Digital Sustainability: Assessing U.S. Newspapers’ Online Readership with the Multidimensional Attention Model

ABSTRACT

More than 20 years into newspapers’ digital experiment, most are still struggling in search of a business model while digital revenue remains a fraction of total revenue. To examine the sustainability of digital journalism, this study assesses the top 50 U.S. newspapers’ digital readership with the Multidimensional Web Attention Model. Empirical analyses using Nielsen and Comscore data identified problems with newspapers’ online readership across multiple dimensions (reach, popularity, loyalty, depth, and stickiness). Seven-day market reach is around 13%. Popularity varies but loyalty is low across the board – an average user makes no more than three visits a month (M = 2.53). Depth and stickiness are also underwhelming, with about two pages viewed per visit (M = 2.21) and slightly more than one minute spent on a page (M = 1.18). While local papers do not benefit from the economies of scale, national newspapers, despite more resources, outperform their local counterparts only on the popularity dimension. Mobile users constitute the majority but fall short on loyalty and depth. Users aged 18–24 remain a small portion of the newspaper audience. These findings parse out the industry-wide failure to engage online readers. At the core of newspapers’ digital sustainability problem is a readership that falls short in multiple ways.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. The “newsfulness” of a device is calculated by dividing the numbers of users who use the device for news access within a particular time frame (e.g., daily or weekly) by the number of owners of that device.

2. The Pew Research Center’s 2019 State of the News Media Report included the top 49 U.S. newspapers by average Sunday circulation for Q3 from 2015 to 2018, with the addition of The Wall Street Journal, which does not publish on Sundays. See a full description of the selection criteria here: https://www.journalism.org/2019/07/23/state-of-the-news-media-methodology/

3. In the Washington, D.C. DMA.

4. Mauchly’s test of sphericity had been violated for the effect of age groups on unique visitors (χ ² (14) = 341.36, p < .001) and views per visit (χ ² (14) = 49.69, p < .001), so the degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity.

5. In contrast, these newspapers’ average print in-market reach is 26%, despite a downward trend.

6. We obtained the same Web metrics data used in our study from ComScore. During September 2019, Google had 242,994,000 unique visitors, 105 visits per visitor, 5.2 views per visit, 0.8 minutes per view. Facebook had 214,769,000 unique visitors, 23 visits per visitor, 8.4 views per visit, 2.9 minutes per view.