How I time-block work, life and ambitions

Why time blocking by the minute breaks for me

I spent a while reading the research on time management, and the conclusion I came out with is not the one productivity videos sell. Working by the clock is not for me. A plan that says work at 9:00, lunch at 13:00, rest at 13:15 and back to work at 13:35 looks great on paper, and I can't feel comfortable living inside it.

I'm active, I'm not patient, and doing a thing at an exact minute is hard for me. When every hour is assigned, I start to feel like a robot and I drop the system - not the uncomfortable part of it, all of it.

Two findings told me this is not only a discipline problem. The first is the planning fallacy: in a 1994 study, 37 students predicted they would finish their theses in 33.9 days and took 55.5, and only about 30% finished by their own date. A minute-level schedule is stacked from exactly those estimates, so it is wrong before lunch. The second is a field experiment with 2,508 Google employees: one group was paid for gym visits only inside a fixed two-hour window, the other for visits at any time. When the money stopped, the share of the fixed-window group still going at least once a week fell by 10 percentage points - double the 5-point drop in the flexible group. The rigid routine lost to the flexible one, and the authors had expected the opposite.

None of this makes time blocking wrong. Cal Newport, who plans every minute of his work day, estimates that a 40-hour time-blocked week produces what 60+ unstructured hours do, and I believe him. It is just not the only structure available. I kept the blocks and took them off the clock.

What gets a clock, and what gets a budget?

The rule came out short. A thing gets an exact time only if my body or other people depend on it. Everything else gets a number of hours.

I still call these time blocks. But a block here is a group of tasks with a number of hours written on it, not a cell in a calendar. My Work block holds three things: the main work and two side projects. Each has its count of hours and its list of tasks, and during the day I decide when to work and how to close those hours.

Sleep is first on the list for a reason. In a study of 60,977 UK Biobank participants, how regular a person's sleep was predicted the risk of death better than how long they slept: the four more regular groups had a 20-48% lower risk of death from any cause than the least regular one. The data is observational and the participants averaged about 63 years old, so I don't read it as a promise. I read it as a reason to protect a bedtime before a workload. My real limit is energy, not hours: after a bad night, five hours of normal work turn into ten or twelve hours of grinding.

How many hours of real work fit in a day?

My budget for the main work is 4-5 hours of real, focused work a day. Not 10-12 hours of sitting near the laptop. If the hours are closed early, the working day is over: the budget is a ceiling as much as a target.

Behind this sits the famous "four hours" figure, and it is weaker than its reputation. It goes back to a 1993 review of deliberate practice by Anders Ericsson and colleagues, which says that practising for more than four hours a day adds almost nothing. In 2019 the sources of that paragraph were rechecked: behind it are small studies of things like typing and memorising digits, nothing that resembles writing code. So four hours is a rule of thumb for me, not a law of the brain.

What is measured better is the shape of the curve. The economist John Pencavel went back to the records of British munition plants in the First World War and saw that past roughly 49 hours a week output stops keeping pace with the hours: at 70 hours workers produced little more than at 56. It is factory labour from a century ago, and I still have not seen a clearer picture of what a 10-12 hour day at the laptop buys.

I had seen the same thing from the other side when I vibe coded a full website in one day: the plan was two weekend days, personal things took one, and the six to seven hours that were left did more for the project than an extra day would have. A fixed number forces the decisions an open day lets you postpone.

Where do ambitions go when the day is already full?

The hardest line in my notes is not about scheduling at all. My real problem was never the schedule. I had eight goals and too few hours. No arrangement of a day fixes that. A budget only shows it sooner, because the numbers refuse to add up.

So the ambitions got a queue instead of a slot each. New goals start one at a time, and the foundation goes first: sleep and training before anything else, then steady main work and getting one side project on its feet, and only after that growing the second one and publishing.

Inside a day, everything long-term lives in one block called Future: 30-90 minutes, not every day, and one thing only. A case study, a blog post, something to learn - whichever it is today, it is not all three. Some of it has no daily box at all, just a weekly budget, so a day can pass without it and the week still closes.

The targets are small on purpose. For a month: one finished case study and 4-8 posts, not daily publishing. I would rather make a few good moves than try to launch twenty projects at once. John Beshears, who ran the Google gym experiment, says the same in plainer words: start with goals you can actually reach, given everything else competing for your time.

Why does Life get its own block?

Life is a block in my system on purpose. It sits next to Work and Health, and it is not whatever is left when the work is done. Games, a walk, friends, a cafe, a film, a trip somewhere, plain rest without guilt. I don't want to optimise my life for productivity, and I don't want it to turn into an endless "work on yourself".

It has a checkbox and no time, and that is deliberate too. Across 13 studies, Gabriela Tonietto and Selin Malkoc showed that when you put a leisure activity on the calendar at a specific time, it starts to feel like work and you enjoy it less, and that scheduling it roughly - "after work" instead of "at 18:00" - makes the effect disappear. That "rest at 13:15" line from the plan at the top is not rest. It is one more task.

The bigger reason is what time management turns out to be for. The largest review in this field, a meta-analysis of 158 studies and 53,957 people, found that time management is only moderately related to job performance, with correlations around 0.25, and that it improves wellbeing - life satisfaction above all - more than it improves performance. An older study by Therese Macan points at the mechanism: the benefit ran through feeling in control of your time - less tension, more job satisfaction - and did not show up in job performance. A system that protects only the work is solving the smaller half of the problem. A budget gives me back the larger half: the day is mine again, and the hours still get closed.

The one line I write before every switch

Three kinds of work and a life in one day mean a lot of switching, and switching is where efficiency leaks. Sophie Leroy gave the leak a name in 2009: attention residue. When you leave a task unfinished, part of your attention stays on it, and the next thing comes out worse.

The cost is easy to miss. In a 2008 experiment by Gloria Mark, people who were interrupted finished their tasks faster and at the same quality - and reported more stress, frustration, time pressure and effort. The day's output looks normal, and the bill arrives somewhere a timesheet never looks.

Part of my day happens where interruptions are not mine to control. Someone needs something, a message arrives. So out of everything I read, I kept the smallest tool: before I turn to whatever interrupted me, I write one line.

stopped here → next step: <what>

It has a name in the literature. Leroy and Theresa Glomb call it a ready-to-resume plan, and they tested it in four studies: people who took a minute or less to note where they stopped and what they would do on return carried less attention residue into the interruption and handled it better - they made better decisions and remembered more.

That result is worth a second read, because it is not about the thing you would assume. The note did not help people go back to the task. It helped them do the interrupting thing better. What got measured was the interrupting task, not the return. So the line I write when someone needs me is for that person: it is what lets them get my whole head instead of the half that is still in the task.

A second study points the same way. Michael Parke and colleagues surveyed employees every day for two weeks and compared two ways of planning a day. Ordinary planning - lists, priorities, what to do when - helped only on calm days: the more interruptions, the weaker its benefit. Contingent planning, where you assume you will be interrupted and decide in advance what happens then, kept working on the noisy days too. A stop note is the same thing in one line.

I use the idea on my tools as well. In Claude Code + Obsidian without the flat log the problem was an agent's work log that could tell me what happened and never where we were. The stop note is the same fix at human size: it records where I am, not what I did.

Six checkboxes instead of a calendar

A budget has to be counted, and I refuse to track my life in 30-minute cells or build a complicated Notion for it. The whole system fits on one notebook page.

In the evening it takes 30 seconds: I look at which blocks got closed today. Six out of six does not happen every day, and it is not supposed to. For bad days there is a minimum day - three boxes: the mandatory work, the personal habit, and not wrecking tonight's sleep. If those are ticked, the day is not a failure.

Two more rules keep the system from lying. A day gets at most three main results: the mandatory work, one thing for the future, and one thing for health or life. Anything on top is a bonus. And the budget for some things belongs to the week, not the day: if I don't touch a side project on a Tuesday, that is fine, and on Sunday I compare the plan with what happened.

What an hour budget buys, and what it costs

The first row is the one that fails loudest. A timetable nags; a budget waits politely until 23:00 and then hands you the bill. My guard is one rule for bad days: unfinished hours move into the afternoon, never into the night. Plus the evening check, which makes a skipped block visible the same day instead of at the end of the month.

Research has a fix for the missing alarm, and it is not a time. Peter Gollwitzer's work on if-then plans - 94 tests and a medium-to-large effect on actually reaching the goal - says an action tied to an event starts more reliably than one resting on intention alone. So a block can start on a cue instead of an hour: breakfast ends, the laptop opens on the first task.

I have run this version of the system for about a week so far, and it suits me: with a lot of ambitions I can still answer for the result in every place I work, and it helps in my personal life too.

If you want to try one piece of this, don't start with the budget or the boxes. Take tomorrow's plan and erase every start time that neither your body nor another person depends on. Then write a number of hours next to what is left.