National products, local reality
What neighbourhood health has taught me about co-design in the real world
This blog post is a write up of a talk I gave at Camp Digital in May 2026
Over the past 18 months I’ve been working on digital neighbourhood health. That has meant using co-design inside a large national organisation, while trying to support local neighbourhood teams doing deeply practical work in complex places. What I’ve learned is that this kind of co-design is not neat, shiny or straightforward. But it is real. And sometimes, that is exactly why it works.
We spend a lot of time designing for users, journeys, services and products. But what happens when what you are designing for is not really a person, a role or even a single organisation? What happens when you are designing for a neighbourhood, and for the integrated neighbourhood teams working within it?
That is a very different thing to design for. Neighbourhoods have history. They have politics, relationships and power dynamics. Integrated neighbourhood teams work across messy organisational boundaries, different local contexts and geographies that do not map neatly to the structures national teams are used to. None of that fits into a tidy persona. But that is exactly what I have been trying to understand through this work: how national teams can support local places without flattening the very complexity that makes them work.
National intent meets local reality
Neighbourhood Health sits at the heart of a wider NHS ambition: moving care closer to home, shifting from sickness to prevention, and making services feel more joined up around people rather than organisations. At national level, that intent is compelling. But once you get into neighbourhoods, and into the places where care is actually being delivered, that clarity gives way to something much more complex.
There are multiple ways people talk about neighbourhood health, and not everyone describes it in quite the same way. But from the work we did, I think of it as an aspiration to deliver more joined-up, proactive and holistic care: keeping people well, reducing avoidable crisis points, easing pressure on general practice and hospitals, supporting people earlier, and working through multidisciplinary teams that extend beyond the NHS alone.
Take a 72-year-old woman living with diabetes, arthritis and mild frailty, and seeing multiple professionals in order to stay well. The problem is not that nobody cares. It is that care can too easily become fragmented across services, sectors and systems. No single professional necessarily has a full view of what is happening. Information gets lost. Effort gets duplicated. The promise of neighbourhood health is something better: more coordinated care, fewer gaps, more confidence for the person managing their health, and fewer avoidable hospital and GP visits.
Understanding the integrated neighbourhood model
We started where I would always advise people to start: by doing research and going out to local integrated neighbourhood teams to understand what was already happening. We expected to find very different models. And we did. Some places were focused on frailty, some on cardiovascular disease, some on COPD, and some on entirely different cohorts. And everyone was doing different tasks to deliver it. But alongside that variation, patterns started to emerge. No one had fully cracked it. Some places were doing parts of it brilliantly. Others were struggling. But reflecting those patterns back helped people describe the model they were trying to build, and what good might actually look like.
Get Rochelle Gold’s stories in your inbox
Join Medium for free to get updates from this writer.
If you zoom in on something as apparently simple as how people access a neighbourhood service, the complexity shows up quickly. People might come in through a professional referral, self-referral, a carer, outreach work, or because someone has identified them through data or local knowledge. Every neighbourhood team may handle that differently. Triage looks different. Contact looks different. And often, the people being supported do not know they are on a neighbourhood health pathway at all. They just know someone is helping. That is part of what makes this work so complex: it is not just about health, and it is certainly not just about the NHS.
Products are peripheral. People are infrastructure.
One of the clearest things from our work was that local teams often did not know what national digital products existed, whether they would help, how to access them, or who to speak to. From a national perspective, the instinct can easily become: here is the technology, here are the tools, here is the answer. But local reality was a fragmented digital estate that no single product could neatly fix. In one place, we mapped the tools being used across the integrated neighbourhood team and Post-its stretched as far as the eye could see. The last thing anyone needed was another tool.
As we mapped the service, something else became obvious. Integrated neighbourhood working does not work because systems, tools and processes are perfectly joined up. It works because people are. Information was not flowing neatly through digital systems. It was flowing through trusted relationships. Care was being coordinated because people knew each other, knew their communities, and stepped in to bridge the gaps. That raises a difficult question: if trusted relationships are what make this work function, should digital always be trying to remove those gaps? I do not think the answer is straightforward. But I am sure of this: in neighbourhood work, relationships are not a nice-to-have. They are part of the infrastructure.
Local teams did not always know what products we could provide or how they might help. But we had our own learning gap too. We did not properly understand local reality, the fragmented estate, the nuance, or the fact that much of what already existed was working, even if imperfectly. That is where co-design mattered. And one of the biggest lessons for me is that co-design is less a method than a behavioural commitment. National and central teams hold power. We own roadmaps, resources and access. We are closer to policy. In many cases, we also are or have historically been the commissioners. So we had to shift from designing for to designing with.
That sounds simple. It is not. Doing with, not to, is uncomfortable when you care deeply about what you are building and believe you can help. I had to hold back brilliant, well-intentioned product teams who were ready to jump in. What they did not always understand at first was that when I said we were going to do this with, not to, I meant them as well. Co-design was being seen as a methodology that we did with others, rather than with everyone as partners and with ourselves as co-participants. We had to create the conditions for local expertise and national digital expertise to sit alongside each other more equally.
Crowdsourcing through co-design
There was also a practical reality: we did not have a fully resourced team for this work. At one point, it was essentially me, a service designer and a user researcher who was about to go on maternity leave. If you want to do this well, you need capacity. So rather than trying to scale a team we did not have, we scaled participation. National product and delivery teams wanted to be involved, so we brought them in not as owners or saviours, but as co-learners. In effect, we crowdsourced the capacity we did not have.
That did three important things. First, neighbourhood teams got direct access to the people behind national products and could build relationships with them directly, not through decks or show-and-tells. Second, national teams were exposed to the real-world complexity their products needed to support. Third, and perhaps most importantly, it became genuinely developmental for staff. People saw more clearly how policy, funding and power land in real lives. They got closer to local delivery. And they were reminded of ‘the why’ of their work and how what they do matters.
This work is ongoing and, I so I can’t say what we will have codesigned in terms of product or service. What we have co-designed is a shared understanding, a way of working, and enough trust to keep moving together. Timelines have stretched. Getting the right people in the room takes time. Keeping local teams engaged takes time. Making sense of what we are learning takes time. But what emerged was more useful than a tidy artefact: a practical model for how national teams can work with local places on messy, cross-cutting problems.
Before better services come better relationships
For me, this is the real lesson. Co-design stopped being mainly about workshops and artefacts. It became about how you behave when you hold more power than the people you are working with. It became about recognising the privilege that comes with being in a national team, and recognising that you are not the expert in the care you are trying to support.
In neighbourhood work, relationships are often the system that makes everything else possible. So yes, I think we can co-design in the real world. But it asks more of us than sticky notes and good intentions. It asks national teams to step into local reality rather than observe it from a safe distance. It asks us to treat learning as two-way, to design for place rather than polish, and to accept that sometimes the most important thing we build is not a product at all, but the trust that makes better work possible later.
