A 4-Hour Workday and +22% Revenue: Parkinson's Law, Ericsson's Focus Ceiling, and an Honest Causality Check on My Own Numbers Author: Дністер Published: 2026-10-18T03:01:45.000Z Language: en URL: https://neurodrift.org/en/blog/skorotyv-robochyy-den-do-4-godyn-na-kvartal-dohid-vyris-22/ Original (Ukrainian): https://neurodrift.org/blog/skorotyv-robochyy-den-do-4-godyn-na-kvartal-dohid-vyris-22/ Tags: productivity, focus, time, work, revenue Original source: https://neurodrift.org/blog/skorotyv-robochyy-den-do-4-godyn-na-kvartal-dohid-vyris-22/ A contrarian case study from real numbers: less time, more revenue. Parkinson's Law, WHO data on 745,000 deaths from overwork, Ericsson's focus ceiling, and an honest interrogation of whether my +22% is causation or coincidence. Zero motivational clichés. ----- The violinist and the clerk hit the same wall Ask someone in their seventies how many hours a day they are truly at it, and they won't take offence at "not many." Three, four — then the body switches the lights off, and that's not laziness; it's the bill arriving for all the overtime of youth. The productivity blogger doesn't know this because they haven't lived long enough to see the invoice: they confuse "being present" with "extracting value" and sincerely believe that hour twelve at the desk is somehow different from hour three. The pensioner stopped confusing them long ago. They know the central fact this article will prove with numbers over the next twenty minutes: a calendar doesn't scale. However many hours you stuff into a day, the day stays a day, and you in it — exactly as long as you last. And here is the science on precisely that ceiling. K. Anders Ericsson spent half his life with a stopwatch beside students at the Berlin Academy of Music, counting how many hours the people we call geniuses actually worked. What he found was inconvenient: when he totalled the genuinely focused playing of the best violinists, it came to roughly three and a half hours per day — spread across several sessions, with a mandatory afternoon nap between them. Not because they were lazy. Because beyond that point the brain stops absorbing notes no matter how long you draw the bow. Ericsson himself wrote plainly in his summary review: more than four hours of concentrated practice per day yields essentially no benefit. The hard biological focus ceiling sits at around four hours a day, and even the elite slams into it. The only difference is that the elite, unlike you, doesn't pretend the ceiling doesn't exist. Now a twin figure from a different world. According to RescueTime's analysis of 185 million working hours, the typical knowledge worker has roughly 2 hours 48 minutes of truly productive work per day — without switching, in focus. That's the hard number from the primary report. How many hours they're physically present at a desk — nine or ten — is harder to pin down exactly: RescueTime doesn't publish a separate "presence" metric; it gets estimated from screen time plus meetings, calls, and phone in hand. So let's hold the honest figure: productive time, roughly 2:48; the rest of the shift, fog. Fog of emails about emails, status updates about status updates, meetings that should have been emails, and emails that shouldn't have existed. Translate that into the language a seventy-year-old uses standing in the pension queue: over a shift you bring home three solid hours of work, but the invoice is written for all nine. The remaining six are that same old man ahead of you in the queue, who showed up an hour before the window opened — not because it's faster that way, but because otherwise the day would feel empty. Busyness as a way of not being alone with silence. Only in an office this queue is also paid by the hour — and called a career. Notice? The violin genius and the open-plan clerk hit the exact same wall somewhere around three to four hours of real work. The difference is one thing only: the violinist then goes to sleep, because they respect the ceiling. The clerk performs work for another six hours in front of witnesses. Because you get paid for presence, and for the ceiling — you get shamed. We built an entire economy that pays people precisely for the hours in which they are objectively incapable of anything. I stopped buying tickets to this theatre. I cut my operational working time at Dnister Strategy to four hours per quarter — not per day, per quarter. And over the next year revenue grew by 22%. This is not a motivational story. It's an investigation. Including an investigation of whether I have any right to say that number out loud — because that's exactly what someone will grab first, to sell you a new life hack. Vasya the Hero and his fourteen tabs I have an antagonist, and he has a face. Let's call him Vasya the Hero — a composite but painfully specific Ukrainian entrepreneur from my circle. At 11:40 p.m. Vasya posts a browser screenshot to his stories: fourteen open tabs, a dashboard, a half-drunk coffee, caption: "who didn't sleep — won." On Sunday at 6 a.m. he replies to an email, and you see that timestamp, and it's part of the message. It actually is the message. The text of the email doesn't matter: Vasya didn't send you information — he sent you proof that he didn't sleep. Vasya is not a bad person. That's the most important thing to understand about him. Vasya is devoted. He sincerely believes my four hours per quarter is a betrayal — of the team, the business, himself. And I couldn't argue with him for a long time, because I used to be him. In 2019 I wore sleep deprivation as a medal too and confused exhaustion with contribution — the only difference between me then and Vasya is that no one stopped me in time; someone just praised me instead. Vasya's problem is not his character. The problem is that he can measure only one indicator — the number of hours spent — and therefore treats it as output. It's like a driver whose only working instrument is the odometer: he's genuinely convinced that the one who logged more kilometres driving in circles in the car park has gone further than the one who took the highway. And what's most frightening is that Vasya shows off this odometer to other drivers. Vasya isn't lying about his busyness. He simply can't tell it apart from output — and so sells one under the guise of the other. First and foremost, to himself. The macro version of Vasya is "996" culture, celebrated to the point where Jack Ma publicly called the regime "9 a.m. to 9 p.m., six days a week" a "huge blessing" for employees in 2019. The same cult of busyness, only scaled to a national level and hoisted on stage to applause. And here the story stops being funny, because this blessing had a price tag — not written in money. The theatre of presence has always existed. But in 2025–2026 it acquired a new stage and new producers. AI assistants are taking over the first four hours of typical knowledge work — routine synthesis, drafts, research, formatting. If your day consists of exactly that, it's already automated; you're just still going to the office as a prop. Meanwhile in Ukraine, wartime operational austerity has made the focus-ceiling question not academic but operational: fewer resources, fewer people, the lever and the empty mileage are clearly visible. What in 2019 looked like a management quirk ("Ford cut the day and nothing collapsed") is in 2025 basic survival arithmetic: either you know which four hours are moving your revenue, or you're just burning fuel that no longer exists. !Vasya at 11:40 p.m. photographs his own fourteen tabs for his stories; through the glass an empty office where the boss times the burning of the lamp with a stopwatch rather than the result… Vasya didn't send information — he sent proof he didn't sleep: outside it's been dark for hours, the output tray is empty, and the boss is still timing the lamp with a stopwatch. The sloth beside him reaches one paw toward a leaf — and will get more done than the whole night shift. 745,000 rows in a table I don't want to be in In 2021 the WHO and the International Labour Organization counted, for the first time, what overwork costs the world. Working 55 or more hours a week killed 745,000 people in 2016 alone: 398,000 from stroke, 347,000 from ischemic heart disease. That's 29% more than in 2000. Working 55+ hours raises the risk of stroke by 35% and death from heart disease by 17% compared with a healthy 35–40 hours. At the time, 488 million people fell in the danger zone. Re-read the word "killed." Not "harmed health," not "reduced well-being." Killed. Vasya's 11:40 p.m. story is not a photo of success. It's, statistically, an application for a place in a table of 398,000 rows where the cause-of-death column is already filled in, and the name isn't yet. When people call me lazy, I no longer argue about ethics. I respond epidemiologically: I'm not lazy, I just didn't volunteer for an overwork mortality study. Between "heroism" and "morgue," the cult of busyness quietly chose the morgue, hung a sign saying "heroism" over the entrance, and kept recruiting volunteers. Someone will say: fine, overwork damages health, but it produces more output — just at the cost of a life. A fair trade. Here's the trap — even that isn't true. More hours stops meaning more output long before it starts meaning a heart attack. You pay with your life not even for productivity. You pay with your life for the performance of productivity. Seventy hours that produce less than forty-eight John Pencavel of Stanford took archival data from the British Health of Munitions Workers Committee during World War I — people whom wartime need was literally pushing to work at the limit — and plotted a productivity curve per hour. The curve is brutal. Up to a point, everything is nearly linear. After roughly 49 hours per week, output per hour drops sharply. And then comes the diagnosis that should make the entire overtime industry blush: total output for those who worked 48 hours over six days was higher than for those who ground away for 70 hours over seven. Not "the same." Higher. The seventh day and the extra twenty hours didn't just vanish to zero — they went negative, because fatigue began eating into what had already been done. Sit with that. Twenty-odd hours a week — a full two and a half extra shifts — don't merely evaporate. They work against you. Vasya, sitting until midnight seven days in a row, doesn't produce more than someone who rested on Sunday. He produces less, and simply pays for that worse result with the most expensive of his non-renewable assets — and calls the overpayment devotion. Forty-eight hours over six days produces more than seventy over seven. The extra time isn't work. It's self-immolation with a nice caption in the stories. And this is not a fresh discovery from the burnout era. It was calculated a hundred years ago. On May 1, 1926, Henry Ford switched his factories to a 40-hour week — without cutting wages by a cent. Not out of love for the worker. Out of arithmetic. Picture the scene: a magnate sitting over the numbers, seeing that the last hour of a shift yields almost zero incremental output, while the exhausted assembly-line worker starts producing defects that then have to be reworked at the company's expense. Ford cut the hours not as a concession to weakness but as a management lever. He didn't gift workers a day off — he simply closed an unprofitable workshop that, a hundred years later, we still proudly keep open and call "overtime." Why compressed time does more: Parkinson running in reverse On November 19, 1955, British naval historian Cyril Northcote Parkinson published a satirical essay in The Economist and formulated a law that outlasted every serious management theory: "Work expands so as to fill the time available for its completion." The illustration he gives is more elegant than any I could invent. An elderly lady is going to send a single postcard to her niece. For a busy person, this takes three minutes between other things. The lady spends an entire day: an hour looking for the card, another finding her glasses, half an hour tracking down the address, and then agonising over whether to take an umbrella to the postbox around the corner. The day was free — the task ate the whole day without choking. Not because the task was large. Because the time was there. And here is the key to my whole story. If work expands to fill available time, the law runs in the other direction too. Compress the time — and the work compresses with it. Give a task a quarter and it will dissolve to wateriness, growing meetings, "let's think some more," and endless versions. Give the same task four hours and it crystallises down to a hard residue of pure levers. An extended deadline is a solvent. A compressed one is a crystalliser. Most "complex" tasks are complex exactly as long as you've generously allocated them time to pretend to be so. My four hours per quarter work not against this law but because of it. I'm not a focus genius. I simply stopped handing work the solvent from the storeroom. !A calm founder with jacket over arm walks out having closed a thin folder of decisions; through the glass — overwork panic, an employee asleep at their keyboard, and a manager… Compress time and work crystallises into a hard residue of pure levers; stretch it and it dissolves into meetings and "let's think some more." One person has already left with a slim folder of decisions. Behind the glass they're still panicking at two in the afternoon. The sloth overtakes the whole shift by moving slowly. Why a fragmented day physically loses to a compressed block There's one more reason — purely physiological. Sophie Leroy of the University of Washington described in 2009 an effect she called "attention residue." When you switch from one task to another, part of your attention physically remains on the previous one — especially if you left it unfinished or it was under deadline pressure. You've already opened a new tab, but the brain is still washing dishes from the previous one. Now picture Vasya's eight-hour day. Email, meeting, call, chat, email again, "just a minute," chat again. Every switch leaves a residue. By lunchtime his attention is a glass of water stirred with the murk of twenty different tasks: nothing is visible in it anymore, but Vasya confidently drinks and praises it. He's present for eight hours, but working through silt — and sincerely mistakes murkiness for a full day. Four focused hours without a single switch beat eight hours in thirty-tabs mode not because that's a beautiful metaphor, but because the water in the glass has time to settle. Fewer inputs — cleaner output. The same principle at the level of whole organisations was stated by Kelly Johnson at Lockheed Skunk Works — the division that built the U-2 and SR-71. His iron Rule #3, verbatim: the number of people involved in a project must be limited "almost cruelly," keeping the team at 10–25% of a "normal" headcount. Fewer people — fewer residue layers between them, just as fewer hours mean less residue inside a head. The classic organisation's instinctive response is the opposite: a project bogs down — throw twenty more people and twenty more meetings at it, so the residue forms faster and more reliably. An eight-hour day is not eight hours of work. It's three hours of work spread thin across nine hours of presence, so no one suspects the ceiling was passed long ago. A tool: cut lever from theatre Before I get to my number — and to why you shouldn't take it at face value — here is the tool without which cutting time produces not revenue but chaos. Run every task through four questions before you start. If even one answer is "theatre" — the task doesn't make it into your four hours. Question for the taskLever (keep it)Theatre of presence (cut or delegate) What changes in revenue in 90 days if I do NOT do this?A specific number or deal falls through"Well, it'll look a bit sloppy" Does this move a lever, or does it merely service a process?Decision, price, deal, hire, productStatus about a status, meeting about a meeting Am I irreplaceable here, or just present?Only the owner can make this decisionAnyone on the team would do it just as well Am I doing this for the result — or to be seen as busy?The result exists even without witnessesThe value disappears if nobody's watching The last row is the most painful. Because it's precisely what separates me in 2024 from Vasya. Vasya's work at 11:40 p.m. mostly won't survive the fourth question: remove the stories audience and the point of the work evaporates, leaving only exhaustion without an audience. My task for those four hours has to exist in the dark, with no witnesses — otherwise it doesn't get in. Theatre is trivially identifiable: it goes dark the moment the camera switches off. Now let's cross-examine my own number I promised honesty, so here it is. I have no moral right to throw "+22%" and walk out with my head held high. Because that's what charlatans do — and I'm about to show you exactly how, using someone else's example first, so you can recognise it later in mine. But first — the strongest argument against my approach. Because if I don't say it, your investor or your manager will, and they'll be right. The cult of busyness is not just mass hysteria. In certain contexts it's rational. In a low-trust organisation where results are hard to verify, presence substitutes for trust: the manager can't assess whether your code is good, but can see you were in the office for twelve hours. Presence as a signal is not Vasya's stupidity — it's a solution to an asymmetric-information problem. There's also a coordination argument: when fifty people need to be simultaneously available for synchronous work, someone has to be on standby — and "busyness" is an acceptable way to distribute that burden. And there's the scale argument: in a zero-to-one startup, a founder who leaves at six genuinely risks more than one who stays — because critical decisions concentrate in one head, and deferring any of them costs more. But here's what those arguments don't explain. First, they describe specific edge cases — not the norm for most office environments, where the task is no longer startup-shaped, the result is already verifiable, and the cult of busyness persists purely by inertia. Second, even in those edge cases Pencavel's data doesn't disappear: seventy hours over seven days produces less than forty-eight over six — regardless of organisational trust level. An exhausted manager in a startup makes worse decisions, not more of them. Presence as a signal solves the trust problem but not the quality-per-hour problem. Those are different problems — and the cult of busyness solves only the first while pretending that is also the second. The British 4 Day Week Global pilot — the largest at the time, 61 companies, roughly 2,900 workers, June–December 2022. The headlines screamed "+35% revenue." Sounds like a verdict for the four-day week. But open the primary report and a nuance emerges that changes everything. Revenue directly during the pilot period rose only 1.4% (weighted by company size). The "+35%" is a year-on-year comparison to equivalent periods in previous years — i.e., the normal growth trend of these companies, not the effect of the shorter week. Between "+1.4%" and "+35%" there's no lie in the numbers — there's merely a quietly swapped denominator. And exactly that swap powers half the business inspiration in your feed. The honest message from the pilot sounds considerably more modest than the headline: cutting time did not collapse revenue, and existing growth continued. And that's actually a strong result — just a different one. Because the real marvels of the pilot weren't in revenue but alongside it: staff attrition fell by 57%, burnout by 71%, and 56 of 61 companies (92%) decided to keep the four-day week. People, it turns out, are remarkably willing to produce the same amount in less time — oddly, it took a pilot to prove it. CaseWhat happened to timeHonest result Microsoft Japan, August 2019Closed the office on Fridays+39.9% sales per employee; −23% electricity; −59% printing; 92% loved it UK pilot, 20224-day week, 61 companiesRevenue during period +1.4% (NOT +35%); attrition −57%; burnout −71%; 92% continued Iceland, 2015–201940 → 35–36 hrs/week without wage cutsProductivity unchanged or higher; stress fell; by 2021 ~86% of the workforce gained the right to a shorter week Dnister Strategy (me)4 operational hours per quarter+22% revenue over a year — causation or coincidence? Honest answer: not proven Now myself, with the same ruler. My "+22%" is from my own archive — the number is real but unverifiable from the outside. And most importantly: I have no control group. I have no second version of myself working 50 hours a week under the same market conditions to compare against fairly. Maybe revenue grew because of the time cut. Or maybe the market was simply good, the previous product had matured, a deal from the year before closed — and I then claimed someone else's credit, like the rooster who believes it raises the sun with its daily crow. The most convenient error in the world: stand next to the outcome and decide you're the cause. The only thing I can state honestly: cutting time to four hours per quarter definitely did NOT collapse revenue — and beyond that begins the territory where I am obliged to say "I don't know" rather than sell you a neat causation. And precisely that — that "I don't know" — is what separates my story from Vasya's stories. Vasya sells you fourteen tabs as output and doesn't even blush. Headlines sell you "+35%" that is really "+1.4%." I could sell you "+22% thanks to the four-hour quarter" and collect the likes in one move. But that would be exactly the same counterfeit coin, just from my side of the table — and it would ring just as convincingly as Vasya's. Four hours without a team is not a lever — it's desertion Here I have to cut my own thesis before someone less friendly does it for me. Because "I cut down to four hours" sounds like a trick repeatable from Monday — and that's a trap worse than Vasya's. My four hours work precisely because of something I've been diplomatically silent about for seven paragraphs: beneath me there is a team and a system that holds everything else. Remove them — and those same four hours turn not into a lever but into desertion from your own post, dressed up as enlightenment. The difference between a lever and desertion is measured by one question: when you walk out the door after your four hours, does the work continue or stop? If it continues — you've built a mechanism and are genuinely standing at the lever. If it stops — you simply stopped doing it and called that a focus methodology. A founder without a team who "switched to four hours" doesn't get rising revenue a week later — they get a queue of clients asking why nobody's answering. Cutting time without delegating is not Ford closing an unprofitable workshop. It's Ford going home and leaving the assembly line running into empty air. Here the lens of a person in their seventies shows what the thirty-year-old optimiser can't yet see. The pensioner has already been through the only honest school of delegation — the biological one. The body no longer allows you to carry everything alone, and the choice becomes bare: either you transferred the work in advance — to children, a successor, a system — and it lives without you, or you transferred it to no one and it simply ends when your strength does. There is no third way. The productivity blogger treats delegation as a time-management technique, an optional life hack for later. The old person knows it's the only way anything outlives the person who did it. My four hours are not about doing less. They're about building a business that breathes without my lungs — while I still have the luxury of making that choice from a desk rather than a hospital bed. So the honest recipe sounds more boring than the headline and carries small print: first — the system and the people who can absorb the theatre, and only then — your four hours of lever. In the reverse order this is not freedom. It's an abandoned post with a nice caption in the stories. Who profits from your Vasya Before I close this conversation — one question the cult of busyness carefully never asks: who won while Vasya became Vasya? Not rhetorical. Concrete, with an answer. Start with the most shameless player. Time-tracking SaaS — Toggl, Clockify, Harvest — costs money for exactly as long as managers believe that counted hours are a proxy for output. At bottom it's a taximeter installed in a horseless carriage: it cheerfully clocks every step the horse takes in place and never once asks whether the distance between cities has shrunk. The manager watches the number grow and pays for it as though it were mileage. Eliminate the cult — eliminate the market, because who pays for a taximeter that finally admits the carriage isn't moving. Next — attention platforms. LinkedIn, Instagram, X monetise not content but exhaustion: a tired person scrolls longer, clicks uncritically, buys a "how to do more" course at 11 p.m. when there's no mental energy left for scepticism. Busy Vasya is not a user. He's advertising inventory who displays himself in the shop window every night. Finally, the managerial hierarchy in corporations where status is measured by the number of subordinates and their logged hours: cut the time — cut the budget — cut the influence. The stake in keeping the theatre running is structural, not personal. Vasya didn't invent any of this himself. He was simply applauded for a very long time by people who benefit from his sleep deprivation — and applause, unlike sleep, the brain accepts as nourishment. The ceiling you either respect or die against Let's return to the violinist at the Berlin Academy of Music. They played their three and a half hours, went to sleep — and while a competitor next door saws through scales for the sixth hour running, the violinist recovers. The competitor next door doesn't get better. They get more tired, leaving residue on the strings and the first signs of a nervous breakdown, certain they're outplaying the rival precisely with that sixth hour. In a year, the one who slept will play the solo. Not because they wanted it less. Because they respected the ceiling instead of fighting it with bloodied fingers. The focus ceiling exists equally for the violinist, the Skunk Works engineer, the Ford assembly-line worker, and you reading this between two meetings. The question is not whether it exists — Ericsson, Pencavel, and the WHO closed that question. The question is what you do with the rest of the day after the ceiling has been passed. The violinist goes to recover. The clerk performs work in front of witnesses. Vasya posts stories. Meanwhile the stroke table calmly adds rows and doesn't ask anyone whether they considered themselves an exception. I cut my operational time to four hours per quarter not because I found a life hack. But because I stopped lying to myself that the fifteenth extra hour was a contribution rather than self-immolation with a nice caption. Revenue grew 22% in the process. Maybe because of it. Maybe despite it. I honestly don't know — and I refuse to pretend I do, even if that honesty costs me half the applause. But one thing I know for certain. When Vasya next posts his screenshot at 11:40 p.m. with fourteen tabs and the caption "who didn't sleep — won," I won't think "what a champion" or "what a fool." I'll think of that old man in the pension queue who came an hour before opening so the day wouldn't feel empty — and of Vasya standing in the exact same queue, only the window at the end of it is different. Above it hangs a sign saying "reward," and in the registry behind the glass there's a column from a 745,000-row table, and the name is already being filled in while he patiently shifts from foot to foot. He's certain he's first in line for the prize. In reality, he's just the first to learn what the person in their seventies knew from the very beginning: a calendar doesn't scale, and no overtime hour buys you an extra day. And I'll put my phone away. Because my four hours for today — I've already worked them. And a ticket to his theatre, I no longer buy. Sources: Microsoft Japan "Work-Life Choice Challenge" 2019, +39.9% sales per employee (August 2019 vs August 2018), −23.1% electricity, −58.7% printing, ~92% positive, ~2,300 employees — CNBC: https://www.cnbc.com/2019/11/04/microsoft-japan-4-day-work-week-experiment-sees-productivity-jump-40percent.html UK 4 Day Week Global pilot 2022 (revenue during period +1.4%, year-on-year +35%, attrition −57%, burnout −71%, 56/61 = 92% continued) — Autonomy Institute: https://autonomy.work/portfolio/uk4dwpilotresults/ Iceland experiment 2015–2019, 40 → 35–36 hrs without wage cuts, ~86% of the workforce — World Economic Forum: https://www.weforum.org/stories/2022/01/four-day-week-work-life-balance-trial/ Parkinson's Law, essay in The Economist, 19.11.1955, illustration with the elderly lady and the postcard — Wikipedia: https://en.wikipedia.org/wiki/Parkinson%27slaw John Pencavel, "The Productivity of Working Hours," The Economic Journal, vol. 125, Dec. 2015 (output at 48 hrs/6 days higher than at 70 hrs/7 days; output per hour drops sharply after ~49 hrs; data from the Health of Munitions Workers Committee). The press formula "70 ≈ 55" is a simplification; the paper's own comparison threshold is 48 hrs. Briefing Royal Economic Society: res.org.uk; Wiley abstract: onlinelibrary.wiley.com WHO / ILO Joint Estimates, 745,000 deaths from 55+ hrs/week in 2016 (398,000 stroke, 347,000 heart disease; +35% stroke, +17% heart disease; +29% since 2000; 488 million people) — WHO, 17.05.2021: https://www.who.int/news/item/17-05-2021-long-working-hours-increasing-deaths-from-heart-disease-and-stroke-who-ilo K. Anders Ericsson, Krampe & Tesch-Römer (1993), violinists at the Berlin Academy of Music; in Ericsson's review — benefit of practice beyond ~4 hrs/day is essentially nil — Wikipedia: https://en.wikipedia.org/wiki/K.Anders_Ericsson RescueTime, "State of Work Life Balance 2019": ~2 hrs 48 min of truly productive work per day (analysis of 185 million working hours). The "9.6 hrs of presence" metric is not published directly by RescueTime — it is a derived estimate (screen time + meetings + phone); noted as approximate in the text — RescueTime Blog: https://blog.rescuetime.com/work-life-balance-study-2019 Sophie Leroy, "Why Is It So Hard to Do My Work?", attention residue, OBHDP 2009, 109(2):168–181 — IDEAS/RePEc: https://ideas.repec.org/a/eee/jobhdp/v109y2009i2p168-181.html Lockheed Skunk Works, Kelly Johnson's 14 rules (Rule #3: team limited "almost cruelly," 10–25% of normal headcount) — Lockheed Martin: https://www.lockheedmartin.com/content/dam/lockheed-martin/aero/photo/skunkworks/kellys-14-rules.pdf Henry Ford, May 1, 1926, switch to a 40-hour week without wage cuts — History.com: https://www.history.com/this-day-in-history/may-1/ford-factory-workers-get-40-hour-week Jack Ma on "996" as a "huge blessing," 2019 — CNBC: https://www.cnbc.com/2019/04/15/jack-ma-defends-the-12-hours-a-day-6-days-a-week-996-work-culture.html Personal Dnister Strategy figures ("+22%," "4 hours per quarter") are from the author's own archive, unverifiable from the outside; in the text they are explicitly marked as unproven causation.