Haute Lumière

Commerce · VI.03 · MMXXVI · daylight

La Bourse  /  Volume VI  /  Nº VI.03  /  Workbook — the Gainshare employee

A small group walking and talking together through a warm office, lamps lit overhead.
Plate VI.03 · Workbook — the Gainshare employeeThe Room Where It Was Already Decided.A nested structure is not a room without authority. It is a room that holds the authority for the thing it is about, and hands nothing upward that it can finish itself.

WORKBOOK — THE LUMINOUS GAINSHARE EMPLOYEE

Chapter VI.03 · Holarchy Without Hierarchy

What nesting looks like from inside, what to measure, and what to ask for.

WHAT THIS CHAPTER MEANS FOR YOUR LEDGER

A gainshare pays you a share of a verified improvement against a signed baseline. Everything about whether that works for you comes down to one question, and this chapter gives you the vocabulary for it:

Is the gain being measured at a level you can actually influence?

If your unit holds its own decisions, the gain it produces is legible and attributable, and your claim on it is strong. If your unit is a pass-through — holding the work but not the decisions — then the gain shows up somewhere above you, gets attributed to whoever signed, and your ledger entry is a rounding.

The chapter's arithmetic is your arithmetic. The containment rate c — the share of decisions your unit closes without escalating — is simultaneously the measure of your autonomy and the measure of how much of the gain can be traced to you. They are the same number. That is the most useful sentence in this workbook and it is worth reading twice.

PART ONE — DISCOVERY

Days 1–30: find where your gain actually comes from

Exercise 1.1 — Your unit's three lists (90 minutes)

Write what your unit decides alone, what it consults on, and what it may not decide. Then take it to two colleagues separately and ask them to mark what they disagree with.

The marks are your real structure. Where the three of you disagree is exactly where a gain will later be disputed, because a gain produced in a domain nobody agreed you held is a gain somebody else will claim.

Exercise 1.2 — The escalation log (four weeks, 10 minutes a day)

Log every decision that needed coordination. One letter each:

Compute c = C / (C + X + P). Then compute P / (X + P).

That second number is the part of your autonomy that is available this quarter without anybody signing anything, and it is usually large. Every P you convert to a C is both a faster decision and a gain you can trace.

State your denominator when you report it: how many decisions you logged, and how many you know you missed. A log that hides what it missed will be attacked on exactly that point at exactly the wrong moment.

Exercise 1.3 — Find the uncounted gain (one day)

Three places it usually sits, and it is usually already there:

  1. Decisions your unit takes quickly that used to take weeks. The saving is the escalation cost avoided: at £270 a level and 0.6266 levels saved per decision under a containment rule, it is real money and nobody is counting it.
  2. Work your unit absorbed rather than escalating. Cover, cross-training, somebody staying late to close something that would otherwise have gone up.
  3. Quality that never became a defect. The hardest to claim and often the largest.

Write each with a figure, however rough. A rough figure in a log beats an exact figure nobody wrote down.

PART TWO — THE ARITHMETIC

Days 31–45: compute what you are owed

Exercise 2.1 — Price one escalation (one hour)

Count the last escalation you were part of. How many people, how many hours, how many elapsed working days. The chapter prices a level at four people for ninety minutes — £270 — and five working days of latency. Yours may be more.

Now multiply by your unit's escalations in a quarter. That is a line item that does not currently appear anywhere, and it is the one your gainshare conversation should open on.

Exercise 2.2 — Your unit's share of the structural saving (2 hours)

The chapter's model, at 10,000 people:

                              HIERARCHY        NESTED
  coordination posts              1,043            50
  posts, cost                   £73.01m        £3.50m
  inside-team coordination       £4.95m       £29.72m
  escalation                    £52.04m       £11.43m
  ------------------------------------------------------
  total                        £130.00m       £44.66m

£85.35m, or £8,535 per front-line person per year.

That last figure is the one to hold, because it is per person and you are a person. It is not what you are owed — much of it is carried by people whose jobs changed and by a coaching cadre that has to be paid for. But it is the size of the pool your share is a share of, and a gainshare negotiated without knowing the size of the pool is a gainshare negotiated blind.

Exercise 2.3 — The mesh cost you are personally carrying (45 minutes)

Here is the cost of self-management that lands on you rather than on the P&L. A team of twelve has 66 possible ties against 11 for a hub — 6.0×. You are keeping some of those alive, in your own hours, and nobody scheduled it.

Count how many people you must stay in genuine contact with to do your job. Multiply by the time it takes. That is an unpaid coordination load and it is legitimate to name it. The per-team crossover is 12.47 people with a band of 10.46 to 14.16; past 18.82 across the organisation, the mesh costs more than the managed version.

If your team is above eighteen, you are not in a flat structure. You are in an expensive one, and you are paying for it in calendar. Ask for the split.

Exercise 2.4 — Break-even, so you can argue both ways (one hour)

Compute c* for your organisation's span and depth. At the chapter's parameters it is 37.8%; for a 3,000-person firm in teams of 10 at span 6 it is 28.2%.

Know it because it cuts both ways. If your measured c clears c*, the nested structure is the cheaper one and your unit is producing the difference. If it does not, the honest answer is that the work is not sufficiently decomposable and a different structure is right — and a person who can say that is trusted with the numbers that matter.

PART THREE — DESIGN

Days 46–70: make the uncounted countable

Exercise 3.1 — Write your peer contract (90 minutes, with each counterparty)

The Morning Star CLOU is the model: a personal commercial mission, the activities you will perform, the metrics you will be judged on, negotiated directly with the colleagues your work touches. The full version runs to about 30 activity areas and some 50 metrics across roughly 10 counterparties, refreshed annually. Start with four areas and six metrics per counterparty.

Four lines each way:

  1. My mission in this relationship.
  2. What I deliver to you.
  3. How you will know whether I did.
  4. What I need from you in order to.

The metrics in line 3 are your gainshare evidence. This is the whole reason to write the contract: it converts a working relationship into a record that a third party can read, and a third party is exactly who decides a gainshare claim.

Exercise 3.2 — Get the baseline signed before anything changes (one week)

The single most important document in any gainshare, and it is the same rule as Chapter I.01: an unagreed baseline is not a baseline, it is a future dispute.

Five things: the metric defined precisely enough that two people compute it identically; the period, long enough to include normal variation; the method, written; the verifier, named; two signatures.

Exercise 3.3 — Get the exception clause written (one meeting)

This protects you specifically. In an emergency, containment goes to zero and L(0, 6) = 6 — 30 working days by consent against 5 under unity of command. Somebody will take command, and they should.

What you need in writing is what happens afterwards: the time box, the automatic reversion date, and — this is the part that concerns your ledger — that the baseline is suspended for the duration and resumes on reversion. Without that clause, an emergency you did not cause becomes a quarter you did not earn.

Exercise 3.4 — Ask for your unit's notional P&L (one meeting)

This is the Haier microenterprise contract, and it is the thing to ask for. Transfer-pricing basis, a notional profit and loss for your unit, and a share of the gain above the signed baseline. Haier runs roughly 4,000 microenterprises this way, about 200 of them customer-facing.

The ask is not "make us a business." It is: give us a line we can see, priced on a basis we agreed, so that what we improve is attributable to us.

PART FOUR — DESTINY AND DELIGHT

Days 71–90: make it hold

Exercise 4.1 — Watch the three drifts (20 minutes, monthly)

The coach re-acquires authority. It starts helpfully and ends with you consulting before deciding, and your c falls with it. The counter is written: the coach holds no budget and signs nothing.

The team grows past the cap. Your unpaid coordination load rises as the square while your gainshare stays the same size. The counter is a split rule written in advance.

The exception is never rescinded. Command was taken and not handed back, the baseline never resumed, and a quarter's gain quietly evaporated. The counter is reversion by date, requiring a positive act to extend rather than to end.

Exercise 4.2 — The quarterly claim, on one page (one hour)

Containment c this quarter — __._%, from ___ decisions logged, ___ known missed. Break-even c* — __._%. Escalations avoided — ___ , at £270 a level. Gain against signed baseline — £____. Share claimed under the scheme — £____. One line on what changed, and one line on what it cost.

Exercise 4.3 — Delight, on purpose (ongoing)

One line a week: a decision you took that you would previously have sent upward, and the hour you would otherwise have spent preparing the case for it.

Add them up at the quarter. That total is your real raise, and it arrived before any money did. People who work this way describe the same thing: not freedom in the abstract, but the specific recovery of the part of themselves that used to be composing an argument in the shower.

WHAT NESTING ASKS OF YOU, HONESTLY

The obligations that come with the autonomy

A workbook that only described what to claim would be doing you a disservice, because the structure has a cost side and you are on it. Four obligations, and each is the price of a benefit named above.

You carry the mesh. Sixty-six possible ties in a team of twelve against eleven for a hub. Nobody schedules that time and nobody backfills it. The honest ask is not to be excused it but to have it counted — and to have the team cap enforced so it stops growing as the square.

You have to say when you cannot contain something. The containment rule only works if escalation is honest. A unit that escalates nothing in order to protect its c is producing a number instead of a result, and that is the failure mode the whole instrument is vulnerable to. Log the X's. Your credibility on the claim rests on the fact that you logged them.

You have to write things down that used to be understood. The peer contract, the domain statement, the baseline. This feels like bureaucracy and it is the opposite: it is what stops the structure being governed by whoever has the longest-standing relationships. A team of twelve sustains about 45.5% of its possible ties; everything else routes informally, and writing it down is how a newcomer gets the same access as somebody who has been there four years.

You have to hand over command when it is genuinely an emergency, and mean it. The clause protects you afterwards. It does not entitle you to argue during.

WHAT TO DO IF YOUR SCHEME MEASURES AT THE WRONG LEVEL

This is the most common problem and it has a good answer.

If your gain is measured at a level three above you, your influence on it is whatever survives four levels of aggregation — and under Williamson's control loss at 0.90 a level, a signal crossing five levels retains 59.0% of itself. You are being paid on a number that has been through a filter.

Do not ask for the scheme to be rewritten. Ask for one additional line: your unit's own containment rate and its own notional contribution, reported beside the group figure, for two quarters, with no money attached.

Two quarters of a clean line with no money attached is the cheapest possible ask and it is almost always granted. It is also how every gainshare that measures at the right level got there: somebody produced the line first, and the scheme followed the evidence rather than the other way round.

KNOW YOUR SCHEME — A CHECKLIST

Ten things to be able to answer without looking. If you cannot answer six, that is this quarter's work.

  1. At what level is my gain measured — my unit, my site, the group?
  2. Is that level one whose decisions I can influence? What is its c?
  3. What is the baseline, who signed it, and when was it last reset?
  4. Who verifies, by what method, and are they independent of the outcome?
  5. What is the share, and is it a percentage of gain or a discretionary pool? These are very different instruments.
  6. What happens to the baseline in a declared emergency?
  7. What happens if my team splits mid-period? Does the baseline split with it?
  8. Can the scheme be changed unilaterally, and with what notice?
  9. What is the transfer-pricing basis my unit buys and sells on, and who set it?
  10. If I left tomorrow, what would be owed to me for a period already worked?

THE CONVERSATION, SCRIPTED

For the meeting where you ask for the notional P&L and the signed baseline.

Open on what is working. "This unit closed ___ decisions itself last quarter — a containment rate of ___%. That is above the break-even of ___% for our span and depth, which means this is the cheaper structure to run and this unit is producing the difference."

Name the uncounted line. "The escalations we avoided are worth about £___ at four people and ninety minutes each. That does not appear anywhere in our reporting."

Make the ask concrete. "Three things: a transfer-pricing basis we agree, a notional P&L we can see, and a baseline signed before we change anything."

Name the cost honestly, first. "The mesh costs us something real — sixty-six ties in a team of twelve against eleven — and it is carried in our calendars. We want the split rule written at twelve so it does not get worse."

Name what protects both sides. "And an exception clause with a reversion date, with the baseline suspended for the duration. That protects the business in an emergency and protects us afterwards."

Close on the comparison, not the sentiment. "c against c, one line, in the quarterly pack."*

APPRECIATIVE QUESTIONS FOR YOUR TEAM

  1. What did we decide entirely ourselves last month that would once have taken three weeks? What made that possible?
  2. Which of the things we improved this quarter can we actually trace to a decision we took — and which one can we not, and why not?
  3. If we had our own notional P&L on one page, what is the first line we would want to see on it?
  4. Who outside this team do we depend on most, and is there anything in writing between us?
  5. When we last handled an emergency well, what did the structure do afterwards that made it safe to do again?
  6. What is the one thing we are carrying in our calendars that nobody has counted — and what would it take to count it?