NPR’s presence on Facebook is growing. Earlier this month, our Facebook page surpassed one million fans.
We recently turned to those fans to get to know them better, to learn what types of stories they like, and to find out what they think of each other (or at least what they think of each other’s Facebook comments). We got an amazing response: over 40,000 people took part in the survey. Just goes to show how amazing our Facebook fans are. Thanks to everybody who participated!
We gathered some interesting results, which will be used by our Social Media Desk to inform and shape NPR’s future on Facebook. You can have a look at our key findings in the slides below.
Among the headlines and observations:
NPR’s fans are Facebook regulars. Almost all respondents (96%) access Facebook at least once per day, and 80% access it more than once a day. According to Facebook, 50% of its users in general are using the service on any given day. (Our result almost certainly reflects a sampling bias, as we recruited via links posted to Facebook: the result being that more frequent users were more likely to see it than others.)
They’re regular consumers of NPR content, especially via broadcast. About three of every four respondents (76%) listen to NPR on the radio. The majority of respondents - 55% - listen to between one and three hours of NPR on air each day. In comparison, the average NPR listener listens to 4.25 hours per week. Almost one in five respondents (18.6%) uses the NPR News iPhone app, while 5% use our Android app, and 2.5% our iPad app. Just under three in ten (28.5%) listen to NPR podcasts. Interestingly, fewer than 1 in 10 (8.4%) follow NPR on Twitter, which was somewhat unexpected, given how NPR actually has more than twice as many fans on Twitter than we do on Facebook. This suggests that these communities are more mutually exclusive than we assumed.
They consume lots of news online – but are more inclined to consume NPR content on Facebook than other news sources. About three of every five respondents (60.7%) said they get most or all of their news online. Almost three-quarters (74.6%) agree that Facebook is a major way in which they receive news and information from NPR. In contrast, just half of them (51%) agree that Facebook is a major way for them to receive information from news organizations in general.
Your friends are your personal news wire. Nearly three out of four (72.3%) expect their friends to share links to interesting information and news stories with them online. In contrast, a 2010 survey by the Pew Internet & American Life Project had 50% of respondents saying they rely on people around them to keep them up-to-date on news. Still a high number, apparently our Facebook users feel even more strongly about this. This potentially bodes well for services such as Paper.li, the Flipboard iPad app and TwitterTim.es, which create personalized news reading experiences based on the links your friends share. These services are popping up quickly on Twitter, and it probably won’t take very long for them to integrate links shared by your Facebook friends as well.
The vast majority NPR Facebook fans regularly read the links we post. More than four in five respondents (84%) indicated that they click through to NPR stories that are posted on Facebook. This response bears out in the actual metrics for our site. NPR receives anywhere from 1.5 to 2 million pageviews per month directly from links we post to our Facebook page.
Our Facebook fans share our content with their friends, though at different levels depending on the platform. Three in four respondents (74%) shared an NPR story with friends via Facebook in the last five months. This isn’t a huge surprise, given how sharing is standard operating procedure on Facebook. Interestingly, though, nearly half of them (49%) also said they’ve shared an NPR story via email in the last five months. A much smaller percentage of them (7.2%) have shared an NPR story via Twitter, but again, this isn’t a surprise given the relatively few respondents who said they followed NPR on Twitter in the first place.
Some of our Facebook fans are NPR.org commenters as well. Three in ten respondents (30%) have commented on a story at NPR.org in the last five months. We intend to track this statistic in the future, given how we’ve just started allowing Facebook users to sign on to NPR.org with their Facebook account. It would make sense for this number to rise over time as more NPR.org visitors discover this option.
Users don’t think the number of “likes” on a Facebook post will make them more likely to click it. A plurality of respondents (44%) say that a large number of “Likes” or Facebook comments on an NPR story does not make them more likely to read that story. In contrast, only 11.8% say that a story with lots of Likes makes it more likely they’ll read it. This response was a surprise, since conventional wisdom suggests that a high number of likes does indeed translate to more people reading the story. Perhaps the disconnect here is how people are exposed to the story in the first place. Even though these users don’t see the number of likes as important, a story with lots of likes will still appear in many more users’ feeds, simply based on how Facebook works, exposing more potential readers to it.
Our Facebook fans consider each other civil, and few of them dislike their comments. Around the office I sometimes hear concerns from staff that our Facebook users are too snarky and swear like sailors. I’ve always taken the position that Facebook is their community rather than ours, so we’ve always treaded lightly when it comes to interfering with their discussions. When we asked Facebook users to agree or disagree with a number of adjectives that could be used to describe comments, they most readily agreed with words like “lively,” “civil,” “polite” and “knowledgeable.” In contrast, they generally disagreed with adjectives such as “vulgar,” “loud,” and “irreverent.” Somewhere in the middle were “useful,” “snarky” and “reputable.” The vast majority of users either liked comments or were ambivalent about them: while 42.8% of respondents agreed they generally like the comments left by other users on NPR’s Facebook posts, 13.7% disagreed and 43.5% neither agreed nor disagreed.
NPR is Goldilocks when it comes to posting frequency. Just over seven in ten respondents (72.6%) said NPR publishes just the right amount of posts to our page each day – around 8-10 posts on average. About one in five (21.6%) said NPR should post even more stories each day. In contrast, only 6% of respondents stated NPR posted too often per day.
Of all the results in the survey, I’m the least surprised by this result. When we first started posting links to Facebook more than a year ago, we admittedly went overboard the first few days, posting more than 20 items per day. And Facebook users gave us an earful, saying it was drowning out their friends in their news feeds. We quickly cut back to a handful a day, then increased it until we reached a point where complaints of too much or too little died down to just a trickle. Meanwhile, users can adjust the number of items they see in their news feed, so we’ve managed to reach a good equilibrium.
I wouldn’t be surprised, though, if many other Facebook page publishers would consider 8-10 items way too many per day. For example, when researchers analyzed the UNICEF-USA Facebook page, they found that its sweet spot was just three items per day – and unsubscribe rates “rose dramatically” with more than three posts per day. At various social media conference I’ve heard three or four as the magic number as well.
Why is it that we can get away with more than twice as many posts per day? My guess is because not all Facebook page owners are posting fully baked original content. When we post to the NPR page, we post links to breaking news, analysis, interviews, reviews, concerts, etc. – and our users expect us to post links with some meat to them. Because we’re not posting press releases, status updates and the like, our users have a much higher threshold of acceptance for how often we should post.
They’re interested in a wide range of topics – except sports and rich people. In the survey, we listed a number of story types and asked users if NPR should post more, less or about the same amount of stories on those topics. Among those topics that topped the list were offbeat stories, hard news, breaking news, international news, interesting people and the environment. Users also expressed interest in more stories about health, music, media, the Internet/social media, politics, food and the economy, though not as strongly as they did about the aforementioned topics. In contrast, respondents wanted to see significantly fewer stories about sports and “rich, powerful or infamous people.”
Among our users, Facebook Open Graph’s “opt-out” system isn’t very popular. Several months ago, Facebook launched a service called Open Graph. The idea behind it is to make it easier for people to continue their Facebook experience across other websites. Among the features they rolled out was a “like” button you can deploy on any Web page you want. (We’re actually in the process of doing so across NPR.org this month.) In addition, though, Facebook created an Open Graph tool that lets you display on your website which pages have been liked the most – and which of your friends have liked them. All of this is displayed automatically, without asking users to opt-in to this particular feature. Many news sites, including the Washington Post, ABC News and CNN have implemented the feature on their sites. At NPR, we were more cautious, suspecting our users might be, well, creeped out by it. So we decided not to implement it until we got the results of this survey.
I expected to see a number of users opposed to seeing what their friends are sharing on NPR.org, but was taken aback by just how many there were. Three out of five respondents (62.2%) said they wouldn’t like to see their Facebook friends’ recommendations when they visit NPR.org. Having said that, of those who would, two-thirds of them (67.6%) would rather the option be opt-in rather than automatically displayed without their permission. Based on these results, though, we are now actively exploring ways to roll out an opt-in version rather than use the default opt-out method.