Tow Report

The Audience in the Mind’s Eye: How Journalists Imagine Their Readers

June 26, 2019
Photo | Adobe Stock



“The most interesting kind of portraiture is that which arises spontaneously in people’s minds.”

—Walter Lippmann[1]


“I’ve been thinking a lot this morning about who my (our) audience is supposed to be. Mainstream media is mostly geared toward an abstract idea that whatever journalism you produce, theoretically anybody can consume it. And there have been a ton of mainstream media journalists, both nationally and locally, who went out and covered the Trump voter phenomenon. Did anyone read their work? Who was the audience supposed to be?”

—Matt Pearce (@mattdpearce)

National Correspondent for the LA Times

Via Twitter – November 10, 2016[2]




A central irony of the newsroom is that while many journalists’ decisions are made with readers in mind, the audiences for their work often remain unfocused, imagined abstractions, built on long-held assumptions, newsroom folklore, and imperfect inference.

This is not particular to journalism. Writing, like reading, is a solitary activity; unlike orators, writers are necessarily separated from their audiences. As a writer works, the decisions they make are based on many factors: their long-term memory, the conventions of their genre, and (conscious or not) an imagined perception of readers’ reactions. Success rests on how accurately they can anticipate how their work will resonate with these imagined audiences; the paradox being that at the time of writing those audiences do not yet exist.

The conventional wisdom of the digital era is that journalists can now know their audiences in far more intimate detail than at any other time in the history of the profession. Previously, journalists based their audience knowledge primarily on their closest social circles. Now, new tools can help them solicit readers’ feedback, analyze and understand readers’ behavior, and open new channels for conversation. These new capabilities promise to shine a light on the abstract audience — making one’s readers present, quantified and real.

Drawing on the existing literature and an original case study, this paper asks whether the new tools of the digital age have indeed influenced the “audience in the mind’s eye.” Our evidence indicates that for the most part, they have not. In reviewing findings from the case study, we were struck by how little seems to have changed since the print era. Although they seemed more open to audience knowledge, the ways in which these reporters thought about their audiences was remarkably similar to those reported in classic ethnographies of the 1970s.

The paper concludes with some hypotheses about why this may be so, and offers some possible approaches to improve audience awareness in the newsroom — in particular, a new perspective on the necessity (and difficulty) of diversity. It is our hope that this paper will inspire future research and experimentation—to narrow the gap between the audiences journalists have in mind and the audiences they serve.


While most newsroom decisions are made with a reader in mind, consciously soliciting and incorporating audience preferences is usually resisted. Journalists recognize their obligations to reach an audience, but they are wary of allowing readers to dictate what is newsworthy. Still, an awareness of one’s potential readership is critical to effective writing. Although this audience thinking is often subconscious, expressing itself as “gut instinct”, the images journalists use to describe their perceived audiences can be quite vivid. These imagined audiences are a critical element of journalistic decision-making.

Studies from the print era indicate that journalists’ audience perceptions were based on four major groups:

  • Archetypes they knew well, such as the institutional audience of the publication they worked for
  • Their professional peers
  • Their sources
  • Close acquaintances, including family members.

Input from strangers was often disregarded or dismissed. This instinctive bias toward “known” readers threatened to exclude other, less familiar reader segments, affecting the choices journalists made while selecting, reporting and crafting stories.

A case study of local education reporters in New York City suggests that while journalists are open to engaging with readers, the ways in which they form audience perceptions remain largely unchanged despite the rise of audience metrics and analytics. These journalists still find it difficult to determine whether their work is actually resonating with the readers they seek.

We conclude that those hoping to bring journalists and audiences closer together must consider the “last-mile” problem of audience perception in the writer’s mind. It is not enough to simply convince journalists that engaging with their readers is important, or provide basic metrics to measure articles’ success. Instead, attention should be paid to how well a journalist’s imagined readership aligns with reality. At the very least, they should know how many of those readers there are, and whether they are actually being reached.

Despite the increasing availability of digital audience tools, this research suggests that personal proximity is a critical factor in influencing one’s audience perceptions. Analytical tools describe numbers, not people; and artificial constructs such as personas are often ignored. Given how deeply one’s peers and sources inform one’s perceived readership, increased newsroom diversity might be the most effective way to ensure that the readers in one’s mind’s eye accurately reflect the audiences for their work.


In addition to my work as a research fellow and adjunct professor at Columbia University’s School of Journalism, I have been employed by the New York Times for nearly fourteen years, spending most of the past six years working on various newsroom-focused analytics projects.

We conducted two interviews with Times journalists as part of the research for this paper, in which my role was clearly communicated as a researcher, not as a colleague. Another interviewee joined the Times after our research had concluded. I have not worked directly with any of those journalists before or since, nor have I used any privileged information from my work at the Times in this project. Aside from those two interviews, any references to Times initiatives found here are clearly cited from publicly-available sources.


We have chosen 1995 as a rough demarcation point for the start of the “digital era”. (By January of the following year, when the New York Times introduced its first website, an industry analyst had announced that “the market is booming for newspapers on the World Wide Web.”[3]) Although print persists well into the twenty-first century, our term “the print era” will largely refer to descriptions of journalism before this date.


1. The Imagined Audience

“The term ‘audience’, old and powerful as it is in the rhetorical tradition, might almost be said to mean too much, to block thought by making us think we know what we are talking about when often we do not.”

—Douglas B. Park[4]


“If some of my assumptions miss the mark about you as a reader, I hope you will nonetheless find a way to temporarily inhabit the roles and attitudes I am inviting you to try on, and in that way become the audience for this chapter.”

—Donald L. Rubin[5]


As much as journalists like to talk about the five W’s of a news story—who, what, where, when, and why—the practice of journalism rests on three other, equally important questions: “Who am I writing for? Why is it important for them to read it? And what will they find interesting?”

These questions may not be asked aloud. They may not even be asked consciously. But they are always answered, in some way, even if merely by habit or instinct. And those answers—a journalist’s hypotheses about one’s readers—influence the many choices they make: the subjects they select, the reporting they do, and the stories they tell.

But knowing one’s audience presents a considerable challenge. Not only are writers separated from their readers by space and time, but the potential audiences for their work are large and diverse. This section draws on a broad range of sources to examine how experienced writers draw on long-term memory, first-hand knowledge, and existing conventions to construct a mental image of their readers, which they use to inform their decision-making. Often, this process is subconscious, forming one’s “gut instinct.”



In September 1760, Ben Franklin penned a letter to his close friend David Hume, who had objected to Franklin’s choice of words in a recently published pamphlet. Reassuring Hume that “we shall always in America make the best English of this Island our Standard,” Franklin concluded the paragraph with an unremarkable but ultimately historic sentence. “It often gives me pleasure to reflect,” he added, “how greatly the audience (if I may so term it) of a good English writer will, in another century or two, be increased, by the increase of English people in our colonies.”[6]

In his novel use of “audience” as a synonym for “readership” Franklin had found a new use for an otherwise familiar word. Historically, the word “audience” — sharing the same roots as the words “audio” and “audible” — had referred merely to those in earshot of a speaker, having nothing to do with the written word at all. Even today, all of its definitions, except for Franklin’s, are based in some way on the act of hearing.[7]

Franklin’s choice of the word “audience” to describe one’s readership is echoed frequently today, especially amongst those in the business of producing and publishing news. Their preference for Franklin’s precedent is no accident. The difference between “readership” and “audience” is something like the difference between hearing and listening. The former is passive, requiring little more than one’s presence; while the latter suggests a glimmer of interest or emotional response — a direct connection between storyteller and recipient; what a modern-day journalist might call engagement.

Audience has become a hot topic in digital media, but its meaning varies from person to person. For publishers, it represents potential value; for advertisers, a specific target group uniquely receptive to particular pitches; for digital specialists, it is a canvas for growth, the basis of much of what is called “audience development.” For journalists, for whom it is mainly a synonym for readership, it remains in many ways an incomprehensible concept.



“The real environment is altogether too big, too complex, and too fleeting for direct acquaintance. We are not equipped to deal with so much subtlety, so much variety, so many permutations and combinations. And although we have to act in that environment, we have to reconstruct it on a simpler model before we can manage with it. To traverse the world men must have maps of the world.”

—Walter Lippman[8]


This aspirational search for an “audience”—consisting of not just readers, but the right readers, compelled to read further/click more/spend more time because the story is just so compelling—is the ultimate quest of any modern publisher, whether for making money or producing news.

The challenge seems even more intimidating in light of Franklin’s problematic use of the term “audience” as a synonym for readership. Speaking and writing are fundamentally different activities, as the philosopher Walter Ong noted in his essay “The Writer’s Audience is Always a Fiction.” And when it comes to telling a story, the writer has far fewer advantages.[9]

An orator benefits from intimacy, immediacy and interactivity. Someone addressing a crowd can see who is listening and gauge their reactions in real-time, and adapt their speech accordingly.[10] But writing is almost always a solitary activity, one that “normally calls for some sort of withdrawal.” Alone at the keyboard, writers must find engaging ways to tell stories — drawing, perhaps, on conventions and past knowledge to do so, but otherwise not knowing exactly who their readers will be.

Considered in this context, the oft-repeated advice to “know your audience” is a paradox. (“How can I know my readers if they do not yet exist?”) It is made even more difficult by our cognitive inability to visualize masses of people. It is one thing for a speaker to “read” a room of even a few hundred people, in plain sight; consider the impossibility of understanding the varied needs and motivations of even a few thousand, all of them hidden from view.[11] And the audiences for news can reach into the millions.[12]

Such large populations are so complex and diverse that reducing them to a set of recognizable personas in one’s mind is virtually impossible. As the eighteenth-century rhetorician George Campbell observed: “the characters of audiences may be infinitely diversified.”[13] From the writer’s perspective, then, an audience is at once important and impossible: an abstract group of invisible readers, critical to one’s craft and mission, but diverse and complex in ways that our brains struggle to comprehend.

If the art of public speaking is to gauge the response of a crowd and adapt accordingly, the genius of writing is in structuring a text that somehow anticipates its eventual readership—ensuring that the story will be resonant and compelling when those readers are made real.

Becoming a proficient writer requires developing this innate sense of audience.[14] Studies of the writing process have found that novice writers focus merely on the topic at hand, whereas more experienced writers are more likely to have “a mental image of their readers”.[15] Noting that successful writers are able to “‘decenter’ from his or her own perceptions of reality to consider the needs of the reader,”[16] Carol Berkenkotter concludes “that the internal representation or mental sketch a writer makes of the audience is an essential part of the writing process.”[17]

But, just as a student must learn to write for audiences beyond their teacher, it is not enough to focus on just a single reader.[18] As Albert Wong reminds us, “a writer’s sense of audience shifts throughout the composing process to address many simultaneous audiences.”[19],[20]



It is important to note that these imagined readers are not simply “caricatured audience specification based on shallow stereotypes.”[21] Instead, they reflect a writer’s aspirations.[22] The choices a writer makes are defined by their ideal readers, and the accuracy of those guesses ultimately determines the value of the finished work. If they are able to craft a story that speaks to their intended audience’s interests and correctly anticipates its reactions, they can consider it successful. Otherwise, this work might not find any audience at all.

Walker Gibson summarizes this dissonance from the perspective of the reader. “A bad book,” he writes, “is a book in whose mock reader we discover a person we refuse to become, a mask we refuse to put on, a role we will not play.”[23]—in other words, an audience that (as readers) we will not join.

Two examples of influential aspirational audiences are particularly relevant in the context of journalism. First, a writer may feel pressure to please a single person whose opinion is particularly important — one’s immediate editor, a more senior masthead figure, or the chair of an awards committee. Second, the context in which one’s work is published may prejudice writers towards a certain type of audience. For instance, the publication commissioning an article may well be targeted to a particular type of reader.

These constraints remind us that all audiences are not created equal. The goal for many journalists is not always to reach as many people as possible; often, it is to reach the right people, whether policymakers or those for whom the story is particularly important. Certain readers have the potential to provide direct value to the writer that others do not, such as prizes, payment, or a promotion.



“Adapting the text to the audience is only possible when the physical task of writing becomes automatic and the writer is no longer absorbed by it.”

—José Brandão Carvalho[24]


Given the complexities of comprehending audiences, it’s not surprising that when writers try to describe “imagined” readers, the mental images they create are often vague and abstract.[25] It is tempting to view this as a flaw in the writer’s ability to know their readers.

But it may be that some audiences exist not in a writer’s mind, but in their muscles; that is, the habits formed through years of writing experience and the structures they have come to adopt.[26] Asking an experienced writer about the audience in their mind, then, is like asking an athlete to describe their hand-eye coordination.

Indeed, the act of invention — even if done privately — relies on external influences. “When we engage in an internal dialogue with another ‘self’,” Donald Rubin writes, citing the work of Karen LeFevre, “that inner conversation is constituted by views and voices we have internalized from social influences and other individuals.”[27]

Often, this instinct is informed by existing conventions, a set of assumptions that dictate a writer’s choices. Think of this basic audience knowledge as the writer’s equivalent of sourdough starter; a reliable catalyst, passed down over time, that guarantees a well-baked loaf.

According to Walter Ong, many of these conventions are learned “not from daily life but from earlier writers who were fictionalizing in their imagination audiences they had learned to know in still earlier writers, and so on.” Douglas B. Park points out that “some of these conventions are so totally accepted as to be invisible; some are more obvious; some explicitly artificial”—nevertheless, “[they] help form the ground on which writers and readers meet with some shared understanding.”[28]

In the context of journalism, this represents what we mean when we talk about “gut instinct.” It is based both on the specific experiences of the journalist as well as the accumulated knowledge embedded in professional practice, a set of conventions developed over years of experience, all intended to ensure that one’s work resonates with an unknown audience.


2. The Audience, At Arm’s Length

“The principle [of journalism] is that stories should matter more to others than to ourselves. They should have impact.”

—Duy Linh Tu[29]


“Journalists are not principally interested in their audience. They are interested in the news.”

—Dwight DeWerth-Pallmeyer, The Audience in the News[30]


To journalists, readers can be both necessary and a nuisance.

Since the ultimate purpose of journalism is to serve the public, it’s gratifying to know that one’s work resonates with readers. But while journalists recognize their obligations to reach an audience, they are wary of allowing readers to dictate what is newsworthy, suspecting that listening to readers too closely might corrode their news judgment. Thus, while most newsroom decisions are made with a reader in mind, consciously soliciting and incorporating audience preferences is usually resisted.

Little wonder, then, that a review of literature on the topic describes journalists as historically wary of incorporating audience research into newsroom decision-making, or using quantitative metrics to evaluate their work. The growth of metrics-driven television news broadcasts in the 1970s, while initially well-intentioned, is a cautionary example of why journalists fear pandering to the crowd in the name of ratings.


The idea of serving an audience is a foundational principle of journalism.

This altruism is not the only reason people become journalists, but it is an important one. A recent study indicates that journalism students still seek “a career that is interesting, creative, altruistic, and speaks to their love of writing.” These aspiring journalists are not entirely selfless (“motivations to be famous are on the rise,” note the researchers) but perhaps there is no greater evidence of their idealism than the study’s finding that “[they] are still not motivated by money.”[31]

Those entering the field quickly learn that it is not enough to tell an important story; the stories they tell must also attract interest from readers. After all, “the journalist’s fundamental job is to create interest—for himself, his editors, and ultimately for his audience.”[32] A journalist’s ultimate goal is to write stories that are both important and interesting.

Dwight DeWerth-Pallmeyer, who authored a study of Chicago news organizations in the late 1990s, refers to the resonance between these two qualities as “impact”—something that “leads the audience to take an interest in that which the journalist deems important.”[33] The true measure of impact, he found, is the audience’s response, preferably in terms of some societal change: specifically, “that the reporting will actually prompt its audience to take action.”[34]

Personal Gratification

To have one’s work noticed—to have that impact, draw that response—is a reward that makes a journalist’s work feel worthwhile. As one WGN news director told DeWerth-Pallmeyer: “People who like this business like it because they know it has an impact on viewers.”[35] An editor quoted in another study put it, “the performers on the stage are after the applause. That’s what we want.”[36]

In a 1959 study, Ithiel de Sola Pool and Irwin Shulman described “gratification” as a key motivator for journalists. Some journalists they spoke with sought to “bestow pleasure on readers, who would reward them for it by admiration and affection”; others found that “gratification came from awareness of the weapon of words which they had in their hands and the damage that it could do to the “bad guys.” Either way, “if it’s read, it’s a success.”[37]

Being ignored, on the other hand, is demoralizing. DeWerth-Pallmeyer quotes one reporter at the Chicago Tribune who recalls, wistfully, her time on one of the paper’s less-glamorous beats: “As someone who covered the Illinois Commerce Commission for three years, [I] learned painfully that only my mother read my stories.”[38] Without a visible audience, a journalist suspects that their work is not interesting. They may also come to feel that it’s not important, either.


Personal fulfillment is not the only reason journalists value reader response. Reaching and engaging an audience also demonstrates the journalist’s value—to themselves, but also to their superiors. Perhaps this is why positive feedback (in the form of letters to the editor, for example) could be welcomed by journalists in the print era, whereas negative feedback was often ignored. With no means to know their exact readership beyond the total circulation of their publication, it was one of the only ways to know that their work was contributing to the bottom line.

Silence from an audience, on the other hand, makes one fearful. As a news producer at WGN said, “Without the audience, you might as well fold up and go home.”[39]. A colleague concurred: “I came from an environment where if you didn’t have an audience, there was a good chance you didn’t have a job.”[40] The Danish media studies scholar Hanne Bruun has called this “the terror of the absent audience”.[41]



Corrosive to News Judgment

One common concern amongst print-era journalists was that listening too closely to readers might cloud their news judgment. DeWerth-Pallmeyer refers to this as “the tension between professional values that prompt them to tell their audience what they ought to know and a marketing orientation that prompts them to tell their audience what they want to know.”[42]

In his study, DeWerth-Pallmeyer detailed how a newspaper might well publish something “important” even if it was not considered “interesting”. One Chicago Tribune reporter called these “oatmeal type stories; they don’t taste particularly good going down, but they’re good for you”[43]—an example of stories that must be told because they serve “a vital responsibility for the greater society.”[44] But in the eyes of many high-minded journalists, publishing something “interesting” but not “important” was to abdicate one’s news principles in favor of the audience’s hunger for mere entertainment.

Higher Ratings, Lowered Standards?

When discussing how audience needs can be incorporated into the news process, many observers point to this tension between what readers judge interesting and journalists deem important. A vivid example of this is how local television news programs shifted their focus to an audience-driven model in the 1960s and 1970s, as described in vivid detail by Craig Allen.[45]

Initially, as Allen recounts, the drive to broaden viewership was well-meaning, a conscious shift away from what sociologists considered an elitist perspective. These sociologists were “baffled that broadcasters felt they were providing public service with newscasts that appealed only to the 25% of viewers with college degrees.”[46] One prominent researcher “equated broadcast journalism’s ‘lousy job’ with the ‘small group of upper-middle class people’ who produced and reported the news.”[47] In other words, they felt that “those in newsrooms, because of their professional associations and college degrees, did not instinctively recognize the ‘average viewer.’”[48]

For these broadcasters, the solution was to “get down to where the mass of people are”—in other words, to let audience research drive the newscast. This conflation of public service and broader reach led to audience-driven newscasts like “Eyewitness News.”[49]

Of course, not everyone agreed with this ratings-driven approach. When a competitor to WGN’s news broadcast started incorporating heavy doses of audience research into their decision-making in the early 1990s, the outcome was indeed higher ratings. But the new programming was seen as a dramatic shift towards sensationalism, horrifying many of their competitors.[50]

To many journalists, the lesson of this parable was that the only way to reach larger audiences was to lower standards. As DeWerth-Pallmeyer points out, when “[ratings] become the ultimate goal voiced at a news organization, that understanding does affect the types of stories that are covered and how they are covered.”[51] Ratings were “implicitly used [by broadcasters] to judge the quality of that which is produced.”[52] In other words, this was a world where “interesting” trumped “important”—the larger the audience, the “better” the story.

To the period’s journalists, news judgment was not just an abstract philosophy; it was an intrinsic part of the value they offered to their reader. As the managing editor of the Chicago Tribune told DeWerth-Pallmeyer: “Subscribers don’t want a role in choosing what they get in the newspaper. That’s why they buy the newspaper. They want me to tell them [what’s important].”[53]

This editor explicitly rejected audience feedback as part of his decision-making process. “How do you satisfy all those readers?” he argued. “Well, the way I know you don’t satisfy them is listening to focus groups and readership surveys. You have to put out a paper that has a good consistent philosophy about what it is you’re going to try to do everyday.”[54]

Suspicions of Audience Research

When audience research was made available to newsroom staff, it was usually resisted. As Herbert Gans found in his seminal study of four major newsrooms in the 1970s, journalists had a great deal of “doubt that audience research can help.”[55]

One reason was that research findings were deemed irrelevant to the regular workflow. After all, Gans noted, “researchers can only report on recurring patterns, whereas journalists must make decisions about individual stories.” As one editor lamented to him, “the studies tell me that the most popular cover is a pretty girl on a red background, but how often can I run such a cover?”[56]

But more often, audience research was rejected because journalists were “reluctant to accept any procedure which casts doubt on their news judgment and their professional autonomy.”[57], [58] Since audience research teams were “under the auspices of non-journalists,”[59] newsroom staff could be suspicious of researchers’ motivations. (Gans does allude to one instance where an NBC research team eventually earned journalists’ trust; but in order to do so, they “restricted themselves to research and did not make recommendations.”[60])

Some skeptical journalists disregarded researchers’ findings—especially when those findings differed from their preconceptions. “I just don’t trust them,” one reporter at the Chicago Tribune told DeWerth-Pallmeyer. “In one of the studies, people said they don’t want to see more news about celebrities and I mean, come on. This is just a case of someone telling an interviewer what he or she thinks they want to hear.”[61]



Being read is an essential goal for any journalist. But to print-era journalists, the example of local television news provided a vivid illustration of how dangerous it might be to value one’s work strictly by audience interest alone.

The deeper lesson, perhaps, is how powerful audiences can be. Mass audiences can be tantalizing; like the sirens of Ancient Greece, they tempt brave journalists to betray their better judgment. Perhaps it is best to lash oneself to the mast and not listen to their seductive call.

But the fearful power of one’s audience can also be harnessed. Readers have direct value to journalists: Their enthusiastic response proves one’s worth; their silence breeds disappointment. This is especially true if the reader has a unique importance to the writer: The opinion of one’s editor, for instance, usually outweighs that of an anonymous reader.

Perception, then, is a critical element in the question of audience. The readers a journalist imagines inform the decisions they make; and the more important the imagined reader, the more influential. The next section will offer a close examination of how journalists perceived audiences in the print era—who journalists considered their most important readers, and what information informed those audience perceptions.


3. Imagined Audiences In the Print Era

“The way in which the world is imagined determines at any particular moment what men will do.”

—Walter Lippmann[62]



When the New York Daily News’ new headquarters was completed in 1930, its exterior featured a literal monument to the newspaper’s audience over the building’s entrance: a huge Art Deco bas-relief engraving of New York City’s “common people.” Bearing the words “He made so many of them,” it represented the News’ target audience — the “ordinary city folk”[63] behind its soaring circulation in the first half of the twentieth century.

While the frieze above the door was one-of-a-kind, the Daily News was not alone: Every publication had an “institutional audience”[64] that reflected its priorities, ethos and approach. It may not have been literally carved in stone, but these audiences formed the foundation of each organization, embodying the very purpose of the publication itself.

Even if the target audience was originally defined by an editorial mission, it was codified on the business side, who defined its literal value—something that could be translated into financial gain, whether by subscriptions, advertising, or both. As such, this institutional audience represented a compact between the business side and the newsroom and as such was highly influential throughout the organization.

“Tell it to Sweeney!”

The history of the Daily News provides an instructive example. In the early 1920s, the paper’s circulation was on pace to become the highest in the country. But it struggled to attract advertisers, who derisively referred to it as “the gum-chewer’s rag”[65]. “Big-space buyers seemed unimpressed by the News’ circulation statements,” noted John Chapman in his 1961 history of the tabloid. “They believed that the paper, plumbing the lower depths, had discovered a subhuman race which was of no account to anybody wanting to sell something.”[66]

This perception did not change until the paper’s marketing chief, Leo McGivena, sent a young researcher named Sinclair Dakin into the Lower East Side—the heart of the News’ immigrant readership. McGivena synthesized what Dakin found into an archetypical reader called “Sweeney”, with a memorable ad slogan that proclaimed: “Tell it to Sweeney! (The Stuyvesants will understand.)”[67]

The resulting trade ad campaign was primarily aimed at convincing advertisers that “even if the News were being bought by ‘ordinary’ people, these ‘ordinary’ people had purchasing power.” But they neatly encapsulated the editorial approach as well. The entire paper was constructed with Sweeney in mind. Its editors and reporters took every opportunity to bring readers’ voices into its pages, incorporating polls and other gimmicks, such as beauty contests and lotteries. In fact, publisher Joseph Medill Patterson had wanted to rename the paper to The Mirror, as “readers were encouraged to see themselves in it.”[68]

Other papers may have looked down at these tactics, but it is telling that, in doing so, they cited the “quality” of their audience as evidence. Take this boast from the Philadelphia Evening Bulletin in 1928 that “never has a single stunt been used to add circulation … never a premium, prize, or contest”—proudly describing itself as a paper that “reflects the sane, moderate spirit of a great people.”[69]

Markets as Audiences

“Throughout much of their history, newspapers had been edited for and marketed to specific classes or socioeconomic groups,” wrote veteran editor James D. Squires in 1993. “There were ethnic papers, labor papers, papers for the masses, and … papers for the ruling classes.” The rise of television, in his telling, led newspaper publishers to adopt the marketing techniques of their new competitors, using qualitative research “to discover what distinguished readers from non-readers, and their readers from those of their competitors.”[70] Their business model was “built on a successful three-way relationship between news content, advertising sales, and target audience.”[71]

A publication’s institutional audience, then, provided a common purpose on each side of the church/state divide, influencing editorial priorities as well as business goals. As DeWerth-Pallmeyer pointed out: “editors and news directors are, of course, keenly aware of their organizations’ goals of making profits. That view also helps shape their understanding of their product, which in turn, shapes their view of the audience.”[72]

These aspirational audiences percolated down through the editorial ranks, taking shape along the way; infrequently informed by audience research (as noted above) but mostly, as with much of a newsroom’s culture, by “osmosis.”[73] A paper’s conventions were self-reinforcing: “The staffer reads his own paper every day; some papers require this,” noted Warren Breed in his 1955 survey of newsroom power dynamics. “Unless the staffer is naive or unusually independent, he tends to fashion his own stories after others he sees in the paper. This is particularly true of the newcomer.”[74] Thus, journalists’ observations of a newsroom’s preferences and habits, informed as they were by a target audience, formed the basis of the their perceived readership.




While the institutional audience defined a journalist’s readership, it remained a somewhat abstract entity. Editors served as their proxies, the most tangible representatives of those unknown readers in the newsroom. And as direct supervisors of journalists’ work—directly influencing a their fortunes—they were perhaps the most important. “When I asked journalists for whom they were writing, producing or editing,” Gans recounts, “they always began with their superiors, and some went no further.”[75]

Responsible for producing a mix of relevant and interesting stories, editors incorporated their paper’s audience into their decision-making. Wielding the power to change or adapt stories in ways they felt would make them more compelling, they could even spike a story that missed the mark. As one journalist told Breed, “I try to figure out what will work best. I learn to try and guess what the boss will want.”[76]

Sometimes, subordinates would push back; and when they did, editor-reporter battles often hinged on who could most plausibly claim knowledge of audience needs and desires. DeWerth-Pallmeyer found that hypotheses about audience preferences were often “used by journalists to convince their editors of a story’s merit,”[77] with editors and reporters each taking on the persona of “audience advocate” in conversations about coverage.[78] “When journalists argue about which stories ought to be covered, they are conveying a tacit understanding about their audience, “ he points out. “They may not explicitly talk about their readers or viewers in their discussions, but understandings about them are clearly cornerstones of their argument.”[79]

Colleagues and Peers

Journalists also instinctively trusted the judgment of their colleagues and professional peers, who served as an important audience segment, if not proxies for other readers. “[Reporters] write for one another as well as for the public,” Robert Darnton recalled in his entertaining account of his brief career as New York Times reporter. “Our primary reference group was spread around us in the newsroom.”[80]

News meetings often served as sounding boards to test one’s audience assumptions. In his 2000 survey of audience attitudes at a large daily newspaper, Robert Sumpter observed that editors saw their task at the daily news meeting “was not to make newsworthiness decisions but to select a set of stories for page one with maximum audience appeal.”[81] In doing so, they “accepted [news] budget meeting conversations about stories as being roughly equivalent to the unknown reader’s reactions.”[82] It was assumed that a good editor was a good judge of what readers wanted. Said one, “If you can sell the story to the editors, you can sell the story to the readers.”[83]

Sometimes, a colleague’s enthusiasm or experience could override one’s own “gut feeling”. Sumpter gives an example in which one editor instinctively rejected a story about sleep deprivation experiments on cats, but when his peers “mentioned their own cats, and pointed out the extreme difficulty of keeping these pets awake,” he changed his mind. Another editor at the same publication “sometimes polled editors from specific gender or age groups” to weigh in on relevant stories, apparently believing they could be a better gauge of those stories for similar, “unknown” audience groups.[84]

This was true of casual settings as well as more formal workplace interactions. Darnton recalled that “clusters of reporters form according to age, life-style, or cultural background … Some have lunch together, buy each other drinks in certain bars, or exchange family visits. A reporter develops trust in his sub-group [and] consults it while working on stories.”[85]

Peers working at other newspapers were also important audiences; their value as readers coming in large part from practical social reasons. “[Reporters] know that the way [external] colleagues judge their work will determine their position in the status hierarchy of the local press corps,” Darton noted. “Professional reputation is an end in itself for many reporters, but it also leads to job offers.”[86]


Sources, too, were a key segment of the reporter’s audience. Not only were they “real people” who journalists encountered frequently, but their opinions had direct influence on a journalist’s success. If one of a journalist’s greatest fears was “getting the story wrong,” negative feedback from those who knew the story best would be especially troubling. If a source was unhappy with the coverage, it may have been evidence that the journalist had erred.

A reporter’s close proximity to those sources, developed over time, could also influence their reporting: “After a year or so on a single beat,” Darnton wrote, “reporters tend insensibly to adopt the viewpoint of the people about whom they write.” As a result, “the Times is so wary of the tendency among its foreign correspondents to develop a bias in favor of the countries they inhabit that it shifts them around every three years.”[87] Darnton asserted that “sources constitute an important element of [the reporter’s] ‘public,’” concluding that “the reporting of news runs in closed circuits: it is written for and about the same people.”[88]

One’s own identity or experience could inform one’s social circles, thereby influencing the types of sources one chose. In his 1981 study of the Washington press corps, Stephen Hess explored how black journalists had different approaches than their white colleagues. “It does make a difference in my coverage that I’m black. My sources will be different,” said one correspondent. “I’ll interview Shirley Chisholm. I know her, while a white reporter might not.”[89]



Friends and Family

Close acquaintances outside of journalism were important influences as well. Most of the journalists Gans met in the late 1970s thought their of audiences as “family members, friends, neighbors and people journalists meet at parties.” As one producer told Gans, “I go after reactions from my neighbors, from my sister in the Midwest, and from my brothers out west.” Another: “You do a show for a cell of people—the office staff, the wife, and the kids.” Especially for works in progress, soliciting reactions from trusted confidantes was easier than approaching unknown strangers.[90]

But although influential, feedback from friends and family was not always terribly incisive. Darnton relates that a typically anodyne comment might run along the lines of: “That was a nice piece on Kew Gardens. I was down there last week, and the place really is going to hell.”[91] He found that remarks like these “carry less weight than the reaction of fellow professionals, but they give reporters a reassuring sense that the message got through.”[92]

If nothing else, they provide feedback that someone is listening: “‘Mom’ may not be a critical reader, but she is comforting,” Darton noted. “Without her, publishing a story can be like dropping a stone in a bottomless pit: You wait and wait, but you never hear the splash.”[93]

Suspicions of Strangers

In contrast to the “well-known” audiences mentioned above, journalists also encountered strangers from time to time—either via phone calls, letters, or casual observation in their everyday lives—and drew some degree of audience knowledge by doing so. As Philip Schlesinger wrote in his 1978 study of BBC News: “Total audience remains an abstraction, made real on occasion by letters or telephone calls, encounters of a random kind in public places, or perhaps more structured ones such as conversations with liftmen, barmen, and taxi-drivers.”[94]

But strangers’ opinions generally did not rate as highly with journalists as those whom they already knew and trusted. Darnton, speaking anecdotally, asserted that in his experience, strangers’ influence was limited: “whatever their subliminal ‘images’ and ‘fantasies’, newspapermen have little contact with the general public and receive almost no feedback from it.”[95]

The most immediate contact with strangers came through phone calls or direct correspondence, although the volume of mail varied widely. Jeremy Tunstall’s 1973 study of newsrooms indicated that journalists’ “most regular experience of members of the general audience is from their letters”[96] although (speaking of the New York Times of the early 1960s) Darnton asserted that “even well-known reporters [did] not receive more than one or two letters a week from their readers”[97]. Gans found that no matter how many letters were received, “journalists ignore or dismiss most of the mail.”[98]

Reader Feedback

Just as the attention was paid to this sort of reader contact varied, feedback from readers could provoke wildly different reactions. Reader response could be welcomed by journalists, as evidence that their work was making an impact. Some took pride in the mail they received, especially when they confirmed a hunch that a story was particularly important. Tunstall noted that “several correspondents spoke with pride of a single piece which had produced, say, 300 letters, and some deliberately ask their readers to write.”[99]

But on the whole, it seems that negative feedback was not seen as particularly valuable. Unknown readers were often discounted as cranks or outliers, especially when their feedback took on a negative tone. Gans says criticism from readers could be easily ignored “because it is ever-present and unchanging, regardless of what is in the news; and because in the end, letter writers can hurt them only when they complain en masse.”[100] The conventional wisdom, he noted, was that “people who object are always readier to write than those who praise.”[101] Those who took the time to write or call were often dismissed as “unrepresentative of the total audience, and therefore need not be taken seriously.”[102]

They may not have been wrong. “Regular citizens’ voices were allowed into the newspaper because they were seen as representing the truth and authenticity of the public,” notes Karin Wahl-Jorgensen in her 2007 study of letters to the editor,[103] despite evidence that “people who write letters [were] often older, better educated and more conservative than the general population.”[104] Pritchard and Berkowitz note that “studies of letters to the editor have reached different conclusions on whether letters are accurate representations of public opinion,”[105] although their own research shows that “newspaper journalists may tend to take cues about reader concerns from letters to the editor.”[106]

More simply, what was received may have just been seen as annoying or irrelevant. Tunstall reported that “the word ‘crank’ occurs frequently in comments about letters; many letters are abusive; others ask for very simple advice or information available in reference books.”[107] Gans himself “judged nine percent of the letters sent to the NBC news program in October 1975 to be partly incomprehensible, and another nine percent to be totally incomprehensible.”[108]

Given this, it’s not surprising that many journalists held a dim view of “unknown” readers as potential participants in their decision-making processes.

Still, DeWerth-Pallmeyer found that some journalists felt feedback from readers to be useful, especially when it was in response to specific aspects of coverage. One example was a “call sheet” circulated by WGN staff that detailed the topic of each call received during the day, and sometimes had a direct effect on newsroom decision-making: when four calls came in objecting to the use of the term “decapitated” in a story, the staff stopped using it.[109] (This openness may have reflected evolving attitudes: one writer at the station admitted that earlier in his career “I would have argued with that and said the viewer is stupid,” but had come to feel that such a response would be “forgetting who we’re serving.”[110]) The number of responses needed to get journalists’ attention was quite small; DeWerth-Pallmeyer quotes a senior editor as saying that fewer than a dozen complaints could be enough to reverse a decision.[111]

Another senior editor admitted she enjoyed answering incoming calls when she got to work in the morning (even the ones she called “nuts”) because “it helps to get feedback from the people you are writing for so it doesn’t seem like you just talk … in a vacuum.”[112] Some editors, realizing the value of getting to know these hitherto unknown readers, tried to encourage more contact with them. Sumpter noticed that one editor “required his reporters to eat lunch once a month with people who were not routine story sources and then to report the conversations to him.”[113]

More subtle feedback came from the small things that journalists saw in their daily lives. One would walk around his neighborhood, noticing that “every bar, every place where you go by, they’re turned to our news.”[114] Another paid close attention to which of her fellow commuters read the Chicago Tribune and who was reading the competition[115] — concluding “you don’t see many blue-collar [workers] or Blacks or minorities reading the Tribune.”[116]



These, then, were the “known” readers that collectively formed the journalist’s conception of their audience: the archetypical readers of the parent publication; their editors, sources, colleagues, and other professional peers; their friends and family; and finally, those strangers who took the time to make their voices heard (and for their efforts, in good faith or bad, were often ignored).

But these segments made up a relatively small portion of a journalist’s actual audience. How could a journalist account for the rest of their audience—their “unknown” readers?

Imagining Masses

Faced with such a mysterious and vast group, journalists sometimes responded with willful ignorance that evokes the incomprehensibility of large groups. As one television news producer told Gans, “I know we have twenty million viewers, but I don’t know who they are. I don’t know what the audience wants, and I don’t care. I can’t know, so I don’t care.” Another editor admitted: “If we had to think about how our readers feel, all twenty million of them, we’d freeze.”[117]

Darnton alludes to this, noting that reporters “have difficulty [imagining] the effect of their stories upon the “mass” public, which probably is no mass at all but a heterogeneous collection of groups and individuals.”[118] Tunstall, too, found a similar dynamic. “The audience of the whole news organization is so general and the direct involvement of one out of 300 journalists on a national newspaper is so specific that there is little contact,” he noted. “The individual audience member—one out of millions—has low marginal utility to the news organization and little power over the individual journalist.”[119]

Another way of avoiding the difficult question of one’s unknown audience was to imagine that it encompassed everyone. “[A]ll story types are assumed to have built-in constituencies,” Gans notes, “but perhaps the principal invention is the assumption that the total audience spans all ages and educational levels. On the one hand, this invention enables journalists to ignore the audience altogether; on the other, it allows them to assume that every story is sure to appeal to someone.”[120]

Imagined Individuals

However, journalists’ audience perceptions were not limited to “everyone” and “no-one”. Faced with these unknowable masses, they often used “known” audiences—real, immediate and tangible—as the basis for vividly rendered reader images.

“Journalists who seek out or take in feedback from the known audience assume that it is a sufficiently representative segment of the total audience,” wrote Gans.[121] “Lacking data about how many people are affected by an event, journalists make impressionistic judgments,” he noted. “Their perception of the population … is derived in part from the people they know best.”[122]

When prompted, journalists could describe their imagined readers in striking detail, sometimes even evoking real people. Take, for example, these detailed images solicited by New York Times public editor Byron Calame in 2005, revealing how some Times editors described their readers:

  • Gerald Mazorati, magazine editor: “I imagine my reader is a late-thirty-something woman, a lawyer or educator or businesswoman. She’s busy with work, and also with family matters, but Sunday morning is a time she’ll allow herself to read something that is not work related, or kids’ homework related. She’s got 45 minutes, an hour. She wants to lose herself in a story, one big story—8,000, 9,000 words. My hunch is she wants to read not something escapist but something substantive—something that holds a mirror up to her own life or opens a window onto a pretty troubled world.”[123]
  • Jill Abramson, managing editor: “Because I feel so urgently about capturing younger readers and interesting them in serious news, I think about one of my son’s best friends from high school in Arlington, Virginia, who enlisted in the Army and is about to deploy to Iraq. Are we telling a younger reader like him about all the dimensions of geopolitics and culture so he can understand all the dynamics relating to the war he is about to help fight?”[124]
  • Andrew Ross Sorkin, business reporter: “I am always trying to make my articles financially sophisticated enough to appeal to my father, while also being accessible and engaging enough to interest my mother.”[125]

Imaginary readers were often based on family members; spouses in particular. One editor admitted to Gans: “I basically edit for my wife.”[126] Another example, from DeWerth-Pallmeyer’s study: “if it’s something you’re going to tell your wife about, that’s what makes it newsworthy.”[127] (These allusions to traditional gender roles, so endemic in classic ethnographies, further highlight the homogeneity of print-era newsrooms.)

Imagined audiences could also be based on prominent individuals — much as Fox News today sometimes feels as if it is “being programmed for an audience of one.”[128] Darnton describes one detailed example:

“When Tom Wicker was covering the Kennedy White House, he not only knew that Kennedy read his stories attentively, he also knew exactly when and where Kennedy read them. The Pentagon correspondent, I was told, knew that MacNamara read defense stories between 7:00 and 8:00 A.M. every day while being driven to the office. Those reporters must have had vivid images of Kennedy and MacNamara scowling or smiling at their prose at certain times in certain places, and those images probably had more effect on their writing than any fuzzy view of the general public.”[129]

Sometimes management developed detailed reader archetypes for their staff, but examples are rare and in any case these personas did not seem particularly effective. Working a reporter at the New York Times in the early 1960s, Darnton recalled that “editors expected [reporters] to aim their stories at an imaginary twelve-year-old girl.” He had trouble taking this seriously: “‘Why twelve years old?’ I used to ask myself. ‘Why a girl? What are her opinions on prison reform and the Women’s House of Detention?’”[130] But “this mythical creature was the only ‘audience image’ I ever ran across in my newspaper work,” Darnton noted, “and she merely functioned as a reminder that we should keep our copy clear and clean.”[131]



These imagined audiences influenced journalists’ decision-making in subtle but apparent ways.

Story ideas often arose from interactions with family or friends. Take this example from Gans’s research: “One magazine decided to run a cover on malpractice partly because an editor, who lived in a suburban community full of doctors, reported that it was a central topic of conversation at the parties he had attended.”[132] Or Sumpter’s description of an editor at a Southwestern paper, who noted: “we’re all kind of the same age … [and] an awful lot of those folks are just now having kids. You can test this. Pick up a paper; there are plenty of stories about how to raise 3-year-olds.”[133] [134]

DeWerth-Pallmayer also notes how reporters also kept the audience in mind while reporting stories. Take, for instance, the choice of questions during an interview; as Michael Shudson points out, such interactions include “at least three parties, not two” — not just the reporter and subject, but also “the always present ‘public’ or audience.”[135]

Sometimes editors would make decisions to protect readers. Sometimes this stemmed from empathy; Werth-Pallmayer gives an example where WGN’s news department decided to spike a story on food poisoning for fear it would cause panic. “What we decided collectively was we’re not going to run it,” for fear it might “scare the living daylights out of the public.”[136] Pool and Shulman likewise tell of a foreign correspondent who “has hesitations about including items which might cause or suggest harm and who wants to include items which will reassure.”[137]

On the other hand, these decisions could also be driven by fear. One nefarious case was cited in a 1956 study by Walter Geiber, who noted that some editors he met “were sensitive to any news that would antagonize local businessmen,” which caused them “to reject or give low display to ‘gloomy’ economic stories.” Otherwise, they rarely thought about their readers at all. One was “blunt: ‘I don’t give a damn for the public’.”[138]

Audiences for Beats

Beats—continuous lines of reporting focused on a particular topic—are of particular interest to this paper. As DeWerth-Pallmeyer notes, the impetus for beats is in fact audience-driven: they “are set up on the tacit premise that the audience is particularly interested in certain topics or is especially affected by those areas.”[139]

Darnton agreed that “the tendency toward specialization within newspapers encourages reporters to write for particular publics.”[140] He describes how the New York Times of the early 1960s “assumes that its readers consist of heterogeneous groups: housewives, lawyers, educators, Jews, suburbanites, and so on. It calculates that certain groups will read certain parts of the paper, and not that a hypothetical general reader will read everything.”[141] (Investing in the labor beat, for example, was seen as by publishers a way to gain the loyalty of dues-paying union members.[142]) As a result, the Times “encourages specialization among reporters. It hires a physician to cover medical news; it sends a future Supreme Court reporter to law school for a year; and it constantly opens up new beats such as advertising, architecture, and folk music.”[143]

Beat reporters operate under somewhat unique circumstances, as Mark Fishman discovered in his 1980 immersive study of two California papers. Journalists on a beat, he found, are given a great deal of autonomy: “Reporters on a beat are rarely assigned stories. The beat reporter is largely responsible for deciding what to cover and how to cover it.”[144] And more than most journalists, they “operate under a continuous work overload.”[145] After all, “the size of a news staff is based on the amount of news space to be filled, not on the size of the community being covered.”[146]

To efficiently cover their assigned area, beat reporters oriented themselves around a well-organized routine that Fishman calls a “round” — the contours of which often date back to previous journalists working the same beat. As a result, Fishman found, “reporters strategically expose themselves to only a few sources of information within their beat territories”[147] — mostly “formally constituted organizations and groups.”[148]

Fishman explains four important characteristics of a beat: (1) it “has a history in the news organization that outlives the organization histories of the individuals who work the beat”; (2) “the reporter is responsible for, and has jurisdiction over, covering the beat”; (3) it has “a domain of activities occurring outside the newsroom”; and (4) it is “a social setting to which the reporter belongs”.

This detailed study, unfortunately, said little about beat reporters’ audience perceptions. But Fishman’s findings suggest that not only might beat reporters develop a unique sense of their own audiences (apart from their institutional audience) but also that those imagined audiences may be heavily influenced by the limited circle of people one interacts with (mostly sources, who have a high degree of familiarity and expertise) and the inherited assumptions of one’s predecessors.

What were the practical effects of this audience thinking? Reporters on these beats often based their decisions on their perceptions of readers’ expertise. As the transportation writer for the Chicago Tribune told DeWerth-Pallmeyer: “Sometimes when I write, I want to make it weighty enough so that a person who is in the profession—a transit professional, somebody in the airlines or whatever—can appreciate the import of it and … the sophistication of it.”[149]

Likewise, a 2012 study of health journalists found that these reporters “likely shape their presentation of health and medical information based on how they interpret an audience’s ability to understand it.”[150] These writers face frequent choices as to what extent they should focus on accessibility (“make information more understandable and easier to act upon”) versus credibility (“practices that signify scientific rigor”).[151] The paper concluded that those reporters who tended to get story ideas from readers tended towards the former, in large part because they were able to “keep their audience’s desires and ability in mind.”[152]



Bias Towards Valuable Readers

“The standard cure for ‘bad demographics’ in newspapers, magazines, radio and television is simple: Change the content.”

—Ben H. Bagdikian, The Media Monopoly[153]


Given that this rich and varied ecosystem of imagined audiences was largely based on readers who journalists knew personally, one wonders how closely they matched the actual audiences for their work.

This certainly seems to have troubled Gans, particularly since those “known” audiences were largely based on the upscale backgrounds of most journalists, whom he called a “national professional elite”[154] Likewise, Sumpter’s study of a large Southwestern newspaper found that “socially constructed readers often resembled the interests and demographics of the people in the newsroom.”[155] One senior newsroom leader he interviewed openly voiced concerns about diversity, worrying that “editors represented a homogeneous economic class that could be insensitive to stories valued by lower class readers.”[156]

Institutional audiences, strongly influenced by a business-side preference for an affluent readership, were constructed with a similar bias. At its extreme, these preferences could be explicit. As Squires notes: “by reducing circulation efforts among low-income, minority readers, newspapers actually improve the overall demographic profile of their audiences, which they then use to justify raising advertising rates.”[157] (As a result of such efforts in the 1970s, according to Squires and Bagdakian, profits soared.[158])

Gans noted how these business-side tactics resonated within the newsroom:

“Business departments compete feverishly to prove to advertisers that each attracts a younger, more affluent, and better educated reader: ‘more upscale’, in business jargon. Story selectors participate in this competition, notably in the back of the book, because they themselves prefer an upscale audience. A better-educated reader is easier to write for and requires fewer space-consuming explanations; more important, journalists are themselves upscale, and can thus write and edit for themselves.”[159]

But what if the actual audiences were much broader? There is evidence throughout the print-era academic literature that journalists’ assumptions about their readers, particularly their interests, often missed the mark. Gans gives this example from a television news program:

“I met only one television journalist who correctly saw the audience as consisting of persons mainly of working-class background. Having himself come from a blue-collar home, he argued that television news should be made more attractive to blue-collar viewers, but he also added that he did not know how to do so, having long ago lost track of the people among whom he grew up.”[160]

Tunstall’s research gives empirical evidence for Gans’s observations. “To test the hypothesis that correspondents would have an inaccurate image of their news organization’s audience,” he writes, “we asked the simplest specific question we could think of —what proportion were manual working class.” Comparing their responses to the best available data on their publication’s audience composition, he found that “the proportion of working class audience members was underestimated by nine out of ten specialists … typically by about 20 percentage points.”[161]

In his 1984 study of television news, K. Tim Wulfemeyer similarly found that “television journalists do not ‘know’ their audience as well as they might.” The journalists he surveyed “thought that viewers would be most interested in stories dealing with sex, crime and violence, but the viewers reported they were most interested in stories dealing with the economy, consumerism, and education.”[162] While sixteen of the twenty journalists thought viewers preferred a male newscaster, three-quarters of viewers felt the anchor’s gender made no difference.[163]


With these examples in mind, one wonders to what extent the “unknown” audience of the print era may have been a “forgotten” audience instead. Does an unimagined reader necessarily represent an audience ignored?

In the print era, this was largely a moot point. The extent to which these proxies represented one’s actual readership was in large part immeasurable, unknowable and ultimately irrelevant. Without the ability to know who was actually reading their articles, or even conceive of a large and diverse audience, journalists generally stuck with who they knew.

For those who find this dynamic troubling, the digital transformation of recent years would seem to offer reasons for optimism. Conventional wisdom holds that the new affordances of the digital era—more efficient means of communicating with readers, and new tools to measure and analyze their behavior—have the potential to expand journalists’ perceptions of their audience. Whether or not this has actually come to pass will be the subject of the next two sections.


4. Changing Audiences in The Digital Era

“Audiences are elusive. [They are] dispersed over vast geographic areas, tucked away in homes, businesses and automobiles; they remain unseen by those who try to know and manage them.”

—James G. Webster[164]


The disruptive changes of the digital era have pushed audience knowledge into the spotlight. On one hand, the business pressures of the digital age — increased competition, lower margins — have made news organizations focus on their audiences as a critical path to survival, if not profitability. Meanwhile, digital platforms offer a potential wealth of data about readers and their behavior; insights that publishers hope to spin into gold.

Although this new audience focus has been driven mainly by financial concerns, it is not limited to the business side. As Jonathan Bright and Tom Nichols rightly point out, “the audience is no longer the ignored quantity it was in offline journalism. It has a clear impact on journalistic practice.”[165] As early as 2006, researchers had noticed that “journalists have begun to see the audience as a critical lifeline for their economic survival and public legitimacy.”[166] Since then it has become clear that “no-one can take their audience for granted.”[167]

Meanwhile, other fundamental changes have affected how publications think about their readers. News no longer needs to be purchased and consumed as a bundle; instead, the entire spectrum of news offerings can be consumed a la carte.[168] As a result, news organizations now must consider a distinct audience for everything they produce, a shattering of content and readership that challenges the very idea of a single, clearly defined institutional audience.


The rise of digital media gives journalists new ways to connect with and understand their audiences. These new sources of knowledge can be divided into two categories: learning from people (new methods for journalists to listen to and talk directly with their readers) and learning from data (use of analytics and metrics to infer audience insights.)

However, as we will see in this section, journalists’ reactions to these new technologies seem to echo the sentiments of the print era. The opinions of strangers are still frequently dismissed; social media may often be an echo chamber of like-minded peers; and metrics — suspiciously similar to ratings — can be mistrusted, perceived as just as corrosive to one’s news judgement.

Learning from People

Journalists have always had the opportunity to hear from and interact with their readers, but the new methods of digital communication are faster, more efficient and in some cases more public and interactive. Moreover, readers are now harder to ignore, since paying attention to one’s audience is increasingly seen as part of the job. “Whereas reporters were once surprised by reader feedback, it seems that they now expect [it],” C.W. Anderson points out, “even if they do not like it, agree with it, or see it as enhancing their ultimate journalistic product.”[169]

Reader Comments and Social Media

In many ways, journalists view conversations on social media positively, especially if they provide thoughtful feedback about specific stories. Jacob L. Nelson noted that Chicago Tribune journalists who “describe routinely communicating with readers … said they do so to answer questions about how stories were reported, or to clarify a point.”[170] But at its best, it is a rich way to interact with previously unknown strangers. As one journalist Nelson interviewed said, “my Facebook page [is] a conversation.”[171]

But social media platforms, especially in the context of journalism, can be something of an echo chamber.[172] Twitter, in particular, has been criticized for reflecting journalists’ real-world social circles; one study described political reporters’ participation on the platform as “deeply insular and self-involved, much like it is off-line.”[173] A 2019 study by the Pew Research Center[174] found that most tweets are produced by a relatively small number of users, whose demographics skew markedly from the rest of the population — findings that have been cited as evidence for this insularity.[175]

Deeply unpleasant conversations are one reason why journalists sometimes try to avoid contact with readers online. Nelson talked with many journalists who “volunteered examples where the audience angered or scared them.”[176] One columnist described it as “almost like fighting a mob.”[177] Another admitted that “it’s still like a tiny number of people in total. But when they all come after you … it kind of gives you the chills.” He cautioned that reader interactions “can lead down some really bad roads. Stuff is misinterpreted and it just becomes hell on earth … a lot of times you’re just dealing with assholes.” [178]

Commenters, in particular, are not held in particularly high esteem; in various studies, they are referred to as “boorish jerks,”[179] “losers,”[180] the “worst of the worst,”[181] and “people that have nothing better to do than surf the internet at 11am.”[182]

These dismissive reactions towards readers, so reminiscent of the print era, are more common when journalists sense criticism. “When their professional roles, reliability and integrity are called into doubt,” write Kees Brants and Yael de Haan, “journalists are often more self-referential (listening to their colleagues) than responsive (listening to their audiences).”[183]

Narrower conversations may be more valuable. Nelson found that “interactions with the audience were less stressful for Tribune staff when confined to specific, niche audiences.”[184] This might take the form of direct conversations over email, a “more intimate” medium which one Tribune columnist described as “less crude than Facebook and Twitter.”[185]

In-Person Contact

Perhaps individual, person-to-person contact is a more fruitful foundation for approaching audiences. Many organizations have started to formalize events in which journalists actually meet the people they are writing for, hoping for more face-to-face contact with their readers. However, many of the examples given are anecdotal; it is hard to know definitively whether these types of interactions have increased in recent years.

One example was Huffington Post’s 2018 “Listening to America” bus tour, intended to hear about story ideas directly from the audience as well as expose editors directly to unfamiliar readers. In the words of editor Lydia Polgreen, the project was spurred by “wanting to get out into the country and hear from people about the things that mattered most to them rather than us deciding from our office in the East Village what are the stories that are most important to Americans and hear what people have to say.”[186]

But it is unclear whether these initiatives have a lasting influence on audience perceptions. (Polgreen herself acknowledged that the tour was a form of “drive-by journalism”.[187]) Nor it is certain that they always introduce journalists to new types of readers. Take Nelson’s study of the Chicago City Bureau, one of a few non-profit digital startups who have experimented with direct outreach to readers by hosting events. Nelson found that these efforts tended to attract “people who are already very engaged,”[188] concluding that “expanding the audience to include disengaged citizens poses a more difficult challenge.”[189]

Qualitative Research

In-person research, such as interviews, focus groups, and other forms of soliciting reader feedback, might seem to be another way to bring audiences into newsrooms. But for the most part, qualitative research has not had the same intensity of attention as analytics, perhaps because its methodology and approaches have not yet been subject to the same radical transformation.

There are some interesting examples of media organizations adopting qualitative research in the newsroom context. ProPublica has experimented with lightweight user testing to counterbalance the preconceptions and assumptions of their journalists (a group described by their design director as “an extremely biased sample set.”) The New York Times, too, has used research to uncover particular tendencies of different audience segments,[190] asking colleagues to hypothesize about the types of users they’d like to reach. (“If the answer is ‘everyone’,” said one researcher, “that answer is wrong.”[191]) This ethnographic research has uncovered interesting patterns of consumption that would not have been evident from analytics alone.[192]

But these types of audience research projects have drawn far less attention than newsroom-based data and analytics initiatives. “Quantitative and qualitative audience data are not treated equally in the journalistic field,” Qun Wang points out. “Quantitative audience data, such as audience metrics, have been given more attention than qualitative audience data [in newsrooms].”[193]

Learning From Data

The biggest change of the digital era, of course, has been the revolution in audience analytics. The explosion of online reader data promises an entirely new source of information and insight: the ability to measure readership for any story, quantify audience behaviors with precision, and eventually transform news products into personalized experiences.

Many journalists find metrics and analytics compelling, but, ever wary of the crowd, they remain suspicious of how these “ratings” might affect their news judgment. Studies have demonstrated that news managers “appear to be more enthusiastic about watching audiences’ behaviors than about altering news decisions in light of these behaviors.”[194] But although some newsrooms (particularly those with roots in the print era) have resisted sharing this data with journalists, there seems to be more acceptance of the practice in recent years.

The Seduction of Analytics

There is something about data that seems to induce giddiness — both among journalists and the academics who study them. Take, for example, this breathless description of the new tools at journalists’ disposal in an academic journal: “‘Audience information systems’ [have] led media workers across all sectors to envision audiences in quantifiable, data-driven ways, thus privileging scientific precision over vague impressions.”[195]

In particular, observations that one topic or story draws more readers than another are often interpreted as a revelatory window into what “the audience” prefers. As one prominent Finnish editor noted in 2014: “Readers say they mostly read political news. Nonetheless, the Web metrics suggest that the mostly read online news often have something to do with sex.” Based on this, he concludes that “the metrics analysis is the one with the greatest integrity. This method gives newsrooms a direct access to what readers are actually interested in.”[196]

By equating raw reach with reader interests, these types of simple analyses risk provoking the do-higher-ratings-mean-worse-journalism tensions of the print era — in some eyes, tilting the scale towards “interesting” at the expense of “important”. As Anderson found in his 2011 study of Philadelphia newsrooms, “the dominant journalistic values of autonomy and ‘writing for other journalists’ were being encroached upon by a new set of occupational values, a focus on raw audience data, and a ‘culture of the click’.”[197]

Metrics as Ratings

Analytics, of course, can be used to reveal sophisticated insights about user behavior. But even when deeper behavioral insights are presented, it is usually done in the service of increasing topline metrics like pageviews or unique visitors. Those totals are often what reporters and editors care most about: “how did my story do?”

Using popularity as a measure of success is nothing new. Adam Kirsch reminds us of “Sophocles or Shakespeare, for whom moving an audience was the immediate criterion for success.”[198] As noted earlier, “being read” has always been an important motivator for journalists — even in the print era, long before the readership of a particular story could be accurately quantified.

So it’s no surprise that counting an article’s total readers or pageviews is often interpreted as a score. Nelson’s observation that newsroom staff “often immediately pointed to pageviews as a primary indicator for story performance” (although “some did so more sheepishly than others”)[199] is typical. As one recent study points out, this doesn’t necessarily align with one’s own personal gratification. Said one journalist quoted in the study, after posting a celebrity fluff piece that drew a large audience:

“It still makes me cringe … you feel yucky posting it … But again, the audience obviously on that day really wanted that story. It’s not something I’m proud of but it happened and we did well.”[200]

These measurements, dictated by institutional goals, revealed nothing about the audience beyond its scale. Wang points out the brutal logic: “If the number is good, then the story is good.”[201]

Likewise, low traffic to an article is interpreted as failure. As one public broadcaster notes, “You can’t get away from the fact that we’re a publicly-funded organization and if no one is engaging, listening or downloading your program, then that’s clearly an issue.”[202] Anderson cites a case in a Philadelphia newsroom where the traffic to an important story was not what was hoped for: “[The writer] will be disappointed more people aren’t interacting with her story.”[203] Another reporter recalled “a powerfully written, extensively researched story [that] ‘just bombed on the website, it just did terrible. You want to throw fear into the hearts of journalism professionals? That’s a way.”[204]

It’s important to call out the inherent assumption when one interprets article traffic in this way — that every story is fighting for the attention of a monolithic audience. This may not be so: there may be some articles that are only of interest to certain readers, and as a result what seem like low traffic totals may actually represent a high percentage of that audience segment. Robert Silvey, audience director of the BBC for many years, perennially banned the publication of a “Top Twenty” list for this very reason:

“[These lists] encouraged an entirely fallacious impression of the real significance of audience size: that every broadcast had the same target — the entire population — and that they were therefore all to be judged by the extent to which their audiences approached that goal. In fact, of course, each type of broadcast had its own target and these targets differed very widely.”[205]

“There was no virtue in size per se,” concluded Silvey. “All that mattered was whether a broadcast attracted the audience which it was reasonable to expect of it.”[206]

Suspicion of Analytics

Nevertheless, metrics are often interpreted as a ratings mechanism for articles. And given journalists’ historic suspicion of ratings and measurement, it should not be surprising that the prospect of incorporating these audience metrics into their decision-making can fill their hearts with dread. In her 2018 study of a news analytics startup, Caitlin Petre noted that “to many journalists, analytics have become a potent symbol of declining editorial standards and deteriorating working conditions.”[207] Valerie Belair-Gagnon and Avery Holton describe this as “an anxiousness driven by inexperience with data and concerns over the loss of autonomy.”[208]

As noted earlier, journalists might “worry that popularity should not be the main determinant of news”[209], wary of a “loss of editorial independence, or ‘slavery’ to the audience.”[210] Some felt “it was dangerous to let a website be run entirely by analytics,” describing the effect as “a rush to the bottom. You would probably find yourself standing for things that you as a journalist, and perhaps your brand, doesn’t see itself standing for.”[211] As one journalist at CNN said, “if I just wanted to chase what people on the Internet wanted to click on, I would do stories about soft porn and football and nothing else.”[212]

A simmering fear amongst journalists is that their news judgement might be further compromised by having topline metrics like pageviews directly determine their compensation. As the inimitable David Carr wrote in 2014, citing some early examples of “pay per click”: “If I were being paid by the click for this column, I might have begun it this way: Will an oppressive emphasis on “click bait” mean that the news ends up imprisoned by transgendered models posing in disgraceful listicles accompanied by kidnapped nude kittens?”[213]

More sophisticated newsroom staff question the accuracy of the metrics they see. “The implications of a particular peak or trough in statistics is sometimes seen as unreliable,” noted Phil MacGregor. “In one case a statistical spike was identified as a rogue result of changes in technology. In other examples, traffic surges occur because of the accidental factor of links to an item of content from other websites. Another reason to mistrust data is that “the number of clicks on a page may actually reflect its position in the design of the site, rather than its readership appeal.”[214] Over a decade later, these issues persist.

Access to Analytics

These concerns have caused some newsrooms to limit access to metrics, although this seems to be changing in recent years. One manager at an analytics vendor noted in a recent study that “these days, [web analytics] tend to be a lot more open and shared within the organization. That sort of internal transparency conversation has changed and moved toward one where there’s a kind of open and free access to analytics.” [215]

In their 2016 survey of newsroom analytics, Cherubini and Nielsen note that the general response from journalists to analytics has in most cases shifted from resistance to curiosity and interest.”[216] They cite the example of the Wall Street Journal, where “the conversation around data and analytics has predominantly involved editors [but is] moving towards including reporters in the discussion.”[217] The rise of subscription models, which tend to value the quality of one’s readership, not just the quantity of readers, may be one factor behind this shift.

To ensure that numbers are interpreted properly, some newsroom managers have started embedding analysts directly into the newsroom,[218] charged with providing interpretation and explanation of the raw numbers. As Sebastian Horn of notes, “We don’t want to overwhelm ourselves with too much. We don’t want to cover our walls with monitors that show random graphs. They will have to be meaningful numbers that people look at and help them make decisions.”[219] However, these analysts are sometimes viewed with wariness, too — viewed by some as “interlopers, or unwelcome guests who may be providing a service to the news process but are not themselves journalists.”[220]


Given journalists’ familiar reactions to these new capabilities, it is worth asking to what extent the digital revolution has actually changed how they imagine their audiences. Unfortunately, there is not much research that addresses that specific question; we are left with inferences from the work presented above.

To explore these questions further, we undertook an original case study, focusing on how journalists covering local education issues in New York City conceive of the audience for their work. In doing so, we hoped to uncover some clues about how much the process of audience perception has actually changed since the print era.

5. Case Study: Local Education Reporting in New York City

“I want to have a good time writing a story. Maybe I’ll just write it for my own little enjoyment and hopefully someone will come along.”

—NYC education reporter,

interviewed as part of this study


Much has been written on the ways in which news organizations and journalists are adapting to the digital era, but very little research has been done on how these changes may have affected journalists’ perceptions of their audience.

To address this gap (and hopefully inspire others to do the same) we undertook an original research project, choosing the local education beat in New York City as our test case. We conducted interviews with beat reporters and editors from eight different news organizations, asking each whom they saw as their audience, what they knew about those readers, and how they knew it.


We chose to focus on education reporting because the topic is of interest to large and distinct audience groups (such as parents, teachers and policymakers) and is covered by journalists representing a diverse range of news organizations, ranging from digital startups to legacy publications. We also felt that the iterative nature of covering a particularly important topic over time would likely encourage audience-focused thinking.

The responses we heard in our interviews echoed our sense of the beat’s importance. One reporter called it a “bread-and-butter” beat while another referred to it “as the real pillar of any place that can claim to report on how New York really works.” It is “one of those beats that has such universal appeal,” agreed a third, adding: “Everyone cares about kids.”

To control the scope of the study, we decided to exclude broadcast (radio and TV) journalists from the study and focus exclusively on print journalism.

We talked with 14 journalists involved in covering the local New York City education beat from eight different news organizations: five legacy newspapers and three digital-first startups. The interviews were semi-structured conversations lasting around an hour each,[221] conducted over an eight-month span between June 2016 and February 2017. All told, we talked with seven reporters as well as six editors, and one other staffer involved in audience work.[222]

Ten of the fourteen respondents were female, almost exactly reflecting overall trends amongst education journalists.[223] (As one reporter in a recent study put it, “when I started in the business, education was very much a girl’s job.”[224]) However, there were no minority journalists amongst our respondents; as best we could tell, there were none covering the (non-broadcast) NYC education beat at the time of the study.[225]

Most journalists agreed to speak on the record, but at least one requested anonymity. So we have decided to obscure the names, genders and affiliations of all respondents to the best of our ability, and avoided direct quotes from any organizations that withheld permission. In any case, the specific identities of journalists and their publications are not particularly relevant to our findings.


Our main research question is whether the digital transformation of the journalism industry in recent years has caused journalists to incorporate a more diverse, and better formed, set of “known” readers into their perceptions of their audience.

We believe that there are two significant changes in particular that might affect how journalists perceive their readers:

  • First, each story can now find its own audience, apart from that of its parent publication. This can be referred to as a change in audience scope. The organization funding the work still exerts a strong gravitational pull on each story, of course, in the hopes that it will draw an audience aligned with its business goals. But now readers can discover that story through off-platform entry points, such as Google, Twitter, or Facebook, whether or not they are aware of the parent publication. The audience for any given story, then, is potentially limitless. No story does, but every story could, conceivably, reach “everyone”.

As a result, one might expect to see journalists relying less on the “institutional audience” to form their imagined audiences, instead starting to discover their “own” audiences—particularly on beats, where a line of coverage is pursued over time. Given the pressure to survive in a much harsher media landscape, we might also expect to see media organizations encourage journalists to take a more active role in seeking out and engaging new types of readers. In sum, this may result in a greater motivation for journalists to learn more about their readers.

  • Secondly, journalists now have far more ways to connect with and understand their readers—what we might call a change in audience proximity. Whether by social media, reader comments, or even email, they now have more ways to connect with readers, quickly and often conversationally. And they have access to analytical tools that promise insights about those readers; not just how many there are, but how they navigate to and interact with each story.

This implies that journalists are now able to better understand their audiences—both qualitatively (through greater and more rapid exposure to audience feedback) and quantitatively (by access to a host of new digital metrics and analytics). Our study examines the extent to which beat journalists have taken advantage of this ability to better know their readers, and thus construct more accurate mental models of their audiences.


Throughout our conversations, we were struck by how little seems to have changed since the print era. Much of what we heard would not have sounded out of place in the classic ethnographies cited in earlier sections of this paper.

These reporters did not strike us as Luddites; nor were they intentionally ignorant. They expressed an openness to communicating with their readers, were curious about audience research methods, and for the most part did not dismiss metrics or analytics out of hand.

Those we spoke with generally agreed that their audiences were important to them. Sometimes that reason is altruistic: “The duty is to the reader,” one reporter told us. Being read also indicates that one’s work was worthwhile. “I’m worried … that I do this in a vacuum,” one reporter admitted, particularly concerned that after choosing what to write about they’d find that “no one reads my story. It would be a waste of time.”

But despite lengthy discussions about these new capabilities, we sensed that audience remained as much of an abstraction as ever. The fundamental components of the imagined audience remain those described in the print era: (1) the institutional audience; (2) colleagues, peers, and sources; (3) friends and family and (4) “vocal strangers”.


Institutional Priorities

The institutional audience still looms large in journalists’ minds. In most instances, the types of audiences they had in mind were aligned with the editorial ethos and/or business models of their parent publication.

One reporter for a paid digital publication aimed at political insiders had no problem defining her audience as “education professionals,” but more specifically, “subscribers,” admitting that “if you’re a parent, you might read 15% of my stories.” Another, employed by a more upscale broadsheet, admitted that she was more likely to reach “wealthier school parents” because her publication’s audience skewed towards more affluent readers. But this institutional audience did not always merely reflect economic value. One legacy reporter conceived of his audience as “working-class people”—directly mirroring the audience of the tabloid he works for: “I’m writing for people in the South Bronx.”

One reporter said the institutional audience of their parent publication remained important, because “otherwise it wouldn’t really work—[you’d have] an education reporter writing for principals and parents in a way that would not be interesting to anyone who doesn’t work in a school building.” She described it thus: “Not every public school parent is gonna be a reader [of my paper], but what’s interesting to parents is going to be interesting to people who read [my paper].”

In our discussions with both reporters and editors, we found a certain tendency to aim for “valuable” readers, often connected directly to revenue. One editor noted that his publication has a “heavy bias towards the smartest possible readers”, pointing out that “we want [our education reporter] to be the reporter that all the subscribers absolutely have to read.” That reporter told us they had adopted this thinking as a measure of success: “Part of [knowing we’re reaching the audience] is just the fact that people are continuing to subscribe.”

Non-profits were just as focused on demonstrating the value of their audiences, if not more so. “We have to very explicitly convince foundations to fund us [by justifying] what we’re doing,” said one editor of a digital start-up, adding: “the activity of seeking funding and grant writing has forced news organizations like us to think more deeply about our audience … you don’t have to do that at the Daily News.” Another digital start-up editor was urged by a board member with a management consulting background to “think about who they’re writing for.” She admits that her interest in audiences is largely driven by her role as a senior leader in the organization with a vested interest in financial success: “I need to understand the audience really well to make sure that I see where the growth areas are and what is monetizable, and come up with a business model to sustain the work.”


An Audience of Peers

Editors, of course, are more likely to focus on the bottom line — but they also are amongst a reporter’s most important readers. As one of our subjects explained,“[t]hey tell you if it’s boring or interesting” and “shape the idea”. Some of their preferences could be idiosyncratic: “I have an editor who lives in New Jersey,” one reporter told us, “so he’s really into New Jersey stories.” But in legacy publications, these “gut feelings” could also be passed down over time, as part of the beat’s institutional fabric. Said one reporter at a legacy paper: “My sense of [my paper’s] audience [is also] based on how previous reporters and editors have seen the beat and what they assumed readers were interested in.”

But with the exception of single-topic education startups, not all of our reporters had specific editors assigned to them; and if they did, those editors may have been responsible for a number of different coverage areas. This gave the reporters we spoke with wide latitude for pitching ideas, and the freedom to pursue the stories they found important.

As noted earlier, this editorial leeway may be a particular aspect of working on a “beat,” dating back decades. As Warren Breed noted in the 1950s, “When we come to the beat story … the function of the reporter changes. No editor comes between him and his beat (police department, city hall, etc.), thus the reporter gains the ‘editor’ function. It is he who, to a marked degree, can select which stories to pursue, which to ignore.”[226] Indeed, one of the reporters we spoke with had previously spent time at a tabloid, where her editors had given us her wide leeway on story selection. “I used some discretion about what was important,” she told us. “Not just what’s popular.”

Many of these reporters drew on personal experiences to inform their news judgement. Some reporters told us that being a parent of school-age children was a particular benefit. “Some of us grew up in the neighborhoods we were writing about; we felt that these could be our kids and it matters what happens in these schools,” one mused, adding that parents “see some stories that others don’t notice.”

Not all of the journalists we spoke with were parents. Those that did not have children in the public school system seemed to value the perspective of peers who did. But a lack of first-hand experience could also be considered to be an advantage: “Nobody can accuse me of either being biased towards one school or another,” one childless reporter told us. “Some of the stuff I’ve written is sort of controversial.”

As in the print era, we saw real people in the lives of the reporters emerge as representative readers, particularly family members: When asked what specific readers she keeps in mind, one had a ready answer: “I picture my family.” Her parents and stepmother all worked in the Department of Education, so her big question is, “are they going to roll their eyes when they read this?”

Sources As Audiences

The influence of one’s sources as important audience proxies seems to have persisted in the digital era. One editor called this dynamic “one of the most basic tenets of beat reporting” but warned that it runs the risk of “a specific kind of myopia,” particularly for younger reporters. Nevertheless, he felt that interactions with sources helped develop a sense of one’s audience: “as time goes on, you just get a keener sense of exactly what the audience wants.”

Reporters pay special attention to sources, in part out of fear; these sources have a special, privileged perspective on their work. “All reporters think about sources as an audience,” one reporter explained. “Are they going to think you got it right?” “I had a medium-size story come out a few weeks ago,” another told us, “and one of my best sources didn’t say anything. I was like, ‘what’s going on?’ and he didn’t really like the story.” One even actively sought out this feedback: “I send my stories to everyone who participated in them … I really want to know if I missed something.”

Another reporter, asked about types of audiences, responded with her observations of her sources: “I don’t have as good sense as I’d like about what kind of stories teachers respond to,” she admits. “Teachers are scared about speaking on the record. Parents are not, generally.”

Personal Interactions

Most of the reporters we spoke with readily talked about the connections they made with readers. As one reporter put it: “Education is very emotional,” adding, “I get more calls on this beat than I’ve ever gotten before. People get very, very into it.” Nobody mentioned that their organizations discouraged reader contact, although some were more passive. “Usually I’ll wait until they come to me,” one reporter admitted.

This contact is mostly person-to-person and ad-hoc. Just as they did in the print era, reporters receive unsolicited contact from readers, not only by phone and email, but also through text messages and on social media. Twitter seemed to be the main platform for reader feedback; none of the reporters we talked with mentioned spending much professional energy on Facebook or other social platforms.

A distinction was made between personal and virtual interactions. While digital communication makes some contact easier, many of the reporters we spoke with still valued personal contact. One reporter talked about “people I’ve written about over the years—parents, kids … I stay in touch with them and they stay in touch with me.” Another concurs: “I’m glad I initiated all these relationships in person. These are people I already know and trust … People are open face to face.”

In fact, one veteran editor was actually wistful at what she considered to be a loss of personal contact in the digital era, specifically phone calls: “I hardly get calls now, but I get a shitload of email,” she says, “I really miss the voices. One way to know who your audience is is to speak with them, and that gets missing in the digital age.”

Direct contact with audiences is valuable not just for getting to know readers better, but also for feedback on their work. “I get calls and emails almost every day from random readers,” said one reporter. “A lot of times they will have story ideas.” Another admitted that “sometimes my readers know a lot more than I do.” She sometimes meets readers face to face (we sensed this was in the course of her reporting) and when she does, she’s most interested in whether they feel she’s “portrayed their issue accurately or fairly.”

Still, some resistance to some forms of reader feedback lingered; particularly with regards to online comments, which seem to have inherited the “cranks” and “crackpot” perception reserved for reader mail in the print era. “The people who are really motivated to go online and rant are often [not representative],” said one veteran reporter, “and they can be really mean and vicious.” Another thought online comments were useless, “a magnet for bigoted racist people … We never thought that our audiences were the people writing the comments.”


Specific Audience Segments

When prompted, reporters generally enunciated specific audiences they had in mind. The segments included parents, teachers, and administrators. (Here we were at the mercy of our methodology; it was hard to know whether these were actually audiences that they focused on as they worked, or if our prompt — “who do you feel are the audiences for your work?” — caused them to consider specific segments in response.) The chart below shows the responses we heard from reporters about who they saw as the primary audiences for their work:

Publication Primary Audience
Legacy newspaper “The parents and families of NYC public school children”
Digital startup “Education professionals [and those involved in] NYC politics”
Legacy newspaper “A general interest readership … parents, teachers, older students, policymakers”
Legacy newspaper “The wealthier school parents … [as well as] anybody who’s interested in NYC, NYC government, how it’s working.”
Legacy newspaper “Parents … consumers of education”
Digital startup [Too new to the job to give a specific answer]
Digital startup “The widest possible audience … a general reader who might be convinced that education is interesting or important”


However, not all of these reporters consistently focused on specific segments. “Education nerds … I sort of take them for granted,” said one reporter for a legacy broadsheet. Instead, she aims for “people who don’t necessarily follow everything closely but are interested in the city—how’s the mayor doing, issues of inequality—and will see in something that I write a window into that.” Does she keep those readers in mind? “I don’t know,” she admits. “Sometimes I just have in mind what I’m interested in.” Another broadsheet reporter admitted, “I want to have a good time writing a story, maybe I’ll just write it for my own little enjoyment and hopefully someone will come along.”

Many of the journalists we talked to felt that focusing on one particular segment could be too constraining. “Niche” was sometimes invoked as something of a pejorative term, because it seemed to imply limiting the potential audience for their work. “Hopefully, there are stories that aren’t just a niche audience,” said one reporter. Another candidly admitted that although his publication was targeted at “a niche market” he himself aimed for “the widest possible audience.” A third argued the inverse, explaining that aiming for more specific audiences would be in conflict with the goals of their publication, which aimed at a broader readership:

“You don’t want to write [just] for policymakers, because that would be a different kind of publication. You don’t want to write just for teachers. We don’t write stories for principals, because that would also be too niche. [But] if you only hit parents, there are a lot of parents, so that’s okay.”

This desire for “broad” audiences seemed to influence the decisions made when crafting stories. “I use little jargon. I use plain language,” one reporter told us. “You need clear journalism so a lot of people can understand.” Another reporter agreed: “You try and write clearly and cleanly in a way that anyone can understand.” Even the most audience-focused editor we spoke with didn’t feel that aiming for an intended audience should influence their writing. If a story is deemed to interest teachers, for instance, they’ll still try to write it in a way comprehensible to others. “Even though we’re a specialty publication, we want to speak to broad audiences,” she says.

However, one editor proudly admitted that a narrow audience is not necessarily a bad thing, as long as they are the “right” readers. “You can really see the power of our stories with the insiders who can make a difference,” she notes proudly, pointing out, as one example, a school district’s decision to double a budget based on a story they published.

Some veteran journalists found audience-focused thinking at the beat level to be something of a new discipline, contrary to what they’d experienced in their previous work. One editor recalled a previous job covering education at a tabloid: “No matter how much time I would have liked to have spent thinking about what my audience was, what drove that story was screaming headlines [and] screaming parents.” What mattered most was “beating everyone and making the front page.” Audience-focused thinking was a new skill — a foreign concept. “[Until now] I have never worked in an organization that was specific about trying to envision audience,” admitted another editor, recently hired by an education startup.

Quantifying Audiences

Most of the journalists we spoke to had only a vague sense of the size of their audience, or whether they were being reached.

One interesting data point showing the murkiness of audience sizing was the various audience estimates for those listing “parents” as a potential audience segment. Virtually all of their guesses were based on the correct count of students in the public school system: 1.1 million,[227] but this led to wildly varied estimates as to the number of parents:

Publication Estimated Number of Parents
Legacy newspaper “I would estimate there are 700,000-1,000,000 parents.”
Legacy newspaper “At least a million parents—somewhere between one and two million.”
Digital startup “Three-some million of them and their parents? [ie 1.9 million parents.] I don’t know. I mean, I have no idea.”
Legacy newspaper “Somewhere between 1-2 million parents. Obviously not all of them read our stories.”
Legacy newspaper “Well over 1.1 million parents.”


No matter which reader segment a journalist was aiming for, their estimates of the audience size was similarly murky; nor did they have any idea of how many of these readers actually read their work. One reporter for a tabloid didn’t even know his paper’s circulation, and professed similar ignorance of digital metrics: “I don’t track how much my stories get read online … I don’t really care … I don’t want to be bogged down in chasing clicks.” “I know how many hits we have and how many visitors, stuff like that,” another editor says, but admits: “I’m not the most numerically minded person in the world. I picture types [of readers]. I guess I don’t think of it in terms of numbers.”

Ambivalence Towards Metrics

Even among journalists at the digital startups we spoke with, audience metrics were a faint input into journalistic decision-making. Most reporters only dealt with the simplest metrics, such as story-level pageviews. And to the extent they are used at all, these metrics are mainly seen as a rough gauge of success, rather than as a path to deeper insights.

“Metrics are valuable,” one reporter at a legacy broadsheet told us. “I’m glad we have access to them.” But the most important thing she uses them for is for identifying stories that only draw small audiences; those “obligatory stories that newspapers felt they should do.” She points out: “now that we know that nobody reads them, we realize maybe it’s not necessary… maybe we should be using our time elsewhere.”

Others consciously avoid reading too much into the numbers. Writing only stories that only got high traffic numbers would be “boring,” says another legacy reporter. “I’d be writing about teacher evaluations all the time … The numbers are swayed by things that have nothing to do with the quality of the story.”

Journalists at the two digital startups we spoke to, while intrigued by metrics, also held them at arm’s length. A reporter at a digital startup admitted, “We’re not the place where there is a real-time monitor of pageviews.” One editor argued that “the audience isn’t really numerable;” even pageview totals “are incredibly imperfect … Some stories do really well and you’re like, ‘really?’ And others don’t do well and you’re shocked.” She admits: “We want our stories to [reach] everyone, but that’s not real. And there are people who just don’t care.”

The reporters we spoke to did not consider their antipathy towards audience size, reach and metrics to be problematic. After all, many of them feel that promoting stories and analyzing audiences is someone else’s job. As one reporter at a legacy tabloid put it, “I write the articles. It’s up to them to get the paper out to the neighborhoods and keep it cheap enough so that people can buy it, or present it on the website in a coherent manner so that people can find it.” Does he care if his work is actually reaching the audience? “I guess I sort of have faith that people are reading my work,” he told us, noting that “a lot of people in my neighborhood have the paper delivered to their houses.”

Many reporters feel that paying attention to metrics is an editor’s job, not a reporter’s. “Usually the editor figures it out with someone else,’ says one reporter at a broadsheet. “I’m tangentially involved but I’m not sitting there.” Another reporter at a digital startup agrees: “I’m not an editor; I don’t like looking at Chartbeat all day or whatever people use now.” Her editor says her lack of attention to story metrics is “typical of everyone here” even though “we don’t hide anything from the reporters.”

Some of this blasé attitude is due to a feeling that metrics are dictated most by things out of the writers’ control, such as a story’s distribution and promotion strategy. One reporter points out that stories about testing tend to do well, because “advocacy groups or unions will spread it around like wildfire. And there are a lot of readers who aren’t unions or advocates.” Or perhaps a story with good art will make it onto the homepage because it’s visually interesting. And so forth. In any case, she concludes, “sometimes ignorance is bliss.”

An editor at a digital startup blames what she calls “the packaging” rather than her news judgment if a story doesn’t draw as many readers as she’d hoped. “I feel like I’ve been doing this for so long now that I really know … what is just interesting,” she says. “I’m talking about the innate sense of what makes any story an interesting story.”

She bristles at the thought that metrics represent audiences. “Numbers matter, but insights matter more,” she says. “It’s an unfair question to ask reporters to think about the estimates of numbers of readers … I don’t think our reporters [have to know], ‘oh, that should have reached 100,000 people.’ [Instead, they should think] ‘I thought it should reach this type of person, like local school principals, and here’s why, and here’s how I wrote about it with them in mind.”

Openness to Audience Research

None of the journalists we spoke to exhibited any instinctive animosity towards audience research. In fact, while somewhat elusive, it was generally seen as potentially valuable — representing a subtle shift from the attitudes of print-era journalists. “We’re trying to figure out who [our readers] are. We don’t know,” one senior editor at a digital startup said. “Wouldn’t it be great if we could do a national survey of a thousand people? But we don’t have enough resources to do that.”

But reporters, too, seemed generally amenable to the idea. One reporter at a (different) startup told us his organization tries to make the audience real by convening a “readers advisory board”, similar to a focus group. “[It] has been super helpful to me to hear actual people,” he said. “It was really cool, because they liked things that I hadn’t thought of. Obviously, it’s a very select group … but it was really invaluable.” Still, like his predecessors in the print era, he is somewhat dubious that readers can actually enunciate what they want: “[People said] ‘we want more positive stories,’ but people don’t actually want more positive stories. They want to read about conflicts.”

Reader Personas

One editor for an education-focused digital startup felt that getting reporters to focus on specific audiences was critical, but wrestled with implementing that type of audience thinking. She decided to ask reporters to “select the intended target audience every time they wrote a story”—a tactic that was, in her words, “not a success”, as “every time, every reporter selected everyone.” So they stopped requiring an audience hypothesis as part of the publication process. “It’s hard to actually decide who the article is for when you’re writing it,” she admitted. “And reporters really resist that—and maybe for good reason.”

Reader “personas” were another tactic this editor used to encourage reporters and editors to think about their audiences. The personas, archetypical readers made “real” with stock photos and invented names, were developed by their “product and growth team” who had responsibility for incorporating this audience thinking into the editorial process. This editor encouraged them to consider “stretch audiences”—“new kinds of readers that we don’t think we’re serving”. One editor at the same publication found them to be a useful way to visualize different reader types, “almost like baseball cards.”

However well-formed these personas, they do not appear to have had the impact of the Daily News’ famous “Sweeney”. The one reporter we spoke to at the publication cited above liked the idea (“It forces you to be more intentional about how you’re writing and the stories that you’re choosing”) but admitted that “I have not adopted them in my head yet.” When we talked with another reporter at a legacy broadsheet, we sat directly under posters featuring similar personas produced by her publication; but even though much of our conversation was on the topic of “typical readers” she did not mention them once.


Compared to the print-era predecessors documented in the studies described earlier, the journalists we spoke to seem far more aware and accepting of audience as a necessary part of their craft, and are remarkably open to interacting with readers. However, the audiences they imagine are still based on familiar sources: the audience of their parent publication and those readers in close personal proximity, particularly their editors, peers and sources.

Contrary to our hypothesis, there seemed to be no greater motivation to know one’s own particular audience—either its characteristics or its scale. The institutional audience still prevails; while journalists often enunciate specific reader types they may have in mind (such teachers, parents, and administrators), they are of secondary importance. The journalists we spoke with were generally wary of writing for “niches” (unless it was a niche that their employer valued) preferring that their work be accessible to a more general audience.

Metrics and analytics have not seem to influence audience perceptions; at best, the reporters we spoke with tend to treat them as little more than rough scorecards that can be as distracting or irrelevant as they are useful. The analytical tools available have not provided memorable insights about particular target audiences: even in single-topic digital startups, journalists have no way of knowing how many potential readers exist in each of these segments, or how many are reading their stories. No wonder that in a recent survey, over a third of education reporters complained that “audience analytics create pressure for me to ignore important education stories that might not appeal to broad audiences.”[228]

Physical proximity remains the most compelling and resonant source of audience knowledge. Many of those we spoke with eloquently described actual readers they had spoken to, and in some cases remained in contact with. Perhaps one of the ironies of the digital era may be that the most persistent and vivid reader perceptions are still based on actual personal contact, a fundamental human connection that virtual communication — be it through numbers, graphs or even email — struggles to replicate.


6. Informing Imagination

“The pervasiveness of past audiences in our heads is one more reason for the difficulty of reaching present audiences with our texts.”

—Peter Elbow[229]


“Any disturbance of [our] stereotypes seems like an attack upon the foundations of the universe.”

—Walter Lippmann[230]


In an era where so much attention has been given to audience growth and development, why have these journalists’ perceptions of their audience been so slow to change? Perhaps imagining strangers is fundamentally difficult; perhaps our analytics tools are not sufficiently compelling to provoke a mental image of actual readers; or perhaps the deep-rooted, subconscious audience knowledge that journalists rely on is simply slow to evolve. All of these are worthy of careful consideration and further study.


Our findings suggest that encouraging informed audience thinking seems in some ways to have a “last-mile” problem. While “the audience” has dominated newsroom conversations in recent years, our findings indicate that these discussions have yet to significantly affect the audience perceptions of those tasked with reporting and constructing stories.

One reason may be that engaging with—and learning from—strangers is fundamentally difficult, no matter what tools or affordances exist. It may naturally be easier to relate to those with whom we feel a personal connection. Or perhaps newsroom and personal relationships hold some tangible professional or social value to the journalist that interactions with distant readers do not.

It is striking how much personal proximity — actual human contact — influences audience perceptions. In both the print and digital eras, the most vivid audience images are often based on people who the journalist has met in person (such as colleagues, peers, sources, and friends) whereas invented audiences, such as personas, are frequently ignored. Given that a writer’s sense of audience is based more on imagined reactions rather than particular demographic characteristics, it makes sense that interactions with “real people” would be the most likely to linger in one’s mind.

The particular nature of a “beat”, with its relative independence from editors and (often) physical distance away from the newsroom, might contribute to our respondents’ indifferent attitudes towards analytics and metrics. This echoes findings from scholars like Mel Bunce, whose study of foreign correspondents in Africa found that distant postings “insulated” foreign bureaus from the pressures of audience data.[231] “I don’t have to think too much about what readers are thinking about,” noted one correspondent. “I find the important things here and pitch those.”[232] Another said candidy: “I’m here, and they are 1000-2000 miles away. So they assume that I know better most of the time, about what the story is.”[233]

But another reason may be that even when analytics tools are available, they do not yet provide deeply resonant audience insights—or at least insights that can be used to inform a mental perception of “real people.”[234]

Analytics tools communicate through charts, graphs and numbers, which some journalists can find intimidating and confusing. A 2001 study found that some journalists saw numbers as a source of influence and power (“we’re not afraid of numbers … we use them to fudge our way around things”[235]). But many suffered from “math anxiety” — a sometimes paralytic fear of numbers[236] that left sufferers too nervous to ask simple questions “without feeling like an ‘idiot’.”[237] Many of these journalists “had chosen a career that they mistakenly thought would allow them to avoid mathematics.”[238]

Moreover, these analytics tools focus almost exclusively on user behavior, rather than intent; revealing little about the emotional underpinnings of engagement that inform the imagined response. Perhaps most importantly, most audience thinking still seems to be unconscious, embodied in powerful print-era conventions. “I’m too busy to think about my audience” was a common refrain amongst the journalists we spoke with, but even this should not be interpreted as a wholesale rejection of the reader’s importance. Rather, it is an admission that expertise depends on deep-rooted, unconscious knowledge, and that a journalist seeking to serve an audience must slowly accumulate and apply insights about those readers until they become second nature. New “gut instincts” may simply take time to evolve.


Of course, depending on existing sources of audience knowledge is only a problem if a journalist’s perceived audiences do not align with their actual audiences — meaning their work is not reaching or resonating with the readers they ostensibly serve.

Absent an analytical framework that could quantify one’s actual and potential audience, there is no way to know this for sure. But this gap can be inferred. For instance, over half of the reporters we spoke to cited parents as a key audience segment — a group numbering well over a million people — and yet the readership for any of their stories was just a small fraction of that total.

If, for whatever reason, a large percentage of public school parents are not being reached, the implications are worrying. Already, there is evidence of fewer resources invested in education coverage, particularly by local publications.[239] Cutbacks in the editorial ranks[240] mean a lack of expertise to guide younger reporters through complicated subjects.[241]

These trends directly affect coverage of important issues. One of our respondents worried that “granular coverage of boroughs and of really local issues that are meaningful to the people … plus the issues that deeply matter to parents, are less and less going to get covered.” And a recent survey revealed that one in four education reporters worry about the scant attention paid to inequality: “by far the most undercovered issue in education.”[242]

A number of difficult questions follow: Who are those underserved readers? What news needs do they have? What audience knowledge might an education reporter need to reach them? And how might they adapt their work accordingly?


What are some specific steps newsrooms might take to improve the accuracy of their journalists’ imagined audiences?

New Approaches

First, journalists must be able to recognize their own assumptions, with the hopes of eventually enunciating which readers they are trying to serve. This may not be as simple as asking reporters what audiences they are trying to reach; this was tried by one organization in our study, to little effect. But since imagined audiences flow in large part from one’s own sense of value, mere knowledge will likely not be sufficient to change journalists’ audience perceptions.

Post-publication evaluation of journalists’ work should include more than just tallies of how many readers did this, and how many did that. Newsrooms should be able to quantify journalists’ intended audiences, and determine whether their work is actually reaching those readers. Otherwise it will be impossible to evaluate whether the choices they make are the right ones.

Already, we are starting to see a change in the landscape: as business models shift from advertising to subscriptions, newsroom analytics are starting to move beyond raw pageview counts to more advanced user segmentation. Why not build out user segments for particular lines of reporting as well? At the very least, such efforts will help journalists understand the scale of one’s audience — providing a basis for measuring success in terms of audience growth, rather than just article-level scorekeeping.

Even better, newsrooms could find ways to blend these quantitative measurements with qualitative insights, providing trusted attitudinal feedback from real readers at a newsroom’s pace.

A reconsideration of roles might also be in order. Reporters sometimes feel that audience thinking is not their job, deferring to the expertise of editors and audience specialists. But as we have seen, reporters (especially those on a beat) have significant autonomy over many journalistic decisions — from choosing what to cover to actually crafting the story. Why not empower (or even expect) them to bring actual audience knowledge to bear on these choices, rather than imagined assumptions based on the conventions of the past?

Like an orator who learns and adapts from listeners’ cues, audience knowledge is best revealed by iteration, not just experimentation. It is not enough to just try new things; having the ability to observe and evaluate what worked is critical — what resonated with one’s readers, and why.

Three particular projects from the New York Times illustrate how journalists might involve audiences as they embark on a line of reporting. Elisabeth Rosenthal’s 2014 New York Times article on the high cost of healthcare solicited reader feedback; subsequent stories in the series drew on the experiences of those responding to the callout as sources.[243] ( In 2016, Deborah Acosta of the New York Times invited Facebook Live viewers to participate as she investigated a mysterious trove of old slides found in a garbage bag.[244],[245] ( More recently, the tech columnist Farhad Manjoo announced that he would have weekly phone conversations with regular readers, since “even opinion columnists get sick of their own opinions.”[246] ( There are surely a host of other examples from other news organizations as well.

Ultimately, however, this knowledge will be impressionistic, rather than deterministic. The goal is not to spend one’s time pandering to readers, but to slowly incorporate these actual responses into one’s imagined audiences, thus improving one’s “gut feeling” to account for new audiences and new forms of storytelling.


Finally, if perceived audiences are largely based on one’s peers, colleagues, and social circle, encouraging newsroom diversity is critical in order to broaden that circle of trusted voices to include people outside the industry’s most common demographic groups. Not only do people of different backgrounds bring varied experiences (and those of their own trusted circles) to bear on their own work, their presence would seem to directly open their colleagues’ minds to new perspectives—and new audiences—in a direct and tangible way.

However, studies have shown diversity alone is not enough. “A diverse newsroom does not always equal better coverage of minorities,” asserts one paper, even though “the public and journalists view newsroom diversity as a good thing.”[247] Our study indicates one reason why newsroom diversity itself may be insufficient: Uprooting deep-seated habits and preconceptions also requires a conscious reconsideration of one’s own audience perceptions.

Diversity of sources is also important, especially on beats, where one’s sources inevitably serve as audience proxies. This is not easy, as stereotypes can be self-reinforcing: journalists can be defensive about confronting their own biases, often believing that their industry’s diversity efforts are more effective than they actually are.[248] “[If] a source does not fit a preconceived mold,” Scott Martindale notes, “a journalist will completely overlook the source.”[249] Quoting a former science writer, he notes that “even when a writer tries to break the stereotypes, editors have an idea in their heads of what a scientist or doctor looks like.”[250]

(Discouragingly, the 2016 State of the Education Beat reported that education journalists were most likely to get story ideas from “public relations efforts”; far fewer got inspiration from parents and students.[251])

It may well be that promoting diversity in the newsroom is as much about encouraging greater self-awareness as it is about counting sources, compiling lists of experts, or enforcing quotas.[252] Those efforts will have limited effect until they shift one’s intuitive, subconscious sense of their readers.

But there are significant obstacles to expanding audience perceptions. Human nature, existing conventions and power structures, and ingrained habits all skew imagined audiences in significant and fundamental ways. To overcome this will require a deeper understanding of what it means to make the unconscious audience “real”—to actively confront, challenge, and develop the audience in the mind’s eye.


7. Appendix


It is not easy to tease out the audience images in a writer’s mind. In researching this paper, we came across a variety of methods and approaches for doing so; we mention them here as a reference and guide for those pursuing similar research.

In their 1959 study, Pool and Shulman interviewed reporters immediately after they finished writing a story. The researcher would “go through the text paragraph by paragraph, asking the reporter to recall in detail all the persons who at the time of writing had come to his mind.” (They note that “the absence of a time gap is crucial to the success of the interviewing technique.”) While recollections varied, “enough material was gathered to enable us to sense certain recurrent patterns.”[253]

They also designed a structured experiment to tease out audience images in a controlled research setting, using a questionnaire instead of an interviewer. “The first question asked the subject to list all the persons who had happened to come to mind while he was writing the story … everybody he thought of, not just persons relevant to the story.” On average, seven people came to mind.[254]

Pool and Shulman admit the difficulties inherent in this methodology. “Except on the psychiatrist’s couch, the flow of mental images has not been extensively used in research,” they write. “We have no illusions that the reports of free associations which we received are either very reliable or complete.” Nevertheless, “at least a portion of the image flow can be recaptured even on a questionnaire, and a sufficient portion to give useful results on at least some topics.”[255]

In her study of how audience awareness manifests itself in the writing process, Carol Berkenkotter employed a slightly different research method. Instead of waiting until her subjects have finished writing, she asked them to “think aloud as they composed. ” This allowed her to “keep track of when and how frequently about audience entered each writer’s mind, and to what extent related considerations guided their rhetorical, organizational, and stylistic decisions.” She asserts that this could be “the best research tool for teasing out the cognitive processes that reveal themselves in what we call audience awareness.”[256]

It’s worth noting that Berkenkotter’s methodology was subsequently challenged, partly because the time allotted (an hour) was not enough for many participants to actually produce a draft.[257] (A subject in one of her later studies said of the pressure to write something meaningful in an hour: “I would have blown up Manhattan to get out of that room.”[258] But he did admit that “I was far more aware of audience than I thought I was during some of the writing… my sense of audience is so strong that I have to suppress my conscious awareness of audience to hear what the text demands.”[259])

Deborah L. Asher, who criticized this “thinking aloud” process, also cautioned that “a writer’s awareness of audience may lie below a conscious level, such that it is not directly verbalized.”[260] She suggested that innate knowledge could instead “be elicited by probing questions in a discourse-based interview or a careful analysis of a written text and its revisions.”[261]

Nevertheless, it seems that any methodology must rely on a degree of introspection and self-awareness. As Ruth Flegel and Steven Chaffee point out, “No one can learn much about what reporters do so long as they cling to a methodological assumption that they are nothing more than units of analysis to be observed from a distance.” They argue that “many of the most interesting intellectual processes can only be worked into empirical theory by asking the person who has done something what he thought about as he did it.”[262]

The best research advice we heard came from Herbert Gans, who had written in Deciding What’s News that “interviewing people is never as productive as watching what they do.”[263] We met with him years later, while researching this paper, and asked if he had any particular tips for getting journalists to speak candidly. “[While] doing field work,” he said matter-of-factly, “I discovered that some of the most important decisions were made in the men’s room.”[264]

9. Acknowledgements

First and foremost, I am grateful for the close collaboration in the research phase of this project with Efrat Nechushtai, a PhD candidate at Columbia University Graduate School of Journalism. Efrat’s smart thinking, keen perspective and diligent assistance helped lay a robust foundation for this project’s success.

My enduring thanks to Emily Bell, Founding Director of the Tow Center for Digital Journalism, for offering me a fellowship to pursue this research, and of course to the John S. and James L. Knight Foundation for their generous support of this and other important projects at Tow.

Claire Wardle believed in this project from the beginning; her guidance in its early stages was invaluable. Elizabeth Hansen shared thoughtful advice throughout, making time for my existential crises whenever needed. Michael Shapiro, as always, was a terrific colleague, full of encouragement through many challenges. Sam Thielman and Pete Brown did yeoman’s work in helping to push the finished manuscript across the finish line.

I am indebted to my many colleagues at The New York Times, particularly Shane Murray. They have not only given me a front-row seat at journalism’s digital transformation over the past fourteen years but also offered me the support and flexibility to simultaneously pursue this research project in addition to my day job.

Nearly every person we invited to participate in the research study took time out of their busy days to talk with us. We are lucky that they did so, with such candor; I found our conversations fascinating and illuminating.

This project introduced me to many amazing people in both journalism and academia, each of whom took the time to share their expertise in some small way. Every one of their contributions—be it an incisive observation, a sensible suggestion, or an interesting recommendation—was extremely helpful. I hope our conversations continue.

Writing this would have taken another three years without the magnificent solitude offered by the Brooklyn Writers Space and the selfless generosity of Alex Wright and Maaike Bouwmeester.

Thank you, all.

This paper is dedicated to my extraordinary wife Tali, whose love, patience and support is immeasurable, and our three amazing sons, all of whom I love dearly.


8. Bibliography

To make it easier for readers to find relevant works for specific areas of interest, this bibliography is presented in four parts, roughly correlating to the major themes of this paper.


Alamargot, Denis, Gilles Caporossi, David Chesnet, and Christine Ros. “What Makes a Skilled Writer? Working Memory and Audience Awareness during Text Composition.” Learning and Individual Differences 21, no. 5 (October 2011): 505–16. DOI link.

Asher, Deborah L. “Response to Carol Berkenkotter, ‘Understanding a Writer’s Awareness of Audience.’” College Composition and Communication 34, no. 2 (1983): 214–16. DOI link.

Atiyah, Michael. “Benjamin Franklin and the Edinburgh Enlightenment.” Proceedings of the American Philosophical Society 150, no. 4 (2006): 591–606.

“Audience, n.” In OED Online. Oxford University Press. Accessed November 4, 2018. OED.

Berkenkotter, Carol. “Understanding a Writer’s Awareness of Audience.” College Composition and Communication 32, no. 4 (1981): 388–99. DOI link.

Berkenkotter, Carol, and Donald M. Murray. “Decisions and Revisions: The Planning Strategies of a Publishing Writer, and Response of a Laboratory Rat: Or, Being Protocoled.” College Composition and Communication 34, no. 2 (1983): 156–72. DOI link.

Butsch, Richard. The Citizen Audience : Crowds, Publics, and Individuals. New York: Routledge, 2008.

Campbell, George. The Philosophy of Rhetoric [Electronic Resource]. By George Campbell, … In Two Volumes. .. London: printed for W. Strahan; and T. Cadell; and W. Creech at Edinburgh, 1776.

Carvalho, J.B. “Developing Audience Awareness in Writing.” Journal of Research in Reading 25, no. 3 (October 2002): 271–82.

Cohen, Moshe, and Margaret Riel. “The Effect of Distant Audiences on Students’ Writing.” American Educational Research Journal 26, no. 2 (1989): 143–59. DOI link.

Ede, Lisa. “Audience: An Introduction to Research.” College Composition and Communication; Urbana, Ill. 35, no. 2 (May 1, 1984): 140.

Ede, Lisa, and Andrea Lunsford. “Audience Addressed/Audience Invoked: The Role of Audience in Composition Theory and Pedagogy.” College Composition and Communication 35, no. 2 (1984): 155–71. DOI link.

Ede, Lisa S. “On Audience and Composition.” College Composition and Communication 30, no. 3 (1979): 291–95. DOI link.

Elbow, Peter. “Closing My Eyes as I Speak: An Argument for Ignoring Audience.” College English 49, no. 1 (1987): 50–69. DOI link.

Flower, Linda, and John R. Hayes. “A Cognitive Process Theory of Writing.” College Composition and Communication 32, no. 4 (1981): 365–87. DOI link.

“The Cognition of Discovery: Defining a Rhetorical Problem.” College Composition and Communication 31, no. 1 (1980): 21–32. DOI link.

Gibson, Walker. “Authors, Speakers, Readers, and Mock Readers.” College English 11, no. 5 (1950): 265–69. DOI link.

Isaacson, Walter. A Benjamin Franklin Reader. Simon and Schuster, 2005.

Kirsch, Adam, and Ayana Mathis. “Does a True Artist Care What His Audience Thinks?” The New York Times, June 23, 2015, sec. Book Review. link.

Kroll, Barry M. “Cognitive Egocentrism and the Problem of Audience Awareness in Written Discourse.” Research in the Teaching of English 12, no. 3 (1978): 269–81.

Lawrence, Joshua Fahey, Melissa Niiya, and March Warschauer. “Narrative Writing in Digital Formats: Interpreting the Impact of Audience.” Psychology of Language and Communication 19, no. 3 (December 1, 2015): 201–21. DOI link.

LeFevre, Karen Burke. Invention as a Social Act. Southern Illinois University Press, 1986.

Lippmann, Walter. Public Opinion. New York: Macmillan, 1960.

Long, Russell C. “Writer-Audience Relationships: Analysis or Invention?” College Composition and Communication 31, no. 2 (1980): 221–26. DOI link.

Lunsford, Andrea A., and Lisa Ede. “Representing Audience: ‘Successful’ Discourse and Disciplinary Critique.” College Composition and Communication 47, no. 2 (1996): 167–79. DOI link.

Magnifico, Alecia Marie. “Writing for Whom? Cognition, Motivation, and a Writer’s Audience.” Educational Psychologist 45, no. 3 (July 2010): 167–84. DOI link.

Minot, Walter S. “Response to Russell C. Long, ‘Writer-Audience Relationships: Analysis or Invention?’” College Composition and Communication 32, no. 3 (1981): 335–37. DOI link.

Mitchell, Ruth, and Mary Taylor. “The Integrating Perspective: An Audience-Response Model for Writing.” College English 41, no. 3 (1979): 247–71. DOI link.

Nelson, Nancy, and Robert C. Calfee. The Reading-Writing Connection. NSSE, 1998.

Ong, Walter J. “The Writer’s Audience Is Always a Fiction.” PMLA 90, no. 1 (January 1975): 9. DOI link.

Park, Douglas B. “Analyzing Audiences.” College Composition and Communication 37, no. 4 (1986): 478–88. DOI link.

“The Meanings of ‘Audience.’” College English 44, no. 3 (1982): 247–57. DOI link.

Pfister, Fred R., and Joanne F. Petrick. “A Heuristic Model for Creating a Writer’s Audience.” College Composition and Communication 31, no. 2 (1980): 213–20. DOI link.

Redd-Boyd, Theresa M., and Wayne H. Slater. “The Effects of Audience Specification on Undergraduates’ Attitudes, Strategies, and Writing.” Research in the Teaching of English 23, no. 1 (1989): 77–108.

Roth, Robert G. “The Evolving Audience: Alternatives to Audience Accommodation.” College Composition and Communication 38, no. 1 (1987): 47–55. DOI link.

Walzer, Arthur E. “Articles from the ‘California Divorce Project’: A Case Study of the Concept of Audience.” College Composition and Communication 36, no. 2 (1985): 150–59. DOI link.

Wong, Albert T. Y. “Writers’ Mental Representations of the Intended Audience and of the Rhetorical Purpose for Writing and the Strategies That They Employed When They Composed.” System 33, no. 1 (March 1, 2005): 29–47. DOI link.

Zimmerman, Claire, and Raymond A. Bauer. “The Effect of an Audience Upon What Is Remembered.” The Public Opinion Quarterly 20, no. 1 (1956): 238–48.



Alamargot, Denis, Gilles Caporossi, David Chesnet, and Christine Ros. “What Makes a Skilled Writer? Working Memory and Audience Awareness during Text Composition.” Learning and Individual Differences 21, no. 5 (October 2011): 505–16. DOI link.

Asher, Deborah L. “Response to Carol Berkenkotter, ‘Understanding a Writer’s Awareness of Audience.’” College Composition and Communication 34, no. 2 (1983): 214–16. DOI link.

Atiyah, Michael. “Benjamin Franklin and the Edinburgh Enlightenment.” Proceedings of the American Philosophical Society 150, no. 4 (2006): 591–606.

“Audience, n.” In OED Online. Oxford University Press. Accessed November 4, 2018.

Berkenkotter, Carol. “Understanding a Writer’s Awareness of Audience.” College Composition and Communication 32, no. 4 (1981): 388–99. DOI link.

Berkenkotter, Carol, and Donald M. Murray. “Decisions and Revisions: The Planning Strategies of a Publishing Writer, and Response of a Laboratory Rat: Or, Being Protocoled.” College Composition and Communication 34, no. 2 (1983): 156–72. DOI link.

Butsch, Richard. The Citizen Audience : Crowds, Publics, and Individuals. New York: Routledge, 2008.

Campbell, George. The Philosophy of Rhetoric [Electronic Resource]. By George Campbell, … In Two Volumes. .. London: printed for W. Strahan; and T. Cadell; and W. Creech at Edinburgh, 1776.

Carvalho, J.B. “Developing Audience Awareness in Writing.” Journal of Research in Reading 25, no. 3 (October 2002): 271–82.

Cohen, Moshe, and Margaret Riel. “The Effect of Distant Audiences on Students’ Writing.” American Educational Research Journal 26, no. 2 (1989): 143–59. DOI link.

Ede, Lisa. “Audience: An Introduction to Research.” College Composition and Communication; Urbana, Ill. 35, no. 2 (May 1, 1984): 140.

Ede, Lisa, and Andrea Lunsford. “Audience Addressed/Audience Invoked: The Role of Audience in Composition Theory and Pedagogy.” College Composition and Communication 35, no. 2 (1984): 155–71. DOI link.

Ede, Lisa S. “On Audience and Composition.” College Composition and Communication 30, no. 3 (1979): 291–95. DOI link.

Elbow, Peter. “Closing My Eyes as I Speak: An Argument for Ignoring Audience.” College English 49, no. 1 (1987): 50–69. DOI link.

Flower, Linda, and John R. Hayes. “A Cognitive Process Theory of Writing.” College Composition and Communication 32, no. 4 (1981): 365–87. DOI link.

“The Cognition of Discovery: Defining a Rhetorical Problem.” College Composition and Communication 31, no. 1 (1980): 21–32. DOI link.

Gibson, Walker. “Authors, Speakers, Readers, and Mock Readers.” College English 11, no. 5 (1950): 265–69. DOI link.

Isaacson, Walter. A Benjamin Franklin Reader. Simon and Schuster, 2005.

Kirsch, Adam, and Ayana Mathis. “Does a True Artist Care What His Audience Thinks?” The New York Times, June 23, 2015, sec. Book Review. link.

Kroll, Barry M. “Cognitive Egocentrism and the Problem of Audience Awareness in Written Discourse.” Research in the Teaching of English 12, no. 3 (1978): 269–81.

Lawrence, Joshua Fahey, Melissa Niiya, and March Warschauer. “Narrative Writing in Digital Formats: Interpreting the Impact of Audience.” Psychology of Language and Communication 19, no. 3 (December 1, 2015): 201–21. DOI link.

LeFevre, Karen Burke. Invention as a Social Act. Southern Illinois University Press, 1986.

Lippmann, Walter. Public Opinion. New York: Macmillan, 1960.

Long, Russell C. “Writer-Audience Relationships: Analysis or Invention?” College Composition and Communication 31, no. 2 (1980): 221–26. DOI link.

Lunsford, Andrea A., and Lisa Ede. “Representing Audience: ‘Successful’ Discourse and Disciplinary Critique.” College Composition and Communication 47, no. 2 (1996): 167–79. DOI link.

Magnifico, Alecia Marie. “Writing for Whom? Cognition, Motivation, and a Writer’s Audience.” Educational Psychologist 45, no. 3 (July 2010): 167–84. DOI link.

Minot, Walter S. “Response to Russell C. Long, ‘Writer-Audience Relationships: Analysis or Invention?’” College Composition and Communication 32, no. 3 (1981): 335–37. DOI link.

Mitchell, Ruth, and Mary Taylor. “The Integrating Perspective: An Audience-Response Model for Writing.” College English 41, no. 3 (1979): 247–71. DOI link.

Nelson, Nancy, and Robert C. Calfee. The Reading-Writing Connection. NSSE, 1998.

Ong, Walter J. “The Writer’s Audience Is Always a Fiction.” PMLA 90, no. 1 (January 1975): 9. DOI link.

Park, Douglas B. “Analyzing Audiences.” College Composition and Communication 37, no. 4 (1986): 478–88. DOI link.

“The Meanings of ‘Audience.’” College English 44, no. 3 (1982): 247–57. DOI link.

Pfister, Fred R., and Joanne F. Petrick. “A Heuristic Model for Creating a Writer’s Audience.” College Composition and Communication 31, no. 2 (1980): 213–20. DOI link.

Redd-Boyd, Theresa M., and Wayne H. Slater. “The Effects of Audience Specification on Undergraduates’ Attitudes, Strategies, and Writing.” Research in the Teaching of English 23, no. 1 (1989): 77–108.

Roth, Robert G. “The Evolving Audience: Alternatives to Audience Accommodation.” College Composition and Communication 38, no. 1 (1987): 47–55. DOI link.

Walzer, Arthur E. “Articles from the ‘California Divorce Project’: A Case Study of the Concept of Audience.” College Composition and Communication 36, no. 2 (1985): 150–59. DOI link.

Wong, Albert T. Y. “Writers’ Mental Representations of the Intended Audience and of the Rhetorical Purpose for Writing and the Strategies That They Employed When They Composed.” System 33, no. 1 (March 1, 2005): 29–47. DOI link.

Zimmerman, Claire, and Raymond A. Bauer. “The Effect of an Audience Upon What Is Remembered.” The Public Opinion Quarterly 20, no. 1 (1956): 238–48.



“A Snapshot of Student Demographics.” Accessed June 2, 2019. link.

Acosta, Deborah. “Fragments of a Life: A Curbside Mystery.” The New York Times. Accessed June 6, 2019. link.

Anderson, C. W. “Deliberative, Agonistic, and Algorithmic Audiences: Journalism’s Vision of Its Public in an Age of Audience Transparency.” International Journal of Communication 5, no. 0 (March 4, 2011): 19.

Brake, David Russell. “Who Do They Think They’re Talking To? Framings of the Audience by Social Media Users.” International Journal of Communication (19328036) 6 (January 2012): 1056–76.

Brants, Kees, and Yael de Haan. “Taking the Public Seriously: Three Models of Responsiveness in Media and Journalism.” Media, Culture & Society 32, no. 3 (May 2010): 411–28. DOI link.

Brown, Pete. “RIP Facebook Live: As Subsidies End, so Does Publisher Participation.” Columbia Journalism Review. Accessed February 17, 2019. link.

Bruun, Hanne. “Conceptualizations of the Audience in Political Talk Show Production.” European Journal of Communication 29, no. 1 (February 1, 2014): 3–16. DOI link.

Bunce, Mel. “Management and Resistance in the Digital Newsroom.” Journalism, February 3, 2017, 1464884916688963. DOI link.

Cassidy, William P. “Outside Influences: Extramedia Forces and the Newsworthiness Conceptions of Online Newspaper Journalists.” First Monday 13, no. 1 (January 13, 2008). link.

Chaplin, Heather. “Guide to Journalism and Design.” Columbia Journalism Review, 2016. link.

Coleman, Renita, Joon Yea Lee, Carolyn Yaschur, Aimee Pavia Meader, and Kathleen McElroy. “Why Be a Journalist? US Students’ Motivations and Role Conceptions in the New Age of Journalism.” Journalism 19, no. 6 (June 1, 2018): 800–819. DOI link.

Domingo, David. “Interactivity in the Daily Routines of Online Newsrooms: Dealing with an Uncomfortable Myth.” Journal of Computer-Mediated Communication 13, no. 3 (n.d.): 680–704. DOI link.

Duffy, Andrew, and Shrutika Mangharam. “Imaginary Travellers: Identity Conceptualisations of the Audience among Travel Journalists.” Journalism: Theory, Practice & Criticism 18, no. 8 (September 2017): 1030–48. DOI link.

Ferrer-Conill, Raul, and Edson C. Tandoc Jr. “The Audience-Oriented Editor.” Digital Journalism 6, no. 4 (April 21, 2018): 436–53. DOI link.

Ferrucci, Patrick. “Exploring Public Service Journalism: Digitally Native News Nonprofits and Engagement.” Journalism and Mass Communication Quarterly; Thousand Oaks 94, no. 1 (March 2017): 355–70. DOI.

“Technology Allows Audience Role in News Construction.” Newspaper Research Journal 38, no. 1 (March 1, 2017): 79–89. DOI Link.

Gans, Herbert J. Interview, February 7, 2017.

Gold, Hadas. “HuffPost Goes out in Search of Middle America.” POLITICO. Accessed October 16, 2018.

Hare, Kristen. “How The New York Times Solved a Mystery with the Help of Facebook Live.” Poynter (blog), July 15, 2016.

Hazard Owen, Laura. “Two Out Of Two News Organizations Recommend User Research.” Nieman Lab (blog), 2015. hlink.

Heikkila, Heikki, and Laura Ahva. “The Mass, The Audience, and The Public: Questioning Pre-Conceptions of News Audiences. In Glowacki, Michal & Jackson, Lizzie (Eds.) Public Media Management for the Twenty-First Century: Creativity, Innovation, and Interaction. (Routledge).” Accessed November 4, 2018.

Heikkilä, Heikki, and Risto Kunelius. “Journalists Imagining the European Public Sphere: Professional Discourses About the Eu News Practices in Ten Countries.” Javnost – The Public 13, no. 4 (January 2006): 63–79. DOI link.

Heinonen, Ari. “The Journalist’s Relationship with Users.” In Participatory Journalism, 34–55. John Wiley & Sons, Ltd, 2011. DOI link.

Heise, Nele, Wiebke Loosen, Julius Reimer, and Jan-Hinrik Schmidt. “Including the Audience: Comparing the Attitudes and Expectations of Journalists and Users towards Participation in German TV News Journalism.” Journalism Studies 15, no. 4 (July 4, 2014): 411–30. DOI link.

Hinnant, Amanda, María E. Len-Ríos, and Hyun Jee Oh. “Are Health Journalists’ Practices Tied to Their Perceptions of Audience? An Attribution and Expectancy-Value Approach.” Health Communication 27, no. 3 (April 1, 2012): 234–43. DOI link.

Lapowsky, Issie. “Twitter Users Are Richer and More Woke Than the Rest of Us.” Wired, April 24, 2019.

Lewis, Peter H. “The New York Times Introduces a Web Site.” The New York Times, January 22, 1996, sec. Business Day. link.

Lewis, Seth C., Avery E. Holton, and Mark Coddington. “Reciprocal Journalism.” Journalism Practice 8, no. 2 (March 4, 2014): 229–41. DOI link.

Litt, Eden. “Knock, Knock . Who’s There? The Imagined Audience.” Journal of Broadcasting & Electronic Media 56, no. 3 (July 2012): 330–45. DOI link.

Litt, Eden, and Eszter Hargittai. “The Imagined Audience on Social Network Sites.” Social Media + Society 2, no. 1 (February 1, 2016).

Loosen, Wiebke, and Jan-Hinrik Schmidt. “(Re-)Discovering the Audience.” Information, Communication & Society 15, no. 6 (August 1, 2012): 867–87. DOI link.

Lowrey, Wilson, and Chang Wan Woo. “The News Organization in Uncertain Times: Business or Institution?” Journalism and Mass Communication Quarterly; Thousand Oaks 87, no. 1 (Spring 2010): 41–61.

Manjoo, Farhad. “Opinion | Office Hours With Farhad Manjoo.” The New York Times, May 17, 2019, sec. Opinion. link.

Martindale, Scott. “Source Diversity in Journalism.” M.A., University of Southern California, 2006. link.

Marwick, Alice E., and danah boyd. “I Tweet Honestly, I Tweet Passionately: Twitter Users, Context Collapse, and the Imagined Audience.” New Media & Society 13, no. 1 (February 1, 2011): 114–33. DOI link.

Matthews, Julian. “A Missing Link?” Journalism Practice 2, no. 2 (June 1, 2008): 264–79. DOI link.

McCluskey, Michael, and Jay Hmielowski. “Opinion Expression during Social Conflict: Comparing Online Reader Comments and Letters to the Editor.” Journalism 13, no. 3 (April 1, 2012): 303–19. DOI link.

McDevitt, Michael, and Patrick Ferrucci. “Populism, Journalism, and the Limits of Reflexivity.” Journalism Studies 19, no. 4 (March 12, 2018): 512–26. DOI link.

McKenzie, Carly T., Wilson Lowrey, Hal Hays, Jee Young Chung, and Chang Wan Woo. “Listening to News Audiences: The Impact of Community Structure and Economic Factors.” Mass Communication and Society 14, no. 3 (May 1, 2011): 375–95. DOI link.

Meijer, Irene Costera, and Hildebrand P. Bijleveld. “Valuable Journalism.” Journalism Studies 17, no. 7 (October 2, 2016): 827–39. DOI link.

Min, Incheol. “Perceptions of the Audience by the Alternative Press Producers: A Case Study of the Texas Observer.” Media, Culture & Society 26, no. 3 (May 1, 2004): 450–58. DOI link.

Møller Hartley, Jannie. “The Online Journalist Between Ideals and Audiences: Towards a (More) Audience-Driven and Source-Detached Journalism?” Journalism Practice 7, no. 5 (October 2013): 572–87. DOI link.

Nadler, Anthony M. Making the News Popular : Mobilizing U.S. News Audiences. Urbana: University of Illinois Press, 2016.

Napoli, Philip M. Audience Evolution: New Technologies and the Transformation of Media Audiences. Columbia University Press, 2010.

Nelson, Jacob L. “And Deliver Us to Segmentation: The Growing Appeal of the Niche News Audience.” Journalism Practice 12, no. 2 (February 7, 2018): 204–19. DOI link.

Nishikawa, Katsuo A., Terri L. Towner, Rosalee A. Clawson, and Eric N. Waltenburg. “Interviewing the Interviewers: Journalistic Norms and Racial Diversity in the Newsroom.” Howard Journal of Communications 20, no. 3 (July 23, 2009): 242–59. DOI link.

Pearce, Matt. “My Current Understanding of the Situation Is That People Have a Lot of Opinions about Journalism They Didn’t Read.” Tweet. @mattdpearce (blog), November 10, 2016. link.

Petrilli, Michael. “Disappearing Ink.” Education Next; Cambridge 9, no. 4 (Fall 2009). “How Twitter Users Compare to the General Public,” April 24, 2019. link.

Rosenthal, Elisabeth. “Paying Till It Hurts.” The New York Times, April 18, 2014, sec. Health. link.

Sherman, Gabriel. “‘No One Is in Charge’: Inside Trump’s New Fox Takeover.” The Hive. Accessed May 10, 2019. link.

Sherwood, Merryn, Angela Osborne, Matthew Nicholson, and Emma Sherry. “Newswork, News Values, and Audience Considerations: Factors That Facilitate Media Coverage of Women’s Sports.” Communication & Sport 5, no. 6 (December 1, 2017): 647–68. DOI link.

Spayd, Liz. “Facebook Live: Too Much, Too Soon.” The New York Times, December 21, 2017, sec. Public Editor. link.

State of the Education Beat 2016: A Field With A Future. Education Writers Association, 2016.

Tu, Duy Linh. Letter to James G. Robinson, October 7, 2016.

Usher, Nikki, Jesse Holcomb, and Justin Littman. “Twitter Makes It Worse: Political Journalists, Gendered Echo Chambers, and the Amplification of Gender Bias.” The International Journal of Press/Politics 23, no. 3 (July 2018): 324–44. DOI link.

Wang, Qun. “Dimensional Field Theory.” Digital Journalism 6, no. 4 (April 21, 2018): 472–91. DOI link.

“Washington Post Ad Media Kit.” Washington Post Ad Media Kit. Accessed June 4, 2019. link.

West, Darrell M., Grover J. Whitehurst, and E. J. Dionne. Americans Want More Coverage of Teacher Performance and Student Achievement. Brookings Institution, 2011.

Invisible: 1.4 Percent Coverage for Education Is Not Enough. Governance Studies at Brookings. Brookings Institution Press, 2009.

Whitehurst, Grover J. “Russ,” E. J. Dionne Jr, and Darrell M. West. “Re-Imagining Education Journalism.” Brookings (blog), November 30, 2001. link.



Anderson, CW. “Between Creative and Quantified Audiences: Web Metrics and Changing Patterns of Newswork in Local US Newsrooms.” Journalism 12, no. 5 (July 1, 2011): 550–66. DOI link.

Belair-Gagnon, Valerie, and Avery E. Holton. “Boundary Work, Interloper Media, And Analytics In Newsrooms: An Analysis of the Roles of Web Analytics Companies in News Production.” Digital Journalism 6, no. 4 (April 21, 2018): 492–508. DOI link.

Bright, Jonathan, and Tom Nicholls. “The Life and Death of Political News: Measuring the Impact of the Audience Agenda Using Online Data.” Social Science Computer Review 32, no. 2 (April 2014): 170–81. DOI link.

Bunce, Mel. “Africa in the Click Stream: Audience Metrics and Foreign Correspondents.” African Journalism Studies 36, no. 4 (2015): 12–29. DOI link.

Carlson, Matt. “Confronting Measurable Journalism.” Digital Journalism 6, no. 4 (April 21, 2018): 406–17. DOI link.

Carr, David. “Risks Abound as Reporters Play in Traffic.” The New York Times, December 20, 2017, sec. Business. link.

Cherubini, Federica, and Rasmus Kleis Nielsen. “Editorial Analytics: How News Media Are Developing and Using Audience Data and Metrics.” SSRN Scholarly Paper. Rochester, NY: Social Science Research Network, February 23, 2016. link.

Curtin, Patricia A., and Scott R. Maier. “Numbers in the Newsroom: A Qualitative Examination of a Quantitative Challenge.” Journalism & Mass Communication Quarterly 78, no. 4 (December 2001): 720–38. DOI link.

Ferrer-Conill, Raul. “Quantifying Journalism? A Study on the Use of Data and Gamification to Motivate Journalists.” Television & New Media 18, no. 8 (December 1, 2017): 706–20. DOI link.

Hanusch, Folker. “Web Analytics and the Functional Differentiation of Journalism Cultures: Individual, Organizational and Platform-Specific Influences on Newswork.” Information, Communication & Society 20, no. 10 (October 3, 2017): 1571–86. DOI link.

Kormelink, Tim Groot, and Irene Costera Meijer. “What Clicks Actually Mean: Exploring Digital News User Practices.” Journalism 19, no. 5 (May 1, 2018): 668–83. DOI link.

Lee, Angela M., Seth C. Lewis, and Matthew Powers. “Audience Clicks and News Placement: A Study of Time-Lagged Influence in Online Journalism.” Communication Research 41, no. 4 (June 1, 2014): 505–30. DOI link.

MacGregor, Phil. “Tracking the Online Audience.” Journalism Studies 8, no. 2 (April 1, 2007): 280–98. DOI link.

Mencher, Melvin. “Young Journalists Are Terrified by Numbers.” Editor & Publisher; New York 128, no. 47 (November 25, 1995): 48.

Neheli, Nicole Blanchett. “News by Numbers.” Digital Journalism 6, no. 8 (September 14, 2018): 1041–51. DOI link.

Nelson, Jacob L. “The Elusive Engagement Metric.” Digital Journalism 6, no. 4 (April 21, 2018): 528–44. DOI link.

Petre, Caitlin. “Engineering Consent.” Digital Journalism 6, no. 4 (April 21, 2018): 509–27. DOI link.

“The Traffic Factories: Metrics at Chartbeat, Gawker Media, and The New York Times.” Columbia University, 2015. DOI link.

Powers, Elia. “Selecting Metrics, Reflecting Norms.” Digital Journalism 6, no. 4 (April 21, 2018): 454–71. DOI link.

Tandoc, Edson C. “Journalism Is Twerking? How Web Analytics Is Changing the Process of Gatekeeping.” New Media & Society 16, no. 4 (June 1, 2014): 559–75. DOI link.

Tandoc, Edson C., and Ryan J. Thomas. “The Ethics of Web Analytics.” Digital Journalism 3, no. 2 (March 4, 2015): 243–58. DOI link.

Tandoc Jr, Edson C. “Why Web Analytics Click.” Journalism Studies 16, no. 6 (November 2, 2015): 782–99. DOI link.

Vu, Hong Tien. “The Online Audience as Gatekeeper: The Influence of Reader Metrics on News Editorial Selection.” Journalism 15, no. 8 (November 1, 2014): 1094–1110. DOI link.

Zamith, Rodrigo. “Quantified Audiences in News Production.” Digital Journalism 6, no. 4 (April 21, 2018): 418–35. DOI link.


[1] Walter Lippmann, Public Opinion (New York: Macmillan, 1960), 8.

[2] Matt Pearce, “My Current Understanding of the Situation Is That People Have a Lot of Opinions about Journalism They Didn’t Read.,” Tweet, @mattdpearce (blog), November 10, 2016.

[3] Peter H. Lewis, “The New York Times Introduces a Web Site,” The New York Times, January 22, 1996, sec. Business Day.

[4] Douglas B. Park, “The Meanings of ‘Audience,’” College English 44, no. 3 (1982): 248.

[5] Nancy Nelson and Robert C. Calfee, The Reading-Writing Connection (NSSE, 1998), 53.

[6] Walter Isaacson, A Benjamin Franklin Reader (Simon and Schuster, 2005), 196 (emphasis mine).

[7] “Audience, n.,” in OED Online (Oxford University Press), accessed November 4, 2018,

[8] Lippmann, Public Opinion, 16.

[9] Walter J. Ong, “The Writer’s Audience Is Always a Fiction,” PMLA 90, no. 1 (January 1975): 9.

[10] Napoli points out that most forms of media in the modern era, rather than incorporating this interactivity, instead are based on a “one-to-many, top-down audience-content provider dynamic”. He cites Butsch 2008 to note that this has caused a “tendency towards passivity”, such as audience behavior at live performances: “for instance, yelling and throwing objects at theater performers and fellow theater-goers has largely gone out of style.” Philip M. Napoli, Audience Evolution: New Technologies and the Transformation of Media Audiences (Columbia University Press, 2010), 13.

[11] See Alice E. Marwick and danah boyd, “I Tweet Honestly, I Tweet Passionately: Twitter Users, Context Collapse, and the Imagined Audience,” New Media & Society 13, no. 1 (February 1, 2011): 114–33.

[12] See, for example, the 2017 Washington Post Media Kit: “Washington Post Ad Media Kit,” Washington Post Ad Media Kit, accessed June 4, 2019.

[13] George Campbell, The Philosophy of Rhetoric [Electronic Resource]. By George Campbell, … In Two Volumes. .. (London: printed for W. Strahan; and T. Cadell; and W. Creech at Edinburgh, 1776), 242.

[14] Denis Alamargot et al., “What Makes a Skilled Writer? Working Memory and Audience Awareness during Text Composition,” Learning and Individual Differences 21, no. 5 (October 2011): 505.

[15] Carol Berkenkotter, “Understanding a Writer’s Awareness of Audience,” College Composition and Communication 32, no. 4 (1981): 388; Linda Flower and John R. Hayes, “A Cognitive Process Theory of Writing,” College Composition and Communication 32, no. 4 (1981): 27.

[16] Berkenkotter, “Understanding a Writer’s Awareness of Audience,” 388–89.

[17] Berkenkotter, 396.

[18] See Moshe Cohen and Margaret Riel, “The Effect of Distant Audiences on Students’ Writing,” American Educational Research Journal 26, no. 2 (1989): 143–59.

[19] Albert T. Y. Wong, “Writers’ Mental Representations of the Intended Audience and of the Rhetorical Purpose for Writing and the Strategies That They Employed When They Composed,” System 33, no. 1 (March 1, 2005): 37.

[20] Also see descriptions of changing audiences in Robert G. Roth, “The Evolving Audience: Alternatives to Audience Accommodation,” College Composition and Communication 38, no. 1 (1987): 48, 49.

[21] Nelson and Calfee, The Reading-Writing Connection, 58.

[22] Lisa Ede and Andrea Lusford draw this distinction as “the audience addressed” vs “the audience invoked”; see Lisa Ede and Andrea Lunsford, “Audience Addressed/Audience Invoked: The Role of Audience in Composition Theory and Pedagogy,” College Composition and Communication 35, no. 2 (1984): 155–71.

[23] Walker Gibson, “Authors, Speakers, Readers, and Mock Readers,” College English 11, no. 5 (1950): 268.

[24] JB Carvalho, “Developing Audience Awareness in Writing,” Journal of Research in Reading 25, no. 3 (October 2002): 271.

[25] See the Appendix for a discussion of how researchers try to get inside the writer’s mind.

[26] See, for example, the discussion of long-term memory in Flower and Hayes, “A Cognitive Process Theory of Writing,” 369.

[27] Nelson and Calfee, The Reading-Writing Connection, 65; summarizing Karen Burke LeFevre, Invention as a Social Act (Southern Illinois University Press, 1986),

[28] Park, “The Meanings of ‘Audience,’” 252.

[29] E-mail: Duy Linh Tu to James G. Robinson, October 7, 2016.

[30] Dwight DeWerth-Pallmeyer, The Audience in the News, LEA’s Communication Series (Mahwah, N.J: Erlbaum, 1997), xi.

[31] Renita Coleman et al., “Why Be a Journalist? US Students’ Motivations and Role Conceptions in the New Age of Journalism,” Journalism 19, no. 6 (June 1, 2018): 812.

[32] Phyllis Kaniss, Making Local News (Chicago: Univ. of Chicago Press, 1991), 86, quoted in DeWerth-Pallmeyer, The Audience in the News, 61.

[33] DeWerth-Pallmeyer, The Audience in the News, 81.

[34] DeWerth-Pallmeyer, 77.

[35] DeWerth-Pallmeyer, 81.

[36] Ari Heinonen, “The Journalist’s Relationship with Users,” in Participatory Journalism (John Wiley & Sons, Ltd, 2011), 40.

[37] Ithiel de Sola Pool and Irwin Shulman, “Newsmen’s Fantasies, Audiences, and Newswriting,” The Public Opinion Quarterly 23, no. 2 (1959): 147.

[38] DeWerth-Pallmeyer, The Audience in the News, 71.

[39] DeWerth-Pallmeyer, 62.

[40] DeWerth-Pallmeyer, 68.

[41] Hanne Bruun, “Conceptualizations of the Audience in Political Talk Show Production,” European Journal of Communication 29, no. 1 (February 1, 2014): 8.

[42] DeWerth-Pallmeyer, The Audience in the News, 60.

[43] DeWerth-Pallmeyer, 82.

[44] DeWerth-Pallmeyer, 70.

[45] Craig Allen, “Discovering ‘Joe Six Pack’ Content in Television News: The Hidden History of Audience Research, News Consultants, and the Warner Class Model,” Journal of Broadcasting & Electronic Media 49, no. 4 (December 1, 2005): 363.

[46] Allen, 364.

[47] Allen, 374.

[48] Allen, 377.

[49] Allen,374.

[50] DeWerth-Pallmeyer, The Audience in the News, 100.

[51] DeWerth-Pallmeyer, 100.

[52] DeWerth-Pallmeyer, 95.

[53] DeWerth-Pallmeyer, 98.

[54] DeWerth-Pallmeyer, 48.

[55] Herbert J. Gans, Deciding What’s News: A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time, 2 edition (Evanston, Ill: Northwestern University Press, 2005), 232.

[56] Gans, 232.

[57] Gans, 232.

[58] Ironically, Gans also pointed out that many business-side executives had the same fear, also believing that “their trained intuition is superior to research.” Gans, 233.

[59] Gans, 232–33.

[60] Gans, 234.

[61] DeWerth-Pallmeyer, The Audience in the News, 89.

[62] Lippmann, Public Opinion, 25.

[63] Aurora Wallace, Media Capital: Architecture and Communications in New York City (Urbana, Chicago: University of Illinois Press, 2012), 92.

[64] Note academic distinctions between similar terms in Napoli 2011, 2-3, who defines “institutionalized audience” as “the audience as socially constructed by media industries, advertisers, and associated audience measurement firms.” But this use is specifically intended to mean the aspirational audience of a given publication.

[65] John Arthur Chapman, Tell It to Sweeney: The Informal History of the New York Daily News, 1st ed. (Garden City, N.Y: Doubleday, 1961), 13.

[66] Chapman, 134.

[67] “Advertisement for the New York Daily News (‘Tell It to Sweeney!’)” (Printers’ Ink, August 17, 1922), 118–19.

[68] Wallace, Media Capital, 92.

[69] Wallace, 157.

[70] James D. Squires, Read All About It! : The Corporate Takeover of America’s Newspapers (New York: Times Books, 1993), 75–76.

[71] Squires, 78.

[72] DeWerth-Pallmeyer, The Audience in the News, 95.

[73] Warren Breed, “Social Control in the Newsroom: A Functional Analysis,” Social Forces 33, no. 4 (1955): 328.

[74] Breed, 328.

[75] Gans, Deciding What’s News, 235.

[76] Breed, “Social Control in the Newsroom,” 329.

[77] DeWerth-Pallmeyer, The Audience in the News, 4.

[78] DeWerth-Pallmeyer, 31–32.

[79] DeWerth-Pallmeyer, xii.

[80] Robert Darnton, “Writing News and Telling Stories,” Daedalus 104, no. 2 (1975): 176.

[81] Randall S. Sumpter, “Daily Newspaper Editors’ Audience Construction Routines: A Case Study,” Critical Studies in Media Communication 17, no. 3 (September 1, 2000): 339.

[82] Sumpter, 339.

[83] Sumpter, 340.

[84] Sumpter, 340.

[85] Darnton, “Writing News and Telling Stories,” 180.

[86] Darnton, 185.

[87] Darnton, 183.

[88] Darnton, 184.

[89] Stephen Hess, The Washington Reporters : Newswork (Washington, D.C.: Brookings Institution, 1981), 73.

[90] Gans, Deciding What’s News, 236.

[91] Darnton, “Writing News and Telling Stories,” 185.

[92] Darnton, 185.

[93] Darnton, 185.

[94] Philip Schlesinger, Putting “Reality” Together : BBC News (London: Constable, 1978), 107, quoted in DeWerth-Pallmeyer, The Audience in the News, 2.

[95] Darnton, “Writing News and Telling Stories,” 182.

[96] Jeremy Tunstall, Journalists at Work: Specialist Correspondents: Their News Organizations, News Sources, and Competitor-Colleagues, Communication and Society (Beverly Hills, Calif: Sage Publications, 1973), 252.

[97] Darnton, “Writing News and Telling Stories,” 182.

[98] Gans, Deciding What’s News, 230.

[99] Tunstall, Journalists at Work, 253.

[100] Gans, Deciding What’s News, 268.

[101] Gans, 229.

[102] Gans, 231.

[103] Karin Wahl-Jorgensen, Journalists and the Public : Newsroom Culture, Letters to the Editor, and Democracy (Cresskill, N.J.: Hampton Press, 2007), 38.

[104] Michael McCluskey and Jay Hmielowski, “Opinion Expression during Social Conflict: Comparing Online Reader Comments and Letters to the Editor,” Journalism 13, no. 3 (April 1, 2012): 306.

[105] David Pritchard and Dan Berkowitz, “How Readers’ Letters May Influence Editors and News Emphasis: A Content Analysis of 10 Newspapers, 1948-1978,” Journalism Quarterly; Columbia 68, no. 3 (Autumn 1991): 390.

[106] Pritchard and Berkowitz, 394.

[107] Tunstall, Journalists at Work, 252.

[108] Gans, Deciding What’s News, 228.

[109] DeWerth-Pallmeyer, The Audience in the News, 91.

[110] DeWerth-Pallmeyer, 91.

[111] DeWerth-Pallmeyer, 97.

[112] DeWerth-Pallmeyer, 92.

[113] Sumpter, “Daily Newspaper Editors’ Audience Construction Routines,” 338.

[114] DeWerth-Pallmeyer, The Audience in the News, 93.

[115] DeWerth-Pallmeyer, 92.

[116] DeWerth-Pallmeyer, 47.

[117] Gans, Deciding What’s News, 234.

[118] Darnton, “Writing News and Telling Stories,” 185.

[119] Tunstall, Journalists at Work, 251.

[120] Gans, Deciding What’s News, 240.

[121] Gans, 236.

[122] Gans, 151.

[123] Byron Calame, “Turning the Tables: What the Times News Staff Thinks of You,” The New York Times, October 9, 2005, sec. Opinion.

[124] Calame.

[125] Calame.

[126] Gans, Deciding What’s News, 238.

[127] DeWerth-Pallmeyer, The Audience in the News, 94.

[128] Gabriel Sherman, “‘No One Is in Charge’: Inside Trump’s New Fox Takeover,” The Hive, accessed May 10, 2019.

[129] Darnton, “Writing News and Telling Stories,” 183.

[130] Darnton, 176.

[131] Darnton, 176.

[132] Gans, Deciding What’s News, 127.

[133] Sumpter, “Daily Newspaper Editors’ Audience Construction Routines,” 338–39.

[134] Also see Ralph K. Martin, Garrett J. O’Keefe, and Oguz B. Nayman, “Opinion Agreement and Accuracy Between Editors and Their Readers,” Journalism Quarterly, 1972, 464 who found evidence of “discernible association between the Wisconsin editors’ perception of their readers’ beliefs and the direction of the stories that appeared in their newspapers.”.

[135] Michael Schudson, The Power of News (Cambridge, Mass: Harvard University Press, 1995), 75.

[136] DeWerth-Pallmeyer, The Audience in the News, 80.

[137] Pool and Shulman, “Newsmen’s Fantasies, Audiences, and Newswriting,” 148.

[138] Walter Gieber, “Across the Desk: A Study of 16 Telegraph Editors,” Journalism Quarterly 33, no. 4 (December 1, 1956): 431.

[139] DeWerth-Pallmeyer, The Audience in the News, xiii.

[140] Darnton, “Writing News and Telling Stories,” 183.

[141] Darnton, 182.

[142] David Witwer, “The Heyday of the Labor Beat,” Labor 10, no. 2 (May 1, 2013): 21, 24.

[143] Darnton, “Writing News and Telling Stories,” 182.

[144] Mark Fishman, Manufacturing the News (Austin: University of Texas Press, 1980), 27.

[145] Fishman, 149.

[146] Fishman, 150.

[147] Fishman, 33.

[148] Fishman, 49.

[149] DeWerth-Pallmeyer, The Audience in the News, 87.

[150] Amanda Hinnant, María E. Len-Ríos, and Hyun Jee Oh, “Are Health Journalists’ Practices Tied to Their Perceptions of Audience? An Attribution and Expectancy-Value Approach,” Health Communication 27, no. 3 (April 1, 2012): 234.

[151] Hinnant, Len-Ríos, and Oh, 235.

[152] Hinnant, Len-Ríos, and Oh, 241.

[153] Ben H. Bagdikian, The Media Monopoly (Boston: Beacon Press, 1992), 110.

[154] Gans, Deciding What’s News, 248.

[155] Sumpter, “Daily Newspaper Editors’ Audience Construction Routines,” 338.

[156] Sumpter, 338–39.

[157] Squires, Read All About It!, 90–91.

[158] Squires, 94–95.

[159] Gans, Deciding What’s News, 219.

[160] Gans, 238.

[161] Tunstall, Journalists at Work, 253.

[162] K. Tim Wulfemeyer, “Perceptions of Viewer Interests by Local TV Journalists,” Journalism Quarterly 61, no. 2 (June 1, 1984): 435.

[163] Wulfemeyer, 435.

[164] James G. Webster, Ratings Analysis: The Theory and Practice of Audience Research (Mahwah, N.J.: L. Erlbaum Associates, 2006), 1, quoted in Heikki Heikkila and Laura Ahva, “The Mass, The Audience, and The Public: Questioning Pre-Conceptions of News Audiences. In Glowacki, Michal & Jackson, Lizzie (Eds.) Public Media Management for the Twenty-First Century: Creativity, Innovation, and Interaction. (Routledge),” 277, accessed November 4, 2018.

[165] Jonathan Bright and Tom Nicholls, “The Life and Death of Political News: Measuring the Impact of the Audience Agenda Using Online Data,” Social Science Computer Review 32, no. 2 (April 2014): 178.

[166] Heikki Heikkilä and Risto Kunelius, “Journalists Imagining the European Public Sphere: Professional Discourses About the Eu News Practices in Ten Countries,” Javnost – The Public 13, no. 4 (January 2006): 258.

[167] Federica Cherubini and Rasmus Kleis Nielsen, “Editorial Analytics: How News Media Are Developing and Using Audience Data and Metrics,” SSRN Scholarly Paper (Rochester, NY: Social Science Research Network, February 23, 2016), 9.

[168] See Rodrigo Zamith, “Quantified Audiences in News Production,” Digital Journalism 6, no. 4 (April 21, 2018): 421; Napoli, Audience Evolution, 54–57.

[169] CW Anderson, “Between Creative and Quantified Audiences: Web Metrics and Changing Patterns of Newswork in Local US Newsrooms,” Journalism 12, no. 5 (July 1, 2011): 558.

[170] Jacob L. Nelson, “And Deliver Us to Segmentation: The Growing Appeal of the Niche News Audience,” Journalism Practice 12, no. 2 (February 7, 2018): 210.

[171] Nelson, 210.

[172] “Beware the Twitter Echo Chamber,” Columbia Journalism Review, accessed June 1,

[173] Nikki Usher, Jesse Holcomb, and Justin Littman, “Twitter Makes It Worse: Political Journalists, Gendered Echo Chambers, and the Amplification of Gender Bias,” The International Journal of Press/Politics 23, no. 3 (July 2018): 327.

[174] Pew Research Center, “How Twitter Users Compare to the General Public,” April 24, 2019.

[175] Issie Lapowsky, “Twitter Users Are Richer and More Woke Than the Rest of Us,” Wired, April 24, 2019,

[176] Nelson, “And Deliver Us to Segmentation,” 210.

[177] Nelson, 211.

[178] Nelson, 211.

[179] Anderson, “Between Creative and Quantified Audiences,” 557.

[180] Anderson, 557.

[181] Edson C Tandoc, “Journalism Is Twerking? How Web Analytics Is Changing the Process of Gatekeeping,” New Media & Society 16, no. 4 (June 1, 2014): 569.

[182] Anderson, “Between Creative and Quantified Audiences,” 557.

[183] Kees Brants and Yael de Haan, “Taking the Public Seriously: Three Models of Responsiveness in Media and Journalism,” Media, Culture & Society 32, no. 3 (May 2010): 411.

[184] Nelson, “And Deliver Us to Segmentation,” 212.

[185] Nelson, 211.

[186] Hadas Gold, “HuffPost Goes out in Search of Middle America,” POLITICO, accessed October 16, 2018,

[187] Gold.

[188] Nelson, “And Deliver Us to Segmentation,” 213.

[189] Nelson, 214.

[190] See also the New York Times case study in Heather Chaplin, “Guide to Journalism and Design,” Columbia Journalism Review, 2016.

[191] Laura Hazard Owen, “Two Out Of Two News Organizations Recommend User Research,” Nieman Lab (blog), 2015.

[192] Hazard Owen.

[193] Qun Wang, “Dimensional Field Theory,” Digital Journalism 6, no. 4 (April 21, 2018): 484.

[194] Wilson Lowrey and Chang Wan Woo, “The News Organization in Uncertain Times: Business or Institution?,” Journalism and Mass Communication Quarterly; Thousand Oaks 87, no. 1 (Spring 2010): 53.

[195] Angela M. Lee, Seth C. Lewis, and Matthew Powers, “Audience Clicks and News Placement: A Study of Time-Lagged Influence in Online Journalism,” Communication Research 41, no. 4 (June 1, 2014): 509.

[196] Janne Kaijärvi, quoted in Heikkila and Ahva, “The Mass, The Audience, and The Public,” 260.

[197] Anderson, “Between Creative and Quantified Audiences,” 555.

[198] Adam Kirsch and Ayana Mathis, “Does a True Artist Care What His Audience Thinks?,” The New York Times, June 23, 2015, sec. Book Review.

[199] Jacob L. Nelson, “The Elusive Engagement Metric,” Digital Journalism 6, no. 4 (April 21, 2018): 536.

[200] Nicole Blanchett Neheli, “News by Numbers,” Digital Journalism 6, no. 8 (September 14, 2018): 1045.

[201] Wang, “Dimensional Field Theory,” 484.

[202] Folker Hanusch, “Web Analytics and the Functional Differentiation of Journalism Cultures: Individual, Organizational and Platform-Specific Influences on Newswork,” Information, Communication & Society 20, no. 10 (October 3, 2017): 1580.

[203] Anderson, “Between Creative and Quantified Audiences,” 557.

[204] Anderson, 559.

[205] Robert Silvey, Who’s Listening? : The Story of BBC Audience Research (London: Allen & Unwin, 1974), 185.

[206] Silvey, 185.

[207] Caitlin Petre, “Engineering Consent,” Digital Journalism 6, no. 4 (April 21, 2018): 513.

[208] Valerie Belair-Gagnon and Avery E. Holton, “Boundary Work, Interloper Media, And Analytics In Newsrooms: An Analysis of the Roles of Web Analytics Companies in News Production,” Digital Journalism 6, no. 4 (April 21, 2018): 505.

[209] Matt Carlson, “Confronting Measurable Journalism,” Digital Journalism 6, no. 4 (April 21, 2018): 412.

[210] Phil MacGregor, “Tracking the Online Audience,” Journalism Studies 8, no. 2 (April 1, 2007): 291.

[211] Hanusch, “Web Analytics and the Functional Differentiation of Journalism Cultures,” 1580.

[212] MacGregor, “Tracking the Online Audience,” 291.

[213] David Carr, “Risks Abound as Reporters Play in Traffic,” The New York Times, December 20, 2017, sec. Business.

[214] MacGregor, “Tracking the Online Audience,” 290.

[215] Belair-Gagnon and Holton, “Boundary Work, Interloper Media, And Analytics In Newsrooms,” 503.

[216] Cherubini and Nielsen, “Editorial Analytics,” 31.

[217] Cherubini and Nielsen, 29.

[218] See Raul Ferrer-Conill and Edson C. Tandoc Jr, “The Audience-Oriented Editor,” Digital Journalism 6, no. 4 (April 21, 2018): 449.

[219] Cherubini and Nielsen, “Editorial Analytics,” 20.

[220] Belair-Gagnon and Holton, “Boundary Work, Interloper Media, And Analytics In Newsrooms,” 496.

[221] In our research, we came across a number of methods for investigating the “audience in the mind’s eye”; these various approaches are detailed in the appendix.

[222] Only one publication we invited to join the study refused to talk with us; it has since folded.

[223] State of the Education Beat 2016: A Field With A Future (Education Writers Association, 2016), 9,

[224] State of the Education Beat 2016, 23.

[225] State of the Education Beat 2016 reported that 22% of education journalists were nonwhite, compared to 9% of all US journalists.

[226] Breed, “Social Control in the Newsroom,” 333.

[227] “A Snapshot of Student Demographics,” accessed June 2, 2019.

[228] State of the Education Beat 2016, 34.

[229] Peter Elbow, “Closing My Eyes as I Speak: An Argument for Ignoring Audience,” College English 49, no. 1 (1987): 61.

[230] Lippmann, Public Opinion, 95.

[231] Mel Bunce, “Africa in the Click Stream: Audience Metrics and Foreign Correspondents,” African Journalism Studies 36, no. 4 (2015): 15.

[232] Bunce, 20.

[233] Bunce, 22.

[234] Or, given understandable privacy concerns, representative archetypes.

[235] Patricia A. Curtin and Scott R. Maier, “Numbers in the Newsroom: A Qualitative Examination of a Quantitative Challenge,” Journalism & Mass Communication Quarterly 78, no. 4 (December 2001): 727.

[236] Curtin and Maier, 727.

[237] Curtin and Maier, 728.

[238] Curtin and Maier, 732.

[239] Michael Petrilli, “Disappearing Ink,” Education Next; Cambridge 9, no. 4 (Fall 2009).

[240] Grover J. “Russ” Whitehurst, E. J. Dionne Jr, and Darrell M. West, “Re-Imagining Education Journalism,” Brookings (blog), November 30, 2001, 2.

[241] State of the Education Beat 2016, 34.

[242] State of the Education Beat 2016, 18.

[243] Elisabeth Rosenthal, “Paying Till It Hurts,” The New York Times, April 18, 2014, sec. Health.

[244] Deborah Acosta, “Fragments of a Life: A Curbside Mystery,” The New York Times, accessed June 6, 2019; Kristen Hare, “How The New York Times Solved a Mystery with the Help of Facebook Live,” Poynter (blog), July 15, 2016,

[245] This approach seemed to be the exception rather than the rule. While the editor supervising the Facebook Live initiative admitted one of its goals was to “help people around the newsroom think more about audience” its participation in the program dwindled relatively quickly after Facebook stopped paying publishers to participate. See Liz Spayd, “Facebook Live: Too Much, Too Soon,” The New York Times, December 21, 2017, sec. Public Editor; Pete Brown, “RIP Facebook Live: As Subsidies End, so Does Publisher Participation,” Columbia Journalism Review, accessed February 17, 2019.

[246] Farhad Manjoo, “Opinion | Office Hours With Farhad Manjoo,” The New York Times, May 17, 2019, sec. Opinion.

[247] Katsuo A. Nishikawa et al., “Interviewing the Interviewers: Journalistic Norms and Racial Diversity in the Newsroom,” Howard Journal of Communications 20, no. 3 (July 23, 2009): 245.

[248] Scott Martindale, “Source Diversity in Journalism” (M.A., University of Southern California, 2006), 20–22.

[249] Martindale, 12.

[250] Martindale, 9.

[251] State of the Education Beat 2016, 17.

[252] Martindale offers some tips for journalists looking to improve their source diversity: see Martindale, “Source Diversity in Journalism,” 36–37.

[253] Pool and Shulman, “Newsmen’s Fantasies, Audiences, and Newswriting,” 146–47.

[254] Pool and Shulman, 151.

[255] Pool and Shulman, 158.

[256] Berkenkotter, “Understanding a Writer’s Awareness of Audience,” 389.

[257] Deborah L. Asher, “Response to Carol Berkenkotter, ‘Understanding a Writer’s Awareness of Audience,’” College Composition and Communication 34, no. 2 (1983): 214–16.

[258] Carol Berkenkotter and Donald M. Murray, “Decisions and Revisions: The Planning Strategies of a Publishing Writer, and Response of a Laboratory Rat: Or, Being Protocoled,” College Composition and Communication 34, no. 2 (1983): 169.

[259] Berkenkotter and Murray, 171.

[260] Asher, “Response to Carol Berkenkotter, ‘Understanding a Writer’s Awareness of Audience,’” 215.

[261] Asher, 215.

[262] Ruth C. Flegel and Steven H. Chaffee, “Influences of Editors, Readers, and Personal Opinions on Reporters,” Journalism Quarterly 48, no. 4 (December 1, 1971): 651.

[263] Gans, Deciding What’s News, 75.

[264] Herbert J. Gans, Interview, February 7, 2017.

James G. Robinson has spent over a decade in a variety of audience-focused roles at The New York Times, leading their web analytics team during the implementation of its digital subscription model and later pioneering the use of audience insights in the newsroom as the company’s first director of news analytics. He is also an adjunct professor and research fellow at Columbia University’s School of Journalism.

About the Tow Center

The Tow Center for Digital Journalism at Columbia's Graduate School of Journalism, a partner of CJR, is a research center exploring the ways in which technology is changing journalism, its practice and its consumption — as we seek new ways to judge the reliability, standards, and credibility of information online.

View other Tow articles »

Visit Tow Center website »