In 2010, the Queensland Government launched what was supposed to be a routine payroll system upgrade. Queensland Health had 80,000 employees, covered by 12 industrial awards, six industrial agreements, and more than 24,000 distinct pay combinations. IBM won the contract with a $98 million bid. By the time the system limped into operation — after ten failed go-live attempts — the State had spent $1.2 billion. The official inquiry head, Richard Chesterman QC, called it "possibly the worst failure of public administration in Australian history."
It was not a one-off. It was a pattern.
Across federal and state governments, and throughout Australia's private sector, the inability to complete complex projects on time, on budget, and to original specification has become so normalised that we've stopped treating it as a crisis. It isn't. It is a systemic failure with a quantifiable cause — and in 2026, for the first time, we have the tools to fix it.
The Numbers Are Worse Than You Think
The global picture has been documented relentlessly since the Standish Group published its first CHAOS Report in 1994 — and across three decades of research, the findings have remained stubbornly consistent.
Estimated annual waste from projects that don't deliver a benefit or are abandoned in Australia alone. With a project success rate of approximately 64%, nearly one in three Australian projects never delivers on its original promise.
Source: CPA Australia / INTHEBLACK research estimateAustralia's Public Sector: A Chronicle of Incompletion
The government sector has the most visible — and most thoroughly documented — record of project failure. PMI's 2025 research identified the government sector as the leading industry for project failure, outpacing even the most complex private-sector environments.
Queensland Health Payroll System (2010)
The most expensive IT project failure in Australian government history. A payroll system for 80,000 health workers, covering 24,000 distinct pay combinations under 12 industrial awards. IBM bid $98 million; the final cost exceeded $1.2 billion. After ten failed go-live attempts, the system went live in March 2010 — and immediately failed catastrophically.
Myki Smart Card System (2005–2012)
Victoria's public transport ticketing system was plagued by delays and cost blowouts from inception. What began as a $494 million project eventually cost an estimated $1.4 billion — more than seven years late.
Australian Customs Integrated Cargo System (2005)
The Australian Customs Service spent approximately $225 million upgrading its cargo processing system — a 600% blowout from an initial estimate of $30 million. An ANAO review documented 13 contracts with four suppliers and 65 contract variations, with few records of the decisions behind them.
Why Projects Fail: The Seven Root Causes
After three decades of research across tens of thousands of projects, the causes of failure are not mysterious. They are well-documented, frequently repeated, and almost never addressed at the process level before work begins.
Unclear or Evolving Requirements
The Standish Group identifies lack of clear requirements as the single most common cause of project failure. PMI research shows 37% of executive leaders cite unclear goals as the primary factor. Requirements that evolve mid-project without formal change control create cascading rework that compounds with every iteration.
Scope Creep and Uncontrolled Change
Scope creep is implicated in the majority of budget overruns. It occurs when initial objectives aren't precisely bounded, stakeholders aren't formally engaged in change decisions, and there is no mechanism to calculate the downstream impact of each addition.
Weak Executive Sponsorship and Governance
The Standish Group consistently ranks lack of executive sponsorship as the top contributor to project failure, cited in 30% of cases. Without genuine sponsor engagement, scope decisions default to project teams.
Underestimated Complexity
Large projects fail at dramatically higher rates. The Standish Group's data shows that projects exceeding $10 million are more than ten times more likely to be cancelled than those under $1 million. You cannot accurately scope what you haven't modelled.
Communication Fragmentation and Silos
Over two-thirds of organisations identify communication breakdowns as a primary factor. When teams are operating from different planning artefacts, misalignment compounds silently until a crisis surfaces it.
Poor User Adoption of New Systems
70% of all software implementations fail due to poor user adoption. Change management is treated as a communications exercise rather than a redesign of the workflows and incentive structures that determine whether people actually use what's been built.
Absence of Real-Time Execution Intelligence
PMI's 2025 research found 54% of organisations fail to track KPIs in real time. Projects that miss deadlines and budgets often do so because the signals were present weeks earlier — but there was no system to detect them.
of enterprise generative AI pilot projects fail to show measurable financial returns within six months. The root causes are identical to traditional IT project failure: unclear objectives, underestimated complexity, and absent process intelligence.
Source: MIT "The GenAI Divide: State of AI in Business 2025"The Finis Approach: Process Intelligence as the Foundation
The evidence points to a clear conclusion: the organisations that consistently finish are not the ones with the most money, the best technology, or the most experienced project managers. They are the ones that have invested in process intelligence — the capability to model how work actually flows, identify where it breaks down, and intervene with precision before failure becomes inevitable.
Process Discovery
We extract event log data from your existing systems and interview your people to reconstruct how work actually flows — not how the documentation says it should.
AI Modelling
We construct a digital twin of your processes, apply constraint analysis and predictive modelling to quantify risk, complexity and automation opportunity before any decision is made.
Intelligent Design
We redesign processes and project governance architecture — simulating outcomes in the model before a single change is committed to in the real world.
Embedded Delivery
We build alongside your team and establish real-time execution intelligence — dashboards, AI alerts and accountability structures that make problems visible before they become crises.
Thirty years of project failure data has produced an enormous body of knowledge about what goes wrong. What it has not produced, until now, is the technology to operationalise that knowledge in real time — to build a living model of a project or operational process that detects the early signals of failure and triggers the right response before the crisis point.
That technology now exists. The question for Australian organisations in 2026 is not whether to use it. It is whether to use it before the next $1.2 billion failure — or after.
Find out what's recoverable in your organisation
In a 90-minute process intelligence session, we'll map your highest-risk process, identify the root causes of incompletion, and give you a quantified estimate of what AI-driven redesign could recover.
Request a Discovery Session