Haute Lumière

Commerce · II.11 · MMXXVI · daylight

La Bourse  /  Volume II  /  Nº II.11  /  Workbook — the Gainshare employee

WORKBOOK — THE LUMINOUS GAINSHARE EMPLOYEE

Chapter II.11 · The Case Against This Book

For the person working inside a gainshare arrangement — where a defined share of verified improvement returns to the people who created it. This chapter is the one that protects your share, because everything that can be softened in a paradigm can be softened in a formula, and by the same hand for the same reasons.

---

WHY THIS CHAPTER IS YOURS

The chapter is about kill conditions: the finding, written in advance, that would retire a claim. It reads like epistemology. It is a clause about your money.

A gainshare has four parts — a baseline, a measure, a share, a period and verifier. Every one of them is a threshold written down in advance, and every one of them is under exactly the same pressure the chapter identifies in kill conditions: the pressure runs one way, towards softening, and it never feels like dishonesty from the inside. It feels like refining a definition.

The chapter's second rule is therefore the most commercially valuable sentence in this edition for anyone inside a scheme:

A threshold may be tightened, never loosened. Amendments are versioned, and the previous version stays visible.

Apply that to a claim register and you get an honest textbook. Apply it to a gainshare and you get paid what you were promised in year four as well as year one. Same rule, same mechanism, and the second use is the one with a number attached to it.

And there is a second thing this chapter hands you. It is the one place in the edition that tells you, explicitly, how to argue with the edition — which means it is the one place that tells you how to argue with your own scheme's assumptions without being the person who is "against the programme." You are not against anything. You are asking what would have to be true, which is the question the register is built to make ordinary.

---

PART ONE — DISCOVERY

Days 1–30: find where the thresholds already hold

Exercise 1.1 — The version history of your own scheme (2 hours)

Find every version of your gainshare document that has ever existed. Ask HR, ask your representative, ask anyone who was here at the start.

For each change, write three columns: what changed · which direction · who signed.

Then count. Of all the threshold changes in the scheme's history, how many tightened the requirement on the workforce and how many loosened it? Write the ratio down.

This is an appreciative exercise before it is anything else, and run it that way. You are looking for the changes that went the other way — the periods where a baseline was held when it could have been ratcheted, where a measure was widened to include work that had gone uncounted, where somebody protected a share percentage in a hard year. Find those and find who did them. Those are the people who will help you with everything else in this workbook, and they are usually not the people you expected.

Exercise 1.2 — The unwritten assumption in your scheme (half a day)

Your scheme rests on claims nobody has written down. Find them.

For each, write the sentence out and then ask the chapter's question: what result would make this false?

You will find at least one where the answer is "nothing could" — and that is a U-claim sitting inside a document that decides your pay. It should be an F-claim, and Exercise 3.2 is where you make it one.

Exercise 1.3 — The appreciative interview with the verifier (45 minutes)

Ask to meet whoever verifies your scheme. Bring one question:

"Tell me about a time you found something in a verification that changed the number. What made it possible to raise it?"

Take notes on the conditions, not the finding.

If they cannot think of an instance, that is your detection probability, and you have just measured it. A verifier who has never changed a number has found nothing, not confirmed everything — the chapter's line about validation units applies here exactly and for the same reason.

---

PART TWO — THE ARITHMETIC

Days 31–45: compute what the softening costs

Exercise 2.1 — The ratchet, priced (90 minutes)

This is the arithmetic that most gainshare members never do, and it is worth doing once properly.

Take your scheme's share percentage and your own realistic annual improvement contribution. Then model two futures over five years:

• Held baseline. The baseline set once, fixed for the term. Your share is computed against the original level each year, so cumulative improvement compounds into cumulative payment.
• Full ratchet. The baseline resets to last year's achieved level each year. Your share is computed only against the incremental improvement, so the same effort yields a flat or falling payment.

Write both five-year totals. The gap between them is what the baseline clause is worth, and it is nearly always larger than any percentage-point argument about the share itself.

Then note the asymmetry that makes this chapter's rule bite: a ratchet is a threshold loosening for the firm and tightening for you, and it is almost never versioned. Ask for the previous version to stay visible. That single request is worth more than a negotiation about the share.

Exercise 2.2 — Decompose your own measure (90 minutes)

The chapter's decomposition test, applied to the thing that computes your pay.

If your measure is a composite — a scorecard, an index, a weighted basket — do three things:

1. Find the weights. If they are not published, that is a finding and it belongs in writing.
2. Recompute last period with the largest component removed.
3. Recompute with its weight halved.

The chapter's worked case is an index reported at 1.71 Earths whose carbon component is 60 percent — 1.026 Earths — leaving 0.684 for everything else. One term carried the whole message.

Ask the same question about your scorecard: which component is actually paying you? If one term carries the result, then everything you and your colleagues do that lands in the other terms is unpaid work, and the fix is a measure change, not more effort.

Exercise 2.3 — Your detection probability (60 minutes)

The chapter prices independent challenge:

   P(claim wrong) x loss if wrong x P(detection)
   ---------------------------------------------  >  1
                verification budget

On the worked example — a £4,000,000 programme, a 15 percent chance the load-bearing assumption is wrong, a 60 percent loss if it is — the expected loss is £360,000. At a 60 percent detection rate, £216,000 is avoided against a £120,000 budget: 1.80×. Break-even detection is 33.3 percent.

Run the same shape on your scheme. What is the total value of your collective share in a period? What is the chance the measure is mis-specified in a way that costs you? What is the chance verification would catch it?

The number you produce is the case for a member-side verifier, and it is a commercial case rather than a grievance. That distinction is the whole reason to compute it.

---

PART THREE — DESIGN

Days 46–70: write the clauses

Exercise 3.1 — Your scheme's register (2 hours)

Build the four-column table for your gainshare, exactly as the chapter builds it for the edition.

Twelve rows. Examples to start you:

• The measure captures the improvement this team creates. F. Killed by: a period where measured improvement and an independently observed operational gain move in opposite directions.
• The baseline period was representative. F. Killed by: the baseline period sitting more than one standard deviation from the five preceding periods.
• Costs netted off the gain are caused by the improvement work. F. Killed by: any netted cost that would have been incurred anyway.
• The scheme rewards the people who create the gain. U as written — nothing could disconfirm it. Rewrite it as an F-claim with a distribution test.

Class each, and count your ratio. Most people's first pass is two-thirds U, and the U rows are precisely where schemes go wrong, because an unfalsifiable promise cannot be shown to have been broken.

Exercise 3.2 — The three clauses to ask for (90 minutes)

Write them out, in scheme language, ready to hand over.

1. The versioning clause. "Any amendment to the baseline, the measure, the share or the netting rules shall be published alongside the version it replaces, with the date and the signatory, for the life of the scheme." Costs nothing. Changes everything, because softening survives only in the dark.

1. The tightening rule. "Thresholds may be tightened by agreement. A loosening of any threshold is a material amendment requiring the same consultation as the establishment of the scheme." This is the chapter's rule two, in the only place it has a cash value.

1. The verifier's standing clause. "The verifier shall not report to, nor be appraised by, any party whose costs are affected by the verified figure, and shall record annually the number of verifications in which the figure was changed." That last count is the detection probability, published.

Ask for all three together and take whichever you get. They cost the firm nothing in cash, which is the reason they are gettable, and they are worth more than a point on the share, which is the reason to want them.

Exercise 3.3 — The second reader (this month)

Find one colleague who will read your register and tell you which row is misclassified. Not someone who agrees with you — someone who reads documents well.

The rule the chapter will not bend applies here too: you may not be the reader of your own register. Give them the credit for whatever they find, in front of other people. That is how a second reader becomes permanent rather than a favour.

---

PART FOUR — DESTINY AND DELIGHT

Days 71–90: make it hold, and enjoy it

Exercise 4.1 — The three conditions (45 minutes)

• In the standing pack. The register and the version history reviewed at a fixed point each period, on the calendar and not the conscience.
• Somebody paid to attack it. The verifier's standing clause, with the changed-figure count published.
• A second owner. One colleague who will notice if the review stops happening. Ask them explicitly.

Write your three. If you cannot name the second owner, that is your ninety-day task and it outranks everything else on this page.

Exercise 4.2 — Watch for the one sign (ongoing)

The chapter names the earliest observable sign that a register has stopped working, and it transfers exactly:

When somebody brings a number and receives a paragraph.

You raise a specific figure about the netting rules and receive an answer about partnership, long-term value or shared purpose. That is the sign, it is free to watch for, and it arrives long before the payment does. Note the date. Ask the question again in writing.

Exercise 4.3 — Delight, which is the point of the whole structure (ongoing)

Here is what this actually buys, and it is worth naming because it is not obvious from the clauses.

A scheme you can check is a scheme you can stop thinking about. The alternative to a register is not trust — it is a low, permanent, unresolvable background suspicion that you are being short-changed, carried into every period, never confirmed and never dismissed. That suspicion costs more than the money does.

Once the version history is published and the verifier's changed-figure count is in the pack, the question is settled each period in about ten minutes and then you are free of it. The pleasure is not in catching anyone. It is in no longer having to wonder, and it is available to everyone in the scheme at once, for the price of three clauses that cost the firm nothing.

Write one sentence: the thing I would stop carrying if this were published is ___.

---

KNOW YOUR SCHEME — A CHECKLIST

Tick what you can evidence. A blank is a question, not a complaint.

• I have every version of the scheme document that has existed
• I know which direction each amendment moved, and who signed it
• I know whether the baseline is held, fixed-term, or ratcheted
• I have computed the five-year gap between held and ratcheted
• I know the weights in my measure, and which component carries it
• I know what costs are netted off the gain, and who decides
• I know the verifier's reporting line
• I know how many times the verifier has changed a figure
• I have written twelve rows of my scheme's register
• I have classified each F or U, and someone else has read it
• I have asked for the versioning clause in writing
• I know what I would have to see to conclude the scheme is not working

The last one is the chapter's whole lesson, brought home. A person who cannot say what would make them conclude a scheme is failing is a person who will conclude it years after everyone else did.

---

THE CONVERSATION, SCRIPTED

Ten minutes, with whoever owns the scheme. Bring one page.

"I have been reading the scheme documents and I would like to ask for three things that do not cost anything.

The first is a published version history — every amendment to the baseline, the measure, the share or the netting rules, shown alongside the version it replaces. Not to challenge any of them. So that everyone can see the shape.

The second is that a loosening of any threshold is treated as a material amendment, with the same consultation as setting the scheme up. A tightening can go through as it does now.

The third is that the verifier records, once a year, how many verifications changed the figure. Just the count.

All three make the scheme easier to defend, not harder. And I would rather spend my attention on the improvement than on wondering about the mechanism."

Then stop talking.

The reason this works is that every one of the three is cheap, defensible and makes the scheme's owner look better to their own auditor. You are not asking for money. You are asking for the thing that makes the money legible, and that is a much easier request to grant.

---

APPRECIATIVE QUESTIONS FOR YOUR TEAM

1. When has our scheme been changed in a way that helped us — and who made that happen?
2. Which part of the measure best captures what we actually do well? What made that part get written the way it did?
3. If everyone here could see every version of the scheme document, what would change about how we talk about it?
4. What would we have to see to know the scheme is working — and could we agree on that sentence today, before the next period closes?