How Do Good Ideas Spread? Why we should support it, when we should not, and why we often talk at cross purposes
by Ruth Puttick
Spreading “ideas that work” is talked about day in and day out by many governments and those who support government improvement efforts. It is taken as a given that sharing and learning can be a better use of resources than always inventing from scratch. And that using better solutions can solve problems, generate value for residents, and help to save money for governments. But what does spreading ideas really mean? How does it happen in practical terms? And do we spend enough time stopping the old, ineffective interventions as much as we advocate the adoption of new ones?
Ideas are in the imagination
First, let’s get precise about what we mean by “ideas”. How to spread things that work is a question that gets asked all the time. It feels obvious. Why not share good ideas? But there is a lack of precision in this statement, and the lack of precision matters.
Under “idea”, the dictionary says, “Idea may apply to a mental image or formulation of something seen or known or imagined, to a pure abstraction, or to something assumed or vaguely sensed”.
But that’s not what we mean when we talk about sharing ideas, is it?
The ideas we advocate for spreading are not often “imagined”. Instead, the examples we can point to, like Care Blocks in Bogotá to better support caregivers, restorative justice that is reducing offending around the world, or the use of phonics to ramp up literary attainment across schools, are not imaginary. These are real, tangible, operationalised solutions backed by compelling evidence of impact.
In other words, what we are often talking about is not an idea but instead something being used in at least one place, maybe more, and can be policies, programmes, practices, interventions, new technologies, or governance structures, such as models for teams or units (for brevity, I’ll just refer to them all as ‘interventions’). You might just ask if this is just a matter of semantics. Does it matter if we call something an idea when it isn’t? In a few ways, yes, it does.
Ideas need to be separated from interventions that are more developed so that we use the most appropriate approaches to spread, deliver, and, crucially, to ensure that they work, are worth sharing, and continue to work as they are adopted more widely.
Is it an idea?
To identify if something is an idea, it can be helpful to look at the innovation process. Nesta, the UK’s innovation agency, uses the innovation spiral to break down the stages of innovation, from problem recognition and idea development to systems change.
The innovation process separates interventions that are at the idea stage from solutions that are more developed and are being used, delivered, or implemented.
Is it an intervention worth spreading?
Establishing if something is an idea and knowing how well developed it is will help guide testing and evaluation to build our confidence in its impacts.
For example, if you are developing an early-stage intervention that has not yet been developed, you will test it differently, using different methodologies and generating different types of data compared to an intervention that is well-established and operational in multiple locations.
This is where standards of evidence can be useful. When I was at Nesta, I helped develop its Standards of Evidence framework, This helps bring interventions being developed everywhere from government, the private sector, or philanthropy, in line with academically recognised levels of evidence, but at a pace that can be tailored to suit different interventions and contexts.
The Standards of Evidence model helps to answer important innovation questions: “Is an intervention doing any good? Is it even harming? Is the status quo just as good as the innovation?”.
Rather than being led by a particular type of data or evaluation method, the Standards of Evidence help guide evaluation design that is useful and proportionate to the intervention’s stage of development.
How can ideas be spread and adopted?
Now this is a question that has occupied practitioners and researchers for decades.
There are a lot of practices and academic theories to look at. Here are a few insights:
- Finding new interventions
- What is trendy? Ideas often take root and spread because of fashion and trends. Sometimes, these interventions are worth sharing and make a real difference. Other times, the promise doesn’t match the reality.
- New ideas are identified from a variety of sources. New solutions can be identified from scanning reports and publications, websites, blogs, engaging with professional contacts, the media and word of mouth, conferences, and site visits. There is then the entire universe of academic research, including individual studies through to reviews that systematically collate all that is known on a topic. There is a lot of information for decision makers to navigate. Often, little is known about which sources or modes of communication are most effective at promoting the best interventions and ensuring they are delivered.
- Governments are under pressure to adopt new solutions. Policy makers are “under increasing pressure to 'get a move on' — to keep up with the latest trends and “hot” ideas that sweep into their offices, to convert those ideas into locally appropriate “solutions” and “roll them out”. A question we should ask ourselves: Do they actually need something new? Do we know enough about the status quo to confidently say that it needs replacing? I’ll return to this point in a moment.
- The spread of interventions is relational: influencers and networks play an important role. The diffusion of innovation theory shows that solution adoption is predicated on the identification and engagement of influential actors and organisations. As well as believing an innovation will have benefits, the potential adopter will draw upon the evaluative judgments of trusted and respected individuals, often called “informal opinion leaders”.As an example, this study shows the impacts of opinion leaders on the spread of data-driven practices across cities in the USA.
2. How to adopt and adapt
- Why an idea is adopted is a continuum, spanning lesson drawing, voluntary adoption, through to coercive transfer where an intervention is imposed upon a government. There is an interesting power dynamic to reflect upon here. When interventions are being promoted and spread, how much control does the recipient have? Is it what is best for their context and residents?
- What is adopted will change and evolve. There is often an implied “implicit literalism” whereby it is assumed “fully formed, off-the-shelf policies” are “imported”. McCann emphasises this point, arguing “Policies, models and ideas are not moved around like gifts at a birthday party or like jars on shelves, where the mobilization does not change the character and content of the mobilised objects”. Instead, there is adaptation and alteration.
- There is a continuum in the degree of transfer. This ranges from an exact “photocopy” of the original intervention, hybrid adoption, selective imitation, or using the intervention as inspiration to create a new approach. Across all of these, there needs to be ongoing testing to ensure the interventions work in their new context and with their iterated design.
- Good adoption balances adapting the intervention to its new context whilst not losing the ingredients that made it successful in the first place. Linked to the point above, how much change and adaptation can there be, and should there be? There is much to be learned from implementation science, which goes beyond just looking at the spread of interventions as a linear process to focus on putting solutions into practice, ensuring they continue to work and have an impact.
Stopping as much as starting
So, we have discussed why and how it can help to spread effective interventions. Yet something that is rarely discussed is stopping doing things.
Some former colleagues wrote a fascinating report, Art of Exit, where they sought examples of the incumbent approach being stopped (or decommissioned, as we’d say here in the UK) to enable a new intervention to flourish. What did they find? That examples of outdated, outmoded, or ineffective services being stopped are rare.
It seems like stopping doing something is hard. Do we do it enough? Are we so fixated on the new and novel that we don’t spend enough time and energy stopping those things that don’t work or are no longer needed? Stopping probably warrants more attention than it gets.
The work never ends
The spread of ideas, or the quest for “what works”, should not be seen as a one-time endeavour. The world’s problems will change, we will change, and new issues will emerge. We should spread interventions for as long as they are better than the status quo and for as long as they deliver impact. And we must stay aware of emerging challenges and continually develop and test new ideas and interventions.
What does this mean for those who support the spread of ideas? To put it simply, the insights summarised here prompt us to think about language. Being precise matters. Ideas, compared to later-stage interventions, require different approaches to spreading, delivering, and testing. We are also prompted to reflect on the power dynamics of transfer. We also need to think why a government is seeking a new solution, if the solution is better than the status quo, and if the solution is adapted and changed, what does that mean for its continued success? If it worked in one place, will it still work elsewhere? And the spread of ideas should not be seen as a one-time endeavour. Nor should it be seen solely as spreading new interventions; we should also concern ourselves with stopping those approaches that are ineffective.
