Why Plans Fall Apart and How to Build in a Time Buffer

You told your manager the report would take an hour. You sat down at nine, sure you'd be free by ten. At midnight you're still fixing the last paragraph, annoyed at yourself and at this report. And it has happened before: with the move, the kitchen remodel, the deck due Monday. Every time you genuinely believe your estimate. Every time you miss in the same direction – short.

This isn't laziness or carelessness. The miss has a name, and it works almost the same way in everyone. Let's look at where it comes from and what to do about it, so your plans stop collapsing halfway through.

What the planning fallacy is

In 1979, Kahneman and Tversky described a stubborn bias: people systematically underestimate how much time, money, and effort a task will take, even when past experience screams otherwise. They called it the planning fallacy. The key word is systematically. The miss isn't random. It almost always leans one way: we think we'll finish faster than we actually do.

The cruel part is that experience doesn't save you. You can build a similar report thirty times and know it eats a whole evening. Yet sitting down for the thirty-first, you hear the same inner voice: "This one I'll knock out in an hour." The previous evenings feel erased. Your brain estimates from a mental picture of the ideal run, not from the record.

It works at every scale. The morning "I'll do it in fifteen minutes" and the construction project that drags on for years instead of months run on the same machinery. Only the numbers change.

Why your brain always lowballs the time

When you estimate, you mentally replay how you'll do the task. And in that imaginary run, everything is flawless. You sat down and dropped straight into flow. Nobody messaged, called, or asked for "just a second." The file opened, the document you needed was right there, the internet held. You never glanced at your phone or stalled rereading a paragraph.

The real evening looks different. Hunting for last year's file, you spotted an email and answered it. Someone asked for help. A sentence wouldn't come, so you rewrote it three times. You got tired, poured a coffee, stared out the window for ten minutes. Each of these steals a little, and together they add up to twice what you planned.

There's a second cause. When you estimate, you think about the work itself and forget everything wrapped around it: gathering your thoughts, switching off the last task, finding materials, waiting on a coworker's reply, checking and redoing. This wrapper often eats as much as the task, yet it almost never makes it into the estimate.

The inside view and the outside view

Kahneman described two ways to estimate. The first is the inside view: you look at this specific task, picture the steps, and add them up. That's what everyone does, and it's exactly the method that keeps failing, because imagination paints a smooth run.

The second is the outside view. Instead of breaking the task into steps, you ask: how long have similar tasks taken me before? Not how long they should have taken, but how long they actually took. Past experience is the most honest forecast, because it already bakes in all the friction you're currently forgetting.

An example. You need to write an article. Inside view: "Intro, twenty minutes. Three sections, half an hour each. Conclusion, ten. About two hours total." Outside view: "The last five articles each took half a day." The second forecast is ugly, but it's the one that comes true. When two estimates disagree, trust the one built on facts, not the one built on imagination.

How to estimate realistically

The simplest working move is to track how long tasks actually take. Not in your head, but at least with marks: started, finished. Within two or three weeks you'll have your own table of truth. You'll see that the "five-minute email" reliably takes fifteen, and the "quick call" takes an hour with prep and cleanup around it.

From then on, estimate off that table, not off a feeling. About to write a report? Recall how long the last one took and use that number. The feeling will argue and promise it's faster today. Don't trust the feeling, trust the record.

If you don't have a history yet, a blunt rule works: estimate as usual, then multiply. For familiar tasks, by 1.5. For new or fuzzy ones, by 2 to 3. It sounds like a cheat, but that multiplier is exactly what closes the gap between the imagined run and the real one.

Estimate the parts, not the whole

A big task viewed whole always looks shorter than it is. "Prepare the presentation" sounds like one evening. But inside it: gather the data, design the structure, write the slide text, lay it out, find images, rehearse out loud, revise. Seven jobs, each with its own start and end.

When you break the task into concrete chunks and estimate each one, the sum comes out more honest. A big lump is easy to underestimate at a glance. Small parts are harder to fool: it's obvious right away that "find images" isn't two minutes but half an hour of choosing and editing.

There's a useful side effect too. Splitting the job into parts, you often discover a step you forgot to include entirely. Get sign-off from a coworker. Wait for access. Those are the invisible chunks that tear the deadline later.

Build a buffer into your schedule

Estimating each task honestly isn't enough. You also can't stack them back to back with no gaps. If you cram nine hours of work into a nine-hour day, any single delay topples everything after it like a row of dominoes. And a delay will come – there's always one.

Leave gaps in the day on purpose. A working rule: fill about six of eight hours with tasks. The remaining two are a buffer for what runs long, for the urgent thing that lands, for a breather between hard blocks. Empty time in a schedule isn't idleness, it's the cushion that keeps the plan afloat.

Same at the week level. Don't cram Monday through Friday to the brim. Keep one free half-day as insurance. If everything goes smoothly, you'll clear small stuff or rest. If something slips, there's room to move it, and the deadline survives.

What eats time without you noticing

Three thieves steal time in a way you don't notice, and they almost never make it into the estimate.

Switches. Every jump from one task to another costs time to climb back in. You answer a message mid-work, then spend a few minutes remembering where you left off. Ten such jumps in an evening is a lost half hour that's in no plan.

Waiting. You send a coworker a question and wait. A file loads. Access gets approved. While you wait, the task doesn't move, but the clock does. That waiting usually isn't in the estimate at all, yet in practice it stretches the timeline several times over, especially when the job depends on other people.

Rework. The first version is almost never the last. You write it, reread it, it's off, you rewrite. You show it, feedback comes back, you redo it. This revision loop doesn't live in the first estimate, because in your imagination you nail everything on the first pass.

Plan for the friction in advance

There's a simple trick against optimism: imagine the plan has already failed. Before you start, ask yourself: if I miss the deadline, what will have caused it? The answers come fast and specific. I'll get stuck on the data. The client will vanish for two days. Something urgent will land from work. I'll get a headache and lose the evening.

Now you have a list of real threats, and you can defuse it. Gather the data ahead of time. Message the client today so the reply arrives by the day you need it. Leave a buffer for the urgent thing. You don't remove the friction entirely – you stop meeting it by surprise, which means it no longer tears the whole plan.

The same trick sobers up the estimate itself. Once you've said out loud everything that could go wrong, you won't write a naive "an hour." The list of friction pulls the number toward reality on its own.

Who needs this most

Freelancers and anyone who quotes deadlines to clients. Your underestimate turns into a missed deadline and a bruised reputation. The rule is simple: figure the time for yourself, then quote the client 1.5 times more. Deliver early, and it's a pleasant surprise. Deliver on time, and you're reliable. Both beat an endless "almost there."

Students. Term papers and theses are the classic planning fallacy. A month feels enormous and the work feels like a week. In reality, that week goes just to getting your bearings and gathering sources. Start earlier than your calm suggests, and slice the work into weekly chunks with slack.

Parents. Your day gets shredded by constant interruptions you can't schedule to the minute. Here a buffer isn't a luxury but the only way to finish anything at all. Plan fewer tasks than feels realistic, and leave big gaps for the unpredictable.

Teams and projects. The error compounds: everyone underestimated, and the whole project slipped. On top of that, tasks wait on each other, so one delay stalls the entire chain. The outside view matters most here – how long similar projects actually ran, not how long the estimate promised.

People with ADHD. "Time blindness" is a known companion of ADHD: an hour and ten minutes feel nearly identical, and switching is especially costly. Your inner sense of time can't be trusted at all here. Only external anchors work: a timer, written-down estimates, and tasks broken into small visible steps.

Common mistakes when padding an estimate

Building in a buffer and spending it instantly. You honestly left two free hours, then immediately packed them with new tasks because "the time's there." A buffer has to stay empty until you actually need it. Emptiness in a plan is a function, not an invitation to fill it.

Multiplying the estimate but not moving the deadline. You can multiply the number in your head all you want, but if you still promise to deliver tomorrow, nothing changed. A realistic estimate only means something when you act on it: quote a different date, start earlier, take on fewer tasks.

Treating friction as a one-off. "Today didn't work out because I got sick, but normally I make it." Friction isn't the exception, it's the norm. Every day has its own snag, just a different one. A plan has to survive a typical snag, not a perfect day without them.

Estimating while rested. In the morning, with a fresh head, any task feels quick. But you'll do it tired in the evening, where everything runs slower. Estimate with a correction for the state you'll actually be in when you sit down.

When a precise estimate isn't needed

An honest note: not everything needs to be estimated to the minute. Some tasks are ones where a precise time is unimportant and even harmful.

Creative work lives badly by a stopwatch. If you're generating an idea or writing text, a rigid timer tends to get in the way. Here it's more useful to set aside a block of time and work inside it, without chasing a specific number at the exit.

Estimating tiny tasks one by one is wasted effort. A dozen five-minute jobs are easier to dump into one buffer hour and clear in a row than to measure each.

And don't turn estimating into a fresh form of procrastination. An hour spent on the perfect plan is an hour not spent on the work. The estimate should be rough and fast: multiplier, buffer, go. Precision to the minute rarely pays off.

Frequently asked questions

Why do I keep underestimating even though I know about this bias?

Knowing doesn't switch the mechanism off. Imagination still paints a smooth run, and the estimate drifts toward it. What helps isn't knowledge but action on top of it: mechanically multiply the estimate by your ratio, even when it feels like this time you'll surely be faster. Trust the record, not the feeling.

How much exactly should I inflate the estimate?

For familiar tasks, start with a multiplier of 1.5. For new ones or ones that depend on other people, 2 to 3. Only your own tracking will give the exact figure: log plan versus actual for a week, and you'll see your average miss. Then multiply by it.

What if the inflated estimate won't fit the deadline?

Then the deadline is unrealistic, and it's better to learn that now than on the final night. Options: cut some of the scope, ask for more time in advance, bring in help. All three are worse to do in a panic the night before than calmly a week out.

How do I estimate a brand-new task I've never done?

Find the closest thing you have done and start from it. If there's no close match at all, split the task into small clear steps and estimate each – that lowers the odds of missing a big chunk. And pad more generously: the unknown almost always costs more than it looks.

Isn't a buffer in the schedule just idleness?

Empty time works even when you don't spend it: it keeps the whole plan from collapsing at the first delay. A day packed to the brim falls over at one task running long. A day with a buffer swallows it and rolls on. The emptiness here is insurance, not downtime.

Team estimates still slip. What do I do?

Add up the outside view, not everyone's optimism: how long similar projects actually ran. Break the work into parts and separately budget time for waiting between people – that eats the most. And keep a shared buffer for the project, not just for individual tasks.

Does a tracker help with the planning fallacy?

Yes, because it turns feeling into fact. With the history in front of you of how long similar tasks actually took, there's nothing left to argue the estimate with. A timer shows the real length of a task, and your logged past days show your personal miss ratio.

You can start today, without rebuilding your whole life. Take one task for tomorrow, estimate it as usual, then multiply by 1.5 and write it down. In the evening, compare it with what actually happened. A couple of measurements like that, and you'll see your miss in numbers.

Do It Today helps at exactly this step. You set the day's tasks and see the history of past days – how much actually fit, not how much you planned. The focus timer counts how many sessions a task took, per day and per week, so the estimate stops being a guess and leans on your own numbers. Recurring tasks and the priority board help you leave slack instead of packing the day edge to edge. The tracker is free, installs on your phone like a normal app, works in light and dark themes, and syncs across devices.