As Women in Tech Week is being marked across government and the wider community, I have been reflecting on my own experiences of leading digital transformation in a complex environment. Over the past year, I have had the privilege of serving as Service Owner for Manage People on Probation (MPOP), helping to take a new service from an early pilot to national rollout. As I prepared to move on to my next role, I wanted to share some of the lessons I have learned about leadership, delivery and transformation.
Manage People on Probation (MPOP) is a new digital service from Probation Digital that provides a single view of people on probation, helping probation staff manage individuals under their supervision, complete tasks, and record case activity more efficiently and effectively. MPOP provides time savings and quality improvements at scale, by making high-volume, everyday tasks simpler and quicker to complete.
Early in 2025, Probation Digital was gearing up to respond to growing capacity pressure in HMPPS, and MPOP was identified as a major potential time-saver for practitioners. The Ministry of Justice also defined our delivery into two kinds: speedboats, which are faster pilots and proofs of concept, and tankers, which are slower transformations of our core systems. MPOP was described as a speedboat and an exemplar, and was set to go live nationally within 12 months. I joined as Service Owner to mature the delivery model and accelerate the rollout. I relied on a steering group of senior leaders who were deeply bought into the vision, and a multi-disciplinary team that worked and deployed alongside our probation users.
Today, MPOP is live nationally, adoption has reached nearly 90% of the target cohort, which is made up of probation practitioners with a caseload, and the team has tripled in size.
I wanted to share my reflections on what it’s like to steer a speedboat and what makes it different from other products. This is what I learned.
Building around what practitioners needed most
MPOP was in Alpha when I joined the team. The team’s hypothesis was that we could help practitioners find the information they need to make effective decisions, reduce re-keying across systems, and join up the journey between them. In a pilot of 20 users, practitioners told us they wouldn't adopt another tool in an already crowded space unless it let them do more. We pivoted by rebuilding the roadmap around the highest-volume tasks practitioners did, and moving to a multi-squad model to deliver features in parallel.
Lesson 1: Scaling should happen alongside delivery
We didn't have the luxury of slow and steady governance gates deciding and assuring whether and when we can launch the service. Our challenge was that we were going to be live with all users within 12 months. To de-risk that timescale, I thought that the safest approach was to scale consistently alongside the building work.
After the initial pilot with 20 users, we onboarded 200 more practitioners to test our early product decisions at a larger scale. We tested our scalability by growing to 700 users and more challenges started coming up - the product load times increased as more people logged in, and we also started seeing the data quirks from the legacy estate that appeared when more people had their casework exposed through the new interface.
Onboarding at least one team from each of the 12 probation regions also meant we can test if the journeys work within the regional variances of practice. A big-bang launch would have shown us all that, but on a national scale with the whole organisation watching, so our decision to roll out steadily proved useful.
Lesson 2: The most complex feature shouldn’t set the pace
We knew early on that non-compliant appointments and contacts will be complex features. They required longer discoveries and more iterations. In order to continue speedily releasing value to users at the earliest opportunity we intentionally decided to not include them as our first set of functionality.
There were plenty of other high-volume, high-value features we could work on – we know practitioners schedule multiple appointments throughout their day and record a very high volume of contacts that require no further action. We shipped those journeys first and we added some “delighters” to them to add value, such as automation with practitioners' Outlook calendars, and SMS notifications for people on probation.
With a speedboat like MPOP it was crucial that different parts of the service moved at a different pace. This helped us ensure that we’re not bottle-necked behind complexity, and that we can ship value with the simpler features upfront.
Lesson 3: A speedboat can’t do everything
The benefits of speedboats mean that we can take people to their destination really quickly! But a speedboat was never intended to move the large complex infrastructure that is required for the future.
Behind the scenes, MPOP still reads and writes to NDelius, the legacy system and database that all probation staff use, which also means that our transformation and blue-sky thinking was limited by the current shape of the technology estate. Our scaled live service performance also demonstrated that the organisation needs larger, and more in-depth transformation of the data domains and core processes to support the new ambitious products in Probation Digital.
Approximately 6 months into scaling MPOP we kicked off additional work to address the foundations and technical architecture of Probation Digital. The Probation Digital Services platform is now responsible for moving the data, creating common capability, reporting and reference, and to unify the journey across services. I intentionally set that programme of work to be bigger and slower, with a 3-year roadmap and multiple dependencies.
Leading this takes optimism and what comes next
Transformation requires optimism. If you ever find yourself in charge of a speedboat, those three points above are my advice for you. We face many challenges and constraints - established ways of working and habits that are hard to break, a highly pressurised workforce, and a policy landscape that shifts constantly. My favourite descriptions of the complex work we do in digital government is "just a cold, dark night on the side of Everest". Nobody who signs up to climb Everest thinks it will be easy.
The platform is the next chapter, and the team building it will continue that story. It's a different kind of delivery: bigger and deliberately foundational, and it will require different lessons that I can’t wait to hear about.
Leave a comment