Diffusion of Innovations
by Everett M. Rogers
Free Press
bookshop.org
Academic but never boring, precise but never pedantic, this book is a masterpiece of expository and pedagogical writing. There, I said it.
Everett Rogers’ Diffusion of Innovations, for a book first published in 1962, remains highly relevant today, more than half a century later. It’s no wonder it’s on its fifth edition, updated in 2003. The diffusion of innovations has generally felt somewhat uncontrollable, even alchemical, to me. So it’s been surprising to learn that while innovations themselves occasionally have a ring of magic to them, their diffusion can very much be planned and coordinated in a controlled (if not quite mechanical) way.
While reading the book, I was constantly distracted, thinking about examples and applications, relationships with other concepts, segments that I wanted to capture for future reference, and bits that I wanted to share with colleagues and friends. (Maybe mine are nerdier than yours.)
I would never finish this if I tried to be comprehensive, so I’m just going to touch on a few ideas that stuck in my head. This post is most definitely not a summary of the book. If you’re even a little bit intrigued, get a copy. You won’t regret it.
Okay, let’s dig in.
Innovation champions: it’s not about power
“The general picture of an innovation champion emerges not as a particularly powerful individual in an organization, but rather as someone particularly adept at handling people, an individual skillful in persuasion and negotiation.”
One of the many easily validated observations from the book is that innovation champions are most often found at lower levels of the organizational hierarchy. In fact, with status at risk, power may even be a negative in some scenarios. How do you spot these individuals who use insights and connections to drive change?
According to Rogers, innovation champions tend to have a strong sense of empathy, which allows them to understand and address the concerns of others within the organization. They are creative thinkers, open to new ideas. Unsurprisingly, they are persistent and resilient, able to push through obstacles to achieve their goals using their technical expertise and connections within the organization. In short: these individuals know the actual work that has to get done, the people that do the work, and the problems they face.
All this suggests that leaders need to pay close attention to individuals at lower levels in an organization to look for good ideas and people to champion them. Moreover, when looking to “influence the influencer,” everyone should fight the urge to look for influencers based on their power.
Communication channels: knowledge and adoption are not the same
“Mass media channels are more effective in creating knowledge of innovations, whereas interpersonal channels are more effective in forming and changing attitudes toward a new idea, and thus in influencing the decision to adopt or reject a new idea.”
Rogers’ analysis of mass-media versus inter-personal communication channels feels so obvious and common-sense. Mass media is useful in spreading awareness about an innovation, making it effective in creating initial knowledge that the innovation exists or is worthy of attention. However, without personal interactions to shape attitudes, one may never make the decision to adopt an innovation. Those personal connections are what ultimately catalyze action.
“Most human communication takes place between individuals who are homophilous, a situation that leads to more effective communication. Therefore, the heterophily that is often present in the diffusion of innovations leads to special problems in achieving effective communication.”
This presents an important complication. Imagine a group of people unaware of an innovation. Mostly, they trust each other because of various social, cultural, or even technical similarities — they are homophilous. When an innovation emerges whose champions are heterophilous compared to that group, bridging the influence gap is a particular challenge. Rogers’ study suggests that individuals with one foot in each group are the ones are best suited to create influence.
Consider a new software library gaining traction in a small corner of Reddit, and a large, established company with careful and detailed processes to make library adoption decisions. The reddit-savvy group may be very different from the group of decision-makers at this large company. Bringing this new library into the company requires an engineer with a foot in both the Reddit group and the company.
The formula, then: when looking to diffuse innovations, find ways to use mass media to share stories of trusted individuals’ adoption, keeping in mind that “trust” is understood from the perspective of potential adopters, not the agent interesting in driving the adoption. If you are a manager trying to convince a group of engineers to adopt a new tool, you may create awareness by talking about it at an all-hands meeting, but adoption will only be triggered by the recommendation of an engineer who collaborates with management and is trusted by their peers.
What makes an innovation successful
“Five attributes of innovations are: (1) relative advantage, (2) compatibility, (3) complexity, (4) trialability, and (5) observability.”
While it’s hard to argue that any of these attributes less important than the others, I’m biased towards trialability and observability as the most critical because they significantly reduce the risk associated with new innovations.
These may well be the most well-understood and widely used ideas form the book. When it comes to trialability, who hasn’t taken advantage of a product’s free trial or received regular (spammy) updates from products they’ve used (“memories” from your photo management app, e.g.)?
Observability, in particular, is hard to implement well for many types of innovations. Take DORA metrics, e.g. A specific action that improves build times which in turn improves deploy frequency is so far removed from concrete and measurable results that if it weren’t also intuitive to most engineers, DORA metrics might never have been adopted. There’s no pat solution, but knowing specifically about observability as a barrier to the diffusion of innovations means we are less likely to fall into a trap of false assumptions.
Early adopters are… different
Earlier adopters have more social participation, are more highly interconnected in the interpersonal networks of their system, are more cosmopolite, have more contact with change agents, greater exposure to mass media channels, and greater exposure to interpersonal communication channels, engage in more active information seeking, and have greater knowledge of innovations and a higher degree of opinion leadership.
Rogers provides the clearest possible definition of what makes for a good early adopter. From the perspective of some looking to diffuse an innovation, this is practically a checklist of attributes to look for in the first users to target:
Greater empathy (like change agents)
Lower dogmatism
Better ability to cope with uncertainty
Better ability to deal with abstractions (seeing the bigger picture)
The points about empathy, dogmatism, and uncertainty seem almost obvious, but the point about abstraction is pretty powerful. Thinking back on my own experience, I can see this particular quality in the folks around me that first adopted VMs, test-driven development, cloud computing, CI/CD, and other tools and practices that emerged.
The change agent playbook
Change agents would not be needed in the diffusion of innovations if there were no social and technical chasms between the change agency and the client system. Their superior know-how actually poses a barrier, making it difficult for them to communicate directly with clients.
Given the challenges of being a change agent, this playbook (my term) can be a great help in guiding action. Knowing the sequence — where are we now, what do we need to do next — allows for a methodical and almost (but not quite) mechanical approach.
The sequence:
Identify and create awareness of the necessity of change
Establish a trust relationship to exchange information with potential users
Understand what issues need to be addressed
Motivate and prepare individuals and groups who want to change
Translate the intent into action
Support the implementation of the change and integration into regular practices
Encourage ongoing adoption and sustained success
With this list in front of me, I’ve certainly been able to look back on failed change efforts and related those failures to missing these steps. In particular, I think 1, 2, and 4 are the most often skipped. Someone (whether or not in a position of power) understands an issue and then, without empathy, proceeds to try to drive or force action. I’ve also seen change agents falter at 6 and 7.
Does this match your experience? I’d be curious to hear about either successful diffusions where the change agent missed steps, as well as efforts that failed despite following all the steps. I’d almost bet money you can’t come up with one!
Not too cold, not too hot
Stable equilibrium occurs when almost no change is occurring in the structure or functioning of a social system. Dynamic equilibrium occurs when the rate of change in a social system is commensurate with the system's ability to cope with it. Disequilibrium occurs when the rate of change is too rapid to permit the system to adjust. Change agents generally wish to achieve a rate of change that leads to dynamic equilibrium, and to avoid disequilibrium.
I’ll wrap up with a final example where Rogers’ clear breakdown of a commonsense point makes it possible to make a near science out of an art.
Organizations in stable equilibrium are likely not accepting of change and don’t know how to deal with change. Forces of inertia are strong. Opinion leaders in such environments are opponents of change.
Organizations in dynamic equilibrium have a culture that accepts change and tools to manage it. Opinion leaders here are likely supporters of change.
Go too far, though, and organizations enter disequilibrium, where things are changing too fast for the culture and processes to catch up. Whether through externally driven change, say a series of major market shifts, or internal change from new leaders, a merger, or simply multiple re-orgs, organizations lose their ability to absorb additional change.
This tells me two separate things. First, that there is such a thing as too much change for an organization, so it is important to throttle the rate of change. Second, and related, is that it may be necessary to precede any attempts at change by first moving the organization from either stable equilibrium → dynamic equilibrium or disequilibrium → dynamic equilibrium.
So there you have it, a tiny little sample from 500+ pages of wisdom. I’m sure there’s criticism out there of some kind; some light googling suggests things like over-simplification and the lack of much follow-on research despite heavy citation. I don’t know what to make of it, because I can see these ideas applied practically everywhere I look. What can I say, the book works for me.
To me, Diffusion of Innovations is invaluable. It is a magnificent and heavily under-appreciated book. Books like Crossing the Chasm by Geoffrey Moore get more attention, despite being so heavily influenced by Diffusion of Innovations. I find it hard to understand this state of affairs for a book as readable, and digestible as this. Now I’m definitely curious if other famous works on innovation and change management, e.g., those by John Kotter, picked up ideas from this book.
I submit that Rogers’ work is essential reading for anyone interested in innovation and change. It offers theoretical foundations, practical guidance, and models and taxonomies that make it a superb reference work for anyone interested in developing and guiding innovations of all kinds.
Which is all of us, isn’t it?