This article is taken from a presentation Andy and Hal gave at the Walkley Storyology conference on August 8 2013.
I. Cats vs dogs
Everyone thinks cats rule the internet.
We started the Share Wars project in 2011 because we believed the distribution of news through social networks was going to change news media.
We built an engine to scan the world’s news sites, to collect stories as they were published and to track how people shared them on social networks. For three months we collected the stories and their sharing histories, and we collated them all in a big database we called Likeable.
1.8 million articles from 118 news sites, gigabytes of data, and at the end of it we laid at least one major existential question to rest. Dogs beat cats hands down.
In Australia and worldwide, dogs are more than twice as likely to feature in a top shared story. Yes, in that three months we had dead stuffed cats flying around, cats biting musicians, cats who survived falls off skyscrapers, cats who didn’t survive obesity, a woman who kept 400 cats. But it turns out dogs run much, much deeper. Not only are they more shareable, they are more newsworthy in general – in the whole data set, they feature in double the number of headlines that cats do. They are interwoven so much tighter into the human story. Why? In part because they are so much better at acting like humans. It turns out that animals acting like people is sharing gold.
Now why would that be?
We think it’s deep, that animal connection. Why do so many children’s stories feature animals acting as humans? There is an archaic, totemic power in animals.
For us, the Share Wars project exposed a universe of stories we’d been living in for years – we’d been living in those stories like fish in water. Andy and I both started in a world of newspaper newsrooms, where you accepted the folk wisdom of the place because there was no other source of story authority. Politics on the front and sport at the back. “If it bleeds, it leads”. “Sex sells”.
Until now it has not been possible to study stories systematically. Traditional news values were values held by journalists, not the audience. They were based on intuition and feedback from editors and colleagues.
If you had asked me before we ran the project, I would have told you cats were a better bet than dogs. I would have been wrong. Now we understand stories better, and we can actually use the research to help predict stories what will be shareable.
II. Shareable words
“Abbott backs gay marriage in Facebook update”
That headline is sharing dynamite.
It contains four of the top 10 most-shareable words from our three-month data capture.
If any of you out there had reason to write that headline, post it to Facebook straight away because it’s going viral and it might as well be your version that takes off. But that’s where the publisher’s control over the fate of the article ends. After that, it’s the power of the narrative that will drive sharing.
The element of surprise also makes this shareable: Tony Abbott coming out for gay marriage … on Facebook. The words are familiar but their sequence is not.
These are the 10 most shareable nouns in Australia during our capture period: Gay Facebook Baby Girl Teen Marriage Mum Abbott Gillard … You.
These words appeared more often in the headlines of the most- shared stories than any others.
How about the sharing shockers? What were the words most likely to appear in headlines of those stories that received zero shares?
I won’t bore you with the full list but here are some:
China, Syria, murder, attack and the worst word in journalism … report.
Versus Baby, girl, teen, marriage, mum … you.
Sharing is in interesting filter isn’t it?
It is a filter for relevance.
III. Sharing leaderboard
Who is winning the Share Wars?
Journalists are a competitive bunch. So we thought we’d show you the sharing leaderboard from our capture period.
When we total Facebook and Twitter shares, News.com.au finishes top ahead of smh.com.au and then ninemsn. You can see how comparatively small the Twitter contribution is – averaging 10% across Australian publications.
This is a total-share figure but does not necessarily show who creates the most shareable content.
When we adjust for audience size, the ABC News is the big mover.
IV. Sexy is not sexy
It’s hard to see things that are not there.
It took a long while for us to wake up to something missing in our top shared stories. Stars, kids, animals, gay rights, zombies, resurrections. Lots of life, lots of death. But not a lot of sex.
Compared to its power in other distribution channels, sex is a dud on social networks. Of the top shared stories in Australia, about 2 percent are about sex in some way. Most of these are not what you’d call sexy. They are about penises drawn on burger wrappers, porn in church presentations and x-rated lollies. Mostly schoolboy humour. Not hot.
Less than 1 percent of stories were overtly sexual, and even then there are other sharing drivers at work. Take the case of the unfortunate German man who called the cops to save him from a nymphomaniac. In this case it’s what we call a reverse – the man begging police to save him from this rapacious woman.
It turns out that almost no one shares purely titillating material.
Smut is on the nose.
If that’s true, then where does that leave the conventional wisdom that sex sells? Have editors all been wrong for years? And if so why are we knee deep in slideshows of Kate Upton?
The anomaly is worth investigating. It hinges on the difference between what people do in the privacy of their own screen-time and what they are willing to share with the world.
Becoming aware of that anomaly was one of the key drivers behind Share Wars in the first place.
v. Sharing is not sharing
So we’ve proven sexy is not sexy. Well guess what? Sharing is not sharing.
We thought most social media sharing would be real sharing – people passing on information for the benefit of the receiver.
But most sharing is people distributing stories to relate to one another, to create or reinforce group identity around a story.
It’s putting a story out there as a line in the sand. If you Like you’re for me, if you don’t Like, you’re against me.
We call this behaviour “Norming”.
Here is an example of norming when the sharer approves of the story: “Obama backs same-sex marriage”.
It’s a big-issue story that people have strong views on. We can tell it is being shared by approvers by the likes-to-total-shares ratio.
Seventy-two per cent of sharers clicked the thumbs up button.
That puts it in the very top approval tier.
Our most shared Australian story was about a young woman who sued Geelong Grammar because she argued the school did not educate her well enough to qualify for Law. The likes-to-total shares ratio puts this in the most disapproving group. Sharers did not like that the student blamed the school for her poor marks.
This is what Share Wars is all about – understanding the things that get under people’s skin.
Personal responsibility. The right to do what you want. Gay rights. Parenting. Life and death. These are things people really care about.
When the audience shares, they’re taking a stand … they’re judging others … they’re indicating the things that matter most.
And we think understanding what matters most to the audience should matter a whole lot more to people like us … the story tellers.