What Does Co-Production Mean in Research?

When Does Working With Stakeholders Become Co-Production?

A research team asks patients to review participant information. Another includes patient partners in recurring advisory meetings. A third develops the research question, methods, governance, interpretation, and outputs jointly with people who have lived and professional experience of the issue.

All three projects involve people beyond the conventional academic research team. But should all three be called co-production?

Probably not. Co-production makes a stronger claim about the relationship among the people producing the research. It is associated with sharing power and responsibility, recognizing different forms of knowledge, and working together rather than merely asking others to react to decisions researchers have already made.

The term is attractive precisely because it suggests genuine partnership. That also makes it easy to overuse. Understanding co-production therefore requires looking beyond whether stakeholders attended meetings and asking who could actually shape the research.

Co-Production Means Producing Research Together

What Co-Production Changes About the Research Relationship

Co-production is about how research is produced, not a particular method

Co-production does not tell you whether to conduct interviews, surveys, experiments, observations, or statistical analyses. Those choices still follow from the research question and methodological requirements.

Instead, co-production concerns how people work together to develop and conduct the research. NIHR guidance describes co-producing a research project as an approach in which researchers, practitioners, and members of the public work together and share power and responsibility across the project, including in generating knowledge.

That places co-production within the broader landscape of participatory approaches to research, although the terminology and traditions are not identical.

Sharing power is central to the idea

Many forms of stakeholder involvement allow researchers to retain most decision authority. Researchers might ask stakeholders to comment on a questionnaire, suggest recruitment strategies, or respond to preliminary findings. Those contributions can be useful and meaningful without necessarily constituting co-production.

Co-production asks a harder question: which aspects of authority can genuinely be shared?

NIHR guidance identifies sharing power as a central principle, alongside including different perspectives and skills, respecting and valuing the knowledge of those working together, and reciprocity.

Power can appear in mundane research arrangements. Who determines the agenda? Who controls funding? Who decides that a methodological disagreement has been resolved? Who has access to data? Who approves dissemination? Who decides whose name appears on an output?

A project can use collaborative language while leaving all of these decisions untouched.

Sharing power does not mean everyone has identical roles

Co-production is sometimes interpreted as requiring every person to have exactly the same authority over every decision. That interpretation is difficult to sustain in real research.

Different people bring different expertise, responsibilities, and obligations. A statistician may carry particular responsibility for the defensibility of an analytical model. A community partner may have knowledge of local circumstances that the statistician lacks. A principal investigator may hold institutional responsibilities attached to a grant that cannot simply be transferred informally.

The point is not to pretend those differences do not exist. It is to avoid treating academic status as automatic justification for concentrating all meaningful influence within the university.

The appropriate distribution of influence among research partners may therefore differ across decisions.

Different forms of knowledge are meant to be valued

Traditional academic structures tend to privilege disciplinary, theoretical, and methodological expertise. Co-production explicitly creates space for other forms of knowledge.

A patient may understand the burden of a treatment in ways that clinical records cannot reveal. A social worker may know why a proposed procedure will fail within routine practice. Community members may understand historical relationships that affect trust in research. Researchers contribute different knowledge about study design, theory, measurement, and analysis.

Valuing these perspectives does not mean treating every claim as equally well supported. Experiential knowledge does not replace empirical evidence, and methodological expertise does not make researchers authorities on another person's lived experience.

Co-production attempts to create a relationship in which these different contributions can inform one another.

Co-production usually requires involvement early enough to matter

It is difficult to claim that research was co-produced if partners arrive after the question, protocol, outcomes, methods, budget, and major decisions have already been fixed.

This does not mean every co-production partnership must exist before an initial idea emerges. It means partners need to enter early enough to influence the decisions that the project claims they share.

That might include helping identify the problem, refining the question, determining priorities, developing methods, establishing governance arrangements, interpreting findings, or planning outputs.

In some projects, partners may participate in deciding what research question should be pursued rather than simply commenting on a question developed entirely by academics.

Co-production requires reciprocity

Research partnerships often generate obvious benefits for researchers: data, publications, grants, career advancement, institutional visibility, and evidence of engagement. The benefits to partners may be less automatic.

Reciprocity asks what different partners gain from working together. That benefit does not have to be identical for everyone. It might include learning, compensation, strengthened networks, useful evidence, influence over services, development of research skills, or outputs that communities and organizations can use.

Reciprocity should not be romanticized. A single research project cannot guarantee that every partner will receive everything they hope for. What matters is that benefits and burdens are considered rather than allowing academic institutions to capture most of the value by default.

Co-production is more than consultation

This does not make consultation inferior. Sometimes consultation is exactly what a project needs. A focused review of participant materials does not need to be relabeled co-production to become valuable.

The problem arises when researchers describe consultation as co-production because the stronger term sounds more collaborative than the actual relationship.

Co-production overlaps with PPI and participatory research

Co-production sits among several related concepts whose boundaries are not universally standardized.

Patient and public involvement can range across different levels and forms of involvement. A particular PPI project may use a co-production approach when patients, members of the public, researchers, and other partners genuinely share appropriate power and responsibility.

Similarly, community-based participatory research has its own intellectual tradition but shares concerns with equitable partnership, different forms of knowledge, co-learning, and power.

The concepts overlap without becoming interchangeable. Researchers should use terminology appropriate to their disciplinary and institutional context and then explain what the partnership actually involved.

Co-production can be difficult within ordinary research structures

There is an uncomfortable practical tension at the center of co-production. Research institutions are not necessarily designed for shared power.

A university may receive the grant and employ the researchers. Ethics systems may recognize academics as investigators while treating community partners differently. Financial systems can make paying partners cumbersome. Publication incentives reward academic outputs. Deadlines may be established before relationships have had time to develop.

These structural constraints do not make co-production impossible, but they can limit what researchers can honestly promise.

Co-production does not allow partners to determine the findings

Sharing power over research does not mean sharing the right to decide what the data should show.

Partners may legitimately influence the question, methods, analytical priorities, contextual interpretation, and communication of findings. They may challenge researchers' assumptions or offer alternative explanations. But conclusions still need to be supported by the evidence.

If results conflict with the expectations of researchers, community partners, funders, practitioners, or patients, co-production does not justify suppressing or rewriting them.

The distinction between influencing research and controlling its conclusions is central to improving research relevance without compromising the findings.

What Co-Production Might Look Like in Practice

The defining feature is not that researchers accepted every suggestion. It is that consequential aspects of the research were genuinely open to collaborative influence and that partners had substantive roles in producing the work.

Common Misunderstandings About Research Co-Production

Use Co-Production When You Can Share Something Meaningful

Before describing a project as co-produced, identify what is actually available for partners to shape. The answer should be more concrete than “they will provide input.”

Plan the partnership itself. Decide how meetings will work, how decisions will be recorded, how disagreements will be resolved, how information will be shared, what happens when someone leaves, and how different contributions will be recognized.

Substantial co-production also has resource implications. Partners may need to be paid for their time and expertise, and appropriate compensation should be planned rather than improvised.

Before Calling Research Co-Produced

Frequently Asked Questions About Co-Production

Co-Production Requires More Than Asking for Input

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.