Chapter 11A New Mix of Hours: Evolving Personal Rhythm

1. The jobs that remain need a different calendar #

In January 2017, Charlie Rose asked Buffett to pass over the paper calendar. He turned to a week in April and counted three appointments for the whole week. Buffett said that by April it might have become four. Gates added a word: "Taxes."

Gates then set his old calendar beside it. He said he had once filled every minute and treated that as the only way to work. Buffett's calendar, by contrast, could leave a whole day empty. Gates put the difference this way: time is yours to command, and sitting down to think may matter more than constantly seeing people; filling every minute does not prove that a CEO is more serious. Buffett added that people will come for your time, and time is the only thing he cannot buy (Charlie Rose, 2017).

The exchange confirms the appointments in one concrete week and what the two men said about them. It does not prove that Buffett arranges every day this way, and it cannot attribute Berkshire's results to a blank calendar. A chairman and CEO also controls the calendar, delegation, and outside invitations in a way that a support agent, an emergency physician, or an engineer in the middle of a delivery does not. Copy the same calendar onto those jobs and you damage the duty to stand watch or to deliver.

Chapter 10 took pure execution and the hauling of context out of core jobs. The jobs that remain have to carry more trade-offs, exceptions, and revision of standards. "Everyone has to judge" still does not mean "everyone's day looks the same." Some jobs mainly deliver. Some mainly coordinate. Some mainly handle problems that do not yet have a standard answer. The density of judgment inside a job differs, and so does the organization's stage, so the reasonable structure of time differs too.

When the first duty shifts from doing more chores to making a small number of important judgments, the calendar has to answer two questions at once: how much time is left for judgment, and how that time improves the next judgment.

2. What a chopped-up day can show #

Set against Buffett's sparse calendar, many knowledge workers live inside continuous communication and frequent switching.

Microsoft's 2023 Work Trend Index put two kinds of material side by side. The first is an online survey of 31,000 full-time employees or self-employed people across 31 markets. Sixty-eight percent of respondents said they did not have enough uninterrupted focus time in the workday. The second is product telemetry from Microsoft 365. In a 28-day window ending in March 2023, identifiable active work by commercial users in Outlook, Teams, Word, Excel, PowerPoint, and OneNote went, on average, 57% to meetings, email, and chat, and 43% to documents, spreadsheets, and presentations. Users in the top quartile of meeting length spent 7.5 hours a week in Teams meetings. Compared with February 2020, the number of Teams meetings and calls each person joined each week was up 192% (Microsoft, 2023).

Those figures cannot be merged into "31,000 people each received person-by-person telemetry." The survey and the product signal are two sets of material. The 57/43 split is also not a complete audit of working hours. It covers only the identifiable active operations inside Microsoft 365. "Communication" is also not a synonym for "waste." A meeting may only repeat information already on hand, or it may complete a material trade-off. A chat may create an interruption, or it may surface a risk in time. The Microsoft report itself says that respondents named "inefficient meetings" as the leading drag on productivity and "too many meetings" as the third, and that the leading reason people still found meetings valuable was information that helped them finish their work. Whether communication is worth the time depends on whether it serves the present duty.

Gloria Mark, in a 2006 on-the-record interview with Gallup, described a field study that first spent half a day getting used to the site, then followed 36 managers, financial analysts, software developers, engineers, and project leads for three days, recording work events to the second. An "event" here is not a whole project. It is a stretch of continuous use of one device, or of continuous interaction with a person. Formal meetings excluded, a single event lasted 3 minutes and 5 seconds on average before a switch or an interruption (Mark, 2006).

Mark also said that 81.9% of interrupted work was picked up again the same day. Looking only at that same-day pickup, the average gap from leaving to returning was 23 minutes and 15 seconds, with about two other tasks in between. That number describes the interval before the original task is picked up again. It is not "the time the brain needs to regain focus." The study did not define the whole interval as a failure of focus, and it did not count the other work finished in the gap as a loss.

Peer-reviewed papers used different samples and definitions. González and Mark, in 2004, followed 14 people for three and a half days each and reported that work events lasted, on average, "a little more than three minutes." A 2005 paper, expanded to 24 people, reported that 77.2% of interrupted work was picked up the same day, with an average interval of 25 minutes and 26 seconds and 2.26 work units in between (González & Mark, 2004; Mark, González & Harris, 2005). The 23 minutes and 15 seconds comes from Mark's own interview text. It is not in the 2008 interruption experiment that is so often cited.

The material is enough for a limited conclusion: frequent switching means the original task often returns only after other items have sat in between, so a complex judgment needs a stretch of time that is not cut at random. It does not prove that every switch is harmful. Mark herself has distinguished helpful interruptions from obstructive ones. New information inside the same problem may help the work. A wholly unrelated insertion is more likely to get in the way. A calendar audit is not hunting all communication. It is hunting the occupancy that does not match the present duty and that keeps crowding out important judgment.

3. Time is a budget, not the only constraint #

Chapter 5 said that, for a person, you look first at time. "First" is the order of allocation. It does not say that time is the only constraint.

An hour already gone cannot be retrieved, so time is a fit first ledger for a personal trade-off. Choose a meeting and you give up the writing, delivery, learning, or rest that could have occupied the same stretch. A task can be handed off. Money can buy tools and help. A day still has only the same set of hours. The calendar therefore exposes a person's real ranking: not what they say is important, but which matters actually received their time.

Time is still not the only personal resource that can run out. Attention decides whether a person can keep a problem in mind in a given stretch. Energy affects how stable a judgment is. Health decides how long the way of working can last. A person may block out six hours and still not get six hours of high-quality judgment if they have not slept, if they keep jumping among five problems, or if they have been living for a long time with illness and anxiety.

The constraints also migrate into one another. When chores overflow, time is what is missing first. When the calendar has already been cleared and concentration still will not hold, the bottleneck may have moved to attention. After several weeks of important judgments with too little recovery, energy or health may become the harder limit. A manager cannot solve every problem by lengthening the working day, and cannot treat rest as the opposite of judgment work. Recovery done well protects the quality of the next stretch. Recovery left short turns even a well-drawn timetable into a draw on the next stretch.

So this mix of hours only answers what the workday is given to. It does not promise that good judgment will arrive if the table is copied. Time is the unit of the budget. Attention, energy, and health together decide what that budget can buy.

4. Two kinds of work, sorted by main purpose #

To reallocate a calendar, first say what "short-term urgent" and "long-term important" mean. They are not money set against value, and they are not good work set against bad work.

Short-term urgent work mainly keeps current cash flow, customer delivery, and safety in place. Payroll and debts coming due, collections in the present period, deliveries already promised, production failures, data leaks, and compliance events may all require a response the same day. They are not only "firefighting." A delivery on time keeps customer trust. A professional handling of a failure may also leave a long relationship. They sit in the short-term urgent bucket because the first purpose of the present action is to let the organization keep running today.

Long-term important work mainly raises future standards, capability, talent, and trust. Writing a recurring judgment as a standard, building the product capability the next stage will need, developing people who can decide on their own, and repairing a customer relationship that has to be tended over a long period all sit here. They also do not have to wait years to show an effect. A clearer standard of acceptance may cut rework the same day. An effective piece of coaching may improve delivery at once. The first purpose of the present action is still to raise the starting point from which similar problems will be handled later.

The two kinds often overlap. The way to handle that is not a false split. Look at the main purpose, and split the task when you have to.

Suppose an online system fails. Restoring service, protecting data, and notifying the customers who were hit mainly keep present delivery and safety in place, so they are short-term urgent. After service is back, root-cause analysis, a change to monitoring, a stop condition, and a recovery drill mainly reduce the chance of a later repeat, so they are long-term important. One incident can occupy both kinds of time. The earlier and later actions should not be poured into the same bucket. If the second half is forever squeezed back into the firefighting hours, the team will restore service again and again without becoming any less likely to fail.

Customer delivery works the same way. Completing the result for a project due today is a present duty. Writing the change in demand that this delivery exposed into a product judgment, or teaching another member to handle a similar customer on their own, is future building. Cash flow as well: chasing a payment that is due keeps the present running; rewriting credit, pricing, and collection standards changes how cash will be obtained later.

Meetings, email, and chat cannot be classified by the medium. A long meeting that only syncs information already on hand may belong to neither kind. A review that settles a material product trade-off may be long-term important. An incident command call is short-term urgent. The way to judge them is not the name on the invite. Ask what this stretch of time mainly protected, or what it added for the future.

With those definitions, 1–2/6 does not collapse into the slogan that money may occupy only two hours and value must occupy six. What it allocates is two main purposes. It does not deny either kind of work.

5. Stage first, then the job's main task #

The five fields of Part IV share dynamic lifecycle management: locate the object, locate yourself, then match and choose. On time, those three steps are the organization's stage, the job's main task, and the mix of hours on the calendar.

The first step is to judge whether the organization is in survival, growth, or a stable stage.

A survival stage first has to hold cash, core delivery, and safety. When collections have broken, when the product has not yet proved that anyone will pay, or when a major incident is still underway, short-term urgency will naturally take more of the day. If the principal-leader is then required, as a hard rule, to take six hours a day for long-term building, cash, delivery, or safety may have no owner. The organization should still leave feedback when conditions allow, so that the same class of problem does not have to start from zero forever. It does not set a minimum number of hours that applies to every crisis.

A growth stage takes both pulls at once. Customers and deliveries rise quickly, and so do hiring, coordination, and the building of capability. A release, a financing, or a channel expansion may tilt a week's calendar toward present matters, and the next week may need to return to standards, product, and people. This stage is better watched by the week than copied as the same timetable every day.

A stable stage is not a stage without problems. It is a stage in which sources of cash, core delivery, and the rhythm of operations have become relatively predictable. The organization does not have to spend all of management time proving that it can still exist tomorrow, and so it has room to put more of the calendar into the next stage's standards, capability, talent, and trust. 1–2/6 applies first in this stage.

The second step is to judge whether the person's present job is mainly delivery, coordination, or judgment.

A delivery-primary job takes its value from finishing promised results steadily. Real-time support, standing watch, safety response, medicine, and much front-line operations cannot stop responding in order to chase long stretches of blank space. They still need improvement and learning. Long-term important time usually appears as periodic reviews, revision of standards, and rebuilding of tools. It will not, by nature, occupy six hours.

A coordination-primary job takes its value from keeping dependencies, information, and commitments moving correctly. Meetings and communication cannot be judged as waste in advance. A project lead who uses an hour to get five teams to confirm the same risk and the next action may be doing more important work than writing a document alone for an hour. What should be cut first is communication with no object of decision, no necessary participants, and no action afterward.

A judgment-primary job mainly handles exceptions the standard does not cover, trade-offs among resources, and decisions that will not be billed for some time. Principal-leaders, heads of a line of business, and some core specialists often sit here. They also deliver and coordinate. If a day is filled entirely with instant response, the most expensive part of the job never has time to happen.

A job title cannot stand in for this locating. The same founder may be both salesperson and delivery lead in the first year, and only later, in a stable stage, turn the main duty toward product, people, and the allocation of capital. The same engineering lead may be delivery- and coordination-primary in a release week, and judgment- and people-primary in ordinary weeks. When the stage or the duty changes, the mix of hours should be counted again.

Only the third step is matching the calendar. In dynamic lifecycle management, 10/10/80 means that a small share of resource is used to see the organization clearly, a small share to see yourself clearly, and most of the resource to actual matching and choice. It is not a ratio of hours in one day. 1–2/6 is the reference mix for one cell: a stable stage, a judgment-primary job, and some autonomy over the calendar. The two sets of numbers answer different questions. They cannot be mixed.

Figure: 1–2/6 is marked only on the cell "stable × judgment." When the organization's stage or the job's main task changes, the calendar changes with it.

6. 1–2/6 is a reference mix, not a uniform workday #

For a person who sits inside the conditions of use, 1–2/6 is a starting point for an ordinary workday: 1–2 hours concentrated on matters whose first purpose is to keep current cash, delivery, and safety in place, and about 6 hours left for work that raises future standards, capability, talent, and trust. Setting the aim by the day, and checking the drift by the week, is more reasonable than demanding an exact punch of the clock every day. A temporary incident can break that day's mix. It should not, without being named, become a permanent state.

To write the reference as a standard that can be executed, four parts have to sit together: the conditions of use are a relatively stable organization, a judgment-primary job, and some autonomy over the person's own calendar; the default action is to arrange an ordinary workday on 1–2/6; the evidence of acceptance is where the hours actually went each week, whether important judgments moved, and whether energy and delivery have clearly worsened; the triggers for revision are that the organization has entered a crisis or a growth window, that the job's main task has changed, or that a run of results shows the original mix does not fit. Only when the numbers sit with those conditions is the line a standard that can be used and corrected.

The 1–2 hours of short-term urgency, when conditions allow, should be handled in a concentrated window. Cash, delivery, and risk can have fixed checkpoints. Only a matter that truly cannot wait inserts itself at once. Batching is not a way of making customers wait, and not a way of hiding risk. It is a way of stopping every ordinary message from holding the same power to cut an important judgment. If a class of problem keeps coming back, the long-term hours afterward have to find the cause, change the standard, or change the system. Otherwise "we respond very fast" is only a way of paying the same bill at a higher speed.

About 6 hours is also not six hours of "thinking strategy." It contains at least five kinds of work, of different intensity.

Some of it goes to judgment: comparing plans, handling exceptions, finding hypotheses that have not yet been tested. Some goes to writing: putting the judgment, the evidence, and the conditions of use into a record other people can check. Some goes to developing people: feedback, joint review, authorization, so that other people step by step carry a more complete judgment. Some goes to learning: looking for a candidate standard, filling in a key fact, checking whether it can migrate. Room also has to be left for recovery: a meal, a walk, a short rest, and the changeover between blocks of work. Those are not the enemy of judgment. They are a condition of the next stretch of quality.

The five kinds of work have no generally optimal sub-ratio. Some people handle complex trade-offs best in the morning; some write more steadily in the afternoon. A high-risk decision needs more investigation. A day of developing people will have more conversation. Six hours constrains the overall direction of the resource. It does not require the brain to hold peak intensity for six hours running.

Bezos's 2018 interview with the Economic Club of Washington can sit here as a supplement. He said that in the morning he reads the paper, drinks coffee, and has breakfast with his children, that the first meeting is at ten, and that meetings with a high cognitive load are, as far as possible, placed before lunch. By five in the afternoon, if he can no longer think seriously, he leaves the hard problem for the next morning. He also listed eight hours of sleep as a priority, because his thinking, energy, and mood all improve with it. His summary of an executive's duty was this: a senior executive is paid for a small number of high-quality decisions, not for making thousands of decisions every day; if the day holds three good decisions, that is already enough (Economic Club of Washington, 2018).

In the same interview he gave the conditions of use: a startup is different, and Amazon was another matter when it had a hundred people. Senior managers in a mature company can "work in the future." An early team still has to solve a large volume of present problems itself. How a calendar is arranged therefore has to judge stage and duty first, and only then decide how much judgment time to protect.

A job that cannot command six hours does not adopt this number. It can first handle the part it does control: put a long-term matter that keeps being crowded out into a fixed window, record what interrupted it, and ask whether those interruptions truly could not wait. What can migrate is the principle of leaving a stable investment in future capability. What cannot migrate is Buffett's or Bezos's clock.

7. Protected time is not a same-day assignment #

Judgment work does need continuous time. A person has to see the aim, the facts, the constraints, the alternatives, and the possible consequences together before an important trade-off can be finished. Mark's research did not determine how long the optimal block should be. It did record that the original task, once left, often comes back only after other items have sat in between. For a problem that has to keep several relations in view, one or several protected blocks are easier to use than seams scattered through the day.

"Protected" is also not absolute quiet. A joint review of a plan with a key colleague may be the judgment work itself. A piece of information that exposes a safety risk should also cut the original arrangement. What is protected is not solitude. It is the problem that currently matters most. Discussion that bears directly on that problem can enter. Ordinary syncs, approvals that can wait, and invitations with no clear object are handled outside the block.

That a stretch needs protection does not imply six hours in a row, and an interruption cannot be written as if everything in the mind had been wiped. Existing research has not shown that two three-hour blocks must outrun several shorter ones by an order of magnitude, and it has not given every job a universal length of block. Complexity, risk, personal rhythm, and the way people collaborate all change the answer. One management requirement can be kept first: important judgment cannot be finished forever on leftover time.

"Thinking must leave an artifact" cannot be executed as a literal rule. Exploring a new problem, setting aside a plan that is not yet ripe, and admitting that the evidence is not enough may leave no document that can be published the same day, and may still be necessary work. If the review requires every time block to hand in a standard, the number of files becomes the easiest metric to hit, and low-quality entries rise with it.

A more checkable practice is to give an important judgment a window of review. A single stretch of thinking does not have to become writing at once. By the agreed time, it should be possible to see where the problem has been moved: whether it has been redefined, which hypothesis still has to be tested, what the decision is, why there is no action yet, who takes the next step, or which standard has to be revised. Thinking does not have to leave a file every time. An important judgment cannot go forever without feedback.

The requirement on reflection is more specific. Because the task of reflection is to change the next action, it has to form at least one candidate output: a new standard, a revision of a standard, a hypothesis still to be tested, or a next action. A "new standard" here is still only a candidate. It is not a clause that takes effect the same day. Keeping exploration, reflection, and the formal issue of a standard apart preserves the time that exploration needs and lets reflection accept a result.

A personal calendar is also shaped by the organization's environment. If a manager sends a dozen "got five minutes?" invitations in a day, members will find it hard to protect time by personal arrangement alone. If the principal-leader is always firefighting, the team's frequency of escalation, the preparation of meetings, and the long-term blocks will move with that. The principal-leader's calendar is not a public good in the strict sense. It is a public signal of the organization's priorities. Meeting budgets, unified context, and company-wide rules of interruption belong to the consensus cycle earlier in the book and to the ninety-day rebuild at the end. Personal arrangement and organizational signal should not cancel each other here.

8. Put reflection and summary on the calendar #

Chapter 7 already defined two habits. Reflection finds, in what has already happened, a wrong standard, a gap in information, and a hypothesis still to be tested. Summary states the judgment that was reached so that another person can understand it and check it. What the mix of hours has to settle is when they happen, and how they join into one round.

Reflection sits best close to an important experience. At the end of a day, after a commute, or after an important decision, pick the judgment most worth looking at again and ask three questions:

1. What standard was the judgment resting on then, and which conditions were only default assumptions?
2. What did the result or the new information change: a bad standard, thin context, or action that drifted?
3. What will change next time, and how will that be verified?

The three questions do not require a formal standard to be manufactured every day. They require that a serious reflection land in at least one of four kinds of candidate: a new standard, a revision, a hypothesis still to be tested, or a next action. When the evidence is not yet enough, "which evidence to collect, and by when to decide" is the action. When a similar standard already exists, revise it rather than start another one. Fragments of time can catch this material. Driving, caregiving, and any other scene that needs full attention should follow safety first, and should not create a new risk for the sake of finishing a habit.

Structured after-action review does have outside evidence. Chapter 5 already cited Tannenbaum and Cerasoli's meta-analysis of debriefs: a correctly run debrief raised later performance by about 20% on average. The layer that belongs with scheduling is this: experience is more likely to improve performance after a structured look-back. That does not prove that the three questions above are the only form, and it does not prove that more reflection is always better.

Summary carries another task.

Summary is not copying out again what was already noted that day. It is pressing a judgment down to a point where a person without the background can still follow it. It belongs after a weekly review, a project milestone, or the first forming of an important judgment, and it uses a stretch of time that is not cut. In writing or in speech, at least four things have to be clear: what problem was originally being solved, what the judgment rested on, under which conditions it holds, and what evidence would force a change of answer.

That check is often called the Feynman technique. In the 1966 lecture What Is Science?, Feynman gave a nearby test: do not use a newly learned word; try to restate what was learned in your own language; if the relation cannot be said once the term is gone, what was learned may only have been a definition (Feynman, 1966). The popular "four-step Feynman method" is a later compilation. It is not a fixed process Feynman formally published.

When the same test is used in a personal standards cycle, one more piece of acceptance is added: tell it to a person who does not know the background, and ask that person to restate it. The other person does not have to agree with the conclusion. They should be able to say the main judgment, the conditions of use, and what new evidence would change it. If the restatement is only a slogan, return to the summary and fill in the cause. If the range of use is restated wrongly, fill in the boundary. An accurate restatement only proves that expression and understanding passed the check. It does not prove that the judgment has already been verified by reality.

Reflection and summary are therefore not two copies of the same move. Reflection turns experience into a candidate understanding. Summary checks whether that understanding can leave the personal record. Reflection alone leaves the candidate untested by restatement. Summary alone still takes as its input an old conclusion that this round of experience has not revised. Both have to enter the calendar before a personal standards cycle has an entrance and an exit.

9. How a personal standards cycle keeps the library from swelling #

The two habits together can be drawn as one personal standards cycle:

Experience and result → reflection → candidate standard / revision / hypothesis / action → summary → outsider restatement → action and verification → a new result returns

The key of the cycle is not "add something every day." It is what happens later to each candidate.

First, deduplicate. If the new thought is revising the same class of judgment, revise the existing entry, leave the reason for the change and the version, and do not build another line in different wording. Second, write the conditions of use. An experience that holds only for a given customer, stage, or class of risk cannot have the conditions stripped off and become a general order. Third, mark the state of the evidence. One success can propose a hypothesis. It cannot, by itself, prove a standard. A counterexample may overturn an absolute claim, and still not show that the method cannot be used in any scene. Fourth, judge whether it is worth reusing. A low-frequency, low-risk, strongly situated judgment can stay in the record. It does not have to enter a formal library of standards. Fifth, write in advance the signal for revision or withdrawal. What result would require a change, and at which node the line should be looked at again, should be said before the success story appears.

After those checks, a candidate has three destinations. If the evidence is not enough, it continues as a hypothesis and is given a small verification. If it affects only one action, it becomes a named task and is archived when the task is done. Only what is frequent, important, already verified, and likely to be reused by other people is raised to a personal or team standard. After a standard is issued, it still has to accept results, and when needed it is merged, revised, or withdrawn. A healthy library of standards is not measured by count. It is measured by whether the judgments in it are actually called, corrected, and replaced.

The cycle replaces two old genes on the front. Cognitive inertia shows up as the same class of result repeating while the candidate standard is never updated; reflection requires that what will change next time be left on the page. Feedback avoidance shows up as recording the conclusion and not recording the counterexample, the missing evidence, and the conditions of revision; summary and restatement turn those gaps into observable results of acceptance. The cycle does not accept subjective confidence. It accepts whether a judgment can be overturned by reality, and whether, when it holds, other people can use it.

The evidence has a clear boundary. Buffett and Bezos supply the calendar and the self-report of high-autonomy judgment jobs. Microsoft supplies a survey and a portrait of product activity. Mark's samples are small and from an earlier period. Together they support the direction that judgment work needs protection and that the structure of time should match the job. They cannot verify that 1–2/6 is the best ratio. Internal data are not yet complete. They are not cited here, and future results are not used to endorse the present claim.

1–2/6 does not claim that "six hours has already been shown by science to be optimal." It is a default starting point that makes public its conditions, its evidence of acceptance, and its triggers for revision. A user can execute from their own stage and job, and then revise on several weeks of results. If later data change the concrete numbers, what changes is the version of the standard, not the principle that the structure of time should match the responsibility to judge.

Once personal time has been reallocated, the next step is not to keep optimizing the calendar itself. Judgment, writing, developing people, and learning finally have to enter the organization's product and action. Which class of demand survives first, when the radius of innovation is enlarged, and how to keep a new capability from being spent on a direction nobody needs, is the product path taken up next.

What to Do Monday Morning (principal-leader and core-job view) #

Use one week for the first round. Do not try to turn the calendar over in a single pass.

1. Locate the organization's stage. Write one evidence-backed judgment: is the company now in survival, growth, or a stable stage? Give the reason from cash, delivery, demand, and how predictable operations are. Do not let a self-label replace the evidence.
2. Locate the job's main task. Look back over the last month. Were you mainly responsible for delivery, for coordination, or for judgment? Do not look at the title. Look at which class of duty you would have to sign for if the result went wrong.
3. Audit last week's calendar. Mark each stretch by main purpose: "keep the present running" or "add to the future." The same project may be split into two stretches. A stretch that belongs to neither may be marked to delete or to explain. Judge meetings and communication item by item. Do not convict them in advance.
4. Set next week's reference mix. Only a job in a stable stage, mainly responsible for judgment, and with some autonomy over its calendar, starts with 1–2/6 as the aim. In every other case, protect one stretch of long-term important time that can actually be held, and record why it cannot go further.
5. Run one personal standards cycle. Pick an important judgment, ask the three questions, and form a candidate standard, a revision, a hypothesis, or an action. Then tell it in ordinary language to a person who does not know the background, and ask that person to restate the conditions and the signal that would change the answer. On that feedback, decide whether to verify, merge, revise, or archive.

A week later, do not first count how many new standards appeared. Look at three results: whether the important blocks were actually used, which interruptions truly could not wait, and whether this round of reflection and summary changed the next action. The mix of hours is there to accept correction by reality. It is not a prettier form for the calendar.

Chapter Acceptance Self-Check (against chapter acceptance standards) #

1. Claim restatable in one sentence ✓: 1–2/6 is a reference mix for a relatively stable stage × judgment-primary job × some calendar autonomy; it is not a uniform hours mandate, and it moves with stage and main task.
2. Nine-section spine and whiteboard figure ✓: Buffett calendar → Microsoft/Mark → time as budget not only constraint → two purposes → stage then job → 1–2/6 as one cell → protected time without a same-day assignment → reflection and summary on the calendar → personal cycle against swelling. Figure is locate / locate / match, then the cycle; inverted 7h/1h bars are out.
3. Evidence caliber ✓: Charlie Rose 2017 as one week plus self-statement, not Berkshire causation. Microsoft 2023 survey and telemetry kept apart (68% vs 57/43, 7.5h, +192%). Mark 2006 Gallup 3:05 / 81.9% / 23:15 as interval before pickup, not time to regain focus; González & Mark 2004 and 2005 listed beside it; 2008 experiment not used as the source. Bezos 2018 as supplement, with the startup / Amazon-at-100 boundary. Tannenbaum & Cerasoli pointed back to Chapter 5, about 20%, not a test of the three questions or of 1–2/6. Feynman 1966 as restatement without new terms; four-step method not his published process. not cited.
4. Case contract ✓: Buffett as positive personal-layer case; Microsoft and Mark as a portrait of a chopped day, not a "loser." Money/value as a false split, "thinking must leave an artifact," "every remaining job is a judgment job," calendar as public good, order-of-magnitude block claim, weekly count of new standards, and as present proof are out.
5. Classification and mix ✓: short-term urgent vs long-term important by main purpose; survival/growth/stable × delivery/coordination/judgment; 10/10/80 and 1–2/6 not mixed; six hours includes judgment, writing, developing people, learning, and recovery.
6. Chapter boundary ✓: meeting budgets and company-wide interruption rules wait; daily three questions are the new three, not the old "was there a standard / dice roll" set; Feynman only checks summary; product path waits for Chapter 12.
7. Fluency ✓: rewritten in English voice from the Chinese authority; current prose-standard.