Why You Always Think Things Will Take Less Time (The Planning Fallacy)
Kahneman and Tversky named it, and a classic study on students' thesis deadlines proved it. Why your own time estimates are the least reliable ones you'll ever get.
You are probably wrong about how long your next task will take, and so is almost everyone else. Daniel Kahneman and Amos Tversky named this the planning fallacy in 1979, and a well-known 1994 study found that students finished their thesis later than even their own worst-case prediction. The fix isn't trying harder to estimate. It's looking at your own actual track record instead of imagining the ideal version of the task.
Even experts get this wrong, consistently
In 1979, psychologists Daniel Kahneman and Amos Tversky described a bias they called the planning fallacy: the tendency to predict that a task will take less time than it actually will, even when you have direct experience showing otherwise. What made the observation interesting wasn’t that people are occasionally optimistic. It’s that the bias holds up even among people with plenty of relevant experience, engineers estimating project timelines, governments estimating construction costs, students estimating how long an assignment will take.
The thesis study that made this concrete
In 1994, researchers Roger Buehler, Dale Griffin, and Michael Ross ran a study that’s become the classic demonstration of this effect. They asked psychology students working on their senior thesis to estimate how long it would take to finish, giving both a realistic prediction and a pessimistic, worst-case prediction assuming things went about as badly as they reasonably could.
Most students didn’t finish by their realistic estimate. Many didn’t even finish by their pessimistic, worst-case estimate. The average completion time ran well past both predictions, even though the students had presumably lived through delays on other projects before.
Why knowing better doesn’t fix it
The leading explanation is that when people estimate a task, they mentally simulate the plan itself: this step, then that step, assuming a clean run with no interruptions. What tends to get left out is the base rate, how long things like this have actually taken before, including the delays, the sick days, the unexpected complications. Kahneman later described this as focusing on the “inside view” (the specific plan in front of you) instead of the “outside view” (how similar efforts have generally gone).
This is also why the fallacy survives repeated failure. Each new task feels specific and different enough that the outside view seems less relevant than it actually is.
What actually corrects for it
- Use your own history, not your intuition. If a habit or project has consistently taken longer than planned before, that pattern is more informative than a fresh guess.
- Ask “how long did this take last time,” not “how long should this take.” The outside view (past performance) consistently outperforms the inside view (this time’s plan) for estimating duration.
- Build in the gap you keep seeing. If you’re reliably off by 30%, that’s not a flaw to fix through effort. It’s data. Pad your estimate by the amount your own history says to. This is also where a specific, measurable goal earns its keep, since a vague one gives you nothing concrete to check your estimate against.
- Track completion times, not just completion. A log of how long things actually took is the only thing that reliably beats the planning fallacy over time.
This is one of the quieter reasons Me Improve’s Tracker keeps a full calendar history rather than just today’s status. Your own record of how consistently (or inconsistently) you’ve followed through is a far better predictor of your future than how motivated you feel right now. See how it works.
Frequently Asked Questions
What is the planning fallacy?
It's the well-documented tendency to underestimate how long a task will take, even when you know from past experience that similar tasks usually run over. The term was coined by psychologists Daniel Kahneman and Amos Tversky in 1979, describing a bias that persists even among experts who should know better.
Is there real evidence for this, or is it just a saying?
There's a specific, well-known study behind it. In 1994, researchers Roger Buehler, Dale Griffin, and Michael Ross asked psychology students to predict how long it would take to finish their senior thesis, under both a "best case" and "worst case" scenario. Most students finished later than even their worst-case estimate.
Why do we keep underestimating, even when we've been wrong before?
The leading explanation is that when we estimate, we tend to focus on the plan itself, the ideal, uninterrupted version of the task, rather than on our actual track record with similar tasks. We picture the best-case path and forget to weight it against everything that's historically gone wrong before.
What actually helps correct for the planning fallacy?
Looking at how long similar tasks have actually taken you in the past, rather than reasoning from the specifics of the current task, tends to produce more accurate estimates. This is sometimes called "reference class forecasting." Tracking your own history is what makes that reference class available to you in the first place.