Why money isn’t enough to solve the problem of the Post Office Horizon replacement

Published: Posted on

WHSmith and Post Office shop front
West Street, Fareham (56) by Barry Shimmon, CC BY-SA 2.0

By Dr Alice Moore, Assistant Professor in Public Management and Public Policy
School of Government, University of Birmingham

One of the headline announcements from last week’s Budget was £1.8 billion for compensation to victims of the Post Office scandal. Yet the money Rachel Reeves allocated to replace the Horizon system and what’s being done to fix the underlying problems that caused the scandal is receiving less attention.

The project to replace Horizon with an in-house system has been plagued with setbacks since it started three years ago. Reeves promised £70 million in the Budget, in addition to around £200 million already spent, but this pales in comparison to the £1 billion the Post Office says it needs.

Five years after the High Court found fatal flaws in the Horizon system, Horizon is still in operation in branches across the country and Fujitsu is still getting paid for it. Why is it taking so long and costing so much to get rid of Horizon? The answer is that the Post Office are failing to take the opportunities of bringing the system back in-house and are replicating many of the problems that led to the scandal.

Unresolved complexity

One of the original problems with the Horizon system was its sheer size and complexity. It manages all transactions, from foreign currency exchange to driving licence renewals, going through all Post Offices, from busy town centres to rural village shops. The constantly growing and changing requirements of the system were one of the first problems Horizon faced when it was being developed and sowed the seeds of many of the disastrous bugs that emerged later. The contract started in 1994 with 366 requirements, which had already been changed 323 times by 1999.

Complexity is typical of government technology. Research shows that public services and organisations are inherently more complex than private ones, because they have more ambiguous goals, less autonomy, and more diverse stakeholders. All of this is reflected in government systems and is part of the reason it has such a patchy record implementing digital technology.

The Post Office has focused on replacing Horizon wholesale and less on understanding, managing, or breaking down the complexity the system must cope with. This is one reason the recent audit report to assess the Post Office’s request for more funding deemed the project “unachievable”. Trying to replace Horizon like-for-like risks recreating many of the same problems that plagued the original system.

Technology in isolation

Managing complexity does not just take clever technical solutions. It needs an understanding of how technology works in and with the real world. By bringing the system in-house, the Post Office had an opportunity to bring technology development and the rest of what it does closer together.

Unfortunately, the focus of the project has still been on fitting the technology around what already happens, rather than changing technology and other processes together. This is perhaps to be expected, as a programme of more widespread operational change might be a hard sell to the postmasters and postmistresses the Post Office has mistreated for so long. Nevertheless, other major public IT projects, including here in Birmingham, demonstrate the dangers of trying to squeeze technology around existing processes and neglecting wider transformation.

Again, the Post Office seems to be replicating the original problems with the Horizon development. It was a lack of understanding between the Fujitsu and Post Office teams that created a poorly designed system and allowed bugs to go undetected and ignored for decades. Yet back in May, sources inside the Post Office told Computer Weekly that the development team had again become siloed from the rest of the Post Office and trust had broken down. Whether they be a contracted supplier or based in-house, hived off development teams are a recipe for disaster.

Knowledge gaps

One of the main reasons that technology built in isolation so often fails, is that few people in an organisation have the expertise needed to spot problems or fix them. Like other public sector outsourcing initiatives, the Post Office lost or never had the knowledge or skills to scrutinise Fujitsu’s work on Horizon.

Developing the new system in-house is an opportunity to rebuild expertise within the Post Office. However, it is still relying on external contractors to manage much of the work. The danger of this strategy was highlighted in the spring, when budget problems meant that those external contractors left, along with their knowledge of the new system.

By sticking with the same model of a single system, designed around not with the organisation, and buying in rather than building expertise, the Post Office are replicating the same mistakes they made back in the 1990s with Horizon. The £70 million Reeves has promised will not solve this problem and may be throwing good money after bad if it’s not accompanied by more fundamental changes to the project.



The views and opinions expressed in this article are those of the author and do not necessarily reflect the official policy or position of the University of Birmingham.

Leave a Reply

Your email address will not be published. Required fields are marked *