May 9, 2011

Navigating News Online

Where People Go, How They Get There and What Lures Them Away

Whatever the future of journalism, much of it depends on understanding the ways that people navigate the digital news environment—the behavior of what might be called the new news consumer.

Despite the unprecedented level of data about what news people consume online and how they consume it, understanding these new metrics has often proven elusive. The statistics are complicated, sometimes contradictory, and often introduce new information whose meaning is not clear.

To shed more light on Web news behavior, the Pew Research Center’s Project for Excellence in Journalism has conducted an in-depth study of detailed audience statistics from the Nielsen Company. The study examines the top 25 news websites in popularity in the United States, delving deeply into four main areas of audience behavior: how users get to the top news sites; how long they stay during each visit; how deep they go into a site; and where they go when they leave.

Overall, the findings suggest that there is not one group of news consumers online but several, each of which behaves differently. These differences call for news organizations to develop separate strategies to serve and make money from each audience.

The findings also reveal that while search aggregators remain the most popular way users find news, the universe of referring sites is diverse. Social media is rapidly becoming a competing driver of traffic. And far from obsolete, home pages are usually the most popular page for most of the top news sites.

What users do with news content, the study also suggests, could significantly influence the economics of the news industry. Understanding not only what content users will want to consume but also what content they are likely to pass along may be a key to how stories are put together and even what stories get covered in the first place.

Among the findings:

  • Even the top brand news sites depend greatly on “casual users,” people who visit just a few times per month and spend only a few minutes at a site over that time span.  USAToday.com was typical of most of these popular news sites: 85% of its users visited USAToday.com between one and three times per month. Three quarters came only once or twice. Time spent was even more daunting: When all the visits were added together, fully a third of users, 34%, spent between one and five minutes on the paper’s Website each month.[1] Even if, as some suggest, online data tend to count some users multiple times, inflating the number of casual users and undercounting repeat visits, casual users till would be the largest single group.
  • There is, however, a smaller core of loyal and frequent visitors to news sites, who might be called “power users.” These people return more than 10 times per month to a given site and spend more than an hour there over that time. Among the top 25 sites, power users visiting at least 10 times make up an average of just 7% of total users, but that number ranged markedly, from as high as 18% (at CNN.com) to as low as 1% (at BingNews.com).
  • Even among the top nationally recognized news site brands, Google remains the primary entry point. The search engine accounts on average for 30% of the traffic to these sites.
  • Social media, however, and Facebook in particular, are emerging as a powerful news referring source. At five of the top sites, Facebook is the second or third most important driver of traffic. Twitter, on the other hand, barely registers as a referring source. In the same vein, when users leave a site, “share” tools that appear alongside most news stories rank among the most clicked-on links.
  • When it comes to the age, news consumers to the top news websites are on par with Internet users overall. This stands apart from news consumption on traditional platforms, which tends to skew older, and may bode well for the industry.

All of this suggests that news organizations might need a layered and complex strategy for serving audiences and also for monetizing them. They may need, for instance, to develop one way to serve casual users and another way for power users. They may decide it makes sense to try to convert some of those in the middle to visit more often. Or they may try to make some of their loyal audience stay longer by creating special content. Advertising may help monetize some groups, while subscriptions will work for others.  And the strategy that works best for each site may differ.

What’s more, with the development of mobile, these layers will almost certainly multiply.

The study builds off of our 2010 NetView analysis of the top roughly 200 news sites in the United States, those that, by Nielsen’s count, averaged at least 500,000 unique U.S. visitors per month. In the new report, PEJ narrowed the focus to the top 25 sites.  Despite the rapid growth of eReader and tablet devices, most online news consumption still occurs on browser. Only between 7-10% of the population currently owns a tablet or e-reader. The study, which examined nine months of consumer data spanning the first three quarters of 2010, sheds light on the significance of search aggregators and social networks, the importance of creating a family of related Websites, and hints at which kinds of sites might have more success with paywalls than others.

This report offers an analysis of data collected by one metric agency, the Nielsen Company. One of the challenges with the current state of most all internet audience data is the differences that arise from one system to another.

Server side data like that of Hitwise, for example, will report different numbers because it uses a different methodology than panel-based data. Google analytics starts from yet another different base.  And all of these are third party measurement systems are going to differ from a single sites in-house measurement system.  Data based on unique visitors will be different than data based on total traffic or total hits. Even the way website domains and sub-domains are broken up makes a difference.

The greatest value within each of these systems, then, is looking at the data relative to itself—to other findings within that same system rather than comparing figures across systems.


Footnote:

1.  Some experts believe that traditional online metrics undercount Website loyalty because “unique visitors” actually count computers, not people. So if the same user visits a site from a different computer, he or she would be counted more than once. Still, these numbers suggest that a good deal of online news consumption involves people arriving casually, often through referrals from search engines or social media. This is discussed more in the section on How Users interact with News.