Introduction

In October 2022, Deputy Prime Minister & Defence Minister Richard Marles declared Defence procurement a ‘complete mess’, citing a raft of mid-range over-runs ahead of the 2022 Budget.[1]

In February 2025, Minister for Defence Industry & Minister for Pacific Island Affairs Pat Conroy noted he had led "eight Projects of Concern summits in less than three years as Minister, bringing industry and Defence together to drive better outcomes for our defence force and Australians... when we came to office, we found at least 28 projects running a combined 97 years late."[2]

In May 2025, the Australian National Audit Office (ANAO) found that ‘Defence has failed to maximise Australian industry participation in its vast procurement program and its arrangements to achieve industry participation were at best only partly fit for purpose.’[3]

In June 2025, the ANAO found that for the sustainment of Canberra Class Amphibious Assault Ships (Landing Helicopter Dock (LHD), ‘Defence did not plan effectively for the transition from acquisition to sustainment. Value for money was not clearly demonstrated, and probity was not well managed… Defence has not managed its LHD sustainment contracts effectively. The LHDs have operated with ongoing deficiencies and have experienced critical failures during operations.’[4]

On 01 December 2025, Deputy Prime Minister Marles and Minister Conroy announced the establishment of a ‘Defence Delivery Agency, which will be an independent agency that will report directly to the Minister for Defence and the Minister for Defence Industry… headed by the National Armaments Director… [and will merge] three existing groups: the Capability Acquisition and Sustainment Group, the Guided Weapons and Explosive Ordnance Enterprise, and the Naval Shipbuilding and Sustainment Group.’[5]

In How Big Things Get Done, Bent Flyvbjerg and Dan Gardner provide seven ideas that may help to forestall and mitigate future repetitions of the above during defence project delivery.

How Big Things Get Done… seven ideas 

Idea 1: Iron Law of Megaprojects

The ‘Iron Law of Megaprojects’: over budget, over time, under benefits is, from Flyvbjerg’s and Gardner’s analysis of 16,000 projects, in 20 different fields, in 136 countries, on all continents (less Antarctica), that only 8.5 percent of projects ‘hit the mark on both cost and time’ [i.e. 1360 projects], 0.5 percent achieve ‘cost, time and benefits’ [i.e. 80 projects].[6] In addition, Flyvbjerg and Gardner find that the ‘mean cost over-run of a major building project is 62 percent.’[7]

Idea 2: Fat tails and thin tails

Fat tails are projects that display bell curve characteristics without normal distribution. This means that the distribution of the project bell curve, to measure risk, retains high risk and extreme outcomes for cost overruns for a greater portion of the bell curve and the tail of the distribution is not thin but remains fat.[8] In a ‘typical fat-tailed distribution in project management, about 80 percent of the outcomes will make up the body of the distribution.’[9]

Fat tailed projects, in priority order of risk, include: information technology; nuclear storage projects; Olympic Games; nuclear power plants; information technology systems; hydroelectric dams; airports; dams; defence projects; big buildings; aerospace projects; tunnels; mining projects; high-speed rail; conventional rail; bridges; bus rapid transport; oil projects; gas projects; water projects; and pipelines.[10]

In contrast, the ‘fortunate five’ are thin tailed projects that due to their ‘considerable degree of modularity’… enabling repetition, scale, speed and standardisation, are ‘faster, cheaper and less risky.’[11] These ‘fortunate five’, ‘may come in somewhat late or over budget but it is very unlikely that they will go disastrously wrong’, and include: (1) solar power; (2) wind power; (3) fossil fuel thermal power (e.g. coal-burning power stations); (4) electricity transmission; and (5) roads.[12]

Idea 3: Think slow, act fast

Projects ‘don’t go wrong, they start wrong. [13] Think first, then do. Flyvbjerg and Gardner argue that this means dividing a project into two phases: (1) planning; (2) delivery.

Planning is ‘a safe harbour.’ Planning is ‘pushing a vision to the point where it is sufficiently researched, analysed, tested and detailed’ to enable a ‘reliable roadmap of a way forward.’ Most planning is best completed with maximum virtual product, based on experience and experimentation, through ‘computers, paper and physical models [as well as ideas; research; exploring options; analysis of risk; scripting; rehearsing; scheduling; testing; and iterating, including simulation through artificial intelligence and machine learning], meaning that planning is relatively cheap and safe.’[14]

Planning ‘requires thinking – and creative, critical, careful thinking is slow.’[15] Other circumstances being equal ‘it is fine for planning to be slow.’ [16] Good planning ‘explores, imagines, analyses, tests and iterates’. This means that ‘slow is a consequence, not a cause, of doing planning right.’ The ‘cause of good planning is the range and depth of the questions it asks and the imagination and the rigour of the answers it delivers.’[17]

Delivery ‘is another matter.’ Delivery is when ‘serious money is spent and the project becomes vulnerable as a consequence.’[18]

Examples of success with think slow, act fast approaches to project management planning and delivery are: Empire State Building; Pixar Animation Studios; Guggenheim Museum Bilbao, Spain, and Heathrow Terminal 5.

Examples of failure to think slow, act fast in project management planning and delivery are: Sydney Opera House; Walt Disney Concert Hall, Los Angeles; and, California High-Speed Rail.[19]

Idea 4: Understand the project goal

Flyvbjerg and Gardner emphasise the importance of ‘developing a clear, informed understanding of what the [project] goal is and why – and never losing sight of it from beginning to end – as the foundation of a successful project.’[20] Flyvbjerg and Gardner signal risk for projects when ‘the goal, which is only vaguely understood to begin with, fades from view…then the project can veer off in unpredictable directions.’[21]

Techniques for keeping the goal in mind include:

• Creating ‘a shared sense of identity, purpose, standards, and maintaining psychologically safe open communications;’[22]
• Project planning or ‘thinking from left to right’;
• Developing a desired future state or ‘backcasting’;
• Defining goals or ‘theory of change’; and, in limited cases,
• Building, testing and seeking immediate real-world analysis or ‘minimum viable product.’[23]

Jeff Bezos keeps Amazon focussed on customers, and the end goal, through mandating that project leaders gain approval from top executives – who ask hard questions and debate ideas presented in the required project completion press release (PR) and frequently asked questions (FAQ) – before a project commences.[24]

To ‘pitch a new project at Amazon, you must first write a PR and FAQ [without PowerPoint], putting the goal smack in the opening sentence of the press release…everything that happens subsequently is working backwards from the PR/FAQ.’ Importantly, the pitch is written in plain English: no ‘jargon, slogans, or technical terms’.[25]  

Idea 5: Anchoring and adjustment

A forecasted project delivery schedule, usually, ‘starts with some fixed point’ or an anchor.[26] Then, through additional analysis, the project delivery schedule is amended or adjusted.

But ‘basing forecasting on anchoring and adjusting is tricky.’ As psychologists have ‘shown in experiments, final estimates made this way are biased toward the anchor, so a low anchor produces a lower estimate than a high anchor does.’[27] This means the quality of the anchor is critical. Use a ‘good anchor, and you greatly improve your chance of making a good forecast; use a bad anchor, get a bad forecast.’[28]

Anchoring and adjustment means that when a project experiences ‘delays or cost overruns, we naturally go looking for things that are slowing the project down or driving up costs.’ But ‘those delays and overruns are measured against benchmarks…. [and we must ask] are the benchmarks reasonable?’ Logically, questioning benchmarks is the first question to ask, ‘but it rarely comes up.’[29]

Once we ‘frame the problem as one of time and money overruns’, we, usually, fail to consider that the real source of the problem may be forecast underestimation and not overruns.[30]

Approaches to mitigating underestimation, from bad anchoring, faulty adjustments, and inaccurate forecasts include: analysing mean actual costs; reviewing relevant data on average adjustments; overcoming uniqueness bias; and, understanding known-knowns through reference-class forecasting.

Idea 6: Black Swan management

Nassim Nichola Taleb defined ‘low-probability, high-consequence events as “black swans.”’[31] Flyvbjerg and Gardner argue that through Black Swan management, some fat-tails are simple to cut. Examples of Black Swan management to ‘study and mitigate risk’ include:[32]

• Tsunamis are fat-tailed, but if you build well inland or erect a high enough seawall, you eliminate the threat.
• Earthquakes are also fat-tailed, but build to an earthquake-proof standard, and you are covered.
• Pandemics require a combination of measures, including a blend of masks, tests, vaccines, quarantines, and lockdowns to prevent infections from running wild.[33]
• Big projects also require a combination of measures including: think slow, act fast through exhaustive planning to enable swift delivery. Finishing a project is the ‘ultimate form of Black Swan prevention.’[34]

Idea 7: Eleven Heuristics for better project leadership

Flyvbjerg and Gardner conclude How Big Things Get Done with Eleven Heuristics for better project leadership. They emphasise that ‘heuristics should never be used like thoughtless paint-by-numbers rules’ and that project leaders should ‘check whether these heuristics resonate with our own experience before using them in practice.’ These Eleven Heuristics are optimally employed as a ‘source of inspiration for investigating, trying new things, and developing [our] own heuristics.’[35]

The Eleven Heuristics for better project leadership are:

1. Hire a masterbuilder – a person with ‘deep domain experience and a proven track record of success.’[36]
2. Get your team right – selecting the team is the ‘job of the masterbuilder.’[37]
3. Ask “why?” – focus your project on ‘what matters, your ultimate purpose, and your result.’[38]
4. Build with Lego – modularity where ‘big is best built from small.’[39]
5. Think slow, act fast – take ‘all the time necessary to create a detailed, tested plan. Planning is relatively cheap and safe; delivering is expensive and dangerous.’[40]
6. Take the outside view – employ reference-class forecasting where you ‘switch the focus from your project to the class your project belongs, to, paradoxically, gain a more accurate understanding of your project.’[41]
7. Watch your downside – for ‘fat-tailed risk, which is present in most projects, forget about forecasting the risk; go directly to mitigation by spotting and eliminating dangers.’ Successful project leaders ‘focus on not losing, every day, while keeping a keen eye on the prize, the goal they are trying to achieve.’[42]
8. Say no and walk away – does the project have the ‘people and funds, including contingencies, needed to succeed?’. If yes, then lead the project. In contrast, saying no to unforecast and unbudgeted project adjustments and expansions ‘is essential for getting projects done.’[43]
9. Make friends and keep them friendly – if something goes wrong with the project ‘the project’s fate depends on the strength of your [personal project] relationships’ … ‘build your bridges before you need them.’[44]
10. Build climate mitigation into your project – build projects that are ‘good for people and to get things done.’[45]
11. Know that your biggest risk is you – project failure results less from surprises, such as ‘price and scope changes, accidents, weather, new management, etc,’ and more through ignoring reference-class forecasting and our own behavioural biases.’[46]

Conclusion

In How Big Things Get Done Flyvbjerg and Gardner offer a range of advice that is applicable across all Defence projects, large and small. 

As Defence implements the Defence Delivery Agency, Flyvbjerg’s and Gardner’s seven ideas – iron law of megaprojects; fat tails and thin tails; think slow, act fast; understand the project goal; anchoring and adjustment; black swan management; and, eleven heuristics for better project leadership – provide guidance and guardrails for project leaders.

In addition, and importantly, Defence planners, in all disciplines, especially when employing the Australian Defence Force – Integration – 5—Decision-Making and Planning Processes may employ Flyvbjerg’s and Gardner’s seven ideas as a check list, or waypoints, in their planning to quality assure their work, their leadership and their professional mastery.