The planning fallacy

A group of psychology students were asked a simple question: when will you finish your thesis? They gave their best honest estimate, on average about 34 days. Then the researchers waited and measured. The real average was 56 days. Almost no one finished on time, and many ran far over. Here is the part that makes it more than a story about students. The same students were also asked to estimate how long other people would take on similar projects, and those estimates were far more accurate. When they judged someone else, they instinctively used the track record, how these things usually go. When they judged themselves, they used the plan, the smooth version in their head where nothing goes wrong. The plan is always optimistic, because the plan does not include the interruptions it cannot foresee.

You predict your own timelines badly and other people's timelines well, and the reason is the method you use for each. For yourself you imagine the ideal run. For others you remember how it actually tends to go. So the fix is almost mechanical: when you estimate your own next project, stop imagining it, and instead look at how long your last three similar projects really took. That number, the boring historical one, is a better predictor than any plan you can build, because it already contains the delays you are about to forget again. Naming exactly when and where you will start is a separate lever and a strong one, which is implementation intentions.

How Big Things Get Done

Flyvbjerg and Gardner, 2023

Built on a database of sixteen thousand projects, of which fewer than one in ten came in on budget and on time. Its method is to stop estimating from inside the project and look instead at what comparable projects actually took.