In November 2019, West Sussex County Council made what seemed like a straightforward decision. The council would spend £2.6 million to migrate from legacy SAP payroll and HR systems to Oracle Fusion. The project was supposed to go live in 2021.
It is now March 2026. The council has just announced another six-month delay, pushing the rollout to October. The actual cost of the programme now sits at £41 million, a fifteenfold increase from the original budget.
On the surface, this looks like a catastrophic failure of planning. Dig deeper, and it reveals something more systemic: a pattern of enterprise software migrations that government and large organisations everywhere have failed to crack.
How a project spirals. The West Sussex project did not blow up overnight. According to documents reviewed by the Register, the council reset its budget in 2021 to £14.07 million, including £7 million that March and another £7.07 million in October 2022. Systems integrator DXC Technologies was brought in under a five-year, £4 million contract in May 2020.
By the time auditors EY conducted their review, the warning signs were unmistakable: "weaknesses regarding budgeting, governance and risk management." In July 2023, the council discovered there was no real prospect of a deliverable plan. DXC's contract was terminated on 1 September 2023, despite the company winning lucrative contract variations.
Throughout its engagement, DXC was paid approximately £6.6 million, more than 50 percent above the original contract price. A council spokesperson explained the approach matter-of-factly: the authority made "a financial choice to use the receipts available" from asset sales to fund the project. In 2025-26, those sales rose from £4 million to £12 million, effectively converting property assets into an IT project that was failing to deliver.
West Sussex is not unusual. Research across the sector shows enterprise software migrations fail regularly. Almost 60 percent of SAP migrations fall behind schedule and budget, mostly due to underestimated complexity and expanded scope. Oracle ERP migrations fail at nearly twice the rate of other enterprise software projects, often spectacularly and expensively. A study of 14 federal COBOL-to-cloud migration attempts found the median project took 4.2 years to complete, cost 3.8 times the original budget, and delivered only 68 percent of scoped functionality.
These failures share common threads. Data from inconsistent legacy systems does not fit the stricter requirements of modern cloud platforms. Business processes documented nowhere except in the minds of long-serving staff disappear in translation. Vendors promise smooth migrations; the reality involves months of manual work discovering what the old system actually does.
The real scandal at West Sussex is not the eventual cost. It is that the council knew things were going wrong as early as 2023, yet continued pouring money into a vendor arrangement that was not delivering. The council is now over 20,000 users dependent on accurate payroll processing, particularly schools and term-time workers. No council wants that system to fail. The pressure to keep spending to avoid losing sunk costs is intense.
Why governance matters. The root cause rarely lies in technology. Rather, it lies in governance: clarity about decision-making authority, acceptance criteria, and who is responsible when things go wrong. According to research by ISG, weak governance rather than technical challenges is the primary culprit behind delays. When multiple vendors, contractors, and internal teams lack "clear decision-making rights, acceptance criteria, and responsibilities", scope creeps and costs balloon.
West Sussex had auditor warnings as early as 2018 about these exact issues. The council did not act decisively until five years and millions of pounds later, by which point the damage was baked in.
The uncomfortable question. Fiscal responsibility demands governments and councils stop treating legacy system modernisation as an inevitable expense to be managed rather than as a strategic decision that can be deferred, scaled down, or abandoned. The £41 million West Sussex will spend has a real opportunity cost. Those funds will not be available for services, maintenance of physical assets, or genuine innovation.
This does not mean councils should never modernise. It means they should do so only when the case is genuinely compelling, with clear governance, realistic timelines, and the discipline to kill projects when evidence shows they will not deliver. West Sussex's slow-motion failure suggests that discipline is in short supply.