Scaling what works: how to turn digital and data into better public service outcomes
During a Global Government Forum webinar, government leaders discussed what citizens will expect from the public services of the future – from hyper-personalisation and targeted intervention to instancy – and how to use digital and data to deliver them
The work of government has never been more complex. Public sector organisations face increasing pressures, not least growing demand for better, more joined-up public services.
During this Global Government Forum webinar, four UK public servants discussed the importance of making a tighter focus on policy and public service outcomes the norm, and how to drive greater use of digital and data to improve decision-making and take iterative action.
Alexis Castillo-Soto is group deputy director, digital missions and transformation group at the UK Department for Energy Security and Net Zero. His opening comment was that using data is critical both for the creation of policy and for the delivery and management of policy – but that it often isn’t harnessed for the latter.
“To come up with a hypothesis and then be able to change it based on the information you’re getting back is really critical,” he said
“If you look back, we have always looked at policy outcomes at the beginning and then tried to measure them at the end, whereas in my mind you should be measuring them on a constant basis and be able to use that information to tweak your policies, so that you get the right outcome.
“It might be [a] slightly different [outcome] to what you started with, but actually that’s the whole point of a digital approach. You’re iterating and trying to find out actually what works.”
A greater focus on outcomes and on the monitoring and evaluation of policy and services was a central part of the discussion and the other panellists agreed with Castillo-Soto’s point.
Michael Lewis, deputy head PD digital, data and analytics at the UK’s Ministry of Defence, highlighted that greater use of digital and data in government is essential to creating a future in which public services are more joined up, more proactive, in which needs are anticipated, and in which decisions “are based on a shared and trusted picture”.
“That way, support can be targeted and from the very start resources… [can be directed] to where they will actually create the most value,” he said, adding that what he expects to see in the coming years and decades is a big shift to a more outcomes-led approach that is based on evidence and insight.
Public Service Data.AI
Public Service Data.AI is the UK’s flagship annual event for civil servants working to unlock the power of data and artificial intelligence across government.
Brought to you by Global Government Forum and hosted by HM Government, the event will take place in London on 15 October 2026 and is free to attend for UK and international public servants.
Find out more about the conference and register to attend
Healthcare perspective: the Greater Manchester Care Record
Such an approach is already being seen in parts of government and the public sector. Dr Gareth Thomas, clinical director, digital and data products at Health Innovation Manchester shared examples of what can be achieved when healthcare decisions are made on the back of intelligently-pooled data.
The Greater Manchester Care Record is a digital resource that brings together information from the National Health Service (NHS) and other care services across all 10 Greater Manchester boroughs – which together serve around 2.8 million citizens – into one joined up record, so that patients’ information can be accessed by frontline health and care workers quickly and efficiently.
“It’s a single source of truth… [that enables us] to understand population health and need”, Thomas said, adding that in some cases data about an individual goes back 20 years “which means we’ve got a really good insight into their longitudinal care journeys”.
He gave an example of the sort of work the care record is facilitating. A research team scanned over 2.3m datasets, identified 5,000 patients at risk of liver disease and increased the number of people receiving what he called “active intervention” by around 75%, with an aim to increase this by a further 10%.
They have also looked at prescribing habits across all of the GP practices in the region – there are about 440 of them – and have been able to identify around 3,000 patients at risk of medication-related harm. Here, too, the focus is on intervention.
“These are the types of outcomes that the use of large datasets across a population can really start to deliver,” Thomas said. “It’s about using data to identify areas of need, intervene and manage patients differently.”
Read more: Maximising the potential of government data: insights from GGF’s Data Summit
The barriers to better data use – and risk vs a ‘do-nothing scenario’
No-one can deny that the use of data to produce better informed policy and services is needed. But as the panellists discussed, there are barriers to overcome before this can be done consistently across government and the wider public sector.
Lewis boiled the barriers down to poor quality data or data that is not standardised or comparable; data that is held within interoperable systems; governance, stewardship and a lack of shared definitions; poorly-designed, or a lack of, measuring criteria; the need to provide the workforce with the skills needed to use data effectively; and security and privacy constraints, guardrails and risk appetite.
He was stark in his assessment of the problem and what needs to change. “We’ve got pockets of excellence but the system is uneven – there are different tools, different data practices, different incentives. Until we standardise what these foundations are and make adoption easy, we run the risk of [continuing to] duplicate effort and argue about whose numbers are right and who owns what.”
Lewis said his team at the Ministry of Defence take an ‘understand, decide, act’ approach to using data for change that involves doing a stock check to identify baseline challenges, adopting a minimum data standard, aligning tools to that standard, and then sense-checking through a community of people who share the same challenges.
The community shares lessons and best practice and helps to coordinate action, acting, Lewis said, as a central point to enable decision making at the tactical, operational and strategic level, and “not just top-down but left to right”.
“Package all those decisions into governance [so] we’re all joined up in our approach [to data] and suddenly we’re all marching to the beat of the same drum”, Lewis said.
He also mentioned programme data standards disseminated by NISTA – the National Infrastructure and Service Transformation Agency, which is jointly managed by the Cabinet Office and Treasury.
Those standards “probably came five years too late”, he said, but he described them as “very, very good” and as creating “smooth sailing” in terms of data use across programmes and agencies.
Continuing the discussion around barriers, Thomas highlighted that these span people, process, culture, tools, and tech.
One of the points he made, elaborating on Lewis’s earlier mention of risk appetite, was the need to compare risks to what he called a “do-nothing scenario”.
This isn’t about being “in a position of equipoise”, he said, but of communicating that while doing new things often creates new risks, efforts should be made to understand the risks and to manage them in the name of progress. Digital and data, he added, should be seen as an “enabler” to new and better ways of working.
Read more: UK launches National Data Library with early years kickstarter project
Good communication the bridge between confusion and clarity
Pamal Sharma is director of digital, data and technology at the UK’s Centre for Environment, Fisheries and Aquaculture Science.
The centre is a public sector research organisation that uses data – it has over a century’s worth – machine learning and artificial intelligence to identify trends and provide analyses for the UK government and for researchers globally.
He reiterated that data should be used more consistently to create the right policies and provide the right services. “It’s the outcome that matters and the outcome should be what the public recognises for their own needs,” he emphasised.
On the barriers, he noted that a lack of communication between data providers and collectors, data owners and data users – who might each view data and its uses in different ways – was a problem.
He gave the analogy of cellular manufacturing, in which “one guy produces one part and then passes it on to the next guy and the [idea] is to do the best you can so that you make the next guy’s job easier”.
“I think that kind of thinking… would be very beneficial. You create standard platforms and there is an understanding that [the data will be] provided in the best form for people’s needs and [agreement] that it will be used in a particular way.”
Castillo-Soto has experience of creating the conditions for a similar approach.
The Department for Energy Security and Net Zero has had a data lake – a centralised repository that ingests, stores and allows for processing of large volumes of data – for five or six years. What was found initially was that data owners tended to have a protectionist view of their data, and not because they didn’t share in the vision of the project, but because “they didn’t want people to see in case their data was wrong”.
To overcome this, the team built a submission portal that checked data before it went into the data lake. “One of the benefits, particularly as we were working with local authorities, was that we were able to help them update their data, and in some cases, speak to their own supply base where they didn’t have the right data”.
This helped government organisations to feel part of the process and to see the benefits and, when the team invited submissions for a second round of data, they found they had so many volunteers that they had to turn some of them away.
The technology, though not easy, “was definitely much easier to control” than the people and culture elements, he said. Once people had been “brought along on the journey”, it was managing their expectations that became the challenge.
Rather than call people and culture barriers, though, he said he prefers to call them definers and enablers. “They are the biggest enablers in digital and data,” he said.
One of the best ways to galvanise people as enablers, he added, is to have “an open door policy” to data owners and specialists – “they’re very passionate about data and you should never dampen that down,” he said.
Then comes the need – as Sharma had noted earlier – to bring data owners and data users together to understand needs and work to meet them.
What Castillo-Soto found, he said, was that data owners and data users “had never spoken [to each other in-person). It was all via emails, or it was all via intermediaries. So, putting people that have different perspectives and different angles in a room was our biggest win”.
Lewis chipped in: “From my experience, people don’t engage with data because we tell them to, they engage with it because it genuinely helps them to do better. So that means agreeing what we’re using data for. We need to go back to first principles and ensure we pass the ‘why’ message to our colleagues and our people, and hopefully that way we get better insights from our data.”
Read more: UK government sets out five AI scenarios to stress-test policy
Demonstrate change, turbo-charge buy-in
Health Innovation Manchester has taken a similar approach to Castillo-Soto and the Department for Energy Security and Net Zero when it comes to enabling data usability and effectiveness through demonstrating the benefits and coordinating interests.
“It’s really important from a healthcare perspective that we’re clear that data is useful, usable and used,” said Thomas. As such, data is transferred between systems through standard coding languages, the quality of data feeds is “optimised and assessed” and organisations are “challenged and incentivised to ensure that those data feeds are live and active”.
Providing Greater Manchester Care Record usage statistics – it is accessed over 500,000 times a month by 30,000 clinicians, a figure that has risen by around 80% in the last two years – and communicating the sorts of inventions being made off the back of it are crucial to getting buy-in across the system, he said.
“The more we can make that case for change, demonstrate that the asset is trusted and used, and then say, ‘there’s a cash releasing benefit on the back of this, in terms of either time saved or new models of care delivery’, then this talks to the fact that actually the data is standardised, it is trusted, and it is fit for purpose”.
Read more: Transformation loading: leaders share insights on unlocking the power of data and AI in government
Looking ahead
Looking ahead to the next 10 to 20 years, the panellists said there is likely to be an increasing expectation from citizens for services that are fast and personalised, and that governments will need to further exploit digital and data to meet those expectations.
Thomas’ view is that citizens will increasingly expect to be “partners” in their public sector care.
“I think there’ll be a focus on co-production of information and individualisation of interventions, and I think that as citizens we’re absolutely right to expect that… So I expect far greater involvement and far greater co-production.”
What’s crucial, he added, is being “absolutely clear that patients and decisions are at the centre of transformation, and that we’re not increasing the inequalities” in healthcare, and that gaining and maintaining the trust of the citizen is “important to have front and centre of your mind at all times”.
Similarly, Sharma said he thinks instancy will be a growing expectation in the coming years. He cited analysis he did in a previous job that showed that those in Gen Z – the digital natives born between 1997 and 2012 – might only engage with a website for five seconds before moving on, demonstrating a trend towards shorter attention spans and a preference for very quick solutions to their needs.
Castillo-Soto sees public services moving towards “hyper personalisation”. Driven by the consumer market, this would see people interacting with government in a completely different way to what is the norm now – with a personalised digital service enabling people not just to interact with government but with other agencies and markets, too.
“This is where we are heading at a very fast pace. That’s what people want, and if [government] isn’t there, I’m sure they’ll will tell us about it,” he said.
The ‘Turning digital and data into better public service outcomes’ webinar was hosted by Global Government Forum and took place on 19 May 2026. Watch the webinar here to hear more of the panellists’ insights on topics such as:
– How to move to ways of working that make partnership/collaboration, shared accountability and a stronger focus on citizen outcomes the norm.
– Whether government organisations are too focused on continually collecting new data rather than using existing data better – and using AI to sift data.
– More on the development of success metrics, and monitoring and evaluation processes.
About Mia Hunt
Mia has been editor of globalgovernmentforum.com since 2019. She has 15 years’ experience as a journalist and editor and specialises in writing for civil and public servants worldwide, including covering sustainability policy and related issues. She has led the Global Government Women’s Network since it launched in 2023. Previously, she covered commercial property having been market reports and supplements editor at Property Week and deputy editor at Retail Destination. She graduated from Kingston University London with a first-class honours degree in journalism and was part of the team that produced The River newspaper, which won Publication of the Year at the Guardian Student Media Awards in 2010.
