The Planning Fallacy and the Outside View
Updated
Knowledge on this page was mainly distilled from Act Like It’s Impossible to Fail.
The planning fallacy is the tendency to underestimate how long a task will take, even when similar work has gone late before. The outside view is the fix: estimate from comparable past projects first, then adjust for your case only after that baseline is clear.
This does not require perfect data. Even a rough reference class, such as your last five launches or comparable projects by peers, is usually more reliable than a story about why this time will be different.
Q&A
What is the planning fallacy?
The planning fallacy is the common tendency to predict that a task will finish sooner and more smoothly than it actually does. People do this even when they have direct evidence from similar past projects. The bias appears in individual work, teams, product launches, and personal goals.
What is the outside view?
The outside view means estimating from a reference class of similar cases instead of relying only on the internal story of the current project. In practice, you ask how long this kind of work usually takes for people in similar conditions. That baseline often gives a more realistic forecast than optimism alone.
How can a solo builder use the outside view without lots of data?
A solo builder can use the outside view by looking at a small set of relevant examples, such as prior launches, side projects, client work, or public timelines from peers. You do not need a large dataset to improve your estimate. Even three to ten comparable cases can expose a pattern your current enthusiasm is hiding.
Does the outside view kill optimism or speed?
No, the outside view is meant to improve planning, not reduce action. You can stay optimistic in execution while still being conservative about timelines, effort, and risk. That combination often prevents avoidable disappointment and makes iteration more sustainable.
What counts as a good reference class?
A good reference class includes projects that are similar in scope, complexity, skill level, and constraints. If the comparison set is too broad or aspirational, the estimate will be distorted. The best matches are boringly comparable, not flattering.
How should you combine an inside view and an outside view?
You should start with the outside view as a default range, then use the inside view to explain only justified deviations from that range. This keeps the project grounded in reality while allowing for real differences. If every estimate becomes an exception, you have slipped back into wishful thinking.