The myth of influencers — how (behaviour) change really spreads
In my work as (service) designer I’ve always struggled with the paradox of scale: how do you create meaningful interactions with large numbers of people? Something strange happens in organisations when the populations they aim to communicate with, exceeds the size of an auditorium. They appear to give up the idea of human relationship building — empathy, love, curiosity, smell, rhythm, synchronicity etc. no longer apply. Instead the focus shifts to distribution. Especially among marketeers, words like ‘exposure’, ‘targeting’ ‘positioning’, ‘total market reach’ etc. take over and the audience becomes a ‘they’-abstraction.
Empathising with an individual is already difficult, with a roomful is even harder, let alone a group that doesn’t fit an auditorium.
This post is skewed by one book occupying my thinking at the moment. Every once in a while, someone presents a model that makes so much sense, that you start to believe that you always knew this was true (which is confirmation bias yelling at me). The book is called ‘Change’, and is written by Damon Centola, Professor of Communication, Sociology and Engineering aan de University of Pennsylvania in 2021. He describes numerous experiments to explain how behaviour change spreads through networks of people in a very particular way, contradicting common beliefs about the spread of information as taught by many marketing and communications students, even today.
With the growth of social technology, the goal to ‘go viral’ has become a household strategy for any self-respecting policy maker or consumer brand. The image of a central, highly connected figure spreading ideas like a virus, infecting everyone who is exposed to the idea, just like real (sick-making) viruses move through populations—plays a key role in this. For simple ideas, which do not involve much effort, nor risk for the person spreading the idea, this is indeed how ideas can go viral very quickly. But when the idea requires more effort, and (social) risks, humans don’t react to new ideas in the same way as our bodies do when exposed to new viruses. To simplify: when presented with a new idea, we check our peers to decide how we will respond to it.. Merely presenting information about the benefits of a new breed of seed corn to farmers who are struggling with declining yields for example, does not mean that they adopt it. Only when their local We know this already for quite some time, but many marketers and policymakers still think that if they just make the information interesting enough, their audiences will change their attitudes and behaviour.
At the root of this lies the idea of small worlds, how everybody is connected with everybody globally in just six steps, if we utilise people in the periphery of our social networks. Mark Granovetter refers to these links as ‘weak-ties’. They represent the outer reach of your network and they are the linking pin to other networks. Which means that they are the most efficient links to spread ideas broadly and quickly. Because they are located at the fringe of your network, they have little mutual connections with your close connections; your strong-ties. A network of strong ties consists of people who are mutually acquainted; each individual has multiple connections to others in the same network. The difference between a strong-tie networks and a weak-tie network can visually be represented with a fishing net shaped pattern and a fireworks-shaped pattern.
Imagine going to a party where everybody knows everybody — or a party where everybody only knows the party face. In the first party, people feel at ease fairly quickly, but in the last party, the host would be quite busy, I guess.
From a marketing efficiency perspective, there’d be no point to focus on the strong-tie fishing net patterns because the ‘redundant’ exposure slows down the spread of the message (‘unduplicated reach’ is the golden standard). However, in order to spread adoption of new and more complex ideas, the fishing net pattern is a key condition because of its redundant connections: repeated messaging from multiple peers, facilitates social proof (see earlier post on Cialdini’s principles of persuasion).
Humans are predominantly group-focused — especially when stakes are high and we experience uncertainty — we determine and adjust our behaviour according to the perceived norms we see our our social peers adopt. Often we do so without being fully aware of it. Randomised experiments have shown how people change their energy consumption if they learn that their neighbours consume more or less than they do. The reuse of towels increases 75% in hotels when occupants learn that the previous occupant did so as well. If you ask people directly about it, they will attribute their behaviour to all kinds of shiny personal virtues and are shocked to learn how social influences affect our daily decisions. We like to think of ourselves as autonomous beings.
In uncertain situations, we desperately seek guidance from others to determine our own behaviour. It’s the reason why people can drown in a canal in broad daylight and in full view of bystanders without anyone interfering. We’d like to think of those people as being heartless and cruel… We should know better.
What this means is that we need our social groups to reinforce the acceptance of a new idea. Influencers typically have large networks, so the chance that at least 20–30% of his network adopts the new idea is very little and the majority will try to maintain the status-quo. The influencers’ reputation is on the line, so taking a bet on something new, which turns out to be a fluke later is risky. Centola found that real change happens first in small, interconnected social clusters. Only when these clusters reach a tipping point, it will spread towards new clusters through people who are connected to both networks, where the same process repeats. This represents what he calls a fishing net pattern. Change ideas spread slowly in networks connected in this pattern, but the adoption of change is far more successful compared to people connected in a fireworks pattern because of the reinforcing effects of the social proof principle which bolsters the adopters’ confidence — “I am not alone”.
Leading up to WW1, the UK military needed to recruit a much larger army to fight the massive war against Germany. In those days, joining the military was seen as something for the gentry. To change this perception, the military introduced “Pals Battalions’. If you’d sign up with your friends you’d be fighting together. This idea triggered social recruitment and drove thousands of social networks to sign up– from stockbrokers to football teams and complete neighbourhoods (a tragedy for social fabrics when the war ended).
Just as complex ideas travel slowly, simple ideas actually travel very fast and adopt what he calls a fireworks network shape. Just like our brains prefer simple heuristics to confirm what we already believe, simple sensations and emotions such as desire, fun/humour or opportunity, and even more so skepticism, anger, mocking, outrage etc, travel very fast. Our social media feeds prove it. This also means that reasons why a new innovation won’t work comes to mind easier and will therefore also travel much faster than deliberate ideas which require consideration and coordination with peers.
Our brain is programmed to confirm our beliefs about the world, so it’s obvious why reasons NOT to adopt a new idea come much faster to mind, especially when it involves any perceived risks.
In advertising, ‘awareness’ is a key ingredient to gain attention for a new idea. But awareness can also backfire. The reason why Google Glass’ viral campaign succeeded to reach a lot of people, but failed to grow adoption, is that the campaign highlighted our awareness that most people like ourselves did NOT have — nor fit the type of person — using Google Glass. The campaign made us acutely aware that the product clearly existed, but wasn’t adopted in our own circles, which means that the campaign essentially highlighted its own failure.
The fishing net pattern and the slightly slow and more protected spread of ideas through pockets of clusters in a larger network has other advantages as well. When it comes to innovation and the conception-process of new and original ideas, it is useful to ‘protect’ the innovators from early exposure to average ideas. Centola presents rigorous experiments showing how teams connected in fishing net patterns outperform teams connected in fireworks patterns by far. The highly connected teams would converge on average solutions early on and would stop exploring any iterating alternative configurations. Paralel to the experiment, they also applied a self-learning algorithm to figure out the best possible solution to compete with the other teams. What they found was that the algorithm would simply select the best solution to each underlying sub-problem. The teams connected in the fishing net patterns, however considered how sometimes inferior solutions to sub-problems combined much better with other solutions to solve the bigger problem as a whole.
When we are brainstorming to solve a complex problem, we are very sensitive to plausible ideas. Somehow we stop exploring to accept the plausible idea as THE best idea–almost as if further exploration is a threat to consensus building.
After reading the book I realised the massive implications for people involved in innovation and organisational change—both of which have seen more failures than successes on average. I firmly believe that the ideas presented by Centola can make a big difference in the way we organise and distribute innovation and change. Insightful here is Damon’s notion of ‘wide bridges’– how social reinforcement, legitimacy and coordination can only function to facilitate adoption of behaviour change if the bridges between different social networks are ‘wide’ enough — meaning multiple people in one group are connected with multiple others in the other group. The more silo’d your organisation is structured, the lower the chance that successful innovation in one department gets adopted in another department. That may sound obvious, but at least we know now that there’s no need to invest in fancy videos or appealing arguments to convince our colleagues. Instead we simply need to identify those change agents on the periphery with multiple connections across the bridge, with similar interests and common goals.
Change agents are not your typical influencer, they have no formal position and are not the connected stars everybody admires and listens to. Most aren’t even aware that they are a linking pin. They simply don’t care much about boundaries and are curious to connect with people. More importantly, at some point people come their way to ask for guidance across the organisational jungle.
Connect and support those agents to form a reinforcing social cluster with mutual interests to spread the change. They are definitely NOT people are seen to be influencers–these are generally non-adopters because change adoption can be risky for their position. Most change agents aren’t even aware of their potential roles. Because of their position at the periphery, maybe they aren’t always taken very seriously, but they hold the key to spreading meaningful change, if only they are supported to shape wide bridges across our silo’d organisational landscapes.
For those who are convinced by now to read Centola’s book; don’t worry about the spoilers in this post; besides the fact that you just have to have read this book, there’s plenty more to learn about.
Take care of your strong ties and listen to those at the periphery of your circle, especially in organisational setting. They may not know it, but they are positioned at important inflection points to carry your change initiatives across rivers of resistance. I wish you wide bridges.
PS, I have no interests nor any investments in The University of Pennsylvania, Damon Centola’s businesses or his publisher. Just a fan.
