Viral content secrets: the science of social sharing
It’s a sign of the times when Rupert Murdoch takes a quiz on Buzzfeed to determine which billionaire tycoon he is.
For Buzzfeed, which recently launched an Australian edition, sharing is the only metric it cares about. “Shares are so much better than clicks,” vice-president international Scott Lamb told a Walkley Foundation event in Sydney in January. “You can’t trick humans into sharing.”
The billionaire tycoon quiz was shared and liked on Facebook more than 9000 times within three days of publication, numbers that mere mortals might be delighted with but are actually only a middling result for Buzzfeed. A similar quiz that lets you discover which Spice Girl you are had more than 187,000 likes and shares on Facebook after just a day.
Sites such as Buzzfeed and Upworthy account for an outsize proportion of what gets shared on social media, but to understand why, you need to look at the reasons people share in the first place. Of course, people have always shared stories – whether gathered around the campfire or by clipping articles from the newspaper and posting them to loved ones as my grandmother does – but social media takes it to a whole new level.
Facebook chief executive Mark Zuckerberg has talked about his site, and specifically the ubiquitous news feed, as being a virtual equivalent of the “potlatch” – referring to a custom among the native tribes of the Pacific north-west of the United States. Chiefs would hold a potlatch festival where they gave their tribes massive gifts of food, skins, weapons, crafts, canoes and so on; the motivation was to enhance their social status rather than altruism.
Social philosopher Tim Rayner says the analogy is apt and competition for reputation and status is a strong driver for sharing good content on social media shared spaces, like the news feed of Facebook and the Twitter stream.
“The experience becomes like a meeting of chiefs,” Rayner says. “Everyone is trying to distinguish themself through what they put in the common space, while supporting each other through shares and likes, which are also kinds of gifts. Social media gift economies give everyone the opportunity to be a chief. This is how social media hooks us.”
Emotional arousal prompts sharing
In a recent interview with BRW, Buzzfeed’s chief revenue officer, Andy Wiedlin, says there are no tricks to make people share – a great headline might make people click, but if the content doesn’t meet expectations they won’t repost.
However, he says that insight into sharing behaviour can help you craft content.
“You have to look at why people share things,” says Wiedlin. “They share stuff because it’s funny; they share stuff because it makes them look clever; they share stuff that has emotional resonance and shows they have a heart; they share stuff like human rights that they believe in and the content says it better than they could in a Facebook post.”
Recent research suggests that the second factor – emotional resonance – is the strongest of all.
Professors at the University of Pennsylvania’s Wharton Business School studied what makes content widely shared. Marketing professor Jonah Berger and his colleague Katherine Milkman studied 7000 articles published in The New York Times in 2008 to analyse what made the most-emailed list and went on to replicate the study in the more controlled environment of a lab setting.
Their research suggests articles that do best are ones that arouse emotions, particularly positive ones. Articles that evoked some emotion did better than those that evoked none, but happy emotions outperformed sad ones. Intensity of emotion trumped everything: a story that made someone feel extremely angry or highly anxious was just as likely to be shared as a feel-good story about a cuddly panda.
In a book called Viral Marketing: The Science of Sharing, published in 2013, Dr Karen Nelson-Field describes how researchers at the University of South Australia found a similar trend with video sharing: videos that provoke an intense positive emotional response are shared more.
For branded videos, sharing was highest if it elicited sadness, followed by the positive emotions of exhilaration and inspiration. For user-generated videos, sharing was most associated with hilarity followed by anger, exhilaration and inspiration.
That does not mean there is a formula. Nelson-Field says a common mistake is to look only at the success stories, when popular videos often have a lot in common with unsuccessful ones. Devices like “dancing babies and juggling kittens” only have viral success if they also arouse emotion. It’s worth noting: videos about personal triumph or weather events did better than expected.
Sometimes people share without reading
Buzzfeed’s Wiedlin says the goal is to create “content that is so powerful that people feel compelled to share it as they build their own brands” and he lists the desire to “look clever” as one of the drivers of sharing.
But in their rush to build personal brand, it seems many people click retweet on Twitter or hit like or share on Facebook without actually reading the content.
In 2012 Dan Zarella analysed 2.7 million tweets with links and found no statistical correlation between the number of retweets and the number of clicks. In fact, 16.1 per cent of the links had more retweets than clicks and 14.6 per cent of the retweeted links had no clicks at all.
In February, Tony Haile, chief executive of web traffic measurement company Chartbeat, tweeted:
Chartbeat’s lead data scientist Josh Schwartzlater clarified to news site The Verge that Haile was talking specifically about Twitter but it would probably apply on Facebook too.
“There is obviously a correlation between number of tweets and volume of traffic that goes to an article,” Schwartz says. “But just not a relationship between stories that are most heavily consumed and stories that are most heavily tweeted.”
While some people might be fake-sharing to look smart or witty, sometimes the explanation is more benign. In some cases the text of the tweet contains all the key information, particularly with news stories, so if your ultimate goal is clicks, an intriguing tweet might do better than a literal one.
It is not that no one is reading. The Verge reports that Upworthy sees a burst of tweets from people who have consumed just a quarter of the article but the biggest boost is once people have consumed the whole thing. Similarly on Buzzfeed, the majority of social media shares happen after people have been on a page for over 3½ minutes on desktop, or over two minutes on a mobile device.
Upworthy is now measuring “attention minutes” and has found visitors from Facebook spend more time watching videos on the site than visitors from any other source.
Assumption that branding should be low key may be wrong
Most content marketing tends to be light on branding compared with traditional advertising. Conventional wisdom is that people will not share something that looks and feels like an ad, so marketers hoping for a viral hit usually avoid this.
A good example is the viral video of a musical flash mob in the town square of Sabadell in Spain, which has clocked up 32.6 million views on YouTube to date.
The video begins with general shots of the town square and then focuses on a little girl giving money to a busker. He starts playing Beethoven’s Ode to Joyon the double bass and gradually more musicians join, until it becomes a full-blown orchestra with a big audience. It is not until the end of the video, after five minutes and 40 seconds, that it is revealed as the work of Spanish bank Banco Sabadell.
Red Bull is another example of the subtle approach: the chief executive of the energy drink brand, Dietrich Mateschitz, is on record saying the approach is to think about the content the audience wants to see and then remove as much branding as possible.
A chapter of Nelson-Field’s book looks at this issue and finds this shyness about branding might be misplaced. It is well understood that branding prominence is linked to advertising recall, so University of South Australia researchers decided to test the hypothesis that the low-profile approach to brand execution gives video a sharing advantage.
In a sample of 200 videos, the average length was almost three times longer than a standard 30-second television commercial. However, the average time of the first brand entry was 32 seconds and fewer than half had branding in the first 10 seconds. Fewer than one in three videos contained a verbal mention of the brand.
Yet the researchers found neither the timing nor the prominence of branding had any real effect on sharing. The dominant factor was emotional arousal and that was unaffected by branding.
“The popular notion of the need for a low-profile approach is a myth,” Nelson-Field writes. “Far greater danger exists in not branding well, particularly if the video achieves a large audience who don’t share and don’t notice the brand.”
She advises that marketers should apply the body of knowledge about effective television execution techniques and apply them to video.
Paying to promote social content could be worthwhile
Nelson-Field also advocates another old-school approach: paying for reach. She says it is a mistake to over-invest in creative and under-invest in distribution, because only one in 10 people who view a video on any given day will go on to share it.
Cat Jones of social video platform Unruly Media says people are more likely to share a video if they come across it as a recommendation from a friend rather than by random browsing. She adds that people are more likely to share if they can see it’s already becoming popular; for example they’ve seen it trending onReddit.
Paying for reach works because a video that is viewed by only a few people is unlikely to be widely shared and because a rapid spike is more likely to lead to a viral effect than a slow and steady rise.
Paying for reach can also help marketers get large numbers of light users, who contrary to popular belief are more important than core users. Light users, who flit between brands, account for half of today’s sales and more of tomorrow’s for any given company, according to Nelson-Field. She says marketers are obsessed with the idea of exclusively faithful consumers who are “passionate” about the brand but they’ve bought into a myth.
*Original Article by Caitlin Fitzsimmons: http://www.brw.com.au/p/marketing/viral_content_secrets_the_science_aUZlhL19viXXTSgwuVMmEK