Variety, not viral spread, is key to Facebook growth

The notion that certain ideas spread like viruses is ingrained in everything from product marketing to political campaigning. It is also wrong, say researchers who have studied the spread of Facebook. They say that people are more likely to accept an idea that has been adopted by several social groups. The result has implications for anyone wanting to promote a message.

The theory that ideas can be "contagious" is an old one. In 1884, for example, US senators attacked companies that colluded to keep prices high, claiming that the practice had "spread like a disease through the commercial system of this country".

Biological research has since deepened the analogy. When a disease spreads, the chances of a person becoming infected increase with the number of carriers they are exposed to. The same rule ? that a person will adopt an idea if exposed to enough people who believe it ? underpins "essentially all current models" of idea spread, say Johan Ugander and Jon Kleinberg at Cornell University in Ithaca, New York.

"There's definitely a strong intuition that the number of exposures to the message is a key," says Ugander. "This is certainly true for biological contagion. Our findings suggest that social contagion is different."

Friend me

To study contagion, Ugander and Kleinberg teamed up with Cameron Marlow and Lars Backstrom, "in-house" sociologists at Facebook. They examined data from Facebook users who uploaded a list of their email contacts to the network. When a user uploads this list, Facebook offers to send email invites to contacts who are not already on the network. The email lists other Facebook users who have the invitee's address in their contact list.

If this list is long enough, contagion theory predicts that the person should accept the invite, since many of their friends are already "infected" with Facebook. But after studying more than 50 million invites, Ugander and colleagues found that size did not matter. More important than the length of the list was the number of different social groups it covered. For example, if all the people on a four-person list came from different social groups, the likelihood of the invitee signing up was more than twice that of when all four people belonged to the same group.

The team found the same effect at work when they studied the use of Facebook. New recruits were more likely to become active users ? defined as visiting the site on at least six days out of seven ? if their Facebook friends come from several distinct social groups. If the results hold true in other contexts, messaging techniques may have to be rethought.

"Our findings suggest that people are much more likely to purchase a book if they are told that both a co-worker and family member have purchased it, rather than two co-workers or two family members," says Ugander. "Public health interventions may also benefit from approaching people from multiple angles.

"When trying to change how people think about things like exercise or nutrition, it may be better to structure messages in such a way that people receive the information in diverse contexts."

"These results have practical implications for marketers and others who seek to amplify their messages via social referrals," says Duncan Watts at Yahoo Research in New York. "Moreover, the paper is a striking example of empirical social science that would have been impossible to perform until a few years ago. Even today it is arguably impossible anywhere but at Facebook."

Journal reference: Proceedings of the National Academy of Sciences, DOI: 10.1073/pnas.1116502109

If you would like to reuse any content from New Scientist, either in print or online, please contact the syndication department first for permission. New Scientist does not own rights to photos, but there are a variety of licensing options available for use of articles and graphics we own the copyright to.

Have your say

Only subscribers may leave comments on this article. Please log in.

Only personal subscribers may leave comments on this article

Subscribe now to comment.

All comments should respect the New Scientist House Rules. If you think a particular comment breaks these rules then please use the "Report" link in that comment to report it to us.

If you are having a technical problem posting a comment, please contact technical support.

hcm loretta lynn gene kelly zoe saldana zooey deschanel and joseph gordon levitt debra messing ayaan hirsi ali