In 1998, the Irish health service set out to modernise its payroll system. The initial budget was €9 million. By the time the project was abandoned in 2005, somewhere between €170 million and €220 million had been spent. The system never worked properly. Parts of it are technically still in use today, which is perhaps the most damning detail of all.
PPARS, Personnel, Payroll and Related Systems, is remembered in Ireland as a public sector embarrassment. It deserves a wider audience. For anyone who has sat across a table from a large systems integrator and tried to hold the line on scope, budget, and delivery, it is one of the most instructive failures on record.
A Mess Digitised at Scale
The project’s central error was not technical. It was conceptual.
At the time, the Irish health service was organised across multiple autonomous health boards, each with its own pay scales, overtime rules, shift allowances, and local agreements. These were not minor variations. They were fundamentally incompatible approaches to calculating what staff were owed. Rather than resolve that complexity before selecting a system, the project attempted to customise SAP, an enterprise platform designed around standardised processes, to accommodate every local quirk.
This is the oldest mistake in large IT implementations. Software does not fix process problems. It freezes them in place, makes them more expensive to change, and forces every future decision through whatever compromises were baked in at the start. PPARS did not digitise a payroll system. It digitised the disorder of fifteen years of fragmented health service governance, and then paid consultants to maintain that disorder indefinitely.
The lesson is not specific to SAP, or to payroll, or to the public sector. Any organisation that begins a system implementation without first standardising the underlying process is not buying a solution. It is buying a very expensive mirror of its own dysfunction.
The Consultancy Trap
More than €50 million of the PPARS budget went to external consultants, principally Deloitte. The arrangement was later described by a government review as one in which the health service had become “addicted” to outside expertise. That is a polite way of saying the organisation lost the ability to manage its own project.
This is a dynamic that appears, in different forms, across large financial services implementations. A bank or insurer engages a major consultancy to deliver a complex programme. The internal team — often thinly resourced and without direct experience of comparable projects — defers to the external advisers on scope, timeline, and cost. The consultants, whose commercial interest lies in the continuation of the engagement, have little structural incentive to simplify. Scope grows. Costs grow. The internal team, never having built the capability to challenge what it is being told, cannot effectively push back.
PPARS became extreme in this regard partly because there was no one on the buying side who understood the system well enough to ask the right questions. When the consultants said the complexity required more customisation, there was no internal voice to dispute it. When the timeline slipped, there was no internal team capable of assessing whether the revised estimate was credible.
The fix is straightforward in principle and consistently underinvested in practice: the buying organisation needs people who have seen these projects before, who understand how scope creep is introduced, who can read a project plan and identify where risk is being obscured. That knowledge does not come from a briefing document. It comes from experience and where experience does not exist internally, it needs to be built or hired before the contract is signed, not after things go wrong.
Accountability Without Consequence
The governance structure around PPARS was, by any standard, inadequate. Oversight was distributed across health boards and government departments in a way that ensured no single person was clearly responsible for outcomes. When the project ran into difficulty, it was passed between bodies. When costs escalated, they were approved incrementally rather than assessed against the original business case.
A single number captures this well: an employee was accidentally paid €1 million by the malfunctioning system. That is not a rounding error. It is evidence of a programme so far outside normal operating parameters that basic sanity checks had stopped functioning.
In financial services, the governance parallel is recognisable. Large transformation programmes, core banking replacements, regulatory reporting overhauls, data platform migrations, often suffer from accountability structures that look robust on paper and fail in practice. Steering committees receive updates designed to manage upward rather than inform. Project directors are incentivised to report green until the moment they cannot. By the time an escalation reaches the level of leadership that could actually intervene, the cost of stopping frequently exceeds the cost of continuing, and the organisation is committed.
The PPARS project was eventually shut down in 2005 by the incoming HSE chief executive, who described it plainly as a black hole. That clarity came seven years and approximately €200 million too late.
What Private Sector Organisations Should Take From This
PPARS is an extreme case. But the conditions that produced it are not unusual.
A complex legacy environment. Inconsistent processes across business units. A vendor with deep product knowledge and a commercial interest in the engagement continuing. A buying organisation without sufficient internal expertise to challenge what it is being told. Governance structures that distribute accountability until it effectively disappears. These are not features of Irish public sector IT in 2005. They are features of large-scale technology programmes in financial services today.
The practical implication is this: the most important investment an organisation can make before a large implementation is not in the software. It is in the people on the buying side who will manage it. That means hiring or developing staff who have direct experience of comparable projects, who know how vendor proposals are constructed, where the risk is typically hidden, and what questions to ask when a timeline is revised for the third time. It means structuring contracts so that knowledge transfer to internal teams is a measurable deliverable, not an aspiration. And it means building governance that gives someone with genuine authority the information and the accountability to make a hard call early rather than an expensive one late.
PPARS was eventually written off as a cautionary tale about public procurement. That framing is too comfortable. It was a failure of organisational capability, vendor management, and internal expertise. Those problems do not respect sector boundaries.
The question worth asking before the next large implementation is not whether your organisation has chosen the right system. It is whether you have the people in place to manage the people selling it to you.