Scaling data foundations for a fast-growing health tech company
Patchwork Health, a healthcare workforce platform needed a modern, reliable data stack that could keep pace with growth. I designed and delivered a production-ready platform while upskilling the internal team to own it.
- Greenfield cloud data platform delivered from discovery through go-live
- Empowered internal analysts and engineers with pair build and knowledge transfer
- Created a foundation ready for machine learning and AI initiatives

Health tech · Workforce management
About Patchwork
Patchwork Health is a clinician founded NHS workforce platform that helps organisations build safe, sustainable staffing. It brings rota management, digital staff banks (including collaborative banks), agency management, and workforce insights together in one platform. At the time of writing this case study, it supports over 200 healthcare organisations to improve fill rates, reduce reliance on agency spend, and give clinicians greater flexibility and wellbeing.
It was a rewarding contract to work on. Patchwork’s mission to support the NHS and improve sustainable staffing is genuinely inspiring, and the team I worked with were dedicated, collaborative, and a pleasure to partner with.
Learn more at: patchwork.health
The challenge
The client's legacy reporting tool had become a bottleneck. As demand grew, the team needed trustworthy, timely data and a platform that could evolve with new products.
They also needed a solution that could support strict compliance requirements in healthcare, while remaining adaptable to the fast changing needs of a scaling product team.
Key pain points
- Legacy reporting tooling couldn't scale with surging data volumes.
- Data refreshes were slow and brittle, slowing down operational decisions.
- Logic was duplicated across sources, introducing errors and rework.
- No path to advanced analytics or ML with the existing stack.
Approach
I began by scoping the requirements and researching suitable tooling options, running demos and reviewing vendor offerings. I then provided the client with pros and cons for each choice, including indicative costs, to support a well-informed decision. Once the client agreed on the design and tooling, I created a proof of concept to validate that the approach worked in practice before delivering the full end-to-end build.
I worked closely with the data team to keep things simple while upholding good practice. Not everyone was technical, so I introduced practical habits, such as version control and code reviews, in a way that was easy to adopt. By showing the benefits of a developer mindset, we embedded habits that will scale.
What I did
- Scoped requirements and clarified future needs with stakeholders in the data team.
- Researched tooling options, ran demos, and gathered vendor insights to evaluate suitability.
- Summarised pros, cons, and indicative costs to help the client make an informed tooling decision.
- Developed a proof of concept to validate the design in practice before full build-out.
- Built the full end-to-end solution once design and tooling were confirmed.
- Worked collaboratively with the team to balance accessibility and good engineering practice.
- Introduced version control, code reviews, and lightweight development workflows to encourage a developer mindset.
Outcomes
The new platform provides a single, reliable source of truth with unified logic across datasets, eliminating duplication and drift. Data refreshes now complete in under an hour (down from around six), giving the team fast, consistent visibility. The platform is robust, observable, and ready for the next phase, supporting machine learning and AI initiatives.
Results at a glance
- A single, unified source of truth with consistent logic that’s easier to maintain.
- Full data platform refreshes in under 1 hour, previously up to 6 hours.
- Stronger alerting, observability, and data reliability across the platform.
- Solid foundations to start exploring machine learning and AI initiatives.
Client voice
Dan has been an all-round data superhero whilst working with us. He has spearheaded the migration to a new data platform, which has involved complex requirements gathering and market appraisal, tool and data environment set-up and building, and lots of complex logic clarification to build and test data pipelines and automation tasks.
He's done all this whilst passing on his experience and knowledge to help upskill the rest of the team (and myself).
Throughout he's just been an utter calm and stable presence in the team, no matter what was thrown at him. He's added so much value to the massive transformational journey we've been on with our data architecture project. But more than anything, we all agree that he is just a genuine pleasure to work with and an all-round good guy!
Head of Data
Patchwork Health
Looking to modernise your data platform?
I can help you shape a practical roadmap, estimate costs and timelines, and work with your team to deliver it. Start small, ship early, and scale with confidence.