Show notes
A $6.9 million estimate. A $1.25 billion catastrophe. How fake deadlines, a rigged bid, and a steering committee allergic to bad news turned a routine payroll migration into one of the biggest project disasters in Australian history. What happens when project leadership treats risk management as optional, decision making gets diffused across committees, and a hard deadline turns out to be completely negotiable? You get one of the most instructive project failure case studies in project management history — and in this episode, Kim and Kate pour it out one painful lesson at a time. The subject is Queensland Health, the public healthcare provider for the Australian state of Queensland, employing roughly 85,000 people across about 300 sites. In 2005, their legacy payroll vendor gave them three years' notice: support ends June 30, 2008. Sounds like plenty of runway — until the organization waited two of those three years to act. By the time a consultant named Terry Burns (remember that name) ran the procurement and IBM signed a contract in December 2007, the original $6.9 million estimate had ballooned to a $98 million contract with just six months to deliver a system covering more than 24,000 unique payroll combinations. Why does this matter for project managers and leaders? Because almost nothing that went wrong here is exotic. Late project intake. Governance where a transition team, a steering committee, and an IT department all had partial authority but no single decision-making structure. Two weeks — total — for requirements gathering on a nine-figure program. A "lift and shift" strategy that preserved every broken process and demanded 2,500 custom software configurations to replicate them. As Kim and Kate walk through it, the episode becomes a masterclass in how ordinary project management failures compound into organizational catastrophe. The discussion highlights are equal parts horrifying and hilarious. There's the program's actual internal slogan, taken straight from the project materials: "Plan A or Die" — no continuity plan, no risk mitigation, no fallback. There's the June 2008 deadline that came and went with a simple change request and some extended support from the legacy vendor, prompting Kate's sharp observation that the date was fake and not fake at once: end-of-life risk is real even when the cliff turns out to be movable. There's the UAT phase in June 2009 that surfaced 2,422 severe defects — after which the steering committee simply reclassified high-severity defects as lower severity so the dashboards would show enough green to proceed. Watermelon project, indeed. And when frontline payroll staff kept failing the system in testing, leadership's answer was to remove the requirement for business sign-off entirely. The go-live in March 2010 delivered the reckoning: 35,000 payroll anomalies, roughly $400 million in accidental overpayments, zero-dollar paychecks for healthcare workers, and government debt collectors hounding nurses for money they never asked to receive. Queensland ultimately hired a thousand staff to manually prop up the new system — a 40 percent increase in labor over the old one — with a projected five-year cost to taxpayers of $1.25 billion. The commission of inquiry that followed, led by the Right Honourable Richard Noel Chesterman, revealed the final twist: Terry Burns, a former IBMer, had leaked a competitor's bid specs to IBM and browbeaten the selection panel into rescoring until IBM won. And when the state sued? A release buried in the fine print of a change order had signed away its right to litigate. Grab a drink and join us. Quotes from the Episode "Everyone's in charge, so no one's in charge." — Kate "If you've got a deadline, question the deadline." — Kim "I've heard the term watermelon project where they're green on the outside and red on the inside." — Kim "And you are paying that company that you're hiring to learn." — Kate "But ultimately, if your project options are plan A or die. Maybe sometimes the best option is die." — Kim Practical Takeaways Governance and decision rights — When multiple bodies share authority without a single decision-making structure, accountability evaporates ("everyone's in charge, so no one's in charge"). Deadline validation — Interrogate whether a project deadline is a true constraint or a negotiable date, and what options (like extended vendor support) exist beyond it. The watermelon project pattern — Recognizing status distortion, including reclassifying defect severity and removing sign-off requirements to force green dashboards. Parallel run as cutover risk mitigation — Why replacing a legacy business-critical system requires side-by-side operation and validation before go-live. Procurement integrity and vendor selection — The risks of conflicts of interest in vendor selection, of choosing a vendor with no prior experience delivering the solution, and of unreviewed contract fine print. Closing Reflection Nobody ever really "accepts" a risk on paper. When risks are realized, someone always pays—taxpayers, frontline workers, or the organization itself. Sometimes the bravest act of project leadership is admitting that Plan A deserves to die. 🔗 Links & Resources Mentioned PM Happy Hour Membership: pmhappyhour.com/membership PM Happy Hour Facebook Page