The mobile news audience is huge. Its habits are… complicated.


The mobile news audience has exploded over the past two years, giving news organizations ample reason to point their resources towards their mobile platforms. 72% of Americans now get news on a mobile device, up from 54% in 2013, and more than half of those who get news from both desktop and mobile devices prefer mobile. This sudden increase has raised concerns for journalism researchers and industry stakeholders. Mobile platforms appear to encourage what scholars refer to as news “snacking” or “grazing”, defined as “shorter, dispersed consumption patterns”. Scholars argue this method of news consumption is negatively associated with levels of public knowledge and civic engagement. Furthermore, news publishers may forgo in-depth investigative reporting for shorter stories more likely to appeal to audiences using this method of news consumption.

Yet prior studies of mobile news consumption have been limited by their reliance on surveys in which respondents provide self-reports of their media use rather than observed data. While surveys are better for exploring audience motivations that underlie media exposure than other research methods, they are less ideal for uncovering media exposure to begin with. And the seemingly endless expansion of media platforms and options is likely to only exacerbate this difficulty in accurately recalling media use. These studies also treat the mobile news audience as homogenous, when there are in fact two mobile audiences: those who consume news via the mobile browser and those who prefer mobiles news apps.

Our study, recently published in Digital Journalism, addresses these gaps by using data from a media panel that tracks multiple metrics across various platforms. These data allowed us us to examine what portion of the digital news audience uses mobile devices as compared to desktop computers, as well as what portion of the audience consumes news via mobile browsers as compared to news apps – an increasingly important distinction.


We analyzed data from comScore, an online audience measurement company used by top media properties in the U.S. comScore has two distinct panels – one that records online activity via desktop computers, and one that records online activity via mobile devices. The same panelists are tracked over time, rather than a different random sampling every month. The desktop panel comprises about one million people ages two and older, who load comScore tracking software on their desktop devices. That software tracks the URLs that the user visits and the time that they spend looking at each address. 

The mobile panel includes about 20,000 people ages 18 and older across iPhones, iPads, and Android devices. Similar to the desktop panel, these participants download software onto their devices that track their online behavior. This software tracks visitation to any website that meets the company’s minimum reporting standard (30 unique visitors per month), as well as all app usage. comScore “deduplicates” users who visit one media source both via mobile website and mobile app. For example, if someone visits The New York Times’ homepage in their mobile browser and then loads the Times’ mobile app in a given month, that visit only counts once. comScore fuses data from both panels with server-based counts of traffic that come from tagging websites. Each month’s sample is separately weighted and projected to reflect independent census estimates of the U.S. population.

comScore divides the sites its panels visit into categories (for example, shopping, social media, news). We chose the sites in our sample using the “News” category in comScore’s Media Metrix menu. This category includes newspapers like The Washington Post and The New York Times, magazines like The Economist, television news channels like NBC, radio stations like NPR, digitally native outlets like The Huffington Post, and politically ideological sources like Mother Jones and Drudge Report. Choosing comScore’s broad “News” category allowed us to examine publishers that cover a range of topics using a variety of methods and mediums. We examined measures of audience size (unique visitors) and audience engagement (average minutes per visitor per month). 


We found that in an average month during our time period (September 2015 - 2016), the total desktop internet audience (the number of people visiting websites from their desktop computers at work or at home) comprised roughly 232 million people, and the desktop online news audience (the number of people visiting news websites specifically) comprised about 160 million people. The average monthly mobile online audience (the number of people visiting websites using mobile devices) during this period comprised about 191 million people. Of that group, 180 million visited news sites. So while, on average, the size of the desktop internet audience was larger than the mobile online audience throughout the year, more people visited news sites from mobile devices than they did from desktop computers.

Image: Desktop and mobile news audience size over time.

In contrast, the audience engagement metric indicated the opposite pattern. The average minutes per visitor to a desktop news site was about 12 minutes per month, while the average minutes per visitor to a mobile news site was about 7 minutes per month. However, this low audience engagement via mobile was the result of an overwhelmingly large mobile browser audience, which was about five times the size of the news app audience. Though much smaller, the news app audience spent an average of 138 minutes per news site per month, which about 10 times longer than the desktop audience, and roughly 20 times longer than the mobile browser audience.

Image: Desktop and mobile news audience attention over time.


Our results confirm prior studies that concluded that mobile has become the primary platform for news consumption. Yet our study goes one step further, demonstrating that, within the mobile platform, there is a subset of the news audience willing to devote enormous amounts of time to news in the form of mobile news apps. More importantly, our findings reveal that the mobile news audience does not comprise one homogenous mass, but two distinct groups – mobile news browsers and mobile app users – with different compositions and behaviors.

These findings have important implications for the future of news production. To begin with, they offer empirical evidence that news publishers and researchers should continue investing efforts into observing and understanding news consumption via mobile devices, as these are where people increasingly turn for news. They also indicate that mobile browser and mobile news app media users spend very different amounts of time when it comes to news consumption. Over the past few years, journalism publishers and stakeholders have advocated a shift away from a revenue model that privileges measures of audience size to one that privileges measures of audience attention and loyalty. This devoted news app audience suggests that a transition from an ad-supported revenue model that privileges measures of audience size to one that prioritizes other traits like loyalty and attention could very well be a beneficial one for commercial news publishers to make.

Read the full research study here.

Image: European Parliament.