Deep dive · A business with soul
Every company already has a charter. Most founders have never read theirs.
It came off a template, it says the company may pursue any lawful business, and one day it will be read out in a room you are not in. That document is the company. Everything else is a preference.
The before
The mission lives in the file with no legal force. The purpose lives in the file with all of it.
Start with what is actually on record. Almost every conversation about values skips this part, and skipping it is why the conversation goes nowhere.
Ask a founder what the company is for and you will get a good answer, usually a moving one. Ask where that answer is written and you will be sent to a slide, a wall, a handbook, a page on the website. Then ask for the articles.
The articles say the company exists to conduct any lawful activity. The operating agreement says the members may distribute cash at their discretion. The bylaws say the board is elected by the holders of the shares, and that the holders of the shares may remove any of them, with or without cause. Nothing anywhere in it mentions the thing the founder just described.
That is not negligence. Those papers were filed in a week when the company was three people and a bank appointment, using a template built to make incorporation fast and cheap for anybody at all. It did that job well. It also, quietly, answered every question the founder still thinks is open.
The document is the company. When there is a disagreement about money, control, succession or a sale, nobody consults the wall. They consult the file. And whatever is in the file has been sitting there since the first month, patient, unread, waiting for the one afternoon it decides everything.
So the gap is not between a company's words and its behaviour. The gap is between two documents — one that carries the meaning and none of the force, and one that carries the force and none of the meaning. Every drift story I know is the second document quietly winning an argument the first one did not know it was having.
A company does not become the other thing by amending its charter. It becomes the other thing by never having written one.
The test
Take the founder out of the room. Take out the good year. Now read it again.
The test is one line elsewhere on this site. Here is how it is actually run. It is a reading exercise, it needs an afternoon and two people, and it is uncomfortable in a specific and useful way.
Put the governing documents on the table. Beside them put one page listing everything you believe about this company — the plain sentences, the ones you would say to a friend. Then take each belief and answer three questions in writing. Not out loud. In writing, because writing is where the vagueness shows.
-
01
Who would have to act? Name a person or a named body. Not the company, not we, not leadership. If the sentence has no actor in it, it is not describing a mechanism, it is describing a hope. Most of the page fails here and it fails fast, which is why this question goes first.
-
02
What would make them? An obligation. A condition attached to money. A quorum that fails without them. A right somebody else holds and can exercise without permission. A policy is not a mechanism — a policy can be rewritten by the same people, on the same afternoon, in the same meeting where it became inconvenient. Nor is a bonus a mechanism; a bonus is a suggestion with money on it, and money can be found elsewhere.
-
03
What happens if they don't? If the honest answer is we would be disappointed, you have written a preference. If the answer is the transaction cannot close, the distribution cannot lawfully be made, the seat is vacated, the trustee has grounds and the funds to sue — now you are holding a provision. The difference between those two lists is the whole of this page.
Almost every page comes back blank in the second column. That is not a judgement on anybody's sincerity — the founders I have run this with are the most sincere people in their industries. It is a description of what a template does. The template answered column one and column three in favour of whoever holds the shares, and left column two empty on purpose, because filling it in costs money the template was designed to save you.
The work is filling in column two. That is all a charter is.
A duty nobody has standing to enforce is a preference in a formal typeface.
The line every clause gets held against
The translation
Every value has an instrument. Most of the work is finding which one.
None of what follows is an improvement on the sentiment. It is the same sentiment, written so that it does something on a day when nobody feels it.
Read the middle column and something becomes obvious. Not one of those instruments is exotic. Every one of them is ordinary commercial drafting, the kind that is done every week for reasons nobody finds inspiring. The instruments are not the hard part. Knowing which sentence needs which instrument is the hard part — and being willing to find out what your sentence actually costs, which is the point where most of this work stops.
The vehicle
You do not need a new legal universe. You need to know what each old one is good at.
People come to this work expecting to invent something. There is nothing to invent. There are five or six well-worn vehicles, each excellent at one thing and poor at another, and the design is mostly a matter of choosing honestly.
- Benefit corporation statutes Good at putting purpose inside the fiduciary duty of directors, so a board may lawfully weigh things other than price. Weak on enforcement: in most versions only shareholders can bring the claim, which means the protection points inward and the community it was written for has no door to knock on.
- Perpetual purpose trusts and steward ownership Good at separating control from economic return, so that the people who benefit are not the people who could sell. Weak on liquidity and on plain explanation. Somebody has to be willing to run a company nobody can cash out of, and a lender will need that explained more than once.
- Cooperative statutes Good at worker and member ownership, one member one vote, and a distribution logic that already assumes surplus belongs to the people who made it. Weak on outside capital and on speed — which matters, because a venture that cannot move is a venture that dies politely.
- Tribal corporate codes Good at the thing every other vehicle has to fake: chartering by the community that lives with the consequences, because there the community already is the chartering authority. Weak on being understood by counterparties, which is a drafting and explanation problem rather than a legal one.
- A well-built operating agreement Good at nearly everything, at small scale, quickly and cheaply, with no special statute required. This is where most of this work should start and where a great deal of it should stay. Weak on outliving the people who signed it — unless the amendment clause is written so that it does.
- The golden share Good at one job only, held by one body only: stopping a defined list of changes. Weak on everything else, and that is the point. It is a brake, not a steering wheel. A golden share held by a body with no appetite to use it is a decoration with a certificate.
None of that is legal advice and none of it survives contact with your jurisdiction unaided — the statute, the filing and the local exceptions all belong to counsel, and counsel should be in the room early rather than at the end. Any page that tells you which vehicle you need before it has met your people is selling something. What this page is telling you is narrower and more useful: the choice exists, it is a real choice, and it is yours. Most people starting something are never told that much, and go on believing the shape came with the paperwork.
The rooms
A charter is never tested where it is written. It is tested in four rooms.
Draft toward the rooms and the language stops being aspirational on its own. Every clause below is written by imagining who will be sitting opposite it and what they want.
-
01
The lender's room A lender is not hostile to purpose. A lender is hostile to ambiguity, and to anything that might stop them being repaid on the date they were promised. Most of what founders assume a bank will refuse, a bank will accept once it is specific and once its edges are drawn on paper. What a lender genuinely cannot live with is a veto with no boundary, a consent right with no deadline, and a governing body that has never actually met. Draw the reach of every consent, put a clock on it, and hold one real meeting before the term sheet — and most of the objections you were bracing for never arrive.
-
02
The buyer's room A buyer will read your charter more carefully than anyone who has ever worked at the company, and they will read it looking for one thing: the seam. Every provision that would survive them is a provision they will ask to have removed as a condition of closing, and they will ask at the hour the founder is most tired and most flattered. What survives that hour is only what is structural — restrictions on transfer, a stake that cannot be diluted, a consent held by somebody who is not the one selling. What does not survive is anything the seller is able to waive alone. Read every clause once as the seller and ask: could I give this away by myself, at two in the morning, if I badly wanted the deal? If yes, it is not a protection, it is a bargaining chip you have not spent yet.
-
03
The bad year's room In a bad year, the question is never whether to honour the commitment. Everyone still means it. The question is which line the commitment sits on. Anything paid out of profit is paid after there is none. Anything that is a cost of production is paid on the way through, before anybody has an opinion about it. That one choice — cost or distribution — decides more about whether an obligation is real than any amount of language wrapped around it. The apprenticeships, the restoration, the fund: if they are generosity they are gone by March, and if they are cost of goods they are still there in December.
-
04
The court's room The last room and the one almost nobody drafts for. A duty needs a person with standing to enforce it. So ask, of every provision: if this were breached on a Tuesday, who could file on Wednesday, and would they actually do it? If the honest answer is nobody — or is a person who would have to sue their own employer, using their own money, to do it — then what you have written is a statement of intent with a clause number on it. This is the entire reason the guardian exists in this design. Not to be wise. Not to advise. To have standing, and to be funded well enough to use it.
Draft for the fourth room and the other three get easier. Draft only for the first three, and the fourth will eventually come looking for you, on a day of somebody else's choosing.
The failures
Seven ways a good charter stops working, and not one of them is dramatic.
These are drafting failures, not moral ones. Every one of them is invisible on the day of signing and obvious about eleven years later.
- The amendment clause Everything else in the document is downstream of who is allowed to change it and by what margin. Lock the mission in beautiful language and leave amendment at a simple majority of shares, and you have locked nothing at all. It is the first clause a buyer's counsel turns to and the last one a founder thinks about.
- The undefined word Community. Surplus. Materially. Reasonable efforts. Every undefined word is a door standing open in the wall you just built, and more money has left through the definition of surplus than through any other clause in any document I have read.
- The trigger with no actor A provision saying a thing shall happen, with nobody named to make it happen, happens when somebody remembers. In year two everybody remembers. In year twelve the people who remembered have retired and the provision is a sentence.
- The obligation with no funding A redemption right the company cannot pay. A first-refusal right the holder cannot exercise. A restoration duty with no line in the budget. Each of these reads as a protection and functions as a delay, and the delay ends the moment somebody calls it.
- The consent with no clock If a body must approve and the document does not say what its silence means, then the first genuinely urgent decision will be taken without it. And after that first time, every decision will be urgent.
- The rule that stops at the parent The charter binds the company. Then the real work moves into a subsidiary, a joint venture, a management contract, a licensing arrangement. Say plainly that the provisions travel down and across, or watch the company become a holding entity with very clean hands.
- The document nobody can read Forty years out, the people bound by it were not in the room and cannot ask anybody who was. If the reasons are not inside the document, only the words survive — and words with no reasons attached get read narrowly by whoever needs them narrow.
Who it answers to
A charter that only protects the people who wrote it is the old document with better adjectives.
This is the failure that is hardest to see from inside, because the document looks careful, principled and thoroughly drafted the whole way through.
There are two questions that sound almost the same and produce opposite documents. How do we protect what we have built? And: who does this thing answer to?
The first question, asked sincerely by good people, produces a fortress. Control provisions. Entrenchment. Consent rights held by the founders, drafted by the founders, exercisable by the founders. Every clause in it is defensible and the whole is a wall around a room. I have read a great many of these and they are always written by someone who genuinely believes they are protecting everybody, because from inside that room the interests really do look identical.
So run one more pass over the document, and for each provision ask a single question: who does this protect? Not who does it mention. Who does it protect on the worst day. If the honest answer, provision after provision, is the founders and the board, then what you have is a shareholders' agreement with a conscience written into the recitals.
A charter has to carry outward and downward or it is not doing the job. Outward to the households that live with what the company does and cannot move away from it. Downward to the person at the loud end of the building, to the manager who has to deliver news they did not decide, to the young person who might take the apprenticeship in nine years if there still is one.
Steadiness that stops at the executive floor is not steadiness. It is a parachute, and only some people are wearing one. A company can be entirely calm at the top and shaking everywhere else, and it will show up in the accounts as a good year. That is the failure worth naming out loud, because nobody names it — the people who would name it are the people the document does not protect.
So: four things to look for in your own document. A stated relationship between the highest pay and the lowest, which is a governance provision and not a payroll one. A redemption right a family can actually call in the month somebody dies. At least one seat at the table held by a person who does not report to the chief executive and cannot be removed by them. And a route by which someone holding no shares at all can raise a breach and have it heard by a body that is obliged to answer.
If none of the four is in there, the document protects the room it was written in. Which is what the template was for.
The recitals
Put the reason in the document, not in the memory of the room.
Ordinary practice treats recitals as throat-clearing — the whereas paragraphs everybody scrolls past. Do the opposite. Write, in plain language, inside the instrument, what each significant provision is for and what it was written to prevent. Not a foreword. Not a preamble that disclaims its own effect. Drafted so that a reader construing an ambiguity is sent to them.
It costs a few pages and it does two things nothing else in the document can do. It gives a judge the purpose on the day the words run out — and the words always run out, because the situation that arrives was not the situation anybody imagined. And it hands the person who inherits the company the one thing that never gets handed over properly: why.
Every clause with its reason written beside it is a clause somebody can defend in a meeting. Every clause without one is a clause somebody will eventually trade away, sincerely, in good faith, for something that seemed more urgent at the time.
The founder's last job is to become optional
The temperament that starts a company is usually the temperament that holds it. For a while this works perfectly and costs nobody anything. It also hides the gap, completely, for as long as it lasts.
Because a founder who is present makes the charter look unnecessary. Every question that would have tested the document gets answered informally, in a corridor, and gets answered well. The mechanism is never reached for, so nobody notices there is no mechanism. This is the reason the drafting has to happen in exactly the season when it feels like paperwork: it is never urgent until it is too late to do it calmly, and a charter drafted in a crisis is a charter drafted by whoever holds the money that week.
And the handover itself is not an event. What actually has to be handed over is not the operation — operations transfer fine, people are good at operations. What has to be handed over is the reasons, and the rehearsed use of the instruments. Pull the brake once, early, on something small. Use the veto on a minor question while everyone is calm and nothing turns on it, so that the first time it matters the room already knows what it looks like, how long it takes and that the company survives it.
A provision used once is a provision. A provision never used is a rumour, and rumours do not hold up in the buyer's room.
Three things to do this month, none of which cost anything
-
01
Read your own governing documents out loud, with one other person. Articles, bylaws or operating agreement, and the shareholder or member agreement. All of it, including the second half, which most founders have never reached. Out loud, because reading aloud is when you hear the sentences that do not say what you assumed they said — and there will be several.
-
02
Find the amendment clause and copy it onto the first page. Who can change this document, by what margin, and with whose consent. Whatever that clause says is your real mission lock, regardless of what any other clause in the document appears to promise. Most people are surprised. A few are badly surprised, and those are the ones who still have time to act.
-
03
Write the one page. What must still be true about this company when you are not in it. One page, plain words, no legal language, no committee. That page is not the charter and it is not meant to be. It is the brief for the charter, and it is the one part of this work that nobody — not your counsel, not me — can write for you.
A charter is what a company does when nobody is watching, written down in advance by the people who were watching. That is the whole of it. The rest is which statute, and which sentence needs which instrument, and both of those are answerable.
The one question
Will this ensure our future generations live in harmony and resilience with each other and Mother Earth?
That is the bottom line of The Council Fire, and it is the last thing asked before a decision is made. Every provision, every account, every council on this page exists to make that question answerable by somebody other than the person who wants the answer to be yes.
The novel
The Council Fire
When the systems running the modern world reach their limits, the keepers of many peoples’ oldest knowledge are drawn to one fire to remember what actually lasts.
The company designed from the bones out is a chapter in that book long before it is a clause in anybody’s charter.