Contagious: Why Things Catch On (12 page)

BOOK: Contagious: Why Things Catch On
13.04Mb size Format: txt, pdf, ePub

The topic? How fluid and gas dynamic theories were being used in medical research.

Grady’s article detailed something called
schlieren photography, in which “a small, bright light source, precisely placed lenses, a curved mirror, a razor blade that blocks part of the light beam and other tools make it possible to see and photograph disturbances in the air.”

Sounds less than riveting, right? Join the club. When we asked people what they thought of this article on a number of different dimensions, the scores were pretty low. Did it have lots of
Social Currency? No, they said. Did it contain a lot of practically useful information (something we’ll discuss in the Practical Value chapter)? No again.

In fact, if you’d gone down the checklist of characteristics traditionally believed to be prerequisites for viral content, Grady’s article, entitled “The Mysterious Cough, Caught on Film,” would have lacked most of them. Yet Grady’s piece clearly had something special or so many people wouldn’t have hit the e-mail button. What was it?

—————

Grady’s interest in science started in high school. She was sitting in chemistry class when she read about Robert Millikan’s famous experiment to determine the charge on a single electron. It was a complicated idea and a complicated experiment. The study involved suspending tiny droplets of oil between two metal electrodes, then measuring how strong the electric field had to be in order to stop the droplets from falling.

Grady read it several times. Again and again until she finally understood. But when she did, it was like a flash going off. She got it. It was thrilling. The thinking behind the experiment was so clever, and being able to grasp it was enthralling. She was hooked.

After school Grady went to work at
Physics Today
magazine. Eventually she worked at
Discover
and
Time
magazine and finally worked her way up to health editor at
The New York Times.
The goal of her articles was always the same: to give people even just a little bit of that excitement that she had felt back in chemistry class decades before. An appreciation for the magic of scientific discovery.

In her piece that October, Grady described how an engineering professor used a photographic technique to capture a visible
image of a seemingly invisible phenomenon—a human cough. The schlieren technique had been used for years by aeronautics and military specialists to study how shock waves form around high-speed aircraft. But the engineering professor had harnessed the technique in a new way: to study how airborne infections like tuberculosis, SARS, and influenza spread.

It made sense that most people thought the article wasn’t particularly useful. After all, they weren’t scientists studying fluid dynamics. Nor were they engineers trying to visualize complex phenomena.

And while Grady is one of the best science writers out there, it made sense that the general population would tend to be more interested in articles about sports or fashion. Finally, while coughs would certainly be a nice trigger to remind people of the article, cold and flu season tends to peak around February, four months after the article was released.

Even Grady was bemused. As a journalist, she’s delighted when something she writes goes viral. And like most journalists, or even casual bloggers, she’d love to understand why some of her pieces get widely shared while others don’t.

But while she could make some educated guesses, neither she nor anyone else really knew why one piece of content gets shared more than another. What made this particular article go viral?

—————

After years of analysis, I’m happy to report that my colleagues and I have some answers. Grady’s 2008 article was part of a multi-year study in which we analyzed thousands of
New York Times
articles to better understand why certain pieces of online content are widely shared.

A clue comes from the picture that accompanied Grady’s
piece. Earlier that October, she had been scanning an issue of
The New England Journal of Medicine
when she came across a piece entitled “Coughing and Aerosols.” As soon as she saw it she knew the research would be the perfect basis for an article in the
Times.
Some of the piece was pretty technical, with discussions of infectious aerosols and velocity maps. But above all the jargon was a simple image, an image that made Grady decide to write her article.

Simply put, it was amazing. The reason people shared Grady’s article was
emotion.
When we care, we share.

MOST E-MAILED LISTS AND THE IMPORTANCE OF SHARING

Humans are social animals. As discussed in the chapter on Social Currency, people love to share opinions and information with others. And our tendency to gossip—for good or ill—shapes our relationships with friends and colleagues alike.

The Internet has become increasingly engineered to support these natural inclinations. If people come across a blog post about a new bike sharing program or find a video that helps kids solve tough algebra problems, they can easily hit the Share button or copy and paste the link into an e-mail.

Most major news or entertainment websites take the extra step of documenting what has been passed along most frequently. Listing which articles, videos, and other content have been most viewed or shared over the past day, week, or month.

People often use these lists as shortcuts. There is way too much content available to sift through it all—hundreds of millions of websites and blogs, billions of videos. For news alone, dozens of highly reputable outlets continuously produce new articles.

Few people have time to seek out the best content in this ocean of information. So they start by checking out what others have shared.

As a result, most-shared lists have a powerful ability to shape public discourse. If an article about financial reform happens to make the list, while one about environmental reform barely falls short, that initially small difference in interest can quickly become magnified. As more people see and share the article about financial reform, citizens may become convinced that financial reform deserves more governmental attention than environmental reform, even if the financial issue is mild and the environmental issue severe.

So why does some content make the Most E-Mailed list while other content does not?

For something to go viral, lots of people have to pass along the same piece of content at around the same time. You might have enjoyed Denise Grady’s cough article, and maybe you shared
it with a couple of friends. But for the piece to make the Most E-Mailed list, a large number of people had to make the same decision you did.

Is this just random? Or might there be some consistent patterns underlying viral success?

SYSTEMATICALLY ANALYZING THE MOST E-MAILED LIST

The life of a Stanford graduate student is far from grand. My office, if you could call it that, was a high-walled cubicle. It was tucked up in a windowless attic of a 1960s-era building whose architectural style has often been described as “brutalist.” A short, squat structure with concrete walls so thick they could probably withstand a direct hit from a small grenade launcher. Sixty of us were clustered together in a cramped space, and my own ten-by-ten fluorescent-lit box was shared with another student.

The one upside was the elevator. Graduate students were expected to be working at all times of day and night, so the school gave us a keycard that allowed twenty-four-hour access to a special lift. Not only did it take us up to our windowless workstations, it also gave us access to the library, even after it closed. Not the most lavish perk, but a useful one.

Back then the distribution of online content was not as sophisticated as it is today. Content websites now post their most e-mailed lists online, but some newspapers published these lists in their print editions as well. Every day
The Wall Street Journal
published a list of the five most read articles and the five most e-mailed articles from the previous day’s news. After scanning a couple of these lists, I was enthralled. It seemed like the perfect data source to study why some things get shared more than others.

So just as a stamp collector collects stamps, I began to collect the
Journal
’s Most Emailed list.

Once every couple of days I would use the special elevator to go hunting. I would take my trusty scissors down to the library late at night, find a stack of the most recent print editions of the
Journal
, and carefully clip out the Most Emailed lists.

After a few weeks, my collection had grown. I had a big stack of news clippings and was ready to go. I entered the lists in a spreadsheet and began looking for patterns. One day “Dealing with the Dead Zone: Spouses Too Tired to Talk” and “Disney Gowns Are for Big Girls” were two of the most e-mailed articles. A few days later “Is an Economist Qualified to Solve Puzzle of Autism?” and “Why Birdwatchers Now Carry iPods and Laser Pointers” made the list.

—————

Hmm. On the face of it, these articles had few characteristics in common. What did tired spouses have to do with Disney gowns? And what did Disney have to do with economists studying autism? The connections were not going to be obvious.

Further, reading one or two articles at a time wasn’t going to cut it. To get a handle on things I needed to work faster and more efficiently.

Luckily my colleague Katherine Milkman suggested a vastly improved method. Rather than pull this information from the print newspaper by hand, why not automate the process?

With the help of a computer programmer, we created a Web crawler. Like a never tiring reader, the program automatically scanned
The New York Times
home page every fifteen minutes, recording what it saw. Not only the text and title of each article, but also who wrote it and where it was featured (posted on the
main screen or hidden in a trail of links). It also recorded in which section of the physical paper (health or business, for example) and on what page the article appeared (such as the front page or the back of the third section).

After six months we had a huge data set—every article published by
The New York Times
over that period. Almost seven thousand articles. Everything from world news and sports to health and technology, as well as which articles made the Most E-Mailed list for those same six months.

Not just what one person shared, but a measure of what all readers, regardless of their age, wealth, or other demographics, were sharing with others.

Now our analysis could begin.

—————

First, we looked at the general topic of each article. Things like health, sports, education, or politics.

The results showed that education articles were more likely to make the Most E-Mailed list than sports articles. Health pieces were more viral than political ones.

Nice. But we were more interested in understanding what drives sharing than in simply describing the attributes of content that was shared. Okay, so sports articles are less viral than dining articles. But why? It’s like saying people like to share pictures of cats or talk about paintball more than Ping-Pong.
That doesn’t really tell us much about why that is happening or allow us to make predictions beyond the narrow domains of cat stuff or sports that start with the letter P.

Two reasons people might share things are that they are interesting and that they are useful. As we discussed in the Social Currency chapter, interesting things are entertaining and reflect
positively on the person who shares them. Similarly, as we’ll discuss in the Practical Value chapter, sharing useful information helps others and makes the sharer look good in the process.

To test these theories, we hired a small army of research assistants to score
New York Times
articles on whether they contained useful information and how interesting they were. Articles about things like how Google uses search data to track the spread of the flu were scored as highly interesting, while an article about the change in the cast of a Broadway play was scored as less interesting. Articles about how to control your credit score were scored as being very useful, while the obituary of an obscure opera singer was scored as not useful. We fed these scores into a statistical analysis program that compared them with the Most E-Mailed lists.

As we expected, both characteristics influenced sharing. More interesting articles were 25 percent more likely to make the Most E-Mailed list.
More useful articles were 30 percent more likely to make the list.

These results helped explain why health and education articles were highly shared. Articles about these topics are often quite useful. Advice on how to live longer and be happier. Tips for getting the best education for your kids.

But there was still one topic that stood out like a sore thumb: science articles. For the most part, these articles did not have as much Social Currency or Practical Value as articles from more mainstream sections. Yet science articles, like Denise Grady’s piece about the cough, made the Most E-Mailed list more than politics, fashion, or business news. Why?

Other books

The Empty Glass by Baker, J.I.
Christa by Keziah Hill
High Fidelity by Nick Hornby
Dark Rival by Brenda Joyce