Much of the conventional wisdom about the diffusion of new ideas, behaviors, and innovations comes from network science, where it tracks back to two sources: Mark Granovetter’s influential theory of weak ties and Albert-László Barabási’s work on network hubs. But in the past decade or so, researchers have revealed some nuances to their ideas that leaders undertaking change initiatives, especially full-scale transformations, should consider.
In 1973, the American Journal of Sociology published a much-cited article titled “The Strength of Weak Ties,” by Granovetter, then an assistant professor at Johns Hopkins University. Based on research Granovetter had done while earning his Ph.D. at Harvard, the article introduced the idea that people with whom you share few connections (your weak ties) are more conducive to the diffusion of your ideas than people with whom you share many connections (your strong ties). That’s because weak ties speed diffusion by extending your reach to more people you don’t know, while strong ties produce a kind of echo chamber in which whatever you are trying to spread reverberates among people you already know — slowing diffusion.
But what if the theory of weak ties, which is now widely accepted, isn’t as universally applicable as it seems? Damon Centola, an associate professor in the Annenberg School for Communication and the School of Engineering and Applied Sciences at the University of Pennsylvania, thinks this is the case. In a new book, How Behavior Spreads, Centola describes his work — and the work of other social network researchers — over the past several years. He explains that strong ties can be better conduits for diffusion than weak ties in some situations. It all depends on what kind of diffusion (or contagion) you are trying to encourage.
What if promoting change through weak ties isn’t as universally applicable as it seems?
Centola says there are two kinds of contagions: Simple contagions that spread effortlessly, such as the measles or the news about your company’s latest quarterly earnings, and complex contagions that don’t spread so easily, such as investing in bitcoins or using a new form of birth control. Adopting complex contagions entails some kind of cost or risk, or it requires that other people start using them before they’re worth your trouble — being the only person you know on Facebook isn’t so great, right? There is, writes Centola, an “important distinction between being exposed to a behavior — and knowing it is an option for you — and actually deciding to adopt that behavior for yourself.”
Simple contagions only require exposure and thus are best spread through weak ties. Reach is the key to success with them. But Centola found that complex contagions require adoption, and so need strong ties to help them gain currency. That’s because redundancy is needed. Complex contagions require multiple confirmations to create enough confidence for adoption. Imagine a manufacturing company contemplating 3D printing: How much evidence of the technology’s utility would senior leaders need before they’d be willing to stop using existing assembly lines and start printing components? How many of the company’s customers would have to own 3D printers before its leaders would have the confidence to make that shift?
Corporate transformations are complex contagions, too: They require changes in behavior, which entail costs and risks. Transformations also feature a high degree of complementarity — lots of people have to get on board if they are to succeed.
This suggests that leaders attempting organizational overhauls should tap their strong ties, which is precisely the opposite of how many leaders approach transformation. They try to reach lots of people with whom they have weak ties at once — bringing large groups together for rousing sessions aimed at describing the burning platform on which they are precariously perched and the visionary lifeline that will save them. The goal: Get the attendees to buy in and then transmit the message far and wide.
Instead, leaders should start smaller and go for redundancy. They should make the case for transformation through their strong ties and let it reverberate among those connections until they convince themselves that it is both real and necessary. And then, leaders can let their strong ties spread the word through their own strong ties. “Viewed from the relational perspective of each individual, strong ties are the most important links in the social network,” writes Centola. “They are proximate, trusted, and familiar, and therefore the most influential for diffusion.”
In 2002, nearly 30 years after Granovetter’s weak ties article, Albert-László Barabási, then a physics professor at Notre Dame, published Linked: The New Science of Networks. The book described the essential role of hubs — websites and other nodes that feature an exceptionally high number of connections — in the expansion of networks. That idea got picked up by Malcom Gladwell and others, and was translated to the spread of contagions through a few super-connected influencers — such as Oprah Winfrey.
Centola’s theory also contradicts the conventional thinking established by Barabási about hubs in social networks. Like weak ties, hubs are great for spreading simple contagions — such as whether or not a new book is worth reading. But when it comes to complex contagions, they may have a stifling, rather than a stimulating, effect.
To see why, Centola says, consider a senior executive in a Fortune 500 company who is extremely well-connected. Such an executive would be among the first to hear about an innovative new technology. But he certainly would not be among the first to adopt it. That’s because of the reputational and career risks involved if the technology didn’t pan out. Rather, the exec would wait for social confirmation. Conversely, someone on the periphery of a company, without so much to lose, would be more likely to embrace the technology. This helps explain why disruption usually happens at the edge of industries and companies, not at their core.
In many transformation efforts, leaders seek to introduce changes in behavior and technology in highly connected hubs — or across the entire company at once. Centola’s thinking suggests that you should instead try to diffuse these changes through less connected areas of the company before rolling them out wholesale. This a good argument for establishing a greenfield organization that operates outside the mainstream of the company. Use it to implement and test changes and generate the confirmation needed for adoption across the company.
Full-scale transformations are fraught endeavors. If the concept and nuances of complex contagion can give you a leg up, I say put them to work.