FutureNHS — the next generation of a national platform
How the NHS’s FutureNHS platform defined its successor — more engaging, more scalable, cheaper to run, and open-source.
How the NHS’s FutureNHS platform defined its successor — more engaging, more scalable, cheaper to run, and open-source.
FutureNHS is the national collaboration platform for NHS staff — GPs, nurses, physiotherapists, department leads — sharing what they know across the thousands of organisations that make up the system. It hosts more than a thousand communities and supports over fifty thousand monthly active users, and it is unusual among staff-facing services in being owned by the people who use it: communities form around their needs and interests with no org chart in the way. During COVID-19 that usage surged, as health and care staff across the country leaned on it to share knowledge quickly and work together under real pressure.
The surge also exposed the limits of the existing platform. To meet the demand it had created, the NHS needed a successor — one that was more engaging, easier to use, more scalable, and less expensive to run.
As Product Director, I worked closely with the NHS's product and service teams to define how the next version of FutureNHS should work, and what it needed to deliver for the people using it.
The work sat across product, delivery, and technical decisions at once, and much of my contribution was keeping those three from drifting apart. We also made an early, deliberate choice about how to build: in the open. FutureNHS would be developed as open-source software, hosted on a public code repository (GitHub) and open to outside contributions, built on existing open projects with fixes contributed back upstream. The goal was a public good, and that stance shaped everything that followed, from how we ran research to how we made decisions with the NHS in the room.
We began with a mission plain enough to hold a programme together: increase collaboration across the health and care system while bringing down the long-term cost of running the platform — and we grounded it in evidence before shaping anything. Discovery ran on three tracks at once: dozens of interviews with the people who run and use the communities, a close read of support tickets and usage analytics, and a platform-wide survey that let every member point to what was missing. What came back became a single, prioritised backlog — every candidate feature scored on the value it would create against the complexity of building it, then sequenced into clear delivery milestones.
From there we designed in the open and tested early. We adapted the NHS.UK and GOV.UK design systems to the context of a staff-facing collaboration tool, and published our design and style guidance openly — in Zeroheight, alongside the code — so the NHS's teams and ours built from one source of truth. Working in Figma and Zeplin, interactive prototypes put navigation in front of real workspace managers and members, so we could watch where the signposting broke down and fix the wayfinding on screen before a line of it was built.
We mapped the wider system and brought Health Education England and NHS Digital into the service architecture we carried into beta. We held ourselves to the GDS Service Standard through discovery, alpha and beta, with governance from NHSX and GDS. We stood up a dedicated FutureNHS workspace where a growing community of beta testers — eighty and counting — shaped the product as it was built.
A public beta on GitHub — forty pull requests in a week, nine contributors — a design system published alongside the code, and eighty NHS beta testers shaping each round. Migration was designed to be phased, so no working community was interrupted. The full launch happened after my time; what I can vouch for is the shape it was given.
“We aim to become a truly digital organisation. We were looking for partners aligned with our vision — partners who challenge our thinking and improve our ways of working. The team showed a real cultural fit and alignment with ours.”
Building in the open taught me what trust as structure looks like at national scale. It is the question I keep working on as AI enters public services: systems people can inspect, challenge, and own.
A national platform earns its successor by being honest about what the current one costs to run. If you are planning that kind of rebuild, let’s compare notes.