Why impact isn’t just a researcher's problem

I ended a recent talk for the impact group within EARMA (European Association of Research Managers and Administrators) by asking a question that I’ve been wrestling with for a while: are we trying to improve the system we have, or design the system we actually need?
 
It’s the kind of question that sounds simple in the moment but quickly becomes uncomfortable when you sit with it. There wasn’t a clear answer — there never is. But the conversation that followed via LinkedIn messages, emails, and in person has made something very clear to me: we are in the middle of a shift in how we think about research and its role in society, and we haven’t yet caught up with what that shift demands of us.
 
Over the past decade, research impact has moved from the margins to the mainstream. It is now embedded in funding criteria, national assessment exercises, and promotion processes. Researchers are expected not only to produce knowledge, but to demonstrate how that knowledge creates value beyond academia—whether that be policy influence, community benefit, or economic outcomes. In many ways, this is a welcome development. It reflects a growing recognition that universities operate governed by an unwritten social contract, and increasingly, that contract demands being able to show relevance, responsibility, and contribution for driving positive social change.
 
And yet, the way we have operationalised impact tells a more complicated story.
 
In practice, impact has largely been framed as an additional responsibility for individual researchers. It is something to be written into grant applications, articulated in promotion cases, and evidenced retrospectively once the research is complete. The implicit assumption is that if researchers are motivated enough, skilled enough, and ethical enough, they will be able to produce meaningful impact alongside everything else that is already required of them.
 
But this assumption rests on a misunderstanding of what impact actually is.
 
In my experience, motivation is not the issue. Researchers overwhelmingly want their work to matter. They want it to travel beyond academia and to be taken up in ways that make a difference. The problem is not a lack of intent but a mismatch between what we ask of individuals and what our institutions are designed to support. We have, in effect, confused individual virtue with systemic capacity.
 
This becomes particularly visible when we start to think about ethics. Research ethics frameworks, across contexts, are highly developed when it comes to protecting participants, ensuring informed consent, and managing risk. They are essential, and rightly so. But they are also largely oriented towards the conduct of research itself. They do not extend very far into questions about how knowledge is used once it leaves the confines of a project, who ultimately benefits from it, or how power is distributed in its application.
 
If we are serious about impact, these are precisely the questions that matter.
 
This is where the idea of ethical impact begins to take shape. It asks us to move beyond a baseline of “do no harm” towards a more demanding expectation of actively seeking to do good, while recognising that “good” is neither neutral nor universally agreed upon. It requires engaging with the reality that different groups may experience the same research in very different ways, and that the benefits of knowledge are not evenly distributed by default.
 
One implication of this shift is that co-produced research, while important, is not sufficient on its own. There has been significant progress in encouraging researchers to co-design and co-produce research with communities, but much of this effort still focuses on the research process rather than its outcomes. Ethical impact pushes this further by asking what it would mean to co-produce impact itself. This involves communities not only participating in research but also helping to define what success looks like, shaping how findings are interpreted and used, and having some degree of accountability over the institutions involved.
 
Another implication is temporal. Much of the current infrastructure around impact is oriented towards measurement, often at the end of a project. We ask what outputs will be generated, how impact will be evidenced, what indicators will be used, and how outcomes will be reported. These are not unimportant questions, but they come too late to influence many of the decisions that determine whether impact is meaningful in the first place. Designing for ethical impact requires attention much earlier in the research lifecycle, at the point where problems are defined, partners are identified, and pathways are imagined. It also requires resources to support this work before funding is secured, which most university systems are not currently set up to provide.
 
All of these point to a broader issue: the growing expectations placed on researchers are not matched by corresponding changes in the environments in which they work. Engagement, partnership development, and knowledge translation are often under-recognised, under-resourced, and treated as secondary to more traditional academic outputs. When impact does not materialise, the tendency is to attribute this to individual shortcomings rather than to the structural conditions that shape what is possible.
 
This is why it is increasingly difficult to sustain the idea that impact—particularly ethical impact—can be delivered by researchers alone.
 
A more useful way to think about it is to see impact as an emergent property of an ecosystem. Researchers are clearly central actors, but so too are institutions, funders, professional staff, and the communities and industries with whom research is conducted. Each of these actors influences the conditions under which impact is generated, from the incentives that guide behaviour to the resources that enable or constrain action.
 
Ethical impact is not something researchers deliver, but something ecosystems produce when the conditions are right. This framing shifts the focus away from individual responsibility and towards collective capability. It invites us to ask not only what researchers should do differently, but what institutions need to change to make different kinds of practice possible.
 
At present, there are examples of this happening, but they tend to be uneven and localised. Certainly, centres, teams, or initiatives have developed approaches that embed engagement, partnership, and impact more deeply into their work, and these pockets of practice are important, but they do not yet amount to a systemic shift. The broader landscape remains characterised by misalignment between expectations and support, ambition and infrastructure.
 
Which brings us back to the original question.
 
If the goal is to rebuild and sustain the social contract of universities, and if ethical impact is one pathway towards that goal, then incremental adjustments may not be enough. Adding new requirements, refining existing frameworks, or improving measurement approaches will only take us so far if the underlying system remains unchanged.
 
Designing a system that genuinely supports ethical impact involves rethinking how we define success, how we allocate resources, how we recognise different forms of contribution, and how we distribute responsibility across the research ecosystem. It also requires a willingness to confront some of the tensions and trade-offs that come with working towards “good” social change in complex and contested contexts.
 
Whether institutions are ready to invest in time, resources and staff for that level of change is an open question.

By Wade Kelly

Bio: Dr Wade Kelly is a Senior Lecturer in the Researcher Development Academy at Deakin University, specialising in developing researchers' impact and engagement capability across the career span. An influential contributor to the impact discourse in higher education, Dr Kelly brings extensive experience advising university leadership, faculties, institutes and funders on strategically embedding impact into policy, practice, and culture.