MPs on the Public Accounts Committee (PAC) were in new territory as they looked for lessons learned from a government IT project that was a resounding success.
At a meeting to discuss the Bank of England’s replacement of its core IT system, MPs made clear this was a break from the usual analysis of project failures.
The UK central bank’s IT transformation project to replace its Real Time Gross Settlement (RTGS) system was likened to an archaeological discovery that could serve as a blueprint for future government IT initiatives.
During the hearing, MPs learned that clarity from the off, an environment of openness, close integration with suppliers, and a “one team” culture were, among other things, key to the project’s success.
The £431m RTGS project achieved value for money and demonstrated good digital transformation practice, according to the National Audit Office (NAO). The spending watchdog said the overall cost increased by 15% from the £375m budget set in 2020.
The NAO said the Bank of England project offers lessons for other government departments.
PAC member Catherine McKinnell, MP, told the meeting: “I have to say it’s quite unusual to be looking at a success story and extrapolating from that why? There will be challenges about how we can do it even better in future across government, where we know there is much digital renewal that needs to happen.”
Clifton-Brown MP said it was important now to consider the question of “what can we all do together to help government?” to provide a blueprint to help “avoid the obvious mistakes”.
How not to manage an IT refresh
One example of a so far unsuccessful IT transformation project, which was fresh in MPs’ minds after being discussed by the committee in a February meeting, is that of the government-owned National Savings & Investment (NS&I) bank.
PAC committee chair Geoffrey Clifton-Brown MP described the NS&I project as “a full-spectrum disaster”. In a report, the committee of MPs said NS&I has no workable plan, lacks the skills to deliver it, and has no idea of the eventual cost of the project.
I don’t want anyone to go away thinking this was easy – there was a lot of work by a lot of people to get there Victoria Cleland, Bank of England
It was cited as a project that could learn from the success of the Bank of England’s RTGS replacement project. But MPs went further, suggesting wider government, notorious for huge overspends and delays, could learn from it.
The RTGS system replacement was no mean feat. It settles £800bn in payments a day through systems including the Clearing House Automated Payments System (Chaps) and bankers’ automated clearing services (Bacs). It has been in use since 1996 and is a part of the UK’s critical national infrastructure which ensures money flows in the economy.
The phrase “it is like replacing the engine of a jumbo jet in flight” is often used in banking IT to describe the complexity of changing systems that must remain accessible, reliable, accurate and efficient. This phrase could have been written for the RTGS project.
Clarity from the start
In terms of the early work done on the project, Cleland told MPs the most important starting point had been deciding what the bank wanted to achieve and setting out a vision for the new system.
“We knew from the very start that resilience was going to be at the very heart of what we were doing because this is part of the critical national infrastructure. But, if we were going to change it, we had to make sure we got the best from what we were doing, so as well as resilience, we were focused on what more we could do,” she said.
The bank recognised that clarity in terms of the end objectives was key. It broke down five key areas of focus: resilience, access, interoperability, risk management and what users wanted. On the latter, it liaised with firms in the finance sector to help it create a blueprint, which Cleland described as “a golden thread”.
Cleland stressed that its focus was “not just on going live, but staying alive”to ensure the system was built for the future. “Taking time and making sure we understood that vision. It was quite daunting, but getting the requirements right at the start was key,” she added.
The project ran for nine years, a length of time not afforded to most organisations. But the RTGS replacement is of national importance. With hundreds of billions of pounds of transactions being processed every day, major disruption is nigh on unthinkable.
The central bank planned meticulously and implemented the system stage by stage, with nearly a decade between its early blueprint in 2017 and its completion in 2025. While it was a long project, there were incremental benefits throughout.
Dave Ramsden, deputy chair at the Bank of England, said the organisation was afforded time because, while it lacked some features, the existing RTGS system worked. “We could keep the show on the road.”
Patience in the procurement process
A major “foundational decision” was made in 2018 when the Bank of England decided it would build its core settlement engine rather than buy one from the market. But it couldn’t build it on its own.
“Although we have a very significant technology function, we didn’t have the in-house capability and were not able to grow it to do all the work,” said Ramsden. “We had to have an effective partnership.”
But there was no rush into procurement either, with planning running throughout 2018 for what he described as a foundational phase of the project.
Although we have a very significant technology function, we didn’t have the in-house capability [for this project] and were not able to grow it to do all the work. We had to have an effective partnership Dave Ramsden, Bank of England
Ramsden told MPs: “We took our time through those early stages in working out how to best set ourselves for success. We didn’t actually embark on the procurement of the key technology partner until February 2019.”
This patience continued, and the Bank of England finally signed a contract with Accenture to build and develop the new platform almost 18 months later, in July 2020.
IT boss Nathan Monk told MPs there was a lot of pre-engagement with suppliers to ensure they were clear about what the central bank wanted.
It also carried out proof-of-concept exercises with multiple suppliers to evaluate their engineering capabilities. “It’s all very well talking about ideas, and another thing actually seeing engineers, hands on, building code,” said Monk.
The three companies in the final bidding for the contract were all incentivised, with those not chosen paid for work they had already undertaken. The bank also bought the intellectual property (IP) that resulted from the early engagements with suppliers.
“We wanted to ensure we kept the best of everyone’s design,” said Monk. “We had some really incredible people from companies that were putting forward really good technical solutions, and we wanted to have the IP for the design, so we could use the best of all the designs.”
The bank also signed a fixed-price contract, described by committee chair Clifton-Brown as “very unusual in a transformational programme”.
Monk said this was possible because there were about 2,000 requirements at the outset of the project, and a lot of work was done up front to ensure these were clear.
Creating a strong ‘one team’ culture
Cleland told MPs it put serious planning into bringing external supplier teams and in-house staff together as one team. “From the start, we said this is one team, and that we couldn’t have a rift from that.”
Key to this was developing the right culture within the project team. “We have been really passionate about creating a culture, especially with a partner coming in,” Monk told MPs. “Multiple things were done over time. Little innocuous things like, rather than giving someone from a third-party supplier a certain coloured lanyard, giving all people the same colour.”
He said they were trying to create an atmosphere of everyone belonging together, adding that the organisation instilled shared visions and outcomes.
“We did a hell of a lot in the early days trying to build the camaraderie, the relationships and people getting to know one another. We had a big team at one point, but it did feel like a family,” he said.
Another important area within the large team was transparency, with team members being made to feel confident that they could call out anything they thought was not as it should be, said Monk.
This approach led to changes to the project plan. “The team sentiment was really key, and our first replan came from listening to teams,” he said, adding: “One of the key criteria for go live was, ‘Are the team happy and are they comfortable?’”
To this end, the Bank of England adopted a “no surprises culture”, applying this internally and to the wider finance sector, which is dependent on the central bank.
Cleland said: “What I would say is red [alert] means we can ask for help – it doesn’t mean you should be hiding it and you should be scared.” The bank had a transparency channel through which staff could raise concerns about the project.
Acting on voiced concerns can avert disaster
A speaking-out culture should not be underestimated as a means of avoiding disaster.
For example, the NS&I project, described as a “full-spectrum disaster” by the Public Accounts Committee, has seen costs increase by £1.3bn since it started, according to a National Audit Office report. One of the criticisms by the PAC was that NS&I has a “good news” culture, which means decisions are not made and disagreements are not resolved.
NS&I is not an isolated case. Government IT projects have been plagued by failures for decades, and the consequences can go far beyond hitting taxpayers in their pockets. The project to introduce the Horizon IT system at Post Office branches across the UK is an example of a disaster that came about after warnings were ignored.
According to documents referred to in a 2022 Post Office scandal public inquiry hearing, feedback from live trials running Horizon in 300 offices in 1999 revealed serious concerns over the software, which was causing “difficulties and trauma” for subpostmasters struggling to balance their accounts.
A National Federation of Subpostmasters (NFSP) executive council meeting report from 1999 said: “The difficulties and trauma being experienced by some subpostmasters were giving rise to concerns for their health and emotional well-being. It was felt by some that a tragedy was not far away if something is not altered soon.”
These warnings were ignored and, as Computer Weekly has reported on extensively, disaster struck.
The final word from the Bank of England’s Victoria Cleland at the end of the committee hearing was perhaps the biggest lesson of all for excited MPs. “The [NAO] report, which is really great, makes it look as if we are a swan, but there was a lot of significant paddling going on underneath. I don’t want anyone to go away thinking this was easy – there was a lot of work by a lot of people to get there.”
