Project Failures

A project failure is not simply a project that ran late or over budget. In the cases examined in this category, failure means outcomes so far adrift from intent that they caused loss of life, destroyed companies, or required public money to unwind the damage. The Boeing 737 MAX programme cost 346 people their lives across two crashes in 2018 and 2019, grounded a fleet worldwide for twenty months, and produced one of the largest corporate settlements in aviation history. Theranos, a diagnostics venture valued at nine billion dollars at its peak, was revealed to be built on technology that did not work as claimed; its founder was convicted of fraud. Queensland Health's payroll replacement, initially budgeted in the low tens of millions of Australian dollars, ended up costing the state well over a billion dollars to remediate after tens of thousands of nurses and staff were paid incorrectly or not at all. These are not scheduling disappointments. They are organisational breakdowns.

PIA investigates project failures forensically. We are interested less in the dramatic moment of collapse than in the sequence of decisions, incentives and ignored warnings that made collapse likely. In nearly every major failure documented by official inquiries, court proceedings or regulator reports, the warning signs were visible inside the organisation well before the event. The JPMorgan Chief Investment Office episode — the so-called London Whale trades that produced losses of more than six billion dollars in 2012 — is a case in point: risk limits were breached, models were altered, and senior management received incomplete information. The failure was not one trader's position but a control environment that allowed the position to grow.

This category covers failures across sectors: aviation, healthcare technology, banking and public administration. Each investigation asks the same questions. What was promised, and to whom? Where did delivery diverge from plan, and when did leadership know? Which controls failed, and which were never installed? And what, concretely, should boards and programme directors do differently? The purpose is not retrospective blame. It is to extract decision-grade lessons from failures that others have already paid for.

Frameworks Applied in This Collection

Project Failures — Frequently Asked Questions