Two of the most talked-about ERP implementation failures in the UK public sector happened at Birmingham City Council and the University of Edinburgh. Both organisations moved to Oracle Fusion Cloud Applications. Both then suffered serious disruption to finance, payroll and procurement. And both commissioned independent reviews that make uncomfortable reading for anyone who runs, sponsors or delivers an ERP programme.
Earlier this year I put together a presentation summarising the two reports and mapping what went wrong to best-practice ERP governance. This post is the written version. You can download both original reports from my Downloads page or from the links below.
Download the reports on both ERP implementation failures
- Birmingham City Council Oracle ERP Implementation – Report in the Public Interest – Grant Thornton, 11 February 2025 (PDF, 67 pages, 9.4 MB)
- University of Edinburgh People and Money Programme – External Post-Implementation Review – PA Consulting, December 2023 (PDF, 41 pages, 0.6 MB)
Birmingham City Council: a Report in the Public Interest
Birmingham City Council‘s programme set out to replace SAP with Oracle Fusion across finance, HR, payroll and procurement. It eventually went live in April 2022. The council’s external auditor, Grant Thornton, later issued a statutory Report in the Public Interest. That is a formal step auditors reserve for matters the public needs to know about, and the council must respond to it. It remains one of the most detailed accounts of ERP implementation failures in UK local government.
Overall, the report’s verdict is blunt. Governance and programme management had fundamental weaknesses that were never fixed. As a result, when the council decided to go live, nobody properly understood the level of risk.
Key findings
- Weak governance and accountability. Roles across the programme board, senior officers, finance leadership and suppliers were unclear. Independent assurance, senior ownership, member oversight and a strong design authority were absent or ineffective.
- Adapt, not adopt. The programme drifted from its principle of adopting standard Oracle functionality. Instead, it customised the system to fit existing processes, particularly in Finance.
- Optimistic reporting. Risks sat in the detail of reports rather than in the key messages. Consequently, senior stakeholders saw a more positive picture than the evidence supported.
- Go-live despite known risks. Payroll, data quality and integrations were still unresolved. In the auditor’s view, had the steering committee assessed testing properly, it would probably not have approved go-live.
- Over-reliance on suppliers. The council lacked in-house skills and capacity. Therefore, interim contractors and suppliers filled key roles, and knowledge walked out of the door with them.
- Business change underestimated. The programme engaged end users too late and too little. As a result, adoption suffered and workarounds multiplied.
The impact
The council went without an adequate financial management and cash receipting system for over two years. In addition, it could not produce reliable financial reports or auditable accounts for 2022/23 and 2023/24. The auditor estimated the cost of the failed implementation, and the fix, at least £90 million over the original budget.
However, it is worth being precise here. The report describes the ERP failure as a contributory factor in the council’s financial crisis, alongside equal pay liabilities. It was not the sole cause of the Section 114 notice. Even so, a working system would have given the council far more grip on its finances when it needed it most.
The story has not finished. By January 2026, The Register reported the forecast cost had reached £144.4 million, and the reimplementation had slipped again.
University of Edinburgh: People and Money post-implementation review
The University of Edinburgh‘s People and Money programme also implemented Oracle Fusion across HR, payroll, finance and procurement. After the Phase 3 go-live in August 2022, the university saw widespread disruption. For example, the university paid suppliers and postgraduate students late. Research finance also became hard to manage. Meanwhile, staff morale and trust fell sharply.
The university commissioned PA Consulting to carry out an external post-implementation review between July and October 2023. Importantly, its focus was lessons learned, not blame.
Its central conclusion is striking, and it echoes many other ERP implementation failures. People and Money worked as a technical system but failed as an organisational change programme.
Principal issues
- Governance. Structures were complex and changed repeatedly. For long periods there was no clearly accountable executive sponsor. Phase 3 went ahead despite “red” readiness assessments.
- Change management. It was under-resourced and applied too late. In addition, the programme rushed training and focused it on the system rather than new ways of working.
- Culture and trust. The programme deepened existing divisions between the centre and the Colleges and Schools. The review calls loss of trust the single biggest risk to future change.
- Strategy and scope. After COVID, the programme shifted from transformation to system delivery. The HR and Finance transformation programmes ended too early to embed the new operating model.
- Planning and delivery. Timelines were unrealistic, so the team re-planned them repeatedly. Meanwhile, the programme squeezed testing, training and change activity to hit fixed dates. There was no full technical dress rehearsal before Phase 3.
- Resourcing. Subject matter experts had little backfill. Similarly, the programme leaned heavily on short-term contractors.
Notably, the review found that most system problems came from design and governance decisions rather than Oracle software defects. Poor data quality, rushed migration, incomplete research integrations, restricted reporting and over-complex approval workflows were all choices or consequences, not bugs.
Mapping ERP implementation failures to best-practice governance
Put the two reports side by side and the same patterns appear. In fact, they are the patterns behind most ERP implementation failures I have seen. Below, I have mapped them to eight governance principles. For each one, I show what was observed and what good would have looked like.
Ownership, go-live gates, risk and change
| Governance principle | What went wrong | What good looks like |
|---|---|---|
| 1. A single accountable Senior Responsible Owner (SRO) | Birmingham: accountability spread across boards and officers. Edinburgh: no consistent executive sponsor, and governance resets blurred ownership. | A named SRO accountable to Cabinet or Court, with written authority to pause or stop delivery. |
| 2. Evidence-based go-live gates | Both went live with unresolved payroll, data and readiness risks. Hard dates beat evidence. | Objective pass/fail criteria, independent assurance and non-waivable thresholds for payroll and financial reporting. |
| 3. Active risk ownership and escalation | Teams logged risks, but leaders tolerated them. Concerns from Schools and workstreams did not reach decision-makers. | A named executive owner for every red risk, mandatory escalation to Audit Committee or Court, and automatic pauses when thresholds are breached. |
| 4. Change management as a core workstream | Business change came second to system build. Training and engagement were late and compressed. | Change readiness metrics reported alongside technical progress, with a protected change budget. |
Design, data, suppliers and culture
| Governance principle | What went wrong | What good looks like |
|---|---|---|
| 5. A business-led design authority | Birmingham customised instead of adopting. Edinburgh imposed contentious designs without buy-in. | Business ownership of process decisions, proven end-to-end scenarios and the ability to adapt designs after go-live. |
| 6. Protected governance for payroll and data | Both programmes underestimated data migration, interfaces and payroll complexity. Failures had direct human consequences. | Separate payroll and data readiness boards, extended parallel runs, and the right to decouple payroll from the wider go-live. |
| 7. An “intelligent client” with independent assurance | Heavy supplier dependence, weak challenge and poor knowledge retention. No clear independent assurance at Edinburgh. | An internal design and architecture authority, supplier payments linked to knowledge transfer, and assurance reporting outside the delivery chain. |
| 8. ERP treated as an organisational risk | Both failures compounded existing stress and damaged trust in leadership. | Wellbeing and capacity metrics in governance reports, visible leadership accountability, and go-live sequencing around organisational resilience. |
Why these ERP implementation failures were not about technology
Technology did not cause either of these failures. In fact, both organisations had the trappings of governance. Boards met, staff wrote papers and teams logged risks. Yet no governance layer consistently used its authority to stop, reset or rethink delivery when the evidence pointed to failure.
That is the lesson I take from both reports. In short, having governance is not enough. Someone has to own the outcome, read the evidence honestly and be willing to say “not yet”. In my experience, that willingness is far rarer than a well-drawn governance chart.
So, if you are planning or running an Oracle Fusion programme, both reports are worth reading in full. They are candid and detailed, and the problems they describe were avoidable. You will find them, along with my other papers and presentations, on the OAB Downloads page. After all, learning from other people’s ERP implementation failures is far cheaper than repeating them.
Need an ERP implementation review?
If your organisation needs an implementation review, whether before go-live, during a troubled programme or after the event, I can help. I am an independent Oracle Applications consultant and solutions architect with over two decades of experience implementing Oracle Financials and Procurement, and I have extensive experience of reviewing ERP implementations. Contact me to talk it through.