Facebook vs. Google, social media vs. SEO: Why BuzzFeed data shouldn't declare a winner
Last week, the latest traffic referral report from BuzzFeed caught Marshall Simmonds's eye. The data indicated Facebook delivered about 3.5 times more page views to BuzzFeed Network sites in December than Google did:
If that observation were broadly applicable to publishers across the web, it would be a game-changer. Simmonds, CEO of Define Media Group, thought it wasn't, so he posted a rebuttal responding to writers who he felt interpreted the chart too broadly.
(Peter Kafka at Re/code first posted the chart, under the headline "The Year Facebook Blew Past Google." Robinson Meyer of the Atlantic seemed similarly enamored in his interpretation: "And Just Like That, Facebook Became the Most Important Entity in Web Journalism.")
In Simmonds's post, titled "Hey BuzzFeed, Search Traffic is Doing Just Fine," he presented referral data for 48 billion page views from 87 sites across 2013 to back up his contention that Google isn't going anywhere. Data came from Omniture and Google Analytics. The findings: Search delivered 2.5 times as many page views as social (and Define added a further stipulation that Google search traffic was actually underreported for part of the year due to misreporting on mobile devices):
Is either dataset representative?
BuzzFeed described its network in November as "a collection of sites including more than 200 publishers such as The Huffington Post, TMZ, The Onion, and Slate, with more than 300 million users each month." Add BuzzFeed itself and you get a list of mostly digital properties that were at the forefront of embracing social media and whose content seems naturally shareable.
Via phone, Simmonds criticized BuzzFeed for what he sees as a lack of transparency regarding its network and methodology, and he criticized reporters for failing to evaluate its numbers critically or put them in context.
BuzzFeed makes that hard to do. The company wouldn’t make anyone available for an interview or provide me with more information about its methodology or network members.
Meanwhile, Define wouldn’t get into specifics about the publishers it works with, either, but clients cited on its website include The New York Times, Cosmopolitan, Rolling Stone, Hearst, Conde Nast, Time, TV Guide, Gannett and AARP.
In his blog post, Simmonds said BuzzFeed's data was "obviously biased towards more socially oriented publishers." So I asked if his data might be biased in the opposite direction, toward publishers who haven't yet realized their social potential. Here's what he said in a follow-up email:
Our publishers have highly successful social media teams invested in community building and outreach. Over the past two years we’ve seen budgets for these departments growing in order to support tools and personnel and in many cases at a much higher rate than that allocated for SEO. As a result our clientele has millions of FB, Twitter, G+, Pinterest, fans/followers.
Overall, Define is in the business of “audience development” — whatever form that takes, including social, which Simmonds didn't hesitate to emphasize is a fast-growing, critical part of any publisher's audience strategy.
“You can’t take your eye off of either ball,” Simmonds said. “We would be remiss if we did not focus on social. However, we know that search is the lion’s share.”
(A recent report from Parse.ly found big gains for Facebook, but said the social media giant remains a distant second to Google in number of referrals to news websites.)
What are the takeaways for publishers?
Obscured when viewing either graphic in isolation is the fact that no two news organizations are alike, so it's not as simple as trusting one dataset over the other and using it as a definitive guide for how to00 allocate resources. Shahzad Abbas, VP of digital media at Define, said there was variance in the data, with the gap between SEO and social narrowing for some categories of content, like video and celebrity news.
"It’s not as if we didn’t see in some of the new media categories that social was maybe ahead of search or equal to search,” he said.
A newer strategy's steep growth curve doesn't mean an old stalwart doesn't still have something to contribute. Simmonds said the social vs. search debate isn't a zero-sum game. "Search has never really been sexy, and certainly social is the new shiny object getting all the attention," he said. "It’s getting the budget but it’s not necessarily getting the results."
It's getting the results for some, but it's worth noting that what Facebook — and Google, for that matter — giveth, it can taketh away. Upworthy, the Facebook flavor of the month in November, has seen a big traffic dip since Facebook's latest algorithm tweaking meant to better emphasize "high-quality content."
(Will Oremus of Slate notes that Upworthy's drop could be attributed to regression from a one-time spike or to changing its editorial goals. Meanwhile, it's not uncommon to see speculation that Facebook plays favorites with certain brands like BuzzFeed, which just keeps growing and loves to talk about Facebook as the reason for the growth.)
Still, assuming no fishy business in Facebook's News Feed algorithm, content can only go as viral as, well, its content allows. That's the crucial, obvious point when determining whether social media or SEO should get more dollars and focus at your news organization: Which strategy makes most sense for my unique brand of content?
So, maybe the Re/code headline should have said, “The Year Facebook Blew Past Google for BuzzFeed and Other Sites It Won’t Identify." And maybe the Define headline should have said, “Hey BuzzFeed, Search Traffic is Doing Just Fine for Our Clients, Whom We Can’t Identify Either."
But surely no social or SEO expert would recommend headlines like that.