178 million hours. The hidden cost in Scotland’s public services
Over the years, our teams at Storm ID have spent thousands of hours inside Scotland’s public services, not just in meetings and workshops, but alongside the people who keep services running day to day.
We’ve watched healthcare professionals act as the human glue between disconnected systems, stitching together information that should already be coherent. We’ve seen experienced case workers re-enter data the organisation already holds, not because they lack skill, but because the system makes it necessary. Time that should go into patients, pupils, families and decisions gets quietly consumed by coordination, records maintenance and routine administration.
And if there’s one consistent truth across services, it’s this. The frontline isn’t short on commitment, ingenuity or resilience. It’s short on headroom.
The fastest route isn’t more AI pilots
Rightly, many public sector organisations are looking to AI to help them.
But AI will only truly matter in public services if it reaches the operational heart of work. That doesn’t happen through isolated demos or dozens of disconnected experiments. Impact comes from identifying repeatable service patterns and operationalising them safely at scale, and that is feasible in Scotland.
Our recent analysis found that 51% of Scotland’s total automation opportunity sits within one pattern: Case and Record Lifecycle Management.
That matters because it points to a practical route forward.
Scotland doesn’t need 50 separate AI strategies. The fastest, safest path is to build reusable, configurable AI components around proven patterns, and integrate them with the systems of record staff rely on. Done well, this approach reduces risk, avoids duplication and concentrates assurance effort where it belongs.
What we’re really dealing with is a capacity crisis
For years, it was easy to talk about this as an efficiency issue. A tech issue. A backlog issue.
But in practice, it’s a capacity crisis, with talented people spending too much of their time doing work that shouldn’t require their skills, chasing information, reconciling records, translating policy into process, coordinating handoffs, because complexity has been allowed to settle into the system.
It’s not that people aren’t working hard enough. It’s that the system is asking them to absorb complexity that should have been designed out.
A defining moment for Scotland
The recent Spending Review and the 2026/27 Budget make the central challenge of the next decade unmistakable: demographic pressures are rising, while public budgets are largely static. There is less slack left to cope, and less room for small improvements to make a meaningful dent.
We’re left with a stark choice. Accept a gradual decline in the scope and quality of public services or deliver a step change in productivity.
Why we wrote Automate Tasks, Not Jobs
That’s why we wrote the whitepaper Automate Tasks, Not Jobs, not to sell a future vision, but to test a practical proposition:
Where is skilled public service time being lost, and can we realistically get it back?
We analysed 50 high priority public service workflows across Scotland, spanning health, education, policing, justice, local government and more.
Across these services alone, we identified a baseline of 178 million staff hours spent every year on routine administration, coordination and records maintenance.
That is time that should be going into care, teaching and decision making. It’s the structural inefficiency that keeps the system under strain and makes the status quo increasingly unsustainable.
And there’s also a credible opportunity to reverse it.
By applying AI to specific, repeatable service patterns, Scotland could release up to 62.1 million hours of capacity annually by 2030, around 35% of the baseline hours we assessed.
We need to be clear about intent. This isn’t about replacing people. It’s about giving the workforce we already have time back, so more effort goes into judgement, care and decisions, and less into copying, chasing and reconciling.
In a system under pressure, nationally and locally, time back is the only currency that protects the frontline.
Making the shift from AI pilots
One reason AI so often disappoints is that it gets stuck in experimental mode, what we sometimes call ‘pilot purgatory’.
In many organisations, early AI adoption concentrates on personal productivity like summarising meetings, drafting documents, producing quick outputs. Useful, yes, but that’s not where the deepest capacity loss sits, and it won’t change the operational maths of public services.
Real impact comes from operationalising repeatable patterns at scale. That means:
- Designing AI components around the work that repeats across services
- Integrating them into systems of record, so staff don’t have to context-switch
- Putting governance and assurance in place so high risk use cases are treated as high risk
- Investing in readiness and adoption so change becomes real, not theoretical
That’s why the ‘51% in one pattern’ finding matters. It tells us the opportunity isn’t scattered evenly across fifty disconnected problems. A large portion is concentrated, and therefore addressable with a more standardised approach.
Trust and sovereignty are non-negotiable
We also can’t talk about AI in Scotland’s public services without talking about trust.
Many of the services we analysed, particularly across health and justice, handle highly sensitive data. Moving from pilots to live operations in these environments requires strong governance, clear accountability, and for some use cases, sovereign infrastructure.
The right approach is not ‘AI everywhere’. It’s matching risk to environment, and building confidence through traceability, assurance and operational control.
Now is the moment to design for time back
This matters now because the system doesn’t have slack left. Demand is rising. Budgets are constrained. And the workforce is already carrying too much administrative drag.
So, the question isn’t whether AI has a place in Scotland’s public services. It does.
The question is whether Scotland can use it to recover capacity, safely, responsibly and at scale, so the workforce we already have can spend more time on what only humans can do.
We should be honest about what won’t work. More isolated pilots, more disconnected tools, more clever demos that never reach the operational heart of services.
The opportunity is in repeatable patterns, implemented properly, integrated with systems of record, governed to match risk, and designed around how organisations, teams and individuals actually work.
If Scotland can release even a portion of the 178 million hours currently spent on routine admin, coordination and records maintenance, the benefits aren’t abstract. It means earlier intervention and more time with patients, better support for pupils and families, faster and more consistent decisions in social work, and more resilient justice and local services. Services that feel reliable again.
Let’s start designing for time back.
If you haven’t read it yet, Automate Tasks, Not Jobs lays out the case for a capacity-first approach, and a practical route to get there without compromising trust. And if you’re grappling with the same constraints, let’s talk about where time back would make the biggest difference.