Building smarter services starts with better data

More than a decade has passed since the establishment of the Government Service Standard, and much has improved in how public services are designed and delivered. We've seen a significant shift toward user-centred, agile, and service-led digital transformation. Yet, across central and devolved government, health, and regulatory bodies, we continue to encounter a persistent, foundational challenge: the quality, structure, and interoperability of public sector data.

Good data is the foundation of digital transformation

At its core, digital transformation is not just about shiny new front ends or migrating services to the cloud. It's about making services better, faster, and fairer for the people who rely on them. And in every case, that transformation depends on the quality of the underlying data.

From health records to building control applications, and from licensing to grant management, the problems we encounter often share a common root: fragmented, poorly structured, or inaccessible data. We’ve worked on projects where sports scientists were gathering unstructured application data on digital forms, where caseworkers had to re-key information across multiple systems, and where decades-old data is effectively held hostage by proprietary vendor formats.

This isn’t just inefficient—it’s risky. Poor data quality can lead to flawed decision-making, barriers to automation, and reduced public trust.

AI is only as effective as the data it consumes

There’s a great deal of excitement around AI and automation in the public sector. At Marvell, we believe in the promise of these technologies, and we’ve helped departments assess how to deploy them responsibly. But we also stress a critical caveat: AI is only as good as the data it’s trained or applied on.

While machine learning models can assist in data cleaning, classification, and insight generation, they cannot make up for fundamental weaknesses in how data is gathered, stored, and governed. In fact, bad input data can generate misleading, biased, or harmful outputs. That’s why we urge public sector organisations to focus on fixing data at the source—embedding better data practices throughout the service lifecycle.

What does "better" look like?

From our perspective, improving public sector data means working across several fronts:

1. Collect more structured data at source

Rather than relying on AI to clean messy data downstream, we should design services that encourage structured data capture from the outset. Good form design, controlled vocabularies, and thoughtful content strategy can reduce ambiguity and increase the value of data the moment it is submitted.

2. Join up data across systems

We regularly see services fall short because they treat data as bounded by the system that collected it. But citizen journeys rarely map neatly onto a single system. We’ve worked with a Welsh public body to implement cross-cutting reference data—linking health, education, and local authority datasets to deliver more accurate, reliable insights and services–and enable more comprehensive data analysis. The key enabler here is adopting data standards and investing in robust APIs and data-sharing agreements.

3. Improve data governance

Good governance enables safe and effective data sharing across departments. One example comes from a cross-departmental team of volunteers, working to support efforts to tackle modern slavery, where data from councils, the Food Standards Agency, and enforcement teams needed to be combined to identify risks and support victims. They highlighted that achieving this required not just technical integration, but careful data stewardship, legal assurance, and clear accountability.

4. Work in the open

Cross-departmental collaboration also improves when teams can see, share, and build on each other’s work. We always promote open standards, open source, and open documentation as defaults. This not only boosts efficiency and innovation but also avoids the vendor lock-in traps that have historically plagued public sector IT.

Avoiding the mistakes of the past

Many of the most frustrating data challenges we’ve helped clients solve stem from decisions made years ago—long-term vendor contracts with closed formats, brittle integrations, and unclear data ownership. As departments embrace new technologies, we mustn’t forget the lessons of the past. Data must be treated as a strategic asset: owned by the public, stewarded by the government, and protected with care.

That means designing for future flexibility: ensuring data is portable, standards-compliant, and decoupled from specific applications. It means investing in upskilling civil servants in data literacy and governance, and in creating the conditions where services can evolve without being locked into inflexible legacy solutions.

Looking forwards

If we want the next decade of public sector transformation to succeed, we need to go beyond service patterns and front-end redesigns. We need to tackle the data challenge head-on: not just with more technology, but with better decisions about how we gather, structure, and use information. Because only with good data can we deliver the great services the public deserves.

About Marvell Consulting

At Marvell, we’re passionate about helping the public sector deliver high-impact, sustainable digital change. We’ve supported departments and agencies across the UK central government, devolved administrations, and healthcare. Our multi-disciplinary teams combine research, design, and technical expertise to help clients build joined-up, user-centred services.

What unites all of our projects is a commitment to building with data in mind—from understanding the people behind the numbers to designing systems that generate and use information more effectively.

Central Government Programme activities

The techUK Central Government Programme provides a forum for government to engage with tech suppliers. We advocate for the govtech sector, evangelise tech as a solution to public sector challenges, facilitate market engagement, and help make the public sector an easier market to operate in. Visit the programme page here.