The diffusion of innovations curve, developed by Everett Rogers, has become a pretty iconic tool in understanding how new ideas, products, or technologies gain traction within a market. Its simplicity is compelling: a bell curve that categorises adopters (or a population) into five groups — innovators, early adopters, early majority, late majority, and laggards. Yet, despite its widespread use, this model is often misunderstood, particularly in the context of technology adoption.
Warren Schirtzinger, a former colleague of Everett Rogers, recently wrote about some of the most common misconceptions about the model, based on some simple research he conducted into how it has been popularly discussed and written about.
One of the most common misconceptions that he describes is that the diffusion of innovation curve applies generically to new technologies when the reality is far more nuanced. What diffuses through the population isn’t just the technology. Instead it’s the use cases or specific applications or problems solved by that technology that follow this pattern. In fact, Shirtzinger goes as far as to say that the model doesn’t apply to generic technologies at all. The original model that Rogers created for example, was based on one particular application: the ‘use of hybrid seed corn by farmers in Iowa’.
Whilst I still believe that the model is a useful way of thinking about the progression in new technologies, looking at the diffusion of specific applications or use cases IS actually a more compelling and nuanced way of understanding technology adoption. Take a general purpose technology like AI as an example. Looking at where AI is on the diffusion model doesn’t tell us a lot. Even going a level deeper and considering the adoption curve in the context of a more specific form of AI like Generative AI doesn’t give us much that we don’t already know.
But if we layer on specific use cases (applying GenAI for content production), and/or discipline (applying GenAI for content production in marketing), and/or sector and geographical region (applying GenAI for content production in UK retail marketing), we have something which is far more applicable and instructive. So as innovation specialist Walter Robertson commented over on LinkedIn it can be helpful to consider use cases in the context of a profession, an industry and a geographical location in order to position them most accurately on the adoption curve.
Applications and technologies are not the same thing and this critical distinction has profound implications for how we approach innovation and for market adoption strategies.
Put another way, the curve doesn’t describe a technology’s journey in isolation but rather the adoption lifecycle of each specific use case the technology enables. A fundamental insight often overlooked when people make this mistake is that a new technology rarely gains widespread adoption in its raw form. Instead, it’s the identification and maturation of use cases that drive the diffusion process. For instance, when the internet emerged, it didn’t immediately revolutionise industries. Early use cases such as email and static websites attracted innovators and early adopters. As new use cases evolved (e-commerce, social media, and cloud computing) the technology itself became indispensable, driving adoption across the early and late majorities.
This insight explains why some technologies, like artificial intelligence or blockchain, appear to linger in the ‘innovator’ or ‘early adopter’ phases for years. It’s not that the technology isn’t valuable, it’s that the use cases haven’t yet matured enough to resonate with broader segments of the market. Looking at the diffusion of use cases therefore gives you a much clearer insight on how a technology is likely to scale.
One of the other big ideas associated with the curve is Geoffrey Moore’s ‘crossing the chasm’ concept, a popular framework for understanding the challenges of moving from early adopters to the early majority. Yet, this framework, too, is often misapplied. Many assume that the challenge lies in convincing more people to adopt the technology. In reality, the chasm represents a shift in focus from selling technology to selling solutions that address specific, widely understood problems.
Organisations that succeed in crossing the chasm do so by identifying a compelling use case that resonates with the pragmatism of the early majority. The initial buzz around blockchain technology for example, revolved around its potential to disrupt finance, but adoption remained limited. When blockchain use cases like supply chain transparency and digital identity began to take shape, they attracted a broader audience, demonstrating the importance of solving tangible problems.
‘Crossing the chasm’ theory does not apply to all innovations and can often lead to the mischaracterisation of those innovations. There is, for example, a big difference between novel innovation which requires new behaviours, or experiences or learning, and incremental innovation which is simply refinements to existing products, services or processes. The latter is far more common and since it relates to evolving existing behaviours there is simply no chasm to cross – it is instead a continuum of change. Chasms only happen when the innovation requires a significant change in understanding or behaviour (which is really defined as ‘discontinuous innovation’). It makes sense therefore that the theory was originally developed for high-tech businesses.
A final, and especially common, mistake related to the model is that people generalise around the psychographic profiles (innovators, early adopters and so on). This means that they often assume that people who are early adopters of one technology application or use case will automatically be early adopters of another (or worse, of ALL new technology applications). Yet it is of course perfectly valid to imagine that an early adopter of short form video and TikTok for example, may be a laggard when it comes to applying Generative AI in marketing.
Misinterpretation of the diffusion of innovations and related concepts matters for several reasons:
Premature Scaling: Companies often assume that initial excitement among innovators equates to market readiness. Without a mature use case, attempts to scale too early can result in wasted resources and damaged credibility. It’s worth remembering the ‘goldilocks zone’ of technology investment.
Misaligned Messaging: Focusing on the technology rather than the problems it solves can alienate the early majority, who prioritise practicality over novelty.
Underestimating the Lifecycle: Technologies often experience multiple diffusion cycles as new use cases emerge. A good example here is the mobile phone. They initially gained traction as communication devices, but successive waves of adoption followed as they became cameras, navigation tools, and platforms for social media. We’re already seeing another example with Generative AI.
To avoid these pitfalls, organisations should shift their focus from technology to use cases and consider a few different techniques:
Segment Use Cases: Map out potential use cases and align them with the adopter groups most likely to embrace them. Not every use case will resonate with every segment.
Iterate and Adapt: Understand that the adoption curve isn’t static. New use cases can reinvigorate technologies that appear to have plateaued.
Solve Real Problems: The more specific and relatable the problem, the more likely a use case will gain traction among the early majority.
Communicate Differently: It’s worth tailoring messaging to emphasise the outcomes enabled by the technology, not the technology itself. Remember the fundamental reasons why people accept/adopt (or don’t accept/adopt) new technologies.
Put simply, the diffusion of innovation curve remains a powerful tool for understanding market adoption, but its application requires a deeper understanding of what actually diffuses. By recognising that use cases—not just technologies—drive adoption, businesses can craft more effective strategies to navigate the complex journey from innovation to mainstream success. Instead of asking, ‘How do we get people to adopt this technology?’ the better question is, ‘What problem can we solve so well that people will want to adopt it?’.
I think there’s some interesting parallels here with common misunderstandings related to applying the adoption curve to organisational and cultural transformation, but I’ll leave those thoughts for another day.
I write a weekly Substack of digital trends, transformation insights and quirkiness. To join our community of thousands of subscribers you can sign up to that here.
To get posts like this delivered straight to your inbox, drop your email into the box below.
Leave a Reply