Doing co-production (well): it’s not rocket science, it’s harder than that – and we need to talk about it.


Kate Tobin | Scotland Director | @katetobin_

When I’ve asked folk what good co-production involves, I’ve received variations of the following responses: ‘It’s just principles of good youth engagement’, ‘It’s not rocket science’ or ‘There’s a toolkit for that’.

These responses do a disservice to good co-production. It requires skills and experience to do well – and we undervalue this when we treat it as easy, or obvious. And when we undervalue these skills, or the importance of a robust process, the people who lose out are the users and practitioners who give their time, energy, and honesty.


Why is co-production important in service design? 

At the Dartington Service Design Lab, we see good co-production – alongside high-quality evidence and an understanding of the wider system – as vital in designing services for children, young people and families that are needed, wanted and likely to make a difference. To unpack that a bit more:

Needed: Are there data or insights to suggest there is a problem to address or a useful gap to fill?

Wanted: Is there demand from users for the service or response, and is it acceptable to those delivering and interacting with it?

Evidence: Is there logic or evidence to suggest that this approach has a reasonable chance of leading to a positive difference?

All these elements are required for good service design.


Co-production Lab-style 

For us at the Lab, good co-production comes down to being specific at the outset about your rationale for using it. In our experience there are five stages at which you can think about involving users – not just service users, but practitioners, and partner agencies too.

  1. To generate insight and understanding about particular challenges

  2. To focus down on challenges that are the users’ priority

  3. To co-create services or adaptations to existing services

  4. To provide feedback to refine early prototyped designs or adaptations (before they are developed further and implemented)

  5. To create feedback loops to inform rapid-cycle testing and improvement efforts, as services are implemented in the messy real world.


“Good co-production comes down to being specific at the outset about your rationale for using it.”

The ‘double diamond’ below is a classic service design framework intended to help folks understand the stages of design. [1]

We use this simple visual diagram of the design process to map out why, when, and eventually how, you would seek to involve users.

The first section or ‘diamond’ goes wide to develop a rich understanding and insight about a given challenge from a variety of perspectives (i.e. practitioners, commissioners, service beneficiaries), narrowing down to focus on a small number of reframed challenges that are most important to users.


The shortlisted ideas will then be worked up into service prototypes. These are road-tested and refined by those who might deliver, interact or use the service to test acceptability. This sometimes lead to rejection outright. Sometimes it provides the information needed to refine it further.


The dotted ‘circle’, which we’ve added to the classic ‘double diamond’ represents the feedback loops and ongoing cyclical nature of implementation, testing and continual refinement.  Key data would be collected at different stages from practitioners and beneficiaries to inform the following functions:

  1. Accountability: Is what’s being delivered, actually being delivered?

  2. Implementation improvement: Is it working as expected, and if not why and what changes can be made?

  3. Proven impact: Is it making a meaningful and tangible difference?

  4. Generalisable learning: Are there insights being gathered through the testing that could benefit practice across the wider sector?


Considering these different functions of co-production, at different stages of a service design process, help determine who you engage to participate and the types of co-production activities. It’s more than asking someone to set up a young person’s forum and assuming it’ll work out. It’s not enough to simply set off with good intentions when it comes to co-production. Before you begin, you have to be clear about why you’re doing it in the first place.

  • If there is a lack of clear purpose – people get frustrated: ‘Why am I here?”

  • Consultation fatigue – being invited to provide views and experiences, which do not result in a meaningful, and visible change can leave people feeling disempowered.

  • In the field there is a lot of talk about embracing and learning from failure. But even this is wanted in a neat little package – in my experience, failure is messy. We need to be able to talk about the mess.


“We need to be able to talk about the mess.”



At the Dartington Service Design Lab, we want to contribute and foster a learning culture with those we work with (and within our team) that allows honest and open conversation about when, and how to do good co-production – and adaptations of changes in practice as a result of past challenges.

So, we’d love to hear from others about when co-production has been a success (and why), as well as those stories when co-production hasn’t gone to plan (and what insights from the experience would benefit others in the future).

In an upcoming blog, we’ll delve into some our own successes and fails, and some of the knottier issues around co-production: who is represented, who has agency, and who has responsibility.


Some useful co-production resources/links: