Introduction
Every year, universities, funders and governments invest enormous amounts of time and money trying to connect researchers with policymakers. There are policy briefings, workshops, fellowships, networking events, secondments and advisory groups. The underlying assumption is simple: if we create more opportunities for researchers and policymakers to meet, better decisions will follow. But is this assumption borne out?
My work focuses on the use of evidence in policy and practice. After several years’ research in the area, I had an opportunity to put some of my lessons into practice when I had the chance to work inside the UK government to support research-policy engagement. Over the course of the last few years, my focus has shifted from ‘how can we get more researchers into policy spaces’ towards ‘how do we know what kinds of knowledge mobilisation initiatives work?
This paper examines the landscape of knowledge mobilisation initiatives, and summarises what we know, what we don’t, and what we need to do next as a research community.
Good intentions aren’t enough
Few people would argue against closer links between research and policymaking. There is clearly a role for evidence in solving complex problems such as climate change, health inequalities and artificial intelligence. We know very little about the most cost-effective ways to support and enable this role.
Over the past decade, the number of organisations dedicated to promoting research-policy engagement has grown dramatically. Universities have established impact teams. Funders have introduced new programmes. Specialist intermediaries such as the University Policy Engagement Network, SpHERE and others have emerged to help bridge the gap between academia and government.
While it is great to see enthusiasm and innovation in this space, in effect we often see a bewildering landscape of initiatives, which are often in competition with each other for limited policymaker resources and time. The initiatives are very varied, deploying different approaches, although predominantly adopting a ‘push’ model which seeks to promote the work of particular research groups or organisations, rather than respond to evidence needs identified by policymakers. Most concerningly, there is very limited evaluation with only 3% identified in our recent review with a publicly-available evaluation. This means the growth in activity is being fuelled by good intentions, rather than based on a robust, theory-informed evidence base about which kind of approaches are most cost effective for which needs. Which is more helpful, a fellowship or a reviews’ facility? What’s better value for money, a match-making scheme or a long-term investment in standing research capacity? At the moment, we can’t answer these questions.
We are also not yet able to say which activities are likely to lead to which outcomes. The goals of these activities are often high-level and diffuse, such as ‘increase in evidence based policymaking’. More realistic outcomes more be ‘upskilled cohort of researchers able to effectively engage with policy’ or ‘analytical capacity available to policy team increased’. But again, we don’t know which activities drive improvements in which outcomes. In many cases, organisations are investing significant resources without knowing whether their activities improve policy, strengthen relationships or simply create more meetings.
We know more than we think—but less than we need
Research into evidence use has expanded rapidly across many disciplines. Health researchers, education specialists, implementation scientists, science policy scholars and social scientists have all been asking remarkably similar questions, often using different language. This inter- or trans-disciplinary research community has surfaced important lessons, as have the related fields of political science, science and technology studies and sociology of science.
We know that evidence on its own rarely leads to policy change. Instead, research influences decisions through relationships, trust, timing, institutional capacity and supportive systems. This may sound obvious, yet many engagement activities still operate as though the challenge is primarily one of communication: if researchers could only explain their findings more clearly, policymakers would act on them.
One of the biggest shifts in our understanding has been moving away from a linear model of impact, which frames the problem as a ‘knowledge gap’ and the solution(s) as louder, more persistent broadcasting of evidence. This model imagines that researchers produce evidence, communicate it effectively and policymakers adopt it. We know that this is not the case.
In reality, academic-policy engagement activities should be designed to reflect the reality of policymaking. Policy decisions are made under pressure, with competing priorities, incomplete information and limited resources. Research is only one input among many. Understanding this does not diminish the value of evidence; it helps us design more effective mechanisms to improve access and absorptive capacity.
From broadcasting evidence to building relationships
One key lesson from the evidence use field is that relational work is critical to improving evidence use. It depends on long-term connections, mutual understanding and ongoing dialogue rather than one-off interactions. There is a large growth in relational interventions such as fellowships, secondments and pairing schemes which seek to give researchers and policymakers the opportunity to form and sustain meaningful partnerships. As expert knowledge brokers know, this is skilled, difficult work. Acquiring these skills can be a long, sometimes challenging process of learning by trial-and-error, with little training and only informal peer support and career development.
As this shows, we need to think about the relational element as being a necessary but not sufficient part of the picture. Increasingly, we should think about research-policy engagement as part of a wider system rather than a series of isolated activities undertaken by an atomised workforce.
Rather than asking, “How do we communicate this research?” we might ask, “What relationships need to exist for evidence to be useful?” and ‘what would it take for universities to identify, support and reward researchers interested in developing these skills’? We might begin to see funders incentivising collaboration between universities, reducing competition and enabling a needs-led engagement workforce.
And instead of celebrating individual examples of impact, we should consider whether we are strengthening the overall system through which knowledge informs public decisions.
A practical example – the Areas of Research Interest
One initiative that illustrates both the opportunities and challenges is the UK’s Areas of Research Interest (ARIs). ARIs are a public articulation of the questions or topics on which government departments and agencies are seeking input. They are a guide for researchers who want to better understand policy priorities, and can help researchers and funders identify opportunities for collaboration and engagement.
In principle, everyone benefits. Researchers gain a clearer understanding of real-world policy needs. Policymakers gain easier access to relevant expertise. Funders can align investments with emerging challenges. In practice, however, it takes analytical work and resources to identify opportunities for knowledge mobilisation and synthesis, as well as research gaps. It takes further, skilled relational work to make the most of these opportunities. Making ARIs effective requires skilled people who can connect departments, academics, funders and intermediaries. It requires organisations with the time and capacity to build relationships, identify existing evidence and coordinate across institutional boundaries.
Engagement has costs as well as benefits
One uncomfortable truth is often overlooked. Every fellowship, workshop or collaboration requires time, coordination and expertise. Government officials invest considerable effort supporting external researchers. Academics juggle engagement alongside teaching and research. Universities and funders dedicate staff and resources to facilitating these interactions.
When engagement is strategic and purposeful, those investments can produce enormous value. When it is poorly coordinated, however, they can also create frustration, duplicate effort and reduce goodwill. More engagement is not necessarily better engagement. This is why it is critical that we start to articulate the assumptions underpinning new academic-policy engagement activities, and to test these assumptions, reporting through publicly available evaluations. Only when we have a robust evidence base can we be sure that we are collectively investing in strengthening, rather than weakening our research-policy system.
The next challenge
If there is one lesson from decades of research, it is that improving the relationship between research and policy is not primarily about producing better evidence. It is about creating better systems for using it.
That means investing in knowledge brokers, strengthening institutional capability, supporting evidence synthesis, rewarding collaboration and recognising that relationship-building is skilled professional work rather than an optional extra.
This means accepting that impact cannot always be planned or measured through simple metrics. The conversations that shape future policy often begin years before a decision is made.
The challenge for the next decade is therefore not to create more engagement for its own sake, but to build healthier systems in which researchers, policymakers, funders and intermediaries work together more effectively. If we succeed, the reward will be more than better research impact.
It will be better public decisions, and ultimately, better outcomes for society.