Planning Fallacy
A project’s timeline is usually built from how it’s supposed to go, not from how similar projects have actually gone, which is why estimates stay optimistic even after years of evidence that they shouldn’t.
What Is It?
Daniel Kahneman and Amos Tversky named the planning fallacy in the 1970s to describe a specific, persistent pattern: people and organizations systematically underestimate the time, cost, and risk of a future project, even when evidence from previous comparable projects says otherwise. The estimate isn’t uninformed. It’s built from the wrong reference point. When someone forecasts a project, they naturally focus on the specific plan in front of them, the steps, the sequence, the intended path, rather than on the outside-view record of how projects like this one have actually gone. Kahneman later called these two approaches the inside view and the outside view. The inside view feels more rigorous because it’s detailed and specific to the actual work. It’s also the view that consistently produces the optimistic estimate. A detailed plan for how something is supposed to go rarely includes the interruptions, dependencies, and complications that made every previous similar project take longer than its own detailed plan predicted. The corrective is often called reference class forecasting: identify a relevant class of past projects genuinely similar to this one, look at how they actually turned out, and use that distribution as the starting point before adjusting for whatever is truly different about the current case, rather than building the estimate up from the plan alone.
Why Does It Matter?
Organizations keep re-deriving the same optimistic estimate from scratch, project after project, because each new project gets planned from its own inside view rather than from the organization’s actual track record on comparable work. A team that has never delivered a major initiative on its original timeline will still produce a confident, detailed timeline for the next one, not out of dishonesty, but because the planning process itself asks “what needs to happen” rather than “how long has this actually taken before.” The record exists. It’s just not the input the estimate is built from. This compounds because commitments get made on the optimistic number. Budgets, staffing, external commitments, and dependent teams’ own plans all get built on top of an estimate that was underestimated from the start, so the eventual overrun doesn’t just cost the original project, it cascades into every plan that assumed the original date would hold.
What Changes Once You See It?
You start asking for the actual outcomes of the last several comparable projects before trusting a new detailed estimate, rather than treating the new plan’s specificity as evidence of its accuracy. You start treating “this time is different” as a claim that needs its own evidence, since it’s also what was said before each of the previous overruns. You also get more comfortable building in the gap between the inside-view estimate and the outside-view track record explicitly, rather than quietly hoping the gap won’t apply this time. You stop treating each overrun as its own unique story requiring its own explanation, and start asking whether the pattern across many overruns is itself the more accurate forecasting model.
Common Misunderstandings
- It isn’t a claim that detailed planning is worthless. A specific plan is still useful for sequencing and coordination, the fallacy is trusting that plan’s timeline over the organization’s actual track record on similar work.
- It isn’t the same as padding an estimate to protect against blame. The fallacy is a genuine cognitive bias that shows up even when people have no incentive to sandbag, which is different from, and a useful contrast to, the deliberate underproduction covered by the Ratchet Effect.
- It doesn’t mean every project runs late. Some genuinely finish on time or early. The fallacy describes a systematic direction of the error, not a claim that every single estimate misses.
- It isn’t fixed by simply asking people to “be more realistic.” The bias comes from which reference class the mind reaches for by default, the specific plan rather than the comparable history, not from a lack of effort or honesty.
- It isn’t the only reason organizations underestimate projects. Sometimes an optimistic forecast is strategic: whoever’s proposing the project knows a realistic number is less likely to win approval or funding. The planning fallacy describes sincere forecasting error, incentive-driven understatement is a different mechanism, and it needs a different remedy, fixing the incentive, not fixing the forecasting method.
Diagnostic Question
If we ignored this project’s specific plan entirely and just looked at how long our last several comparable projects actually took, what date would that history predict?
Explore Further
Field Notes
None yet.
Related Field Guide
- Escalation of Commitment
- Second-System Effect
- Ratchet Effect
- Overconfidence Effect
- Base Rate Fallacy
Origin
Daniel Kahneman and Amos Tversky introduced the planning fallacy in “Intuitive Prediction: Biases and Corrective Procedures” (1977); Kahneman’s later distinction between the inside view and outside view, developed further with Dan Lovallo, remains the standard framework for correcting it. Bent Flyvbjerg’s subsequent research on megaproject cost overruns further separated genuine forecasting bias from deliberate strategic misrepresentation.