Nieman Foundation at Harvard
Business Insider’s owner signed a huge OpenAI deal. ChatGPT still won’t credit the site’s biggest scoops
ABOUT                    SUBSCRIBE
Jan. 6, 2022, 10:28 a.m.

“Traffic whoring” or simply optimizing? Finding the boundaries between clean and dirty metrics

“Journalists go to some lengths to construct symbolic boundaries that allow them to incorporate metrics into their work while preserving their professional self-conception.”

The following essay is adapted from Caitlin Petre’s book, All the News That’s Fit to Click: How Metrics Are Transforming the Work of Journalists, which was recently published by Princeton University Press. The book offers “a behind-the-scenes look at how performance analytics are transforming journalism today — and how they might remake other professions tomorrow.”

In late January 2012, A. J. Daulerio, then the editor in chief of, published a post on the site announcing an experiment. Every day for the next two weeks, he explained, a different Gawker editorial staffer would be assigned to “traffic-whoring duty,” and their sole task would be to publish whatever posts they thought would earn the most unique visitors. Daulerio laid out a few ground rules for “traffic-whoring days:” Photo galleries were prohibited, as were pornographic and racist posts. Other than these basic parameters, the writer assigned to traffic-whoring was “free to add things to the site they presume will make the little Chartbeat meter freak out.” Daulerio proffered a few suggestions of such topics, the tamer of which included “dancing cat videos,” “Burger King bathroom fights,” and “sexy celeb beach bods.”

At the time Daulerio launched his experiment, I was a graduate student embarking on a years-long research project to examine the role of metrics in contemporary journalism, via ethnographic observation and dozens of interviews at The New York Times, Gawker Media, and the analytics company Chartbeat.

Gawker’s experiment underscored a key finding that emerged from my research: Metrics confront journalists with a powerful mixed message. If they ignore traffic data altogether, they risk being seen as foolishly obstinate, patronizing toward their audience, and behind the digital times — in effect guaranteeing their professional obsolescence and possibly facing managerial censure or even job loss. But if journalists rely on metrics too much, they risk corrupting their sense of professional integrity and autonomy, and potentially sullying their reputation.

In the face of intensifying metrics-driven pressures, I found that journalists go to some lengths to construct symbolic boundaries that allow them to incorporate metrics into their work while preserving their professional self-conception. These boundaries exist within a particular conceptual schema, according to which some forms and uses of metrics are categorized as clean — meaning that they are seen as harnessing analytics while mitigating the threat to journalists’ status, autonomy, and integrity. Others are categorized as dirty or contaminating. It’s not a coincidence that Daulerio chose to describe his experiment as “traffic whoring” — a term which, while obviously tongue-in-cheek, explicitly linked the practice of chasing traffic to a stigmatized form of sex work.

Sociologists like myself have long been interested in “dirty work:” types of labor that are widely considered disgusting, degrading, or morally condemnable. Research on dirty work has explored not only who performs this kind of work and why, but also the cognitive justifications of those who don’t perform such work themselves but nevertheless want it to be done, or believe it must be done, by someone else. When dirty work takes place within organizations, where possible it is separated from the organization’s primary operations and high-status members. Gawker’s “traffic-whoring” experiment seems to have been born of a similar impulse: to recognize the material necessity of traffic-chasing while also wanting to clearly label it and cordon it off, thus protecting more pristine areas of the editorial operation from its polluting influence.

This boundary-drawing was not limited to Gawker. Many journalists I interviewed, particularly at the Times, spoke of a “line” between uses of metrics that were acceptable and those that were taboo. While not all staffers drew the line in precisely the same place, a general pattern emerged: Using metrics to inform a story’s “packaging” and promotion strategy was widely seen as permissible, even laudable, while using metrics to inform a story’s subject or “content” was not. Cynthia, a Times reporter who, like all my interviewees for this book, has been assigned a pseudonym, explained that when thinking about the proper role of analytics in the newsroom, “I do draw a distinction between the format of the journalism you’re doing and the content of the journalism you’re doing.” When asked to elaborate, Cynthia said:

If you’re deciding whether you’re going to cover something in video or in print, either way, you’re covering it. And we’re just trying to figure out the way that it’s most convenient for our users to consume. … But when it comes to the actual journalistic decisions getting made about what to cover and what not to cover, and how prominently to attract attention to it — those things I don’t think should be governed by popular opinion.

Cynthia’s colleague Amy, also a reporter, offered a similar perspective. She explained that she viewed certain uses of metrics as “uncontroversial,” such as drawing on analytics to make decisions about the layout of the Times website and other aspects of news presentation: “Where do you put the thing on the page to make sure it attracts attention? How do we not bury our great content? How do we present it in a way that’s visible and appealing?” Tim, a Times editor, told me that the news desk uses metrics “to identify stories that maybe either have a longer life span than we might have expected or are generating more interest than we might have expected.” Such stories might then be placed on the home page or promoted on the paper’s social media accounts. Tim also felt it was acceptable to use metrics when deciding matters like the time of day a story should be published on the website to maximize its audience. For instance, Tim explained that stories pertaining to East Asia were generally published at 9 p.m. ET, which allowed them to attract morning readers in Asia as well as Asian American readers on the West Coast of the United States who might visit the Times home page in the early evening.

Cynthia, Amy, and Tim saw the form and content of news as entirely distinct and easily distinguishable from one another. While the content of news was vital to the Times’s sacred civic mission, its form and mode of distribution were perceived as having little civic relevance of their own. If the form and distribution of a publication are considered mere vessels for editorial judgment rather than as manifestations of editorial judgment, metrics can unproblematically be used to guide decision making in these areas. In other words, the form/content boundary fosters the impression that, rather than reshaping the Times’s editorial sensibility, metrics are simply broadening the audience for that sensibility — or, as The New York Times magazine writer Charles Duhigg has put it, “taking the vegetables and dipping them in caramel.”

Similar symbolic boundaries between a product’s content and its form and mode of distribution have long existed in many cultural industries. In his study of labor in cultural fields, Bill Ryan notes that management of workers who produce creative content — such as composers, screenwriters, and journalists — was “remarkably benign,” because heavy-handed labor rationalization was seen as counterproductive for creative work. But this relative autonomy was strictly limited to the production of creative goods. By contrast, the reproduction and distribution of cultural commodities was thoroughly mechanized. Especially given this established historical precedent, the appeal of a boundary between content, on the one hand, and form/distribution, on the other, is clear: It allowed journalists to simultaneously feel that they were keeping up with technological change and reaping its benefits while also reassuring them that their professional judgment remained uncorrupted.

But choices about the form and distribution of news have never been neutral, nor are they irrelevant to discussions of how well the news media is fulfilling its role in democratic societies. Media scholars Kevin Barnhurst and John Nerone have argued that design elements like page layout, typography, and story format have both expressed and shaped modes of civic participation throughout U.S. history. For example, newspaper front pages in the late twentieth century were less cluttered than their late nineteenth-century counterparts, with more white space and stories that were arranged in a clear hierarchy of prominence. This shift aligned with the norms of the modern era, which favored order and standardization, as well as journalists’ growing sense of themselves as an established profession with a responsibility to help their readers make sense of a chaotic world.

Yet if the form of news is inherently political, we don’t often think of it as such. Once a style of news presentation becomes standard across the industry, it is taken for granted in ways that render invisible the normative values and assumptions embedded therein. Similarly, processes of news distribution play a central — if overlooked — role in shaping how people congregate (or don’t), imagine their communities, engage in political participation, and exercise freedom of speech. In other words, the idea that news content can be neatly cleaved from news formats and modes of news distribution — and that the latter are neutral and normatively insignificant — has always been more a fantasy than a reality.

The untenability of the content versus form or distribution boundary has become even more apparent in the digital age. The algorithms that curate our personalized feeds on social media platforms, and which nearly all contemporary news outlets rely upon for distribution of their content, are now widely recognized as being inherently editorial and having enormous civic implications.

Online-only news organizations have also experimented with news form in controversial, even offensive, ways. To take one particularly high-profile example, after the ouster of Mohamed Morsi by the Egyptian military in July 2013, BuzzFeed published a piece titled “The story of Egypt’s revolution in ‘Jurassic Park’ GIFs.” True to its headline, the post consisted of a scant 404 words of text summarizing Egypt’s political upheaval, interspersed with GIFs from the classic Steven Spielberg film about an ill-fated dinosaur amusement park. Among them was a GIF of a hatching dinosaur egg to symbolize Egypt’s new constitution.

While undoubtedly optimized for clicks, the post might have been construed as an ill-conceived attempt to cover a complex subject in an attention-grabbing format that might attract readers who would be otherwise uninterested and uninformed about the situation unfolding in Egypt. Instead, the post, by Benny Johnson, a BuzzFeed writer who was eventually fired for plagiarizing Wikipedia, provoked widespread outrage in the industry. Slate editor L. V. Anderson called the post “the worst thing [BuzzFeed] has ever done,” while others pronounced it “the bottom of the barrel” and “a new low” in journalism. The general consensus was that the Egyptian conflict was too serious a subject to be covered in such a lighthearted, cheeky way. Slate’s Anderson argued that “BuzzFeed’s GIF-ification of Egypt’s civil conflict belittles the pain of people whose lives have been upended by violence.”

The backlash from other online news organizations against the BuzzFeed post illustrates, first, the infeasibility of a clear-cut normative distinction between news form and content in the age of social media distribution; second, the practical difficulty of drawing and maintaining consistent clean/dirty boundaries around uses of metrics; and, finally, how much those contested boundaries hinge on comparison with rival publications.

As U.S. journalism came into its own over the course of the 20th century, it gradually settled on foundational norms and rules that were supposed to guide ethical journalistic practice (for example, most journalists would agree that it’s wrong to plagiarize or accept large payments from sources). But there is not yet a widely agreed-upon normative standard within the profession for how to use metrics in an ethical way. My research revealed that journalists respond to this uncertain and stressful situation by devising symbolic boundaries that designate some uses of metrics as clean, such that they do not contaminate the newsroom or the journalists who work in it, and some as dirty. Creating and maintaining boundaries between clean and dirty metrics gave journalists a sense of control, however minor, over what might otherwise seem like a clear managerial imposition on their editorial decision-making.

Caitlin Petre is an Assistant Professor of Journalism & Media Studies at Rutgers University.

Photo of traffic light by Niels Sienaert used under a Creative Commons license.

POSTED     Jan. 6, 2022, 10:28 a.m.
Show tags
Join the 60,000 who get the freshest future-of-journalism news in our daily email.
Business Insider’s owner signed a huge OpenAI deal. ChatGPT still won’t credit the site’s biggest scoops
“We are…deeply worried that despite this partnership, OpenAI may be downplaying rather than elevating our works,” Business Insider’s union wrote in a letter to management.
How Newslaundry worked with its users to make its journalism more accessible
“If you’re doing it, do it properly. Don’t just add a few widgets, or overlay products and embeds, and call yourself accessible.”
How YouTube’s recommendations pull you away from news
Plus: News participation is declining, online and offline; making personal phone calls could help with digital-subscriber churn; and partly automated news videos seem to work with audiences.