The Planning Fallacy: Why Every Project Costs 90 Percent More Than Quoted

The Reliable Overrun: Across more than 500 megaprojects analysed by the Oxford Programme on Major Infrastructure Risk, the average cost overrun was 90 percent, the average schedule overrun was 80 percent, and the average benefit shortfall was 40 percent. The pattern holds for IT projects, kitchen renovations, dissertation completion times, and the lunch you confidently said would take 30 minutes. The bias is not stupidity. It is one of the most reliable departures from rational forecasting in the human cognitive repertoire.

The planning fallacy was first formally described in 1979 by Daniel Kahneman and Amos Tversky, who observed that university students estimating their own thesis completion times systematically underestimated the actual duration, even when explicitly told that the average student took longer than their estimates. The pattern held when students were asked to account for “unknown unknowns,” when given access to historical completion data, and when explicitly warned about their own past underestimates. The bias was, in this sense, immune to most of the obvious corrections.

The most rigorous quantification at industrial scale has come from Bent Flyvbjerg’s Oxford Programme on Major Infrastructure Risk, which has spent two decades cataloguing cost and schedule overruns across more than 16,000 megaprojects globally. The findings are unsettling. Roughly 9 out of every 10 megaprojects deliver over budget, over schedule, and under benefit — with the magnitude of the overrun showing no improvement over the past 70 years of project management discipline.

1. The Three Cognitive Drivers Behind Reliable Underestimation

The planning fallacy operates through three convergent cognitive mechanisms, each well documented in the decision-research literature. Understanding them individually is the first step toward designing forecasts that the bias cannot subvert.

Three operational drivers appear consistently in the data:

• The Inside View: Planners construct estimates by walking through the specific steps of their project, imagining a smooth execution path. The inside view systematically ignores the base rate of similar projects, which would predict the overruns the inside view cannot see.
• Optimism Bias: Self-perception research consistently shows that humans rate themselves above average on competence, luck, and adversity tolerance. The same bias drives planners to assume their project will encounter fewer obstacles than average, despite no evidence supporting the assumption.
• Strategic Misrepresentation: In organisational contexts, plans are often subject to deliberate underestimation to secure approval, with overruns absorbed later as “unforeseen.” The pattern, documented in Flyvbjerg’s work, is so reliable that it has become a structural feature of major project approval processes.

2. The $5.4 Trillion Annual Planning-Fallacy Cost

The economic translation of the planning fallacy is staggering. Project management researchers at the Project Management Institute have estimated the global annual cost of project cost-and-schedule overruns at approximately $5.4 trillion — a figure dominated by IT projects, construction, infrastructure, and complex product development. The cost is not concentrated in obvious mismanagement; the vast majority of overruns occur in well-managed projects with competent leadership, planning teams, and oversight.

The individual translation is similarly large. The average homeowner who undertakes a major renovation pays roughly 43 percent more than the original quote by completion, and the project runs roughly 31 percent longer than originally scheduled. The same pattern affects dissertations, novel-writing, business launches, and almost any non-routine personal project. The cost compounds across a lifetime into a substantial personal financial drain and a meaningful share of the chronic stress that comes from chronically running late on commitments.

3. Why “This Time Is Different” Is Almost Always Wrong

The most uncomfortable feature of the planning fallacy is its resistance to past evidence. Project managers who have personally experienced overruns on five consecutive projects routinely predict that the sixth project will be different, because the specific causes of the previous overruns “have been addressed.” The pattern is so reliable that Kahneman, in his 2011 book Thinking, Fast and Slow, used it as the paradigmatic example of how the inside view systematically defeats accurate forecasting.

The solution that actually works — reference class forecasting — is an explicit override of the inside view. Instead of asking “how long will this project take?” the planner is forced to ask “how long does this type of project typically take?” and to use the historical distribution as the primary forecast, adjusted only modestly for the specifics of the current project. The technique consistently produces forecasts that are 30 to 60 percent more accurate than inside-view estimates, but it is rarely used because it produces unflattering numbers that are harder to defend in approval processes.

4. How to Defeat the Planning Fallacy in Personal Projects

The protocols below convert the academic findings into a defensive routine that any individual project planner can apply. The framework is uncomfortable to apply but consistently produces forecasts that survive contact with reality.

• The Reference Class Audit: Before estimating a project’s duration or cost, identify 5 to 10 comparable projects you or your organisation have completed previously, and use the actual outcome distribution as the starting point. The base rate beats the inside view almost universally.
• The Multiply-by-Two Heuristic: For personal projects without a meaningful reference class, take your honest estimate and multiply both cost and time by two. The heuristic is unflattering but consistently produces forecasts that survive completion.
• The Pre-Mortem Exercise: Before starting, imagine the project has failed disastrously and write a brief explanation of what went wrong. The exercise routinely surfaces obstacle categories the optimistic forecast had not considered.
• The Buffer Reserve Discipline: Reserve a 30 to 50 percent buffer of time and budget that is not allocated to specific tasks. The buffer is the line item that will be consumed by the unanticipated obstacles that the planning fallacy systematically hides.
• The External Reviewer Default: Have someone uninvolved in the project review the timeline and budget before commitment. The outsider routinely surfaces optimism the insider has internalised. The cost of the review is small; the calibration benefit is large [cite: Buehler, Griffin & Ross, Journal of Personality and Social Psychology, 1994].

Conclusion: The Estimate You Trust the Most Is Probably the Most Wrong

The planning fallacy is one of the most reliable cognitive biases ever documented, and its commercial cost — running into trillions of dollars annually — reflects the consistency with which it defeats even sophisticated planning processes. The professional who treats their own time and cost estimates as fundamentally biased forecasts that require external correction — through reference classes, buffers, pre-mortems, and outside review — consistently outperforms peers who treat their inside-view estimates as accurate. The wealth, the schedules, and the credibility built across a working life are decided not by how confidently you predict but by how systematically you correct your predictions before they become commitments you cannot keep.

What is the most recent project you completed on time and on budget — and what specific feature of how you planned it might be worth replicating against the planning fallacy in your next undertaking?