Why Government Projects Fail, and How Thinking Slow, Acting Fast Can Fix Them

Happy Tuesday, Transformation Friends. Another week, another opportunity to go Beyond the Status Quo.

In government, urgency is constant. Political leaders face electoral cycles. Citizens expect timely services. Scrutiny from the public and the media pushes departments to show quick progress. The reflex is to “get started” immediately: launch the program, break ground, cut the ribbon.

Side note: This tendency to focus on visibly doing new things is one of the central humorous themes of the excellent Australian Broadcasting Corporation series Utopia, which follows a fictional government agency. The show captures, with sharp wit, many of the frustrations and ironies familiar to public servants. If you work in the public sector, you may find it both entertaining and uncomfortably relatable.

Yet decades of global research on large-scale projects show this instinct often backfires. Bent Flyvbjerg, the Oxford University professor behind the world’s largest database of megaprojects, calls it the Iron Law of Megaprojects: over budget, over time, under benefits—over and over again (Flyvbjerg, 2023). Whether it is a digital transformation, infrastructure build, or policy overhaul, the pattern is consistent.

In his book How Big Things Get Done (with Dan Gardner), Flyvbjerg offers a simple countermeasure: think slow, act fast. This principle, grounded in data and behavioural science, is especially relevant for public sector leaders managing high-stakes, politically sensitive initiatives.

It also reflects Daniel Kahneman’s Nobel-winning insights into decision-making. Kahneman’s research shows that we are overconfident, underestimate risks, and rarely use the “outside view” when planning (Kahneman, 2011). The “think slow, act fast” approach operationalizes these findings.

Today, we explore what “think slow, act fast” means in the context of the public sector, why it works, and how project teams and senior leaders alike can apply it in practice to achieve stronger, more reliable outcomes. We will draw on the work of Flyvbjerg and Kahneman to translate research insights into practical steps for government decision-makers.

Grab your morning coffee, and let’s get started.

Why Public Projects Start Wrong

Before getting into the method, it’s worth asking: why do projects fail? While there are many factors, Flyvbjerg keeps hammering on three recurring issues seen across projects worldwide that deserve closer attention. These are exactly the issues the think slow, act fast approach is designed to help address.

1. The Planning Fallacy
Coined by Kahneman and Amos Tversky, this describes our tendency to underestimate costs, timelines, and risks, even when past experience suggests otherwise (Kahneman and Tversky, 1979). Political optimism amplifies it: the belief that with enough determination, constraints can be bent to fit promises.

2. Strategic Misrepresentation
In some cases, proponents deliberately understate costs or overstate benefits to get approval, knowing the estimates are unrealistic (Flyvbjerg, 2009). The calculation is political: if you present the truth, the project might not be funded. This often reflects short-term incentives to secure political or financial backing at the expense of long-term viability. It can erode trust, distort priorities, and set the stage for chronic overruns or underperformance once delivery begins.

3. Premature Start
Flyvbjerg (2017) warns against launching projects before designs, governance structures, and procurement strategies are ready. Starting too early often leads to reactive firefighting: scope changes multiply, bottlenecks emerge, procurement slows, and costs rise. This premature start undermines delivery speed and budget discipline, especially compared to projects that invest more time upfront in thorough planning.

These problems are not inevitable. They are predictable and preventable when leaders change the sequencing of effort.

The Core Idea: Think Slow, Act Fast

This approach reverses the common rhythm of public projects. Instead of a rushed start followed by slow delivery, it calls for deliberate preparation followed by fast, disciplined execution (Flyvbjerg, 2023).

Thinking slow means:

• Understanding the problem, desired outcomes, and constraints in detail.

• Using data from past projects to produce realistic estimates.

• Identifying and mitigating risks before they stall delivery.

• Designing governance, procurement, and sequencing before signing contracts.

Acting fast means:

• Delivering in parallel work streams rather than one long sequence.

• Freezing scope and controlling changes tightly once work starts.

• Empowering teams to make quick, informed decisions within clear boundaries.

• Monitoring progress closely and correcting course immediately when needed.

The Behavioural Science Behind It

Kahneman’s System 1 / System 2 model explains the value of this approach. System 1 is fast, intuitive, and prone to bias. System 2 is slow, deliberate, and analytical (Kahneman, 2011). Most large public projects begin in System 1, a fast rush to action driven by optimism and political momentum.

Complex, unfamiliar projects require System 2. The “think slow” phase forces this shift, countering bias and preventing commitment to flawed plans.

Once the project is well understood and execution is designed, leaders can return to faster decision-making, with System 1 operating within a stable, predictable environment.

Thinking Slow in Practice

Thinking slow is active preparation, not delay. In the public sector, it involves:

1. Define “Done” in Measurable Terms
   Broad aspirations must be translated into specific, testable outcomes, such as service times, cost savings, or satisfaction scores (Flyvbjerg, 2023). In practice, this means agreeing on a short list of clear performance indicators at the outset, ensuring that all stakeholders share the same definition of success and that progress can be tracked objectively over time.

2. Use the Outside View
   Reference-class forecasting is a method for predicting outcomes by looking at a broad set of similar past projects and using their actual results as a benchmark (Kahneman and Tversky, 1979). Put simply, it means comparing your project to a "reference class" of completed projects to see what really happened, instead of relying only on your own assumptions. In practice, this involves gathering reliable data on the cost, schedule, and benefits of comparable cases and using these to set a realistic starting estimate. For example, if a similar project elsewhere took 36 months and cost $200 million, that becomes your baseline, which you then adjust for your own local context.

3. Pilot and Prototype
   Small-scale tests expose hidden risks before major commitments (Flyvbjerg, 2023). In practice, this could involve trialling a single service channel, building a minimum viable product, or testing a new procurement method with a limited set of suppliers, allowing you to uncover issues and refine your approach before scaling up.

4. Map the Delivery System
   Define governance, decision rights, procurement strategy, and sequencing before launch. In practice, this means creating a clear delivery roadmap, assigning accountable owners for each decision point, agreeing on escalation paths for resolving issues, and sequencing activities to avoid bottlenecks. Ambiguity here is a major source of execution failure.

5. Run a Pre-Mortem
   Imagine the project has failed, then identify why and address those risks now (Klein, 2007). A pre-mortem is essentially a structured "failure rehearsal" that helps teams think ahead about what could go wrong. In practice, this exercise should bring together the full delivery team and key stakeholders to brainstorm plausible failure scenarios, explore the chain of events that could lead to them, and record preventive actions. This technique often reveals vulnerabilities that traditional planning overlooks and encourages candid discussion of risks that might otherwise remain unspoken.

Acting Fast in Practice

When planning is complete, execution becomes focused acceleration.

1. Break Down the Work
   Structure the project into independent modules that can be delivered in parallel. In practice, this might mean dividing a large project into smaller, self-contained components with clear interfaces, so that if one part hits a problem, others can still move ahead. This reduces interdependencies and makes progress more resilient.

2. Freeze Scope, Manage Change
   Avoid adding features or requirements mid-stream. Every change is a risk to cost and schedule. In practice, this means having a formal change control process where all requests are logged, assessed for their impact, and approved only if they add significant value.

3. Push Decisions to the Front Line
   Allow delivery teams to handle day-to-day choices without escalation. In practical terms, this involves giving clear decision rights, authority, and accountability to those closest to the work, while reserving senior decision-making for major shifts in scope, cost, or policy.

4. Maintain Real-Time Insight
   Track progress against the baseline forecast using timely, reliable data. In practice, this could mean dashboards updated weekly, key performance indicators tied to milestones, and rapid escalation protocols when metrics show signs of slippage.

5. Protect the Critical Path
   Identify the most time-sensitive tasks that dictate the project’s overall schedule. In practice, this means prioritizing resources, attention, and problem-solving on these activities, ensuring they are unblocked quickly to keep delivery on track.

Common Pitfalls

Even with the right sequencing, traps remain:

• Analysis Paralysis: Thinking slow must be purposeful and time-boxed with clear decision gates. In practice, this means setting defined periods for planning, with milestones where decisions must be made, and resisting the urge to endlessly gather more data or options before moving forward.

• Political Impatience: Manage expectations by explaining why early preparation leads to better outcomes. This involves transparent communication with political sponsors and stakeholders about the risks of rushing, supported by examples from past projects where planning paid off.

• Scope Creep: Guard the execution phase from politically motivated or opportunistic additions. In practical terms, this means enforcing scope controls, having a clear process for assessing change requests, and being willing to say no when additions threaten delivery.

• Weak Kill Criteria: Define the conditions for stopping or pivoting in advance. This includes setting measurable thresholds for cost, schedule, and benefits, agreeing on who has the authority to halt work, and ensuring these criteria are understood and accepted by all key decision-makers before execution begins.

Leader’s Checklist: Making “Think Slow, Act Fast” Work in Government

Here’s a practical summary you can apply directly in your work.

During the “Think Slow” Phase

• Define Success Clearly: Translate political or strategic goals into measurable outcomes.

• Adopt the Outside View: Use data from past projects to set realistic targets.

• Identify Kill Criteria: Set thresholds for pausing, pivoting, or cancelling.

• Pilot Before You Commit: Test on a small scale to uncover risks.

• Design the Delivery System: Establish governance, escalation paths, procurement, and sequencing.

• Run a Pre-Mortem: Identify and address possible causes of failure in advance.

During the “Act Fast” Phase

• Break the Work into Modules: Deliver smaller increments to reduce risk and speed results.

• Freeze Scope and Control Changes: Review all scope changes formally.

• Push Decision-Making Down: Let teams handle operational calls without delay.

• Monitor in Real Time: Track progress against plan and act on early warnings.

• Protect the Critical Path: Keep the most time-sensitive work moving.

• Report Transparently: Keep stakeholders informed with clear, honest updates.

Wrap up

Speed without preparation is recklessness; preparation without speed is stagnation. The challenge is balance.

For senior leaders, this means resisting the urge to launch too soon while avoiding bureaucratic drift. It means investing in the discipline of thinking slow: defining “done,” forecasting realistically, and stress-testing plans, before unleashing the urgency of acting fast.

When you feel the pressure to “just get started,” remember: a project that starts slowly and finishes quickly is far more valuable than one that starts quickly and drags on for years.

Take a moment to consider how the principles discussed apply to your own work and projects. These prompts will help encourage honest evaluation and highlight where the think slow, act fast approach could strengthen your results:

1. In my current or upcoming initiatives, have I taken enough time to reduce uncertainty before moving into execution?

2. Am I using evidence from similar projects or relying mainly on internal assumptions to set expectations for cost, schedule, and benefits?

3. Do I have clear, agreed-upon criteria for when to stop, pivot, or accelerate a project?

Until next time, stay curious, and I’ll see you Beyond the Status Quo.

---

References

Flyvbjerg, B. (2009) Survival of the unfittest: why the worst infrastructure gets built—and what we can do about it. Oxford Review of Economic Policy, 25(3), pp. 344–367.

Flyvbjerg, B. (Ed.). (2017). The Oxford Handbook of Megaproject Management. Oxford University Press.

Flyvbjerg, B. and Gardner, D. (2023) How Big Things Get Done. London: Penguin Press.

Kahneman, D. (2011) Thinking, Fast and Slow. London: Penguin.

Kahneman, D. and Tversky, A. (1979) ‘Intuitive prediction: Biases and corrective procedures’, Management Science, 12(3), pp. 313–327.

Klein, G. (2007) ‘Performing a project premortem’, Harvard Business Review, September, pp. 18–19.