Unlocking Popu­la­rity: Deco­ding the Guar­di­an’s Game-Chan­ging Approach to Enga­ge­ment Evolu­tion


AI provided the title of this article, although with some guidance from the author, to attract the reader’s attention.

Being popular is now “being noticed by many”, right? (Hecken 2006: 85). The popularity of a song, social media account, art piece, consumer product, politician, academic journal, or journalistic piece is based on the attention that people pay. But here is the question: what’s popular? Really, what is the most attention-grabbing, or let’s put it this way, what is the most engaging? We are in a period in which popularity is measured more rapidly, simultaneously and quantitatively than ever before, and according to these measurements, which are also in competition for popularity, the popular are becoming more popular (Werber et al. 2023). Seeing that popularity can be measured more quickly and distinctly, those seeking popularity are now trying to master not only the measures of popularization but also the techniques. While it is perhaps easier to conclude that songs that are popular on Spotify are the most listened to, does the fact that a news item is very popular really mean that it is read or that it is of real interest? The Guardian’s new ranking, which was shared with its readers at the end of February, is an exercise that will deepen the reflection on popularity, first in journalism and then in all industries where popularity is a factor.

At the end of February, the Guardian introduced a new feature called the “Deeply read” list to differentiate between the most engaged and the most viewed news (2024). Traditionally, the Guardian, like many other sites, has displayed a “Most viewed” list, which tends to highlight popular articles. However, the Guardian has informed their audience on their explain page, that they have long been concerned with understanding how readers engage with its journalism. While they agree that clicks on an article indicate its potential importance or popularity, these lists may overlook excellent journalism on less mainstream topics. Basically, the “Deeply read” list, on the other hand, prioritizes articles based on the amount of time readers spend engaging with them.

The most striking point in Ümit Alan’s article on the subject was that these lists were quite different, almost completely different (2023) Following Alan’s observation, as a reader and academic, I have been regularly comparing the lists for a week and have observed that, in most cases, the most viewed news items and the most deeply read items are completely different. Only a few times, I have observed a maximum of one or two identical stories appearing in both the most viewed and most deeply read lists. It is clear from the words of Chris Moran, the Guardian’s head of editorial innovation, in his interview with Owen in NiemanLab, that the Guardian’s editors had recognised this difference for a long time and therefore felt the need to share it with the audience. “We’ve had this for a number of years internally to help us see less reach-y pieces that really work with a smaller audience,” he added. “And for many years I’ve wanted to share it with readers because it highlights such great journalism and little off the beaten track of trending topics. To be clear it still matters to show people what is popular, but we love showing them something more.”

Before we share our insights on how this new metric and other changes may affect the way news and popularity are viewed, let’s take a look at how this metric works. As we predicted, the popularity metric for news sites is simply the number of clicks on a story, but other factors need to be taken into account to measure the reader’s attention to read it carefully: but a long story and a short story will, of course, have different read times. In Ophan (The Guardian’s internal analytics system), the so-called “attention metric” assesses active reading time, taking into account the length of the article. Moran summarises the workings of the metric as follows: a news story is assigned a score ranging from one to five hours, with five indicating excellent engagement and one indicating sub-par engagement relative to the length of the article.

The Guardian’s new metric has a lot to say about both journalism and popularity. While subscriptions and advertising still play an important role in financing journalism, news publishers have come to rely more on subscription revenues than advertising (Zandt 2022). News websites have therefore become more concerned with metrics and engagement rates at the expense of the wider ‘public interest’ dimension of journalism (Rourke 2019). On the one hand, this may help them to offer a wider range of journalism that resonates more deeply with readers and emphasises quality journalism, as in the case of the Guardian, but on the other hand, this increasingly widespread subscription system may also lead to a ‘news divide’ between the audience that prefers and can access quality news and the audience that does not emphasise and/or cannot afford it. This news fragmentation may deepen the polarisation in public opinion and create a more fragmented public opinion. As we escape from the filter bubbles and echo chambers imposed by social media algorithms, we may find ourselves in our own class-based news bubbles where everyone is personalised according to their financial situation on their own popular platform for economic reasons.

At this point, it would be appropriate to take a look at developments in social media. Social media companies, which are living their golden days with the pandemic, seem to be the victims of their algorithms and ambitions so far, while dealing with AI. Cordilia James of the Wall Street Journal’s “We Aren’t Posting on Social Media as Much Anymore. Will We Ever?” (unless you are a WSJ subscriber, in which case you will be able to access part of this story), a survey conducted by Gartner in the US found that more than half of respondents believe the quality of social media has declined over the past five years due to misinformation, toxicity and the rise of bots. Gartner predicts that in the next two years, nearly 50 per cent of users will reduce or completely abandon their interactions with social media. Users’ declining trust in social media brands is leading to a less satisfying experience and a reluctance to share personal views and opinions. Moreover, intrusive adverts and suggested posts contribute to user dissatisfaction. Likewise, in a report published by Morning Consult in October 2023, 61% of US adults with social media accounts admitted to being more cautious about what they share. Reasons for this selectivity range from feeling overwhelmed by uncontrollable content to a desire to protect their privacy to reduced enjoyment of social media. “The algorithmic spotlight on creators and their hyper-curated content has made some users feel insecure and less likely to share their own photos and videos”, says Kevin Tran, media and entertainment analyst at Morning Consult. This so-called “lurking mentality” is said to extend to almost all social media platforms, including Instagram, Facebook, TikTok and others. This suggests that social media users are increasingly behaving like traditional media viewers and leaving less and less data. This data has, of course, galvanised social media companies, and they have already begun efforts to create new spaces that they claim are more secure, where privacy is more user-controlled, and which direct users to their immediate communities. For these reasons, by 2024, social media platforms are expected to become more decentralised and fragmented, with smaller-scale communities and less interconnectivity (Zuckerman 2024).

These trends may actually be a boon for journalism and for unpopular values that have failed to attract attention, as popularity and popular metrics may make the popular more popular and give those who are lost amidst so much data and noise a chance to stand out. Natalya Anteleva and Ümit Alan describe this insidious but dominant problem as noise. Social media’s vast, multi-content platforms and view-based popularity algorithms have so far created noise in journalism and information gathering that is more hidden than disinformation and the production and dissemination of fake news, but which hinders truth and good journalism. The truth is still there in the noise, but it is increasingly difficult to hear the voice of truth. Anteleva describes this situation from a journalistic point of view:

This deafening noise of the information ecosystem, I believe, is a huge reason why journalism — and by journalism I mean high quality insightful, inclusive reporting — lost its role as a curator of a public conversation. It’s the element of the disinformation crisis we, journalists, should have been combatting all along. Noise is the new censorship. It is what builds the walls that split millions of communities around the world, polarizing our societies beyond recognition.

And, of course, around the corner is a revolution that can provide all these trends and corresponding solutions, and that will profoundly affect popularity and popularisation techniques, and that is artificial intelligence. Ümit Alan says that in a media ecology where artificial intelligence is involved in the production and distribution of content, popularity by clicks is no longer sustainable because now artificial intelligence will compete with clickbait headlines that succeed in attracting attention and will be trained at this point and will be able to make this tiring race more equal (2023). It can also better regulate curation and offer much better practices in personalisation, which could be very important contributions at a time when social media is fragmenting, and news sites are moving towards engaged audiences. Of course, as we have seen with the recent controversy over Princess Kate’s photo, which has become the talk of the world, there are significant challenges ahead for journalism. The increasing realism, accessibility, and dissemination of AI images and deepfakes will have a deep impact not only on journalists but also on social media and popularity.

The image was created by DALL-E and Duygu Karatas

So, what do all these discussions and trends tell us about popularity and its constant transformation? Is it maybe not as difficult to stand out in the noise as it used to be? What is popular is likely to become more important on which social platform and for whom. With artificial intelligence, advertisers and those who want to be popular will now be able to identify their target audience more easily and easily predict who will act and buy more (Kaput 2024). The Guardian’s sharing of the “deeply read” list is important in terms of popularity in every field because advertisers and sponsors can now question how much their ads are seen by those who browse, but for effective interaction and return ads, it will be likely to turn to those with high engagement rather than those viewed on that page or content. Would you invest in a popular for one-second views of millions or an engager (a new word perhaps between French and English) that tens of thousands spend minutes? Maybe we will have different definitions like niche popular … Here, of course, who succeeds in attracting limited attention and how will be important. In these days of social media algorithms, I think we will need to look deeper into popularity and attention in relation to practices and developments. As popularisation transforms, let’s see how our popular will transform.


Alan, Ü. (2024): “The Guardian’ın yeni Listesi Yeni Medyanın Geleceğiyle ilgili Değerli Bir şey söylüyor”, in: 10Haber, 3 Mar. 2024. URL: https://10haber.net/yazarlar/umit-alan/the-guardianin-yeni-listesi-yeni-medyanin-gelecegiyle-ilgili-degerli-bir-sey-soyluyor-374048/ (Accessed March 14, 2024).

Alan, Ü. (2023): “Yeni medyayi ve hayati degistirecek asil buyuk yenilik ne olacak?”, in: 10Haber, 10 Dec. 2023. URL: https://10haber.net/yazarlar/umit-alan/yeni-medyayi-ve-hayati-degistirecek-asil-buyuk-yenilik-ne-olacak-310963/ (Accessed March 14, 2024).

Antelava, N. (2023): “Noise is the new censorship: How journalism got disinformation wrong”, in: JSK Fellows, 30 Nov. 2023. URL: https://jskfellows.stanford.edu/noise-is-the-new-censorship-b64b8c50e7e8 (Accessed March 14, 2024).

Hazard Owen, L. (2024): “The Guardian’s new ‘Deeply Read’ article ranking focuses on attention, not just clicks”, in: Nieman Lab, 28 Feb. 2024. URL: https://www.niemanlab.org/2024/02/the-guardians-new-deeply-read-article-ranking-focuses-on-attention-not-just-clicks/ (Accessed March 14, 2024).

Hecken, Th. (2006): Populäre Kultur: Mit einem Anhang ‘Girl und Popkultur’. Bochum.

James, C. (2024): “We Aren’t Posting on Social Media as Much Anymore. Will We Ever?”, in: The Wall Street Journal, 23 Dec. 2023. URL: https://www.wsj.com/tech/personal-tech/social-media-nobody-posting-f6c2fd3e?mod=tech_feat4_personal-tech_pos4 (Accessed March 14, 2024).

Kaput, M. (2024): “What Is Artificial Intelligence for Social Media?”, in: Marketing Artificial Intelligence Institute, 22 Jan. 2024. URL: https://www.marketingaiinstitute.com/blog/what-is-artificial-intelligence-for-social-media (Accessed March 14, 2024).

The Guardian (2024): “What is the ‘Deeply read’ list?”, in: The Guardian, 28 Feb. 2024. URL: https://www.theguardian.com/info/2024/feb/28/what-is-the-deeply-read-list (Accessed March 14, 2024).

Werber, N., D. Stein, J. Döring, V. Albrecht-Birkner, C. Gerlitz, T. Hecken, J. Paßmann, J. Schäfer, C Schubert and J. Venus (2023): “Getting Noticed by Many: On the Transformations of the Popular”, in: Arts 12, 39. DOI: https://doi.org/10.3390/arts12010039.

Zuckerman, E. (2024): “Social media is getting smaller-and more treacherous”, in: Wired, 24 Jan. 2024. URL: https://www.wired.com/story/social-media-is-getting-smaller-and-more-treacherous/ (Accessed March 14, 2024).