There's an experiment everyone who plans should know. In 1994, psychologists Roger Buehler, Dale Griffin and Michael Ross asked a group of students to estimate when they'd finish their senior thesis: a realistic date, an optimistic one, and a pessimistic one —their worst possible scenario. Then they waited.
The result was devastating. On average, the students finished later even than their own pessimistic estimate. The "worst case I can think of" turned out to be, for most, the optimistic case. And the most interesting part: those same students were perfectly able to remember that in the past they'd always taken longer than expected. They knew their track record. And even so, for this task, their minds promised speed again.
The pattern already had a name. Daniel Kahneman and Amos Tversky had christened it in 1979 as the planning fallacy: the systematic tendency to underestimate how long your own task will take, even with the proof in front of you that similar tasks took longer. It's not absent-mindedness or one-off bad luck. It's a factory-fitted bias, and it feeds especially on whoever builds something new and depends on their own deadlines to survive.
The default reading is moral: "I take longer than I say because I lack discipline, because I get distracted, because I don't try hard enough." From there comes the usual prescription: next time, push harder, promise less margin, demand the ideal plan of yourself.
And that's exactly what perpetuates the problem. Because if the cause isn't your character but how your mind estimates, demanding more of yourself only makes you promise even more unrealistic deadlines —and miss them with more guilt.
This isn't about you being lazy or disorganized.
It's about planning from inside the task, watching it go well, when you should be looking at it from outside, as one more in a series you already know.
1. When you plan, you imagine the best day. As you estimate, your mind builds a concrete film of how it'll go: you sit down, it flows, no interruptions, nothing goes wrong. That film almost never includes the client who writes in, the bug that appears, the day you wake up with no energy. It's not that you lie: it's that the unforeseen, by definition, can't be imagined. And since you execute on the real days —with interruptions, errors and variable energy— the plan made for the best day breaks against the average one.
2. Your track record knows what your imagination ignores. The reason you fail even while remembering having failed before is that, when planning, you look forward (the optimistic film) instead of backward (how long it really took last time). The cure isn't to imagine better: it's to stop imagining and look at the data. How long did the last similar thing take? That, and not your optimism, is your best prediction.
3. That's why margin isn't weakness, it's realism. Doubling your initial estimate isn't giving up or settling: it's correcting a known bias with a known factor. Whoever promises the ideal deadline and misses it looks ambitious but delivers late and burns trust. Whoever promises the real deadline and meets it looks prudent and turns out reliable. In business, reliable beats ambitious almost every time.
Think of the last thing you finished that you had planned: a delivery, a launch, a task with a date.
How long did you say it would take, and how long did it really take? And the one before that —same thing?
Now take what you're estimating today. Are you calculating it by looking at that real figure from your past, or by looking at the film of how you'd like it to go?
If the deadline always fools you in the same direction, that's not a problem: it's a predictable bias, and the predictable can be corrected. The layers:
Before calculating how long you'll take by imagining the task, ask how long the previous three similar tasks took. That outside view predicts far better than your inner film. If you don't have the data, start saving it today: note the estimated and the actual for everything you close.
Almost everyone has a stable multiplier: you take one and a half times, or double, or triple what you think. Compute it from your record and apply it without negotiating. Your honest estimate is what your mind says multiplied by your factor. Stop arguing with a bias that always wins.
Outward —clients, partners, yourself— commit to the already-multiplied figure. Delivering when you said builds a reputation that delivering late destroys. Reliability is an asset that accumulates one met deadline at a time.
Anything you estimate yourself, about yourself, is biased toward optimism. So the recurring question isn't how long I think I'll take. It's how long I took the last few times on something like this, and by how much I multiply to be right. The data of your past is smarter than your plan of today.
You don't take twice as long because you lack willpower. You take twice as long because you plan with the best day you imagine and live on the ordinary days you get. The good news is that a bias that always fools you in the same direction is the easiest to correct: you just have to stop believing yourself and start believing your track record.