Media — To sell a crisis


Understanding the incentives and control system behind sensationalist news and misinformation

The Cambridge Analytica incident in 2015, where the data analytics firm harvested Facebook data to influence voter choices for the Brexit and 2016 Donald Trump presidential campaign[1], reaffirmed the influence of mass media and revealed some of its vulnerabilities. While prevailing discourse, as represented by Ruha Benjamin’s Race After Technology from 2019 and Meredith Broussard’s Artifical Unintelligence from 2018, explore the role of technology in contemporary media problems, Ryan Holiday’s Trust me, I’m lying – Confessions of a media manipulator considers the interlocking incentives that shape the media industry. Although this by no means suggest digital media platforms should not be held accountable for our current predicament, Holiday’s book is a sobering reminder that our condemnation of technology is its own form of technochauvanism.

Trust me is the product of Holiday’s desire to “expose the worst of the web’s marketing and publishing practices [as someone who] created and perfected many of them.”[2]. The first edition of the book was published in 2012 and a third edition, “revised and updated for the fake news era”[3], was published in 2017. Without the resources available to academics, Trust me presents an industry perspective to a mass audience that while valuable is not rigorous. His sincerity is at times undercut by the lack of reliable documentation on his claims. I remain skeptical of the reproducibility of some of his marketing tactics in part one of the book, and question whether there is a good faith effort to account for the many actors as well as variables in the events he used to illustrate the repercussions of abusing the media in part two of the book.

Nevertheless, Trust me, as an anthology of personal observations from a practitioner, serves as a good starting point towards understanding the incentives for creating sensationalist news and misinformation, as well as the antiquated control system that enables their dissemination. This essay expands upon Holiday’s notes on the industry by connecting his views to established literature.

Trust me’s most poignant takeaway is that the current business model in the media industry incentivizes online authors such as journalists and bloggers to produce sensationalist news and misinformation because web traffic directly translates into revenue: publisher’s advertising revenue from authors equals to the cumulative CPM (cost per thousand) multiplied by the number of page views[4]. This interpretation offers a more nuanced understanding of Holiday’s claim that “Advertising x Traffic = Revenue”[5], where he dismissed online publishers’ other sources of revenue. Publications nowadays derive an increasing amount of their revenue from events[6], and a few news companies such as Thomson Reuters (News Analytics) derive additional revenue by providing news data to financial companies for trading and fund management[7]. The two pieces of supplemental information, however, does not negate Holiday’s observation that “Traffic is money.”[8] The paragraph below, quoted from Trust me illustrate the pressures facing authors. It is based on back of the envelope calculations at $10 CPM (cost per thousand impressions) by Henry Blodget first published as a string of Tweets[9].

“Henry Blodget, the founder of Business Insider once explained that his writers need to generate three times the number of page views required to pay for their own salary and benefits, as well as a share of the overhead, sales, hosting, and Blodget’s cut, to be worth hiring. In other words, an employee making sixty thousand dollars a year would need to produce upward of 1.8 million page views a month, every month, or they’re out.”[10]

The paragraph below illustrates how the advertising revenue model complicates day-to-day publishing.

Gawker writer Hamilton Nolan titled ‘New Rules for Media Ethics.’ He said it plainly: ‘Media people — reporter, commentator, or otherwise – shouldn’t have a financial stake in what they’re reporting on.’ But then I realized how hypocritical it all was, since Nolan is being paid by how many views his posts do. […] In the pay-per-pageview model, every post is a conflict of interest. [emphasis added]”[11]

Holiday argues that the integrity of the author is further compromised by the type of content that becomes viral by quoting a study conducted by John Berger and Katherine Milkman in 2009. “According to the study, ‘the most powerful predictor of vitality is how much anger an article evokes’ [emphasis by Holiday]”[12].  His claim “Nothing [Berger and Milkman] say contradict their earlier point”[13], with reference to the revisions made to the article in 2012, is dubious. Firstly, the 2009 version he cites states that their findings apply to “anger or anxiety inducing articles”[14] rather than anger alone. Secondly, the 2012 version finds that “while more positive or more negative content is more viral than content that does not evoke emotion, positive content is more viral than negative content”[15], which directly contradicts his argument that authors write anger-inducing articles because they are the most viral.

Acknowledging the new findings would in fact strengthen his broader argument that the feedback mechanism in the media industry incentivizes the production of sensationalist news and misinformation. Berger and Milkman’s finding: “positive and negative emotions characterized by activation or arousal (i.e. awe, anxiety, and anger) are positively linked to virality.”[16] reveal the positive reinforcement in the form of increased traffic authors will receive when they publish sensationalized news and misinformation.

A self-interested media industry challenges the conventional notion of the media as a gatekeeper that will present issues accurately, fairly, and completely[17]. First theorized by Kurt Lewin in 1943, gatekeeping was used to describe the phenomena where not all family members have equal weight in making household food decisions[18]. David White in 1950 applied the theory to understanding newspaper publication, and was elaborated upon by Bruce Westley and Malcolm MacLean in 1953, who together introduced the notion of multiple gatekeepers[19]. Research on the selection mechanism has been furthered most notably by Shoemaker between 1996 and 2009, noting organization-level, story-level, and professional factors that contribute to biases in news selection[20], with a skew towards stories that are “more sensational, and/or conflictual, and/or geographically proximate”[21]. To study gatekeeping, researchers generally “(a) explore, through interview or surveys, the decision-making processes by journalist and editors, and (b) examine mass media content itself”[22]. Despite Trust me’s controversy and occasional lapses in reasoning, it is important to recognize that few have attempted Holiday’s ‘show me the money’ approach towards examining the incentives to produce sensationalized news and misinformation.

As Holiday points out, the most critical control system in place to stop the dissemination of misinformation was the “delegation of trust”[23]. Traditionally, publications have an implicit contract of trust extended to third party publications as a result of commonly understood editorial standards[24]. However, since “every person in the media ecosystem (with the exception of a few at the top layer) is under immense pressure to produce content under the tightest deadlines”[25], online publications have become increasingly dependent upon smaller publications with lower editorial standards.

“There are thousands of content creators scouring the web […who] must write several times each day. […] Above them are hundreds of midlevel online and offline journalists […] who use those bloggers below as sources and filters. They also have to write constantly – and engage in the same search for buzz, only a little more developed. Above them are major national websites, publications, and television stations [that compete to ‘confirm’ the story]”[26]

Holiday’s observation[27] is supported by journalist Nick Davies’ findings. Davies commissioned researchers from Cardiff University, who surveyed 2,000+ UK news stories from the four qualities dailies (Times, Telegraph, Guardian, Independent) and the Daily Mail. The study found that 80% of the stories “were wholly, mainly or partially constructed from second-hand material, provided by news agencies and by the public relations industry”, and only 12% of the facts have been fact-checked when they were reproduced[28].

The dissemination of misinformation is exacerbated by the atomization of news. The atomization of news describes the process of breaking stories into smaller parts for dissemination on online platforms, which decouples it from the source[29]. It negates attempts at iterative journalism. Any misquotes or errors will continue to propagate online even if the original author corrects the original error[30]. The short form content also poses significant challenge to unbiased authorship. Quite simply, even a comprehensive perspective, when re-shared and re-quoted in the length of a tweet, can easily be misrepresented and misconstrued[31].

The antiquated control system as well as overwhelming incentives for sensationalized news and misinformation leaves the media industry vulnerable to exploitation by bad actors such as Holiday. Media manipulators, lobbyist, state actors use qualifiers such as “‘We’re hearing…’; ‘I wonder…’; ‘Possibly…’; ‘Lots of buzz that’; ‘Chatter indicates that…’; ‘Sites are reporting…’; ‘Might…’; ‘Maybe…’”[32] to instill doubt, spread rumors, while avoiding lawsuits. In more extreme cases of coordinated PR attacks, tactics such as ‘Astroturfing’, the act of using fake accounts to propagate an opinion, and ‘shitposting’, the act of overwhelming an account with unhinged rants, compounds the misinformation problem[33].

To address the current challenges and combat the new wave of conspiracy theories being disseminated online[34], the online publishing industry require new business models[35]. What responsibility do online platforms have over our current predicament? How do we reign in malicious actors? What can a well intentioned author do to reconcile the inherent tensions between their values and the logic of the market[36]? How should consumers respond when they are being sold sensationalized crisis stories “designed to make you stop whatever you’re doing and consume because the world is supposedly ending.”[37]?

Technology will be part of the solution, but technology alone will not be sufficient.

by Marvin Cheung, Head of Research and Strategy at Unbuilt Labs


[1] Emma Graham-Harrison and Carole Cadwalladr, “Revealed: 50 Million Facebook Profiles Harvested for Cambridge Analytica in Major Data Breach,” The Guardian (Guardian News and Media, March 17, 2018), https://www.theguardian.com/news/2018/mar/17/cambridge-analytica-facebook-influence-us-election.

[2] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. XIV.

[3] Ibid. Cover copy.

[4] Ibid. Pg. 36.

[5] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 36.

[6] Innovation Media Consulting, “Events Are Becoming a Substantial Source of New Revenue,” Innovation, February 17, 2016, https://innovation.media/magazines/its-showtime-folks-and-payday.

[7] Gautam Mitra and Leela Mitra, The Handbook of News Analytics in Finance (Chichester: Wiley, 2012). Pg. 25.

[8] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 37.

[9] Henry Blodget, “More Than You Ever Wanted To Know About The Economics Of The Online News Business — A TWEETIFESTO,” Business Insider (Business Insider, March 27, 2010), https://www.businessinsider.com/henry-blodget-more-than-you-ever-wanted-to-know-about-the-economics-of-the-online-news-business-a-tweetifesto-2010-3.

[10] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 47.

[11] Ibid. Pg. 50.

[12] Ibid. Pg. 73.

[13] Ibid. Pg. 301.

[14] Berger, Jonah A. and Milkman, Katherine L., “Social Transmission, Emotion, and the Virality of Online Content,” 2009, https://www.scribd.com/document/67402512/SSRN-id1528077. Pg. 2.

[15] Berger, Jonah A. and Milkman, Katherine L., What Makes Online Content Viral? (December 25, 2009). Available at SSRN: https://ssrn.com/abstract=1528077 or http://dx.doi.org/10.2139/ssrn.1528077 Pg. 5.

[16] Ibid. Pg. 8.

[17] [Author removed at request of original publisher], “14.3 News Media and Ethics,” Understanding Media and Culture (University of Minnesota Libraries Publishing edition, 2016. This edition adapted from a work originally produced in 2010 by a publisher who has requested that it not receive attribution., March 22, 2016), https://open.lib.umn.edu/mediaandculture/chapter/14-3-news-media-and-ethics/.

[18] Ghulam Shabir et al., “Process of Gate Keeping in Media: From Old Trend to New,” Mediterranean Journal of Social Sciences, January 2015, https://doi.org/10.5901/mjss.2015.v6n1s1p588. Pg. 589.

[19] Ibid. Pg. 589.

[20] Stuart N. Soroka, “The Gatekeeping Function: Distributions of Information in Media and the Real World,” The Journal of Politics 74, no. 2 (2012): pp. 514-528, https://doi.org/10.1017/s002238161100171x. Pg. 516.

[21] Ibid. Pg. 514.

[22] Ibid. Pg. 516.

[23] The original web article from JournalismEthics.info is no longer available, and further research suggest that this is a concept more commonly used to understand the ethical risks of using anonymous sources, but I present the concept here as I believe it merits further discussion

[24] Rogério Christofeletti, “Ethical Risks, Informers, Whistleblowers, Leaks and Clamor for Transparency,” Brazilian Journalism Research 12, no. 2 (2016). Pg. 55.

[25] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 30.

[26] Ibid. Pg. 23.

[27] To support his observations, Holiday originally cited a national survey conducted by Cision and The George Washington University in 2009. Unfortunately, the link for the complete survey results is no longer active. As a result, it is unclear what the sampling method and sample size is.

[28] Nick Davies, “Nick Davies: Our Media Have Become Mass Producers of Distortion,” The Guardian (Guardian News and Media, February 4, 2008), https://www.theguardian.com/commentisfree/2008/feb/04/comment.pressandpublishing.

[29] Wilding, D., Fray, P., Molitorisz, S. & McKewon, E. 2018, The Impact of Digital Platforms on News and Journalistic Content, University of Technology Sydney, NSW. Pg. 37.

[30] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 208.

[31] Barnard, S.R. 2018, Citizens at the Gates: Twitter, Networked Publics and the Transformation of American Journalism, Palgrave Macmillan, London.

[32] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 189.

[33] Ibid. Pg. 203.

[34] Zack Stanton, “You’re Living in the Golden Age of Conspiracy Theories,” POLITICO (POLITICO, September 4, 2020), https://www.politico.com/news/magazine/2020/06/17/conspiracy-theories-pandemic-trump-2020-election-coronavirus-326530.

[35] Wilding, D., Fray, P., Molitorisz, S. & McKewon, E. 2018, The Impact of Digital Platforms on News and Journalistic Content, University of Technology Sydney, NSW. Pg. 43.

[36] Ibid. Pg. 40.

[37] Ryan Holiday, Trust Me, I’m Lying: Confessions of a Media Manipulator (New York, NY: Portfolio / Penguin, 2017), Pg. 177.