Across government, the call to do more with less has become a familiar refrain - especially amid tightening public finances and recent spending reviews. Yet this ambition often clashes sharply with the reality on the ground: outdated, inflexible legacy systems are making meaningful transformation nearly impossible. Legacy technology accounts for an estimated 28% of the government’s IT estate, and outdated systems cost the public sector up to £45 billion annually in lost savings due to inefficiencies, duplicated effort, and missed opportunities to modernise services (DSIT, 2025). These legacy systems aren’t just expensive to maintain; they slow down processes, limit data sharing, and hinder the government’s ability to respond quickly to emerging demands. In short, legacy tech acts as a hidden handbrake on government transformation. While the State of digital government review clearly highlights the problem and sets clear intentions, the real challenge remains: how to deliver change at scale.
We all recognise that legacy is a significant challenge, and many departments are genuinely committed and working hard to tackle it. There’s no shortage of clear guidance, solid principles, and best practices to support this work - so why are we still struggling? The answer lies in the fact that legacy isn’t just a technical problem; it’s a systemic one. Solving it isn’t just about replacing code - it’s about changing the system that shapes how we prioritise, fund, govern, and deliver change. Until we acknowledge and address these systemic issues, our solutions will fall short. Truly fixing legacy means rethinking the entire ecosystem around how change happens.
In this article, we’ll explore three key shifts to help us make real progress on tackling legacy IT.
Funding transformation is one of the biggest levers we have in tackling legacy IT - but traditional funding models often get in the way. These models prioritise certainty: fixed costs, rigid timelines, and predefined solutions, all justified through long-term business cases. But legacy transformation isn’t a predictable journey - it’s complex, evolving, and often heading toward an undefined better state because legacy systems are deeply intertwined with outdated processes, multiple interdependent technologies, and shifting user needs. Moreover, the full scope of technical debt and hidden complexities often only become apparent as improvements are made, requiring continuous adaptation. This means the destination can’t be precisely mapped out from the start; instead, the path unfolds through iterative discovery and learning. The tendency to seek certainty comes from a rational place: how can anyone be expected to approve millions of pounds of public spending without knowing exactly what they’ll get? Historically, that certainty has come from tightly defined outputs. To move forward, we need a shift. We must be brave and draw confidence from using the right methods - adopting product-centric funding at the system level. That means embracing uncertainty, building a culture and leadership that supports iterative delivery, and most importantly, delivering - showing progress through integrated, incremental improvements that build both confidence and momentum.
Breaking big problems into smaller moves - is vital for addressing legacy IT effectively. A common challenge is the lack of appreciation for the legacy system’s strengths, which leads to isolated, big-bang replacement attempts that fail because they underestimate the scale and complexity of what’s already been built. Legacy systems’ failings are easy to spot, giving project teams confidence that a wholesale replacement will be better. But the system’s strengths - often developed over decades through subtle, nuanced changes - are overlooked, causing teams to miss critical context. Building a new system in isolation can seem simpler than integrating small, incremental improvements, while the complex task of migrating users often gets pushed to the end and becomes a costly hurdle. To shift this, we need to lean into the complex technical work required to improve legacy systems, celebrate small but meaningful changes that deliver real value like continuous improvement and tech debt, whilst getting better at telling the story - showing how incremental improvements contribute to tangible benefits, reduce material risks of IT systems becoming legacy IT and importantly help realise the broader service vision.
Disassociating big problems from big programmes is essential to making real progress on legacy IT in central government. The scale of the challenge can create a sense of urgency - and with it, a temptation to launch large programmes that promise speed and impact. But the evidence tells a different story: major IT initiatives routinely overrun on cost, miss deadlines, and deliver limited long-term value. This is often driven by a well-intentioned but flawed assumption - that big problems need big, powerful solutions. While that may apply to infrastructure projects, it doesn’t hold for complex public services shaped by policy, layered with interdependencies, and full of uncertainty. Still, a systemic mindset persists: the bigger the problem, the bigger the programme, and the more senior the leader. It can feel easier to reassure sponsors with scale than to be honest about what meaningful progress really takes. Instead, government must be realistic about its delivery capacity. It’s far more effective to invest in a small number of skilled, empowered teams than to stretch resources thin across a sprawling, high-risk programme. Though this may look slower on paper, it’s the path to real, sustained change - and to delivering better outcomes and value for taxpayers.
So what next? Legacy IT isn’t just an obstacle - it’s a systemic failure that demands a systemic response. Continuing with outdated funding models, conflating a programme's size with its likelihood of success, and assuming tech that predates this very calendar year as "broken" is dangerous. We risk the cycle of programme delays, overspends, and under-delivery continuing. Real progress means embracing complexity, accepting uncertainty, and committing to sustained, incremental change backed by leadership willing to challenge the status quo. Anything less puts taxpayers’ money at risk and delays the move to modern, agile public services which the country urgently needs. The legacy problem won’t solve itself - and frankly, we can no longer afford to wait.
Principal Consultant
Natasha joined Clarasys in 2022 as an experienced hire specialising in public sector transformation. She is passionate about improving the public sector and is driven to work on meaningful projects that make a real impact on society and people’s lives. She enjoys working collaboratively with public sector clients to solve complex problems with innovative solutions which support service transformation. She has worked on a range of projects from complex programme management, process improvement, service design and digital product ownership. Recently, she led a multi-disciplinary product team of 16 to design and implement a digital service for a central government department. This included leading senior stakeholder engagement with ministers, no10 policy unit, the treasury and C-suite leaders across UK businesses.
Principal Technologist | Cloud & Platforms
Department for Science, Innovation and Technology
Chris has spent over 25 years in large scale government IT programmes in the UK and abroad. He has been employed as both a Civil Servant and a consultant. Having created legacy, failed to replace legacy and more recently tackled legacy better, it is something really close to his heart.