Anyone who runs a website knows it: Not all visitors are the same. A loyal reader who’s bookmarked your site and returns twice a day isn’t the same as a social media flyby who didn’t even catch your site’s name on the trip back to Twitter. And a new study from the Pew Research Center makes it clear that readers who come to your site via different routes act differently once they get there, too.

“Facebook and search are critical components for any news organization’s distribution strategy in today’s society, and they give you real opportunities to reach new audiences and are critical for bringing eyeballs to any individual story,” Amy Mitchell, Pew Research Center’s director of journalism research and the study’s lead author told me. “But what these data suggest is that it’s hard to hold relationships with people who are coming through Facebook or coming through search.”

Pew found that users who arrive at the 26 news sites studied directly spend about three times as long there as opposed to users who arrive via Facebook or a search engine. (By “directly” the study means arriving by typing in the URL or going to a bookmark — or, technically, with arriving without a referer.) Direct visitors also view about five times as many pages per month and visit a news site about three times as often as users coming from Facebook or search.

“For news outlets particularly that are looking to building a loyal audience, perhaps, and gain revenue from paid subscribers, for example, figuring out how to get those social or search referrals to also come to you directly is critical to their strategy,” Mitchell said. “In this digital realm, strategy in terms of content distribution and in terms of economics may well be tailored to the various spaces within the digital realm and to the various types of audiences that are going to come into contact with your news.”

Pew compiled the study by examining three-months’ worth of comScore data for a group of 26 news sites, spanning both pre-web and online-native outlets (from NBC News to BuzzFeed) and including a number of outlets with political leanings (from The Blaze to The Huffington Post). The study was limited in its scope because it only focused on Facebook (not other social platforms like Twitter) and did not analyze mobile traffic since comScore’s mobile panel is smaller than its desktop one. does not collect as much data on mobile usage. Still, Patrick Cooper, NPR’s director of web and engagement, told Pew the desktop findings also held lessons for smartphones and tablets.

“The big things publishers should take away from the desktop data, even if desktop is going away, is that: 1) method of entry matters to the experience and 2) they can’t control method of entry,” Cooper told Pew.

(One other caveat: “Direct” traffic can mean a lot of different things — anything without a referer, really, not just bookmarks and direct typing. Links in email or chat clients often show up as “direct” in traffic logs, for instance — the looming mystery of “dark social.” Pew acknowledges the category can be fuzzy on the edges.)

Despite the limitations on the study, here are three points that stuck out:

You’re most likely to only enter a site one way

If you visit NYTimes.com directly every morning when you get to work to check the latest news, you’re likely not also clicking on that latest Times trend story on monocles in your Facebook News Feed later in the day.

Most people who visit these news sites do so only through one of the methods — directly, via Facebook or through a search engine — the Pew study found. The percentage of people who visited a site directly and through Facebook ranged from 0.9 percent to 2.3 percent, with one exception: BuzzFeed, where 11.3 percent of direct visitors also accessed the site through Facebook. And between 1.3 percent and 4.1 percent of direct users also came to a site through a search engine — the one exception being Examiner.com at 8.6 percent.

“Looked at another way, for the majority of sites studied (19 of the 26), more than three-quarters of their direct visitors only visited the site directly over the course of a month,” the study found.

Engagement vs. sharing

In a talk last month at the Nieman Foundation, BuzzFeed editor-in-chief Ben Smith emphasized the importance of sharing in BuzzFeed’s editorial strategy. “Each story has a potential audience, and if it’s a story about Ukraine or a story about lobbying in D.C., there are maybe tens of thousands or hundreds of thousands of people who might, in an ideal world, share and read that story,” Smith said. “If it’s a feature about rebuilding a house in Detroit, there may be millions. If it’s a list of cute animals or something that’s about a universal human experience, there may be tens of millions. We think: What’s the possible audience for this piece? And let’s try to hit that whole audience.”

And BuzzFeed’s penchant for sharable content is reflected in the fact that fully half of its desktop traffic comes from Facebook, while only 32 percent of desktop visitors arrive at the site directly. Though the percentage of Facebook referrals outpaces direct visits, users who navigate directly to BuzzFeed content are still more engaged. They average 5.6 minutes per visit and 17.8 pageviews per visit, compared to 2.3 minutes per visit and 2.7 pageviews per visit for Facebook referrals.

“Their Facebook engagement numbers were higher than the average and we also saw a greater percentage of their audience that was coming to their site both through direct and through a Facebook referral,” Mitchell said of BuzzFeed. “So it’s suggesting that people are finding them in both of these pathways as opposed to just one, which is what we saw for a lot of the other sites.”

Contrast that with The New York Times, which sells and increasingly relies on revenue from digital subscriptions. It wants to see higher levels of engagement to support that subscription business. In all, 38 percent of its traffic comes from direct visitors ,who spend an average of more than 9 minutes on the site per visit. A relatively minor 7 percent of its traffic comes from Facebook.

“There are only about one-third of Facebook users that follow a news organization or an individual journalist,” Mitchell said. “So those referrals are often coming from friends who are passing along links to news stories. So that might be giving you a different mix of outlets then those that you have on top of mind to turn to to find out about a particular event, or just to get your news fill for the day.”

Even direct visitors have varying degrees of engagement

While referrals from Facebook and search engines consistently have low levels of engagement across a vast majority of the sites in the study, there is a varying range of levels of engagement among direct visitors:

Take, for example, the 10 sites with the highest percentage of visitors coming directly to the site. CNN tops the list with 60% direct traffic, whose visitors view on average 13 pages per month and return 8.7 times a month. But they stay for just 1 minute and 30 seconds per visit. This length of time is essentially the same as visitors coming to CNN.com from search or Facebook, and is much less time than direct visitors spend on BuzzFeed (5 minutes and 36 seconds), NYTimes.com (9 minutes and 24 seconds) or Foxnews.com (9 minutes and 6 seconds).