication refers to the ability of networked platforms to render into data many aspects of the world that have never been quantified before. Mass media uses polling, audience research, and ratings to understand their audiences, adjust their programming accordingly, and help advertisers target them. Social media platforms enable a more granular understanding of audiences, add new layers of information, and do so in real time. Finally, and perhaps most significantly, while in mass media data are the byproduct, in social media they are the product.
Programmability refers to the move from a human editorial process (mass media logic) to an automatic process in which algorithms determine the content users receive (social media logic). However, tracking programmability on social media is all but impossible (van Dijck and Poell, 2013). First, algorithms are in constant flux: they evolve and develop in part in response to user behavior. Second, algorithms are proprietary and therefore hidden and protected. In terms of media logic, Altheide and Snow (1979) highlighted mass media’s ability to shape reality, filtering, and framing social reality. With social media, the power is transferred to algorithms and to users who participate in steering content by sharing and commenting.
When considering the last dimension, popularity, the differences between mass media logic and social media logic are less abrasive, and the two complement one another. Under both types of logics, pursuit of public attention is a key factor. Mass media once enjoyed the power of setting the public agenda. Social media, in their early stages of development, held out a more egalitarian promise: that all users would be equally capable of engaging public attention. Yet, as they developed, social media weaponized their ability to present popularity instantly and at the same time to manipulate it. This ability to measure and present popularity indices is a unique property of social media which trickles over into other environments, like mass media. In this way, van Dijck and Poell (2013) argue, mass media logic and social media logic are entangled, creating a feedback loop between mass and social media.
The model of networked media logic (Klinger and Svensson, 2014) relates in many ways to that of social media logic (van Dijck and Poell, 2013). In the context of the current work, we borrow the aspect of Production, highlighting the massive change in content creation. While mass media relied on professionals creating content according to given guidelines and norms, social media are primarily driven by user-generated content – so much so that ‘the people formerly known as audience’ (Rosen, 2006) are now referred to as Produsage (Bruns, 2008). 2
Empirical evidence of the penetration of social media logic
Although social media logic has clear characteristics distinguishing it from mass media logic, this does not mean that the former supersedes the latter (Couldry, 2008). In fact, the two logics intertwine (Klinger and Svensson, 2014; Klinger and Svensson, 2015; Thimm, 2018).
Social media have the ability to ‘transport their logic outside of the platforms that generate them’ (Thimm, 2018: 130). One example is the ways news editors (both offline and online) rely on audience metrics, which is based on audience activity (i.e. popularity indices) on social media, in order to plan future content production and make editorial changes (Vu, 2014). Yet there are also some examples in the opposite direction (Klinger and Svensson, 2014). For instance, Hermida (2010) suggests that in some cases journalistic norms are upheld in social media, creating a hybrid form he calls ‘ambient journalism’.
Indeed, the idea that media logic is dynamic and overlapping underscores the notation of a hybrid media system in which there is both conflict and interdependence between older and newer media logics (Chadwick, 2013). In such a system, we can find blurred boundaries between the roles, identities, and norms, and changes in organizational structure and news production (Smith et al., 2015: 15).
‘One major challenge, both for theoretical and empirical perspectives on the concept, is to reflect on the diversity of several media logics at play instead of focusing on one single, mass media-related logic’ (Thimm et al., 2018). Indeed, while there is evidence of the emergence of hybrid media and its effect on journalists’ work (Chadwick, 2013; Hermida, 2010), a comprehensive analysis which follows the guidelines put forward by the social media logic model is less prevalent. 3 This is despite the fact that such analysis is vital in order to unpack the influence of social media on mass media routines and, more broadly, in order to demonstrate how different media logics intertwine (Couldry, 2008).
For example, Enli and Simonsen (2017) suggest that the use of Twitter hashtags is an example of the connectivity principle. Poell and van Dijck (2014) offer an overview of the influence social media logic has on journalistic work, critiquing the idea that social platforms are neutral technologies which simply enable user activity. Rather, they argue, these platforms undermine press independence since they have positioned themselves as a technological layer that structures the relationship between news organizations and their audience. They offer a variety of examples, demonstrating how the news industry is being reshaped by social media: the use of Facebook Insights, for instance, a tool which offers a wealth of information to Facebook page owners, analyzing their audiences’ characteristics and behavior.
Building on this scant work, the aim of our study is to offer a consistent analysis directed at identifying traces of social media logic within the work of mass media organizations. The distinctive dimensions provided by the models reviewed above offer analytical tools that can be employed empirically to recognize and classify the rules, norms, and processes of production and dissemination of media content according to their logics. By classifying observed behaviors, we wish not only to identify if and when social media logic appears but also the interaction and negotiation between logics. Such an analysis can shed light on the dynamic processes shaping hybrid media. To this end, this study focuses on SMEs, arguing that their role demonstrates how the modus operandi of hybrid media comes into being. As will be reviewed below, SMEs can be depicted as in-house negotiators or translators, whose role is to translate media content produced under one logic into another logic. As such, their position offers a firsthand account of the processes shaping hybridity: is the process of translation smooth, or are there professional clashes? If there are clashes, how do they resolve them? Can these processes be described as ‘clean’ translation, whereby each language stays intact, or rather does a mergence between languages occur which creates some kind of pidgin?
Social media editors
SME is a relatively new function in media organizations: the first recruitments for the position were around 2008–2009 (Gleason, 2010). For a relatively long period of time, the position was not clearly defined at the organizational level. Indeed, some major organizations soon dissolved the position, arguing that social media needed to be a part of every staff member’s job description. Ananny (2014) argues that ‘Such debates suggest that news organizations are trying to figure out whether social network sites are integral or ancillary to news work’ (p. 944). This organizational tension may highlight the conflict between media logics.
In recent years, as part of the move to a multi-platform media organization (i.e. specific editors with unique skills for different platforms, Schlesinger and Doyle, 2014), the role of SME was stabilized. The literature mentions four uses for social media (Enli and Simonsen, 2017; Sacco and Bossio, 2016; Tandoc and Vos, 2015; Wasike, 2013): cultivating brand loyalty by encouraging user interactions with the organization over social media; creating traffic to the organization’s website, using posts which include links to news stories; sourcing information, tracing trending issues that might yield news stories; and ‘frontline’ customer support, dealing with complaints posted on social media.
While most of the research thus far has focused on the work of journalists and web editors (Chadwick, 2013; Gleason, 2010; Hermida, 2010; but see: Tandoc and Vos, 2015), examining the work of SMEs could yield beneficial results when deconstructing media logics, for two reasons. First, by definition SME’s role is to connect and negotiate between the two media logics. While they belong to organizations based on mass media logic, they need to comply with the rules of social media logic. In that sense, they are ‘servants of two masters’: they are the embodiment of the convergence of these two worlds. Second, contrary to more traditional positions which have already been studied, this is a relatively new position, whose professional norms are not yet well defined (Ananny, 2014). In that sense, the study builds on the importance of studying how tasks are assembled, bundled, and amalgamated into a role (Arazy et al., 2016).
Method
In previous studies, SMEs have been interviewed alongside other news professionals (Tandoc and Vos, 2015). To the best of our knowledge, this is the first study to focus solely on SMEs. Following Smith et al. (2015), we opt for in-depth qualitative work, to obtain deep and comprehensive information about SMEs’ considerations and work practices (Lindlof and Taylor, 2011).
Thirty-one SMEs, from every major Israeli outlet, were contacted, of whom 20 editors from 18 media organizations participated in the study. They represent a large portion of the Israeli media: 8 TV stations (6 private, 2 public; 5 broadcast, 3 cable TV), 6 newspapers, 2 news websites that have no print papers, and 2 radio stations.
The interviews were conducted face-to-face from mid-November 2017 to mid-January 2018. We used semi-structured interviews which followed a list of 32 questions. Assuming SME is a new role implementing new media logic (Dahlgren, 2014), we based the questions on the work of Weaver et al. (2009), which mapped the functions, duties, and traits of American journalists, and enriched their inventory with items specific to the work of SMEs, such as media production (Bloom et al., 2016), brand and marketing (Rohn and Baumann, 2015), interaction with the audience (Wasike, 2013), and the influence of web analytics (Hanusch, 2017; Tandoc and Ferrucci, 2017). The editors were asked about their professional background, role, editorial decisions, and communication with the audience. 4
All interviews, totaling 788 minutes, were transcribed. In a process similar to the one suggested by Berger (2015), we then detected main themes.
Findings
Following Zeller and Hermida (2015), our findings address two levels of analysis. First, we will introduce the findings on the organizational level (meso level). The professional practices of SMEs will then be described, according to the four dimensions of social media logic (micro-level). This approach was chosen since previous studies have demonstrated that the introduction of social media into mass media organizations occurs at both the meso level (Ananny, 2014) and the micro level (Tandoc and Vos, 2015). Furthermore, as a translator of media logics, the work of SME takes place on both levels: at the meso level editors are the point of contact between media organization and social media, and at the micro level they are in charge of daily practices that are inherent to the function of translation.
The role of SMEs and the organizational structure
We begin our organizational analysis by addressing the job description of SMEs as defined by the interviewees, and comparing it with previous findings, thus highlighting the ways the practices tied to the position were established. Next, we discuss their position within the organizational structure as a reflection of the organizational perspective with regard to the connections between platforms and thereby between logics. We conclude our discussion by focusing on one organization that stands out as it has applied a unique attitude to this issue.
Job description: ‘A salesman of news stories’ (editor in a newspaper)
The commercial interest has been part of mass media logic for decades and cannot be ascribed to social media penetration. Yet the position of SME, in and of itself, reflects a novel merge between marketing positions and editorial positions which, as argued by others (Poell and van Dijck, 2014; Tandoc and Vos, 2015), has been attributed to the embracing of social media. While mass media logic held to the allure of neutrality or independence (van Dijck and Poell, 2013), the conectivity of social media strives against it, blurring these boundaries to a new degree.
Commercial interests are very present in the professional background and responsibilities of the SMEs. In contrast to what has been suggested in the literature (Wasike, 2013), only two thirds had journalistic or editorial backgrounds. The other third came from marketing backgrounds and are therefore fluent with the platforms but not with mass media. Obviously, the decision to position marketers in such roles underscores the understanding that their function is to market the stories (Tandoc and Vos, 2015), as is demonstrated by the quote which opens this section. Unlike journalists, who generally side with their professional norms (Gulyas and Gulyas, 2016), it seems that SMEs tend to side with the prevailing commercial interest (see also: Tandoc and Vos, 2015). For instance, when asked about resources, an editor at major news site said, ‘Why do we have such limited resources? Well, it’s following an old-fashioned approach and the illusion that our consumer affairs reporter will win a Pulitzer prize… At the end of the day, you can’t eat a Pulitzer Prize’.
In 2014, Ananny observed that the position of SME is not clearly defined, but as the consensus between interviewees demonstrates, their position has since crystallized. As an editor with an elite newspaper noted, When I got here, they treated it like it was a non-job. Now I receive feedback and demands from my boss and from my boss’s boss. At first, it was like, do whatever you want, we don’t know what this can be used for.
The areas of responsibility mentioned align with those previously described: traffic, branding, tracking trending issues, and customer support (Sacco and Bossio, 2016; Tandoc and Vos, 2015). In addition, about a quarter also said they actively operated as marketers (by means of product placement, commercial collaborations, etc.) as a form of revenue generator. For instance, one news organization ran Facebook posts offering a daily deal. The editor was not especially concerned with the product placement or with journalistic ethics (mass media logic), but rather with popularity and engagement (social media logic, as will be reviewed below). He said, ‘With all due respect to “xxx” [promotion of the day] … it lowers They [her bosses] stress that they prefer traffic over “shares”. So we won’t forget that they [the social media posts] aren’t meant to be viral, [but] to attract users to the website.
Traffic and exposure are terms that are well embedded in mass media logic: they represent mass media ability to foster connectivity. Social engagement, which is the force that drives social media (Gerlitz and Helmond, 2013), represents the platforms’ ability to foster connectivity. In that sense, a preference for engagement over traffic can be seen as the triumph of social logic over mass logic, albeit only among SMEs. On a deeper level, the understanding that traffic harms engagement is an illustration of acceptance of the ‘walled garden’ ideology that Facebook is often accused of promoting (Gerlitz and Helmond, 2013); SMEs are well aware of this tension. Thus, there is an evident friction presupposed within the definition of SMEs’ responsibilities, and it is their job to reconcile this clash.
Organizational structure
The ways in which SMEs are positioned within mass media organizations can be seen as a structural manifestation of the ongoing changes and negotiations between the two logics. The level of organizational cohesiveness is an indicator of the level of merging and hybridity.
In most organizations, a division between logics is evident in the organizational structures, since social media teams distinguish themselves from other parts of the media organization. All (but one) of the SMEs interviewed described working in very small teams of one to three people. They consider themselves to be something of an elite unit, with unique knowledge, since they are the only ones who understand the social ‘language’ and are able to act as translators. For instance, ‘Some are not destined to “do” social’, one editor said. ‘They come from journalism and they were told to do social too. They don’t get it’. This editor makes a clear distinction between the two types of logics and claims that other members of the organization do not necessarily understand the grammar of social media (as will be further discussed in the programmability and popularity dimensions).
Interestingly, despite this professional separation, SMEs have attempted to facilitate collaborations with other staff members like journalists and talents, trying to convince them to be active on social media, and helping them translate their work to these platforms. For instance, several interviewees tried to influence the egos of journalists by using data from analytics to create informal competitions about whose work attracted more traffic and engagement (a practice clearly linked to the dimensions of datafication and popularity). This use of analytics demonstrates how the principles of metrics-driven news production (Poell and van Dijck, 2014) manifest itself in everyday routines.
The unusual organizational structure – The case of parallel logics
One media organization, a new public broadcaster established in 2015, has a unique model which was referenced by editors in other organizations as the highest standard in the Israeli market. Due to legislative issues, when this media organization was first established it still did not have access to traditional channels, so its directors adopted an approach which we term here ‘Digital First’. Before recruiting other teams, the organization established its social media division, which to this day has 10 times more people than any other social media team in Israel. This ensures that almost all the content they post is uniquely produced for social media. As their editor said, ‘All the creative was built around “Social”, that is the things that users would like to share and comment about. Complete formats were built that way’. This was because ‘Facebook is another channel for the brand … the vision is a video channel that is managed on Facebook’. This approach is present in the organization’s website, which has different tabs for TV, radio, and digital. This is the only organization where mass media logic and social media logic were presented to their fullest extent and co-existed side by side as equals. It is unclear whether this unique approach of parallel logics will prevail in years to come, or remain as an untypical example; or whether is it the result of the inherent differences between commercial media organizations and public broadcasters (due to the limited number of public broadcasters in Israel we cannot be conclusive on this issue).
To summarize, aside from the case of parallel logics, the organizational structure and job description of SMEs in media organizations reveal tension between social media logic and mass media logic. This tension is a constant presence in their work and remains unresolved.
The dimensions of social media logic
In this section, we will address the four dimensions of social media logic and add the production dimension from network logic.
Programmability – The algorithm image
Facebook’s algorithm is a very powerful entity in the editors’ professional life. However, they see the algorithm as mysterious as it is powerful. Much like regular users (Bucher, 2016), they feel in the dark with regard to its operation and priorities. However, while regular users invent folk stories as a way of explaining Facebook’s mechanism (Rader and Gray, 2015), editors – who also devise theories – can put these theories to the test, much like a ‘naïve scientist’.
Be that as it may, as hard as they try to figure out the mechanism, they are aware that it is constantly evolving, leaving them with a sense of uncertainty and lack of control. As one editor noted, ‘You are operating blindfold’. Framing it as an all-powerful, mysterious entity, SMEs personify the algorithm. They talk about its preferences, using terms like, ‘The algorithm loves visual’. As an editor with a leading TV station said, ‘Because of the change in the algorithm … we are in their hands … He [the algorithm] is like a person … but, it can’t be … it is only an algorithm. So we are constantly pausing [to ask]: are we trying to please it?’
To conclude, it is clear that SMEs are well aware of, and preoccupied with, the programmability dimension. Perceiving themselves as an ‘elite unit’, they describe their work as a constant battle, trying to crack the code that will reward them with popularity. This state of uncertainty is a byproduct of the algorithm’s invisibility, described by van Dijck and Poell (2013). As a result, their routines are shaped and reshaped by their perception of the technology, highlighting the massive importance of ‘Human action associated with appropriating those structures during use of technology’ (Orlikowski, 2008).
Connectivity
Connectivity highlights algorithm-mediated connection as the key organizing principle of social media. This idea was identified in two recurring themes. First, the connection with the audience. Many SMEs mentioned using the platform to respond to users on issues ranging from complaints to follow-ups on tips for stories (increasing engagement, as noted above). In addition, SMEs reshape connectivity in a manner that helps bring the two media logics together. For example, TV hosts respond live on air to comments from the show’s Facebook page. Thus, connectivity actually occurs through the move from one (social) platform to the other (mass), in a form of cross-platform connectivity.
The second aspect of connectivity touches upon editors’ perception of Facebook, which serves as a ‘technological layer’ (Poell and van Dijck, 2014) between users and media organizations (as was also evident in the programmability dimension). Editors feel resentment toward Facebook, fully aware that they are ‘building their house on someone else’s land’ (Bell et al., 2017), while this ‘someone’ cashes in substantially. For instance, an editor with a news site said, ‘Half of the time we are working for Facebook, and [the use of] TL; DR [too long; didn’t read]. People don’t understand how destructive it is, how much power it gives to Facebook’. Another editor with a TV channel said, Facebook cuts exposure to posts with links [to outside websites]. So we have to think what content we are willing to sacrifice in order to get exposure on Facebook. And we use video, which we give to Facebook for free … If I make a Facebook referral to the video, I don’t earn traffic – the person who earns it is [Facebook Israel CEO] Ms. Soffer-Teeni.
Many even go further, blaming Facebook and its algorithm for cutting exposure to their content because they are unwilling to use ‘sponsored posts’. One said, ‘it’s not a lot of money; it’s the principle’, meaning it contradicts their norms – of mass media logic – to pay for exposure. Thus, editors’ perceptions of Facebook indeed echo the connectivity dimension. Seen as a layer that masks their relationship with the audience, Facebook limits their independence which, according to mass media logic, ought to be a key principle (Poell and van Dijck, 2014). In the light of this conflict between logics, editors tend to resent the manifestation of connectivity in their daily routines and try as much as they can to outwit it.
Popularity and production
This is what interests the editors the most. Editors described how after posting on Facebook they will count ‘Likes’ to determine if the post was successful. One editor recalled how happy she was when a meme she created, based on ‘Game of Thrones’, went viral. Importantly, editors’ strong interest in popularity echoes the tension between traffic and engagement previously discussed. Unlike engagement, traffic is not visible to other users; thus, it is not an indicator of popularity. The dominance of popularity as an organizing principle might explain why editors prefer engagement over traffic.
According to van Dijck and Poell (2013), programmability and popularity go hand in hand, since as well as the algorithm the diffusion of content is also dependent on user behavior (similar to the distribution dimension; Klinger and Svensson, 2015). However, it seems that, to a certain degree, editors distinguished between the two. They describe the algorithm as guided by what they consider arbitrary criteria, intended to benefit Facebook, yet they do not mention that it is also influenced by user behavior. Maybe this perception helps them make sense of their interaction with technology, alongside their professional achievements (i.e. ‘elite unit’). As already mentioned, the algorithm is often blamed for the lack of exposure. Yet, when they do secure popularity, this is due to their hard work – ‘the content is still the king’, as one interviewee said.
So what content works on social media? Most editors noted that eliciting a variety of emotions is a sure path to success: ‘Annoying, upsetting, or inspiring things, people like to share’. Consumer stories (how to find cheap flight tickets, for instance), celebrity gossip, funny memes, and viral content drawn from international sources have proven themselves to be ‘like’ and ‘share’ magnets. This type of content is sometimes deemed unworthy of publication under mass media logic but has the potential to go viral on social. One editor said, ‘If [British singer] Lily Allen posts about the arrest of [Israeli parliamentarian] David Bitan it’s not an item, not for the news nor for culture. But on Facebook, it’s cool’. Posting such content, then, demonstrates editors’ preference for engagement (over traffic), since this post does not link into the website, thus can only snag users’ attention on social media. At the extreme, this preference clears its way to the editorial room when SMEs ask the newsroom to create content for the sole purpose of social traffic. One editor recalls, There was an ad for a new digital condom. And I said ‘wow, this will bring loads of traffic’. So I went to the [online] editor and he said ‘I will arrange it for you’, and he wrote an item, and I posted it on Facebook.
Here, content is produced by mass media, according to social media logic, for social media. In cases like this, social media logic devours mass media logic.
In contrast, newsworthy content does not work on social media. An editor of a large news site said, ‘Politics? No. Another investigation of [Israel’s Prime Minister] Netanyahu won’t get any attention on Facebook’. The editor of another large news site said, We would like a person who reads … forms an opinion and says ‘this is a wonderful article’ and then shares it and likes it, but this is not what happens … we see … I’m trying to find the right term … They throw poop at each other … it’s a raging zoo … whoever shouts the most gets the most. Interviewer: And what is your role in this zoo? To throw bananas, and sometimes to say ‘look at what he is doing’, or ‘come to me, my bananas are the best. I can show you the banana in a cooler way’.
Still, this approach is not all-consuming. Like previous literature (Ananny, 2014; Klinger and Svensson, 2015; Tandoc and Vos, 2015), our interviews provide evidence that the content posted on social media is bound by journalistic norms. However, the editors do not speak about norms, but instead appeal to branding in statements like ‘Memes work, but we try not to use them. We are expected not to fool around’ (large newspaper). Thus, it seems that SMEs do uphold certain standards; however, they also demonstrate more of a marketing perspective – thinking about the brand’s values rather than journalistic values. Only one news-site SME spoke about journalistic values. When asked about the tension between what interests the audience and what is important, he replied, We try to keep a balance. It is not even what’s interesting, it’s what works … Sure, I want items that will work, like the Lights Festival in Thailand. But on the other hand, if a terror attack happens, it’s important to post it.
Thus, whether driven by the notion of newsworthiness or the need to maintain the appearance that suits the prestige of a news organization, editors sometimes forgo popularity and put their social media logic to one side.
As SMEs are biased toward the type of content that ensures virality, many technical decisions are also a part of this process, like length of videos (no longer than a minute), subtitles (people watch without sound), and surveys (encourage engagement). Importantly, some technical elements go against the principles that guide content production on mass media. For instance, while the first few minutes of a story presented on television are used for introductory purposes, on Facebook content is judged in the first 10 seconds, after which the user decides if she’ll watch it or move on. Likewise, videos must be oriented vertically and not horizontally because ‘people don’t tilt their phone, because they are driving, god forbid, or on the bus, or walking with the phone as they are accustomed to [holding it]—vertically’.
Other non-traditional technical preferences relate to the idea of production (Klinger and Svensson, 2015): the move from professional content creation to user-generated content. 5 Editors mentioned the value of authenticity on social media: content created by users, or that looks like it has been. This is why they prefer cellphone footage to professional camerawork. One media editor noted that it is a no-no to use a brand’s logo on social media unless it is small and unobtrusive, displayed ‘incidentally’ on a microphone, for instance.
To summarize, as suggested by Tandoc and Vos (2015), popularity is the Holy Grail. It has a very different set of rules on social media; rules which SMEs fully embrace and at the expense of the logic of mass media. In the instances where mass media logic is still upheld, the editorial decision is explained not through the prism of journalistic norms, but through branding; this highlights, yet again, the dominance of the marketing approach in the world of social media that has reshaped gatekeeping processes (Tandoc and Vos, 2015).
Datafication
Data and analytics are present in the everyday life of SMEs, for two reasons. The main reason is to measure popularity; it is also used to characterize the audience. That said, both are done superficially, not exhausting the full potential of analytics. An editor with a large news website said, Do we use it the way it should be used? All the way? No … They came from journalism [he came from marketing] … So they don’t fully understand it and resent it.
In the case of audiences, most of the editors made little mention of using analytics to learn about their audience. Instead, the audience remains somewhat imagined. For example, an editor with a TV channel said, ‘I mostly look at variables like women, men, ages and things like that … but this will not make me post something’. The majority referred to the characteristics of their audience as unchanging, and determined by their brand identity (akin to mass media logic). Only one editor demonstrated a more sophisticated understanding, noting that there is no such thing as audience, but that each post has its own audience – demonstrating his awareness of programmability and connectivity.
Where popularity is concerned, the variety in use of analytics is greater. Most editors use the simplest and most basic indicators, mainly counting ‘likes’, ‘shares’, and comments on their posts. At the other end of the scale, just four organizations rely on help from in-house research departments or external companies. Only one editor fully embraced the use of analytics, declaring he is ‘addicted to analytics’.
When asked about the uses of these tools, a conflict between logics emerged. Editors talked about success on social media as a mixture of intuition and trial and error, saying that ‘it’s not an exact science’. For instance, when asked about analytics, one editor said, ‘There is no formula but I know stuff, I can tell if something is going to work or not … I don’t know whether or not to call it a sixth sense’. This argument resembles the discourse around journalistic work as a craft rather than profession, and the concept of ‘journalistic instinct’ (Tumber and Prentoulis, 2005). It is very different from discourses around social media and other digital platforms, which highlight the ability to employ big data to achieve prediction (Mackenzie, 2013). Only when directly asked about the use of analytics did editors state they use it to see ‘what works.’ It seems, then, that SMEs look at data as a way of testing themselves after the fact. For instance, an editor with a TV corporation related to how she launched an application using a network talent, based on her gut instinct. When she examined the data she realized that another talent attracted more attention and downloads, so she changed the promotion accordingly.
To conclude, although the discourse around social media and datafication suggests a world in which data are used to develop a granular understanding of users and their preferences, it seems that in reality their use is limited. These findings coincide with analysis of the European news market, suggesting that only market-leading news organizations employ a sophisticated use of analytics (Cherubini and Nielsen, 2016). Be that as it may, as one interviewee said, ‘For many years we worked according to our gut instincts, but it got to the point that it is too much to handle… and the tools are more sophisticated and helpful’. Thus, it is possible that while this trend has yet to take hold on the media industry, it may in future.
As the analysis thus far has demonstrated, reviewing the four dimensions of social media logic suggests a complex relationship whereby logics intertwine or even clash. In all dimensions, traces of both logics are evident, yet the balance between them differs. Popularity could be seen as one pole of the spectrum, where the influence of social media logic is clear, and this has been embraced by the editors. This could be attributed to the inherent reciprocal relations between logics where popularity is concerned (van Dijck and Poell, 2013). At the other end, datafication demonstrates a different balance between logics, where for the most part social media logic did not penetrate deeply into work routines. Between these two poles lie connectivity and programmability: social media logic is clearly evident here, but conflict and tensions between logics remained unresolved.
Conclusions
Social media are becoming a dominant force, shaping not only human interaction within the platforms but also shaping other societal constructs, among them mass media. Through platformization – a technological process through which the web’s data is transformed in order to be distributed by social platforms (Helmond, 2015) – social media absorb mass media content. However, such a process necessitates adaptation: the transformation of content from one platform to another involves a process of translation from one media logic to another (D’heer and Verdegem, 2014). This process is at least partly in the control of SMEs.
As our research demonstrates, this transformation is not smooth. While some principles of social media logic are embraced due to their alignment with mass media logic (popularity), others are resented, yet their overwhelming power suppresses any resistance (programmability). Still others are intertwined, woven into mass media logic. As such, media organizations present a hybrid system, where neither social media logic nor mass media logic is perceived as an absolute. Rather, social media logics reflect the practices and relationships that are negotiated through the role of the editors, also affecting old logics practices.
What are the broader implications? As Poell and van Dijck (2014) argue, the appearance of media independence from state or commerce cannot be maintained. Social media technology is far from neutral, and adopting their logics compels mass media to construct their relations with their audiences in new ways. In addition to considerations regarding newsworthiness or virality, SMEs are forced to shape their messages in the light of the external limitations imposed by the technological layer, even at the price of compromising their journalistic norms and practices.
As others have noted, surrendering to metric-based journalism harms news organizations’ ability to perform their gatekeeping role (Tandoc and Vos, 2015). The current study serves as a clear illustration to that. Being editors, SMEs are enjoined to function as gatekeepers yet, as the interviews show, newsworthiness was of little concern to them. The growing dependency of mass media upon social media could accelerate the blur between journalistic norms and marketing considerations to the point that media’s role as the fourth estate will erode. In their search for a new sustainable business model, struggling media organizations produce content that will gain users’ engagement. Such practice marks a significant difference from the well-established professional tradition, which highlight journalism’s role in democracy.
SMEs’ role as translators naturally explains that tension between the dependency in social media and journalistic practices. It is also manifested in SMEs’ position as somewhat estranged from their organizations: they are part of the newsroom team, yet watch it from the sidelines. They share mass media logic, but also embrace their own, so there is somewhat of a professional divide. Yet even their world is not truly theirs. SMEs are painfully aware of the fact that they have minimal control and that the power is in the hands of social media corporations. This is another reminder that mass media’s autonomy is at risk. When large parts of the production, marketing, and distribution processes are carried out by outside companies or in accordance to their preferences, news slowly slip outside of the grasp of news organizations.
As mentioned earlier, the SME’s role is new and its logic is still being designed. A follow-up study would be able to capture longer trajectories and more specific aspects like the effects of changes in the algorithm on media logic. It would be useful to compare our findings to media organizations in other countries. Future studies should consider the heavy invasion of market logic into social media logic. Financial logic might trickle deeper and wider than social media logic as media organizations increase their reliance on audiences as a source of income, to compensate for their shaky financial state (Newman, 2018).
Footnotes
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the ERC (grant number 680009).
