The business runs through you. A business that lives in people's heads cannot be measured, and one that cannot be measured cannot be handed to anyone: not a manager, not a lender, not a buyer.
You are not five businesses. You are one person with capital deployed across several of them, and the questions you ask at the end of a quarter do not respect entity boundaries.
A business that lives in people's heads cannot be measured, and a business that cannot be measured cannot be handed to anyone: not a manager, not a lender, not a buyer. That argument is now well understood, and there is an industry answering it. It does not stop at the entity boundary. Everything an owner holds outside the operating companies also lives in one head, has never been written down, and cannot be handed to a partner, a child or an executor for exactly the same reasons.
Nobody sells the second half. The professional facilitation canon and the consumer family-software market do not touch each other. Facilitators charge thousands for a retreat and serve families above a couple of hundred million. Family apps charge eighty dollars a year and manage chores. Nothing in between runs a household through a values elicitation, a five-capital balance sheet, a written charter, or an annual gate that is allowed to say no.
The name is Acequia. The argument is oikos. The offer is one instrument across two estates, and one lamp per person.
The condition has a familiar half and an unnamed half. The familiar half is why owners buy operating systems for their companies. The unnamed half is the reason those same owners are no better off at home than they were twenty years ago.
The business runs through you. A business that lives in people's heads cannot be measured, and one that cannot be measured cannot be handed to anyone: not a manager, not a lender, not a buyer.
Everything else does too, and nobody ever built you a system for it. The estate lives in one head, has never been written down, and cannot be handed to anyone either: not a partner, not a child, not an executor.
The second one has a specific binding form. Two owners make the same decision repeatedly, because nothing recorded why they made it the first time. The balance sheet has never been consolidated at the level things are actually owned. The succession problem is larger than the one inside the companies, because the companies at least have managers. And the whole of it is carried by memory — usually one person's, and usually not the one who gets credit for it.
Which is why this is buildable rather than aspirational. Acequia is not a second product. It is a second set of views on the same record, plus the integrations a household needs and a library of frameworks that has never been written as software.
An acequia is the shared channel that carries water from where it is stored to where it is used. It is also, and this is the part that matters, a self-governing institution: an elected mayordomo who allocates and enforces, an elected commission that sets policy, proportional rights by membership, written bylaws, and shortage shared out rather than seniority taking everything.
The word is Andalusi Arabic, and the institutions it names are the most credentialed commons on earth. Elinor Ostrom built the eight design principles of Governing the Commons from four case families, and the Spanish huertas are one of them. The Valencia water tribunal and the Murcia council are UNESCO intangible cultural heritage, inscribed in 2009 and dated to al-Andalus between the ninth and thirteenth centuries. They have been running on written rules, in public, for a thousand years.
Four of Ostrom's eight principles are a family charter already, which is why this is the one name where the thing named and the thing sold are the same object:
Who is a member, and of what. The first question a family charter has to answer and the one most families never write down.
The setting almost every family lacks — something between ignoring a breach entirely and expelling someone from the family.
Oral, public and free. The design target for a family assembly, as against a mediation retainer nobody calls until it is too late.
Governance at each level matched to that level. That is the section drawing on the next page, described eight hundred years early.
The deeper argument is oikos — the classical household as a single productive unit of family, land, stores, people and the enterprises that fed them; and oikonomia, its management, from which the word economy descends. Aristotle splits acquisition into a natural kind, which is part of household management and bounded by what the household needs, and an unlimited kind pursued as an end in itself. We are not inventing a category. We are naming an amputation and closing it.
One drawing, cut vertically through everything a principal owns. Above the waterline, the household. Below it, the enterprise. The same three instruments are named on both sides — that repetition is the entire product claim, and it should be the first thing you notice. The second thing to notice is what sits on the line rather than above or below it: the lamp is issued to a person, and a person is rarely confined to one estate.
| Layer | Instrument | What it holds | Available today? |
|---|---|---|---|
| The Charter | POSEIDON | Who decides what, where, and on what grounds. The assembly, the shared story, the refusals list, the conflict protocol. | Rarely. Fewer than half of family offices have formal governance at all. |
| The Horizon | GLAUCUS | Where the family points itself. Planning across money and everything that is not money. Succession, the next generation, philanthropy, entry and exit. | Rarely, and expensively. Roughly a third have a succession plan. |
| The Lamps | PHAROS | One instance per person, on either estate — the same instrument the enterprise side runs, pointed at a different record and a different set of goals. Five items a day, each binary, each citing the goal it serves or disqualifying itself. | Nowhere. No analogue exists in the category. |
| The House | TRITON | Calendar, health, travel, education, household staff and vendors, family communication — with checklists and cost decomposition, so the household has a real P&L. | Split between staffing agencies and eighty-dollar calendar apps. Never instrumented. |
| The Ledger | PROTEUS · NEREUS | One balance sheet across property, side ventures, investments and the group. Liquidity, liabilities, entities, tax, the member portal, purchase and travel requests. | Partly. This is what the outsourced-family-office category actually sells. |
| The Group | PROTEUS · NEREUS | Consolidated close, intercompany, related-party pricing, one compliance calendar, one canonical record. | Yes. This is the ordinary enterprise sale. |
| The Operations | TRITON | SOPs, checklist battery, crews, training cascade, cost decomposition per product. | Yes. This is the ordinary enterprise sale. |
| The Gate | KEEL | The annual double loop across the whole section — the one licensed place to question the governing variables rather than the plan. | Nowhere, at either estate. |
This is the module that makes Acequia a product rather than a better ledger. The design brief is exact: encode what professional facilitators do, so that a couple can run it themselves, with a facilitator available but not required. Every framework below is real, published, attributed, and shaped like software. None of it is ours, and that is the point — the work was to turn a canon into a runnable sequence with prompts, artefacts and a decision record.
A fixed sequence, after George Kinder. If you were financially secure, how would you live. If you had five to ten years, what would you do with them. If you had one day, what did you miss and who did you not get to be. And then the fourth: you are the elder of your family — what world do you want your great-grandchildren to live in? Run privately by each principal, then revealed side by side. The third question produces the material, because regret is where the real goals live.
Eighty-three cards, three piles, built on the free and validated University of New Mexico personal values card sort rather than a licensed deck. The facilitation move is not the sort. It is the aggregate reveal: everyone sorts alone, then overlap and divergence are shown at once. Overlap becomes the charter. Divergence becomes the agenda.
After James E. Hughes Jr. Financial capital is the thumb; human, intellectual, social and spiritual capital are the four fingers, and they are the growth targets. The artefact is a balance sheet whose assets are the four qualitative capitals, shipped on the same page as the financial one.
Owner Room, Board Room, Management Room, Family Room, after Lachenauer and Baron. Each has rules about who is admitted, what is decided there, and how. As software it is a routing rule: every item on the register is stamped with the room it belongs to, and items in the wrong room are flagged rather than resolved.
Tagiuri and Davis. Family, ownership, business, and the seven overlapping sectors people occupy. A conflict-diagnosis tool rather than a planning one: map each participant to a sector and a dividend argument stops being a personality clash and becomes structural.
Six components, after the IFC handbook: values and mission; the family institutions; the board; senior management; the authority between them; and policies on employment, share transfer and succession. The employment policy is the most encodable document in the entire literature, and its governing line is that employment is neither a birthright nor an obligation.
A pool from which family members borrow on terms unavailable commercially, to fund enterprise or education. Foundation, then evaluation where the next generation pitches to a board that scrutinises and mentors, then maturation with quarterly review. An application form, a rubric, a decision log and a cadence — the most literally software-shaped artefact in the canon.
Twenty minutes, assumed-failure framing, after Gary Klein: it is a year from now and the decision failed — write down why. The best conflict-surfacing instrument available. And three alternative five-year lives, after Burnett and Evans, run in parallel by two partners and overlaid: nine cells, and an artefact no facilitator currently produces.
A retreat is the beginning of a process, not an end in itself. Which means the software's job is not the retreat — it is the six weeks before it and the eleven months after. Four cadences, each in its own room, each producing a decision with an owner and a date.
What we are describing has existed for about a century. It is called a family office, and until recently it cost roughly nine hundred thousand dollars a year below a quarter of a billion in assets, and needed a floor of staff to run. The cost stack is published, and it is lopsided in one specific direction.
A family office is a nineteenth-century staffing answer to an information problem. What that ratio actually buys is consolidation, reporting, document handling, a compliance calendar, and coordination between people who do not share a system. Those are precisely the four things that got an order of magnitude cheaper in the last two years — and they are, verbatim, the four things family offices already report using AI for successfully. The overlap is not a forecast.
The best-documented democratisation event in wealth management is a robo-advisor cutting its account minimum from five thousand dollars to five hundred. Reher and Sokolinski studied it in the Journal of Financial Economics. Middle-class participation rose 107%. The welfare gain was three times larger for people over fifty-five. And the finding that matters most here: participation in the bottom quintile did not change at all.
The service moved down exactly one tier and stopped. So the correct claim is not a family office for everyone. It is families one or two tiers below the traditional threshold — and that is the claim the evidence supports rather than contradicts.
When family offices hire, they rank a candidate's personality at 73% and whether the family will trust them at 72%. Qualifications rank last, at 52%. Trust and temperament matter more than credentials, and the industry says so in its own surveys.
Software substitutes for the attribute they rank last. The small expert team is doing more work in this thesis than the technology is. Anyone selling you the opposite is selling you the easy half.
The household module is the one that attracts the most nonsense, so the argument for it is built from the most defensible material available and stops exactly where the evidence does.
A family's affairs have no production process. Nobody ever designed how a household gets run; it accreted. So it runs on ad-hoc coordination, and coordination overhead scales with the number of open commitments rather than their size — reaching one decision across five or six exchanged messages can generate dozens of extra inbox checks. Unresolved loops then degrade the next block of work: switching away from an incomplete, time-pressured task leaves attention residue that measurably damages performance on whatever comes next. So the cost of an unresolved household item is not the ten minutes it will eventually take. It is the quality of the hours next to it.
And more sensing makes it worse, not better. Alert acceptance falls sharply with each additional item in a session, and the cause is per-session cognitive overload rather than desensitisation. Which is why this is a subtraction engine with a hard five-item cap, and why an empty brief is a valid brief. Sustained load of this kind has a documented physiological cost with an honest name — allostatic load, the cumulative wear across neuroendocrine, cardiovascular, immune and metabolic systems.
A calendar with no holes has failed. Scheduled parties and scheduled laughs are a golden jail, and a person aiming to be perfectly regulated at all times ends up looking violently regulated rather than flexible.
Unstructured time is a first-class deliverable — scheduled and defended, not residue. Every module removes obligations before it adds structure. One change a week, and retiring something counts as the change while adding one does not.
If the experience survives the removal of one element of the discomfort, that element was a feature. A system that removes all friction removes the thing the principal came for.
The mental-load research is unambiguous that one partner does the setup. A product that becomes another surface on which she carries the load has reproduced the problem it was sold to solve. Conception, planning and execution move together or the module is not installed.
No children on scorecards. No performance instrumentation of anyone who did not consent to it. This tooling is legitimate for chores, money and calendars, and corrosive applied to a child's inner life or a marriage's emotional register.
No graph in the world can do full justice to the unexpected moments. That is a design constraint we accept, not an objection we rebut, and it is printed inside the product.
A family of any complexity already employs a bench: an accountant, a tax adviser, an estate lawyer, a notary, a banker, an insurer, and periodically an auditor or somebody's diligence team. None of them share a record. Each holds a partial, private, dated copy of your affairs, assembled at your expense — and none of the copies agree with each other.
Every engagement opens the same way: a request list. The entity documents, three years of statements, the deeds, the trust instruments, the share register, the insurance schedules. Somebody in the household then spends four to six weeks finding them, and you pay senior professional rates for that search. Next year you pay it again. When the next professional arrives you pay it a third time, because the artefact the archaeology produced stayed on their server rather than in your house.
Administrative burden has three costs and only one of them is invoiced: learning what is required, complying with it, and the psychological cost of carrying it. The bill captures the second. The family absorbs the first and the third, usually through one person, and usually the same one every time.
A new adviser is granted a scope on day one and reads themselves in. The first meeting is a judgment conversation rather than a document conversation, which is the conversation you were trying to buy.
The obligation register is not a private worry list. Your accountant sees the same filing calendar you do, with the same dates and the same evidence attached — and a row closes on a document, never on somebody's word.
The opinion, the filing, the deed, the amended trust: each is filed against the entity it concerns and the decision that prompted it. The next professional inherits it instead of rebuilding it.
Every counterparty carries an engagement letter, a scope, a fee basis, a renewal date and an owner. You can see what each one costs and what each one produced — a conversation almost no family is equipped to have.
A family constitution is morally binding, not legally binding; the shareholders' agreement, the articles, the trust deeds and the buy-sell agreements are what enforce it. The charter work is the input to that drafting. It tells your lawyer what you actually decided, so they draft the instrument rather than interview you into one.
Not a search. The scope is defined in advance and opens on a documented trigger — which is the only real answer to the question everybody at this level is quietly asking.
A data room is normally built under pressure, for one counterparty, at the worst possible moment: a sale, a refinancing, an audit, a death. It is assembled from whatever can be found in six weeks, it is incomplete in ways nobody discovers until it is being read adversarially, and it is discarded afterwards. Here it runs the other way around. The room is the record, permanently, and an engagement is a scope granted against it.
Each counterparty sees a defined slice. Nobody is handed the whole estate by default, including us.
Superseded documents stay, marked superseded. What was known, and when, is the question that decides disputes.
Every access is timestamped. You can see who opened what, and when, without asking anyone.
Access ends on a date by default. Renewal is a decision somebody makes, not an oversight nobody notices.
The whole room leaves with you in open formats, on request, without a conversation.
The licence to the configured instance is perpetual and, for a family estate, it outlives the principal. So does the room.
We do not replace them. Judgment, opinions and signatures stay with licensed professionals, and so does the professional liability. The system carries the record and the cadence; your accountant still signs the return and your lawyer still drafts the deed.
We do not sit between you and them. Direct access in both directions. A layer that becomes a bottleneck has failed at the one thing it was installed to remove.
We take no referral fee, introduction fee or commission from anybody on your bench. Not from the bank, not from the insurer, not from the firm we like. Ever.
And we do not run a beauty parade. We are not going to tell you your accountant is the wrong accountant. We are going to make it possible for you to see that for yourself, which is a different and better service.
The same mechanism runs on the enterprise side of the section, where the counterparties are auditors, lenders, regulators and eventually a buyer. One record, two estates, and a room that was never struck in a hurry. The long version is The Archaeology Tax.
Operating proof is a list of refusals, not a list of inclusions. Each of these costs us something, which is the only reason to believe any of them.
A refusals list costs nothing if it costs nothing, and the only test is whether operating decisions actually differ. So price it: the assets-under-management refusal alone gives up the most profitable fee structure in the category.
An owner with an operating business and a life that has outgrown the way it is being run. Usually two countries, two currencies, several entities, property, some side ventures, and two owners rather than one. Usually somebody who has already fixed the company and can see, precisely because of that, how much less instrumented the rest of it is.
That in ten years you will be doing exactly this, just more tired. Most owners do not have language for it. When you name it, they recognise themselves immediately.
Not knowing what you do not know — and the sharper version of it, which is looking foolish in front of the partner who said it seemed like a bad idea.
Somebody who has done the branded-luxury circuit and found it hollow, whose objection is never price. It is whether this will actually be different from the last thing.
You pay for the state, feel the effect, come home, and lose it within a week — because the life you return to has no system in it. The translation is the product.
There is one sentence that matters more than the rest of this page. In a company, we bring the institutionalisation and you bring the domain expertise. Here you bring something we cannot supply at all, which is what your family is for. We do not know what you want your grandchildren to inherit, we are not going to learn it in ninety days, and anyone who says otherwise is selling you something. We hold the instrument. You hold the reason. Every framework here is built to elicit rather than to prescribe.
Three stages. The first is sellable on its own and most families should start there.
The values work, the four questions run privately and compared, the five-capital balance sheet, the decision-rights map across the four rooms, and the first written family constitution. A facilitator hands you a summary of a retreat. This hands you the instrument the retreat was supposed to produce, plus the eleven months after it.
One balance sheet across property, side ventures, investments and the group. The household P&L, the reserve floor written down as a number with a date, the entity map, the filing calendar, and the member portal.
One fee. No per-seat licence, no per-member charge, no uplift for the second entity or the fourth property. The House and the Lamps install inside it, and the annual gate is included.
Fees are on request, and they are never indexed to assets, per seat, per family member, or hourly. A fixed build fee terminating in a named artefact, then a monthly indexed to complexity: entities, jurisdictions, properties, principals carrying a lamp, and liquidity events a year.
The roster is capped at six families. Published rather than implied, because a cap that is stated is a cap and a cap that is hidden is a waiting list.
What you own at the end is a perpetual, irrevocable licence to the configured instance. The data is yours in every scenario, exportable, without a conversation. The licence transfers on a change of control and, for a family estate, it survives the principal — which is the only answer to the only question that really matters here. We earn the retainer by maintaining and improving the system, never by holding the off switch.
There is a gate before there is a proposal. It is a conversation with both principals present and a written statement of what you actually want, and it exists because this product cannot be installed into a household where only one person wanted it.
The full written argument is The Principal’s Dilemma, white paper II. The rest of the library is at Writing.
Nothing in the framework library is ours. It is a published canon, and the work was to encode it. Everything below is attributed on purpose, and no proprietary mark is used or implied.
A family constitution is morally binding, not legally binding. The shareholders' agreement, the articles, the trust deeds and the buy-sell agreements are what enforce it, and nothing in this product will ever imply otherwise.