---
title: "Chapter 13: The Collapse Itself"
book: "Collapse or Collapse"
author: "Vinay Pasricha"
url: https://vinaypasricha.com/books/collapse-or-collapse/chapter-13-the-collapse-itself
status: "On Kindle: https://www.amazon.com/dp/B0HNL53J5R. The full text is shared on vinaypasricha.com with the author's permission."
---

# Chapter 13: The Collapse Itself

## 1. Revenue first, and whole

Twelve chapters have given the reader the parts. The vector and its cockpit, the gate, the agents, the master’s control room, the Keeper’s command centre, the twelve tenets, the pay, the selection. What they have not given is the order. A firm cannot collapse everything on one morning, and it cannot collapse one thing a quarter for five years and call the result a collapse. This chapter is the order. It has two rules before it has a calendar.

The first rule is revenue first. The Outcome that stands closest to the invoice is collapsed before any other, because its result is the one nobody in the building can argue with. When a quote that took nine days goes out in four hours and the order comes back, the owner holds a number that no department can explain away. Every department collapsed after that is collapsed in the shadow of that number. Collapse the plant first and the owner spends a year defending a cost saving. Collapse the revenue chain first and he spends the year explaining a growth figure. Chapter 10 warned that Tenet II is broken first by the functions kept at the top. Revenue first is the answer to the warning. It puts the result in the owner’s hands before his own desk comes up for the test.

The second rule is that the revenue chain goes whole. The chain from a customer’s first word to cash in the bank runs today through sales, the order desk, contracts, dispatch, invoicing, collections, and the forty people who carry messages between them. A chain with one department left standing in the middle of it is not half collapsed. It is dead. The vector on either side of the surviving department waits on it, and a vector that waits is a department with a new name. So the revenue chain is collapsed on one Monday, every department in it, into every vector it needs, and nothing else in the firm is touched that day. The rest follows one department at a time, each proven before the next, by the only test there is: the Outcome’s result for the week, with no department left in the chain.

## 2. The firm

The firm is imaginary, and has to be. No firm has finished what this chapter describes, and the one I know best has not finished either.[^45] Call it Pravaha Pumps. It makes industrial pumps and valves in a plant two hours from Delhi. It has a thousand people, counting the owner, a turnover of four hundred and eighty crore, and ships two hundred units a month against order, at an average of twenty lakh a pump. It is owned and run by one man, who is its board and its managing director at once, which is common and is this chapter’s last problem. It has ten departments, which is also common, and a hundred people in a corporate office whose work is keeping the ten departments informed about one another. Chapter 2 explained that office.[^46]

| Department | People | What it does all day |
|---|---|---|
| Sales and marketing | 90 | Enquiries, visits, quotes, the order desk |
| Production | 420 | 300 at machines and benches; 120 planning, supervising, chasing, recording |
| Procurement | 35 | Buying against the plan, and chasing the buying |
| Quality | 60 | Inspecting at inward, in process, and at dispatch |
| Finance and accounts | 45 | Invoicing and collections 20; treasury 10; statutory accounts, tax and audit 15 |
| Legal and secretarial | 10 | Contracts 6; secretarial and litigation 4 |
| HR and administration | 40 | Hiring, payroll, the law of employment, the canteen |
| IT | 25 | The systems, and the people who keep the systems talking to each other |
| Logistics and stores | 110 | Inbound stores 70; dispatch and outbound 40 |
| R&D and design | 65 | New products, changes, drawings for the plant |
| Corporate office | 100 | The owner, his directors, their assistants, the PMO, the MIS |
| **Total** | **1,000** | |

The plant runs one shift of eight hours, six days a week. Inside the shift its machines cut metal a little over half the time. Put together, the machines are working sixteen per cent of the hours in a week and standing for the rest.[^47] The reader should hold that figure to the end of the chapter, because four times sixteen is sixty-four, and sixty-four per cent of a week is what a machine does when it loads itself and nobody goes home. The fourfold output this chapter ends with is mostly not bought. The machine hours are standing in the plant today, switched off; what is bought is the ability to use them.

## 3. Month zero: the sort

Before one person moves, the ten departments are sorted on paper by the one test. Does the work produce an independent result, or is it a skill used in producing one? There is no third answer, and the sort takes an afternoon.

Sales produces a result: an order won. Outcome. The order desk, dispatch, invoicing and collections together produce a result: cash received against an order shipped. Outcome. Production, with stores, buying and inspection inside it, produces a result: a finished unit in stock from raw material. Outcome. Design produces a result: a first unit of a new or changed product, shipped against an order. Outcome. Treasury produces a result: capital on the balance sheet, on terms. Outcome.

Everything else is a skill. Buying is a skill the plant vectors use; it becomes a buying agent in each of them, raising purchase orders against the plan, with the aligner’s signature as the gate above a threshold. Inspection is a skill; it becomes the inspector agent in the line, and the human gate is batch release. Bookkeeping, invoicing, the credit check, tax on an invoice: skills, agents inside order-to-cash. Contracts: a skill, an agent inside the winning vectors, with the signature as its gate. Payroll, selection and the law of employment: skills; the selection vector of chapter 12 lives inside every vector, and payroll is an agent that runs the formula of chapter 11. Systems, software, the machines’ own software: skills, held by the agent maker and its tools in every vector. Statutory accounts, audit, filings, litigation: skills, agents inside the two Outcomes whose results they serve, cash and capital. The whole of HR, IT, quality, procurement, legal, and finance but for treasury, is agents. Not one of them keeps a department, a head, or a seat beside anyone.

The sort gives five Outcomes.

| Outcome | Its result | Vectors | Humans |
|---|---|---|---|
| Enquiry-to-order | An order won on the firm’s terms | 8, by market: utilities, process industry, OEMs, export, spares and service, three regions | 16 |
| Order-to-cash | Cash received against an order shipped | 14, by order type and value | 28 |
| Raw-material-to-finished-stock | A finished unit in stock against the plan | 16, by cell and product family | 32 |
| Design-to-first-unit | A first unit of a new or changed design, shipped | 6, by product family | 12 |
| Capital-to-balance-sheet | Capital raised on terms | 1 | 2 |
| **Five Outcomes** | | **45** | **90** |

![Picture 11: The collapse itself](https://vinaypasricha.com/assets/images/collapse-or-collapse/pictures/11-the-collapse-itself-1600.webp)

*Picture 11. The collapse itself: ten departments above, five Outcomes below with their vector counts; a thousand people to a hundred, output four times.*

Ninety humans in cockpits, an aligner and a tuner to a vector. Five masters. The Keeper. The two beside him. Ninety-eight. The book calls it a hundred.

Month zero does four other things. The five masters are chosen by the four layers of chapter 12, in four days each. Six of the ten department heads apply and one is chosen; the other four are found among the directors of the corporate office, selected like anyone else, and the two who will stand beside the Keeper in month five are found there too. Every one of the ninety seats is designed on the ten-line worksheet of chapter 9, and its agents are built: a quartermaster, the line agents, an agent maker, for each vector. The agents of the revenue chain run beside the departments for the last two weeks of the month, on live enquiries and live orders, with nobody acting on them, so that on the Monday the firm switches a department off rather than an agent on. And on the first day of month zero, every one of the thousand is told, in one room, what will collapse in which month, what the ninety seats pay, that the seats are open to anyone through the four layers, and that most of the room will leave. The firm pays the leavers above what the law requires and says so in the same breath; what that costs, and the law that governs it, is the subject of section 10. What this section’s subject is, is that nobody learns the order of the collapse from a rumour.

## 4. The first Monday

On the first Monday of month one, sales, the order desk, contracts, dispatch, invoicing, collections, and the forty in the corporate office who served them, stop. A hundred and ninety-six posts. Twenty-two vectors start, eight winning and fourteen fulfilling. Forty-four humans, an aligner and a tuner in each cockpit, most of them from the building, selected on intention, some of them a month ago clerks.

The enquiry that arrives at nine is answered at nine by the vector’s conversation agent, as in chapter 9; the aligner reads the transcript before lunch. The quote that took nine days, because it passed through sales, design and finance for three signatures, takes four hours, because the pricing agent prices it against the plant vectors’ live capacity and the aligner signs. The order that was keyed three times, into sales, the order desk and the plant’s planning sheet, is keyed once by an agent and read by three vectors. The invoice goes out the day the unit ships, because the invoicing agent sits in the fulfilment vector and not in a department that batches on Fridays. The collections agent calls on the first day past due, not the fortieth.

By the last Friday of month one, the win rate on quotes, eighteen per cent for three years, is twenty-three. Order-to-invoice has gone from eleven days to one. The plant has not changed and ships two hundred and five units, because the order book, not the plant, had been the thing standing still. The owner has his number. He has spent nothing, and nobody in the building can argue with a quote that went out the same morning it came in.

Two of the twenty-two vectors fail their Monday test in week three: an export vector whose contract agent was built on the wrong trade terms, and a fulfilment vector whose dispatch agent could not read one transporter’s documents. Neither is given a person. Both are rebuilt by their agent makers inside a week, by chapter 11’s rule for a vector that fails: mirror, rest, rebuild, and no improvement plan.

## 5. Month three: the plant’s workaround

Month two is proof. Nothing is collapsed; the twenty-two vectors run, receivables fall from ninety-four days to seventy-one, and the master of order-to-cash shows the Keeper, on a Monday, a week’s result with no department anywhere in the chain. That is the test the plant must wait for, and it is the only reason the plant waits.

The plant is then collapsed in two steps, and the order of the two is the most important sequencing decision in the year. The hands at the machines go second. The people between the machines go first.

Of the four hundred and twenty in production, three hundred stand at a machine or a bench. The other hundred and twenty plan, schedule, supervise, chase and record. Around them stand thirty-five buyers, sixty inspectors, seventy storekeepers, and thirty from the corporate office. Three hundred and fifteen people whose work is the workaround chapter 2 described: keeping the parts of a department informed about one another. On the first Monday of month three, all three hundred and fifteen posts stop, and sixteen plant vectors start. Thirty-two humans, each cockpit holding a cell or a product family from casting in to finished unit out, its buying agent buying against the plan, its line agents scheduling the machines and reading the gauges, its aligner holding two gates: first article, and batch release.

The three hundred hands stay, for now, inside the plant vectors, and the chapter says so plainly, because the doctrine has one human per vector and this is the one place in the year that begins with more. They are a cost the vector carries and the calendar shows. They are not in the chain of decision. The aligner is.

What changes in month three is not the machines. It is that the machines are told what to do by an agent that knows the order book, the stock and the maintenance record at once, instead of by a planning sheet three days old. Lead time falls from fourteen weeks to nine in the first month, without a shift being added, because the queue in front of each machine, which was the sum of ten departments’ uncertainty about one another, is gone. Units shipped in month three: two hundred and thirty.

## 6. Months four to nine: design, the command centre, and the hands

Month four collapses design. Sixty-five engineers, and the ten in the corporate office who tracked their projects, become six vectors, twelve humans, each holding a product family from brief to first unit shipped. The agents draw, analyse and cost; the aligner holds the release. The firm’s third new product of the year ships in month seven, which is two more than it shipped in the three years before.

Month five collapses the last three departments that are only skills: HR, IT, and what remains of the corporate office. Seventy-seven posts, into agents that were already doing the work inside the vectors. The same month the Keeper takes his two humans, a Chief of Staff and a Head of Outcomes, and the command centre of chapter 8 is open. He has five masters reporting to him and nobody else, and the five masters have about forty-five vectors between them and no staff. There is no quality head beside the master of the plant; there is an inspector agent in each of her sixteen cockpits. There is no accountant beside the master of order-to-cash; there is an invoicing agent and a collections agent in each of her fourteen. The reader who has been waiting for a staff tier to appear at the master’s elbow should stop waiting. It does not come, because it was never a result.

Months six to nine are the hands. The plant vectors’ agent makers, not a capital committee, ask for machines by name: robotic loading and pallet pools on the forty principal machines, in-line gauging that stops the cut when the number is wrong, two assembly cells with collaborative arms for the small and mid-range pumps, and an automated test rig. Fifty-two crore between month three and month twelve, paid from the margin on the extra units, and from the term loans of month ten. None of it is new. Machines that load themselves and measure as they cut have been sold since the nineteen-eighties, and plants where machining and assembly run unattended through nights and weekends have been in the trade press for twenty years.[^48] Pravaha had not bought them because its machines stood idle eighty-four per cent of the week and no department could see a reason to make an idle machine work at night.

The hands fall as the cells commission: two hundred and twenty in month six, a hundred and sixty in month eight, a hundred in month nine. The second shift starts in month six and the third in month eight, and nobody is on them. Units shipped in month nine: five hundred and twenty. Lead time: three weeks.

Month six is also the month of the year’s one decision about price, and it is the decision that makes the rest of the arithmetic work. A plant that will soon make four times the pumps needs four times the orders, and the orders do not exist at the old price: Pravaha’s market bought two hundred of its pumps a month at twenty lakh before the collapse; the new revenue chain has taken that to under three hundred, and the plant will soon make eight hundred. So when the first cells commission and the cost of a unit begins to fall, the pricing agent proposes, and the master of enquiry-to-order signs, a list price a fifth lower: sixteen lakh for the pump that sold at twenty, for every order quoted from month six. A price a fifth below its old one, a lead time on its way from fourteen weeks to three, and a quote that comes back the same morning open doors the old Pravaha could never open: the OEMs that buy pumps as components, the export buyers who had priced it out, the process plants that had kept a second supplier for speed. The fourfold output is still the plant’s own idle hours switched on. The price cut is what lets the market take them.

## 7. Months ten to twelve: capital, and the owner’s own desk

Month ten collapses treasury into the fifth Outcome. One vector, two humans, whose result is capital on the balance sheet on terms: the bank lines, the term loans for the cells, and, in month twelve, the firm’s first outside equity, raised on the year’s numbers. Ten posts stop.

Month eleven is the owner’s. For ten months he has kept nineteen people outside the sort: statutory accounts, tax and audit, and the four in secretarial and litigation. He kept them for the reasons every owner keeps them. They are his. They report to him. The auditor knows them by name. A firm is a legal thing before it is a producing thing. And he is the board and the Keeper at once, so nobody above him applied the test to his own desk, which is exactly the failure chapter 10 named. In month eleven his own Monday test does it for him. The Head of Outcomes shows him the week’s result for the two Outcomes his nineteen people serve, cash and capital, and asks which part of either result the nineteen produce. They produce none. Statutory accounts are an agent inside order-to-cash. Tax is an agent in two vectors. Filings and the secretarial record are an agent in the capital vector. Litigation is an agent in whichever vector holds the contract. The nineteen posts stop in month eleven, and the owner’s desk is the last department in the building to go.

Month twelve commissions the second assembly cell, and the last hundred hands leave. A firm that cannot buy the second cell inside the year keeps them a quarter longer, and its count at month twelve is a hundred and ninety-eight, not ninety-eight. The order of the collapse does not change. Only the calendar does.

## 8. The year in numbers

| Month | Collapsed | People | Units/mo | What the number says |
|---|---|---|---|---|
| 0 | Nothing. The sort; five masters; ninety seats designed; agents built and run beside the departments | 1,000 | 200 | Everyone told on day one |
| 1 | Revenue chain: sales, order desk, contracts, dispatch, invoicing, collections, 40 corporate | 848 | 205 | 22 vectors; win rate 18% → 23%; order-to-invoice 11 days → 1 |
| 2 | Nothing | 848 | 215 | Two vectors rebuilt; receivables 94 → 71 days |
| 3 | Plant workaround: planning, supervision, buying, inspection, stores, 30 corporate | 565 | 230 | 16 vectors; lead time 14 → 9 weeks; 300 hands remain |
| 4 | Design | 502 | 260 | 6 vectors |
| 5 | HR, IT, rest of corporate office | 425 | 290 | Command centre open: Keeper, two humans, five masters, no staff |
| 6 | Nothing; first cells commission | 345 | 320 | Second shift, unattended; list price cut by a fifth; hands 220 |
| 7 | Nothing; cells run in | 315 | 370 | Hands 190; OEM and export enquiries arrive |
| 8 | Nothing; more cells | 285 | 430 | Third shift; hands 160 |
| 9 | Nothing | 225 | 520 | Hands 100; lead time 3 weeks |
| 10 | Treasury → capital Outcome | 217 | 600 | 1 vector; term loans for the cells |
| 11 | The owner’s statutory finance and legal | 198 | 700 | The last department |
| 12 | Nothing; second assembly cell; last hands leave | 98 | 800 | Run rate Rs 1,536 crore at Rs 16 lakh a pump; machines cutting 64% of the week |

A thousand people to ninety-eight. Two hundred units a month to eight hundred. Four hundred and eighty crore to a run rate of fifteen hundred and thirty-six crore: four times the pumps at four-fifths of the price, a little over three times the money. The payroll at target pay fell by more than a third, from sixty crore a year to a run rate of about thirty-eight, while pay per head went from six lakh to nearly forty: ninety seats at a base of eighteen lakh and a share equal to it at target, five masters, the Keeper and the two beside him. Shares earned above target come out of the result they are earned on, not out of this sum, which is Tenet V, intention very highly paid, by the formula of chapter 11 and not by a budget. Receivables thirty-eight days. Lead time three weeks. Win rate thirty-four per cent. And a corporate office of a hundred replaced by the Keeper, the two who sit beside him, and the five who run the Outcomes.

## 9. A firm that is not a pump maker

The order does not change with the trade. Sort every department by the one test. Collapse the revenue chain first and whole, on one Monday, after its agents have run beside it for two weeks. Collapse the Outcome that costs most next, its workaround before its hands. Collapse the departments that were only skills when their agents are already doing the work. Collapse treasury into capital. Collapse the owner’s own desk, which the test said should have gone in month one and which goes last because he is the one who holds the test.

A services firm has no hands and its year is shorter; the whole of it is workaround, and the machines it needs it already has. A firm with three plants collapses three plants in the same month, by the same vectors, because a plant is a vector family and not a department. A firm whose owner has hired a Keeper and sits above him on a board applies the test to the board in month zero, and chapter 8’s seat above is empty by month one, which is the one way this chapter’s last problem does not arise.

## 10. The nine hundred

Chapter 2 promised that the doctrine would say what it owes the people a collapse releases, and this is the place to say it, because this is the chapter in which nine hundred of a thousand leave. I have put it after the numbers and before the rules on purpose. A reader who has followed the year month by month knows exactly who the leavers are: the sales department of ninety, most of whom were not chosen for the seats; the three hundred hands whose cells now load themselves; the hundred and twenty who planned, supervised, chased and recorded; the corporate office of a hundred; the owner’s nineteen. They are not an abstraction by now. They have desks the reader can picture.

The first thing the doctrine owes them is the truth, early, in one room, which is why the sort is announced on the first Monday and not discovered from a rumour. The second is the seats. Every one of the ninety is open to anyone in the building through the four layers of chapter 12, and nothing in a person’s grade or department counts for or against her. At Pravaha, as the table shows, most of the pairs who held the revenue chain in month one had been in the sales department the month before, and the plant aligners had run cells. The people who hold intention were already inside the company. The collapse does not import them; it finds them, pays them far more than they were ever paid, and asks the rest to leave.

The third thing it owes is money, and here the doctrine has a rule of its own that goes beyond the law. The law first, as it stands as this book goes to print, because an owner who collapses a plant in India will meet it on the first Monday and should not meet it unprepared. Since November 2025 the Industrial Relations Code governs retrenchment. For a worker with a year’s continuous service, it requires a month’s notice or pay in its place, compensation of fifteen days’ average pay for every completed year of service and any part over six months, and fifteen days’ wages credited by the employer to the worker’s re-skilling account within forty-five days. For an industrial establishment that has averaged three hundred or more workers over the past twelve months, which Pravaha has, retrenchment needs the prior permission of the appropriate government and three months’ notice or pay in its place, and lay-off and closure have procedures of their own; permission is a decision the government may refuse. The Code’s word is *worker*, and it reaches further than the shop floor: clerical, technical and supervisory staff can be workers, and only those employed mainly in a managerial or administrative capacity, or supervising above a wage line the Code sets, fall outside it. So the hundreds on the floor are the hardest part of the year, but much of the office is covered too. This is a sketch and not advice. The thresholds and the procedure vary with the state, with the kind of establishment, and with the year, and the owner takes counsel before the first Monday, not after. But the shape is this: the hands need permission, notice and a paid season; the office needs its contracts honoured; and a collapse that forgets the difference stops in month three with a notice from the labour commissioner, which is not a stall the cockpit can show.

The doctrine’s own rule is simpler than the law and larger. The payroll saving, at the rate it runs by month twelve, belongs to the people who leave before it belongs to the owner. At Pravaha the payroll runs at sixty crore a year before the collapse and about thirty-eight at target pay by month twelve, and the owner sets aside a year of that difference, twenty-two crore, as a fixed transition allocation, paid out, over and above what the Code requires, to the nine hundred, in proportion to their years, and the owner’s gain from the payroll saving begins in year two. There are three reasons for the rule, and none of them is kindness. The first is that the money was theirs: the department was a workaround, but the people inside it did the work, for years, in a structure that was not their choice, and the year the machine holds the skills is the year their work is finally measured and found to have been real. The second is that a collapse paid for by its leavers poisons the ninety who stay. The pairs will hold the vectors for a decade; they will not hold them well in a building whose former occupants were sent out with the legal minimum by an owner who booked the saving the same quarter. The third is that the rule makes the owner honest about the arithmetic. A collapse that only pays if the leavers are paid badly is not a collapse; it is a layoff with new vocabulary, and the doctrine has no interest in it. Four times the output from a tenth of the people, even at four-fifths of the price, pays for the year many times over. If it does not, the sort was wrong, and the owner should read section 3 again before he reads this one.

What the doctrine does not owe them is the pretence that the jobs will come back, or that an equal number of new ones will appear in the same town. Chapter 2 said that the manual frontier that absorbed four men in 1989 is smaller now and shrinks every year, and nothing in this chapter changes that. Some of the nine hundred will hold vectors in other companies, because the four layers select for intention and intention is not rare only at Pravaha. Some will run the small businesses that a town of pump makers has always needed and that a severance can start. Some will not find anything as good, and the book does not pretend otherwise. That is the honest account, and the owner should give it in the room on the first Monday, with the number, because the alternative is that the nine hundred learn it from the rumour the doctrine was built to replace.

The last thing to say is the thing a recruitment company’s founder is obliged to say. GoodSpace sells hiring, and this book argues that a company needs a tenth of the hires it makes. Those two sentences are both true and I hold them together without discomfort, because the hires a company of vectors makes are the ones that matter: the ninety, chosen for intention at a price that reflects it, selected through four layers that cost more and refuse more than any interview ever did. The hiring that disappears is the hiring that was never selection at all, the filling of seats in a workaround, and I do not mourn it. If this book costs GoodSpace nine mandates in ten, the tenth will be the one worth having, and the firm has organised itself on exactly that bet.

## 11. Rules for the order of collapse

1. Sort every department on paper before touching one: skill → agent, result → Outcome. No third category.
2. Name the Outcomes by their result, from the customer’s first word to cash. Five, give or take two.
3. Revenue first. The chain from enquiry to cash collapses on one Monday, whole.
4. Never collapse half a chain. A department left standing in the middle of an Outcome kills it.
5. After the revenue chain, one department at a time, each proven on its Monday test before the next.
6. The plant’s workaround before the plant’s hands. Planners, chasers, checkers, buyers go first; machines replace hands second, bought from the margin on the extra units and from term loans.
7. Choose the masters in month zero, by the test of chapter 12. No department head keeps a seat by right.
8. Every seat is filled by selection for intention. Nobody is reassigned.
9. Tell everyone on day one what collapses when, and what the seats pay.
10. Pay the leavers above the law and do not argue about it.
11. Keep no function at the top. Finance, legal, HR, IT, quality: agent or Outcome, like everything else.
12. The master has no staff. A head beside her is a department that survived.
13. Beside the Keeper, two humans. No committee runs a collapse.
14. Build the agents before the Monday, run them beside the department for two weeks, and switch the department off, not the agents on.
15. Measure each vector’s result from its first day. The department’s old numbers are not the baseline; the Outcome’s result is.
16. Buy no machine a vector has not asked for by name through its agent maker.
17. A vector that fails its Monday test is rebuilt by its agent maker, never staffed up.
18. The owner’s own desk goes through the same test, and if he will not apply it, his Head of Outcomes does.
19. Twelve months, or a quarter more if the second cell is late. Longer is a reorganisation, not a collapse.

## 12. The collapse worksheet

One page, signed by the Keeper before month zero begins.

1. The firm: people today, output today, turnover today.
2. The departments, each with its headcount.
3. Each department sorted: agent, or Outcome with its result in one line.
4. The Outcomes, three to seven, named by result.
5. The revenue chain: which Outcomes, and the date of the Monday.
6. The order of the rest, with the proof each must show before the next.
7. The masters: names, each selected by the four layers.
8. Vectors per Outcome, each with its ten-line worksheet from chapter 9.
9. The seats: aligner and tuner per vector, plus masters, Keeper, two; the total at month twelve.
10. Output and price at month twelve, and the pay formula for every seat.
11. Machines to buy, by vector, by month, paid from margin and term loans.
12. The owner’s own functions: sorted, with their month.
13. The Keeper’s signature, and the date of the first Monday.

[^45]: Pravaha Pumps is invented. Its figures are a composite, built from firms the author has seen and from the GoodSpace numbers of chapters 6 to 9, scaled. No real firm is described, and GoodSpace’s own collapse is not finished, which is why it is not the example here.

[^46]: Gary Hamel and Michele Zanini, “Excess Management Is Costing the U.S. $3 Trillion Per Year,” *Harvard Business Review*, 5 September 2016: 23.8 million managers, first-line supervisors and administrators, one for every 4.7 employees. Cited in chapter 2.

[^47]: The firm’s own arithmetic. One shift of eight hours, six days: 48 of the 168 hours in a week. Inside the shift the machines cut metal 55 per cent of the time. 48 × 0.55 ÷ 168 = 16 per cent. For the benchmark the plant is measured against, Seiichi Nakajima, *Introduction to TPM* (Productivity Press, 1988), sets world-class overall equipment effectiveness at 85 per cent.

[^48]: FANUC’s plant at Oshino, below Mount Fuji, has run with robots assembling robots, and machining unattended through nights and weekends, and has been described in the trade and general press for two decades.