How a broad hope for a flourishing world becomes a small number of well-reasoned, coordinated actions — each with a named owner, a matched resource, a dated milestone, and evidence that decides the next move.
Earlier drafts were stress-tested against composite family perspectives — an internal editorial exercise, not interviews with real families and not an endorsement by anyone. The failure it exposed most consistently was that the draft described intentions in the grammar of accomplishments. This version labels every claim, separates what exists from what is proposed, and names what we do not have. Method on page 48.
In service today. Real users, real transactions, real payroll behind it.
Built and shown working. Not yet in routine service with clients.
Partially built. Some elements near release; others months away. No fixed date.
This document's proposal. Nothing agreed, nothing committed, nothing approved.
A capability we need and do not have. Named deliberately, on page 33.
Status tells you how far along something is. This axis tells you what kind of check stands behind it. Most capabilities in Part IV were shown to us on screen or described by the people who built them. Shown is not audited, and a missing check may change a claim rather than confirm it.
Organizations named in Part IV are independent. Diagrams show intended division of work and coordination, never ownership or control. Any partner may ask to be described differently, or not at all.
A resilient and flourishing future for all life.
Move capital in service to life. We create the conditions for that future by seeing clearly, finding where effort changes the most, finding work that can actually deliver, and bringing the right people and resources behind it — so that humanity, nature and machine can exist in a healthy relationship.
There is more money available than there are good places to put it. The shortage is not capital; it is clarity, coordination, and work that is ready to use capital well. Families hold extraordinary means and face an unreadable world. We exist to improve the choices made where those two meet.
Six steps, in order. Each produces a written artefact you can hold.
Independent organizations we intend to work with are already operating: a partner family service priced at €180k per family per year (that partner's price for its own service, not Third Attractor pricing), a UK asset manager reported to hold an FCA licence, a coordination platform carrying cap tables and data rooms, and a volatility model built on open-source indicators. None of this is yet an integrated Third Attractor service, and we have not run one complete case ourselves. Partner-reported · checks outstanding, see p.29–31
There is no single ask. Four separate requests exist, with different recipients, instruments and approval routes, and none with an agreed amount: a 90-day pilot, a 2027 operating envelope, a platform seed, and catalytic project capital. Unreconciled, on page 41.
Dashed = contribution and coordination. Solid = commitment. No line implies ownership.
This is the long document. It is not the required first read. A twelve-page companion, The Family Brief, carries the same argument at family length and treats this volume as its evidence appendix, alongside a working business model covering membership economics and partner terms. If you have not read the brief, start there and use the routes below to come back here.
| 06 | Three attractors, one choice |
| 07 | The asymmetry we exist to close |
| 08 | The scale we are actually operating at |
| 09 | What is actually scarce |
| 10 | What flourishing means |
| 12 | Vision, mission, and what sits under them |
| 13 | What Third Attractor is, and is not |
| 14 | The honest intersection |
| 15 | Who we serve, described honestly |
| 16 | Four places we look |
| 17 | The operating compass |
| 19 | The whole route, end to end |
| 20 | Mission before mechanism |
| 21 | How deep to intervene |
| 22 | The family journey, with deliverables |
| 23 | The smallest system that can do the work |
| 24 | The gates: what must be true to proceed |
| 25 | Not every problem needs the same money |
| 26 | The Catalyst Pool |
| 27 | One North Star, two balance sheets |
| 29 | Evidence: intelligence and software |
| 30 | Evidence: capital and regulated perimeter |
| 31 | Evidence: people, network, operating base |
| 32 | The readiness matrix |
| 33 | What we do not have |
| 34 | What is still unresolved between us |
| 35 | Who does what |
| 37 | Six workstreams, and what we fund first |
| 38 | Four quarters, four gates |
| 39 | The first 90 days |
| 40 | The team we would build |
| 41 | The money, and what it is not for |
| 42 | Trust, conflicts, and how we are paid |
| 43 | Worked example: food resilience |
| 45 | What success looks like |
| 46 | Hold us to this |
| 47 | Decisions we need from you |
| 48 | What seven families said about v5 |
| 49 | Sources, status and contact |
The family route — nine pages. 06 the choice, 14 the honest intersection, 22 your journey, 32 readiness, 33 what we lack, 39 the 90 days, 41 the money, 42 conflicts, 47 the decisions.
The adviser appendix — the pages built to be checked. 29–31 evidence and its gaps, 34 what is unresolved, 35 who does what, 41 full budget workings, 42 fee routes, 49 sources and status. Read these before advising.
The sceptic's route — four pages, worst first. 33 what we do not have, 34 what we have not agreed, 48 what the simulated review said we got wrong, 46 how to hold us to it. Nothing here is verified by a named external checker.
We are moving life out of the biosphere and into the technosphere, at an alarming rate. And there is no shortage of money in the world — what is scarce is clarity, coordination, and work that is ready to use resources well. Everything that follows is an attempt to produce those three things reliably.
Polycrisis, metacrisis, metamorphosis. The first two are descriptions of what is happening. The third is the only one that is a choice.
An attractor is a state a system tends toward without anyone choosing it. Water finds the valley. Capital finds the familiar return. Institutions find the pattern that protects them. Ahead of us sit one future led by chaos, one led by control, and a third which is neither — coherence. The first two are already well formed and well funded. The third has to be built, deliberately, and with enough gravity that things can actually fall into it.
A family like yours brings a great deal: capital, operating businesses, customers, relationships, influence, experience, land, networks, time, and a horizon measured in generations.
The world you are trying to serve is hard to read. There is too much information. Expertise is scattered. The stories people tell about it conflict. Coordination is weak. Cause and effect are unclear. And there are more apparent opportunities than anyone could act on well.
So capital accumulates on one side of a gap, and work that could absorb it well is thin on the other. The gap is not a funding gap. It is a readiness gap, and almost nobody is paid to close it.
A mechanism is chosen before the problem is understood. The question "where should we invest?" is answered before the question "what actually needs to become true?" is asked.
What we do insteadOne of us was asked by his mother, years ago: why don't all the most successful people just get together and solve the world's problems? This page is the honest answer — and the work is to remove it.
Both sentences are true at the same time, and holding them together is the whole reason this organisation exists. If the constraint were money, the answer would be fundraising. It is not money. It is coordination — and coordination is, in our partner's words, extremely underrated and underfunded.
All four figures are partner-reported. None has been independently checked and no checker or date is named. Not yet confirmed
Find the small number of places where modest capital changes the rules under which much larger capital behaves.
If most effort is lost to redundancy, recovering even part of it is worth more than any single new fund.
At this scale the highest-return move is often policy, standards or index design — what one of us calls lobbying for good.
Investing company by company cannot reach critical mass. The target is the tissue connecting an ecosystem.
Each of these is a distinct failure, with a distinct remedy. Treating all five as a money problem is why so much well-intentioned capital produces so little durable change. We organize our own capability around closing them in order.
Too much information, scattered expertise, conflicting narratives, and incentives that reward confident storytelling over accurate description.
An independent view with its sources attached, its uncertainties named, and alternative explanations considered before a conclusion is drawn.
Beyond a handful of people in a room, coordination degrades. Good actors working on the same problem rarely know one another, and almost never share a decision record.
Shared coordination objects — actors, needs, offers, assets, commitments, signals — held in common infrastructure rather than in individual heads.
Important work is routinely not investable: no anchor customer, no diligence pack, no governance, no measured baseline, no owner who can be held to a milestone.
A pool and a practice dedicated to making work ready — before anyone is asked to invest in it. See page 26.
Families are offered products that suit the seller. The question of where a particular family has genuinely unusual leverage is rarely asked, because answering it often reduces the sale.
A written account of where your intent and capability honestly meet real leverage — including where they do not. See page 14.
Venture economics need outcomes most systemic work cannot produce. Grants rarely build durable capability. The instrument is chosen by habit rather than by the shape of the work.
Six forms of response — give, catalyze, invest, build, influence, wait — selected after the work is understood, not before. See page 25.
When we ask what needs to become more true, we ask it in four places. A gain in one layer purchased by damage to another is not a gain. This is the test we apply to our own work as much as to anyone else's.
Soil, water, forests, oceans, atmosphere, biodiversity, the energy and material flows that everything else rests on. Not sustained — regenerating.
Ownership, incentives, rules, information flows, the design of money and markets, and whether institutions are capable of learning.
Trust, shared sensemaking, legitimacy, the capacity to hold conflict without fracture, and to make collective decisions that hold.
Agency, health, meaning, development, the capacity to bear responsibility. No system design survives contact with people who are not developing.
A family's own coherence sits inside this frame too. A family that cannot make a decision together cannot steward anything across generations — which is why family development precedes capital deployment in our sequence.
Stated plainly enough that you can tell when we are failing to live up to it, and specific enough that you can tell when we are wrong.
A vision names the world we are working toward. A mission names what we do about it. Everything downstream — the gates, the instruments, the costs — is accountable to these two.
A resilient and flourishing future for all life.
It does not change. We cultivate the conditions in which all life can flourish — and in which humanity, nature and machine can exist in a healthy relationship.
Move capital in service to life — creating the conditions for a future where all life can flourish, by seeing clearly, finding where effort can change the most, finding work that can actually deliver, and bringing the right people and resources behind it.
Put most plainly from inside the group: shift the claims on the natural world from serving power and wealth to serving life. It is a shared mission — not theirs that we serve, nor ours that they fund — held to account by evidence.
Maintain an independent view of reality. Identify where leverage genuinely lies. Find or build credible work. Match the right form of capital. Assemble the smallest system that can deliver. Measure. Let the evidence change the next decision.
Sensemaking · family relationship and development · supply-side origination and diligence · structuring and capital strategy · coordination infrastructure · measurement and learning.
A grant. An investment. A company. A research programme. A piece of software. A convening. A procurement commitment. A policy intervention. Or waiting. A fund is not a strategy. A foundation is not a strategy. Software is not a strategy. They are tools.
An organization that starts at level five builds a fund, then searches for a mission the fund can serve. The mission becomes a description of what the mechanism can already do. Our structure is designed so that we are never forced into that position — which is why our intended revenue is designed to come from subscription and service rather than from deploying your capital. No commercial arrangement is yet confirmed, and Third Attractor is not yet earning any such revenue. See page 42.
Families told us "orchestration layer" meant nothing to them. Fair. Concretely: we hold an independent view of where change matters; we help you see where you can genuinely contribute; we find and strengthen the work; we assemble the people, governance and capital around it; and we measure what happens. We do not own the pieces. We make them work together.
Two names have been offered from inside the group: enlightened capitalism, and a family legacy operating system. Both useful. Neither is the mission — that is the first line of this document.
The mission. Our own independent reading of how the evidence fits together. The setting of priorities. The shape of family relationships. The work of bringing people and resources together. Discipline where interests conflict. And learning.
Deep family development work. Software. Specialist research. Regulated investment management. Local implementation. Venture building. Each is better done by people who do only that.
Every product, expert, operator, software tool, fund, institution or service a family might need. An orchestration layer that tries to own everything stops being one.
Every engagement, commitment and experiment has a defined point at which it can be ended, with a documented reason. A machine that cannot stop is not being steered.
Third Attractor is for families thinking in generations. It holds an independent view of where change may matter most, helps a family see where its intent and resources can genuinely contribute, brings together the right people and capabilities, and moves the appropriate forms of capital into high-leverage action — in service of a resilient and flourishing future for all life.
Coherence without control. We bring a map before we bring a product. The aim is for the whole to hold together without Third Attractor controlling it.
The commercial temptation is to draw this overlap large, because a large overlap justifies a large engagement. In our experience the genuine overlap is narrow — one or two places, not eight. Naming it narrowly — one concern, and one place the family can actually act: its capital, its operating company, its land, its convening power or its name — rather than a long list of worthy causes — is the single most useful thing we do, and the thing most likely to cost us revenue. We have designed our economics so that it does not. See page 42.
Every family brings different responsibilities, experience and views. Nothing below is a profile of you, a diagnosis, or an assumption we will act on. These are patterns reported by people in our group who sit inside such families; we begin any relationship by asking what matters to you, who should be involved, and what you would and would not consider doing.
Strong analytical and operating capacity is usually abundant. Experience of open-ended, multi-decade, contested questions is less common in any organisation, including ours. Whether a family wants to do any development work at all is entirely its own call — it is an option we offer, never a prerequisite we impose.
People arrive constantly with ideas, pitches and asks. Strong filters are a rational response to that. Anything we build has to survive those filters rather than attempt to bypass them.
They can see where this is going. The distance between the dominant narrative and the state of the world keeps shrinking. What they lack is not conviction about the problem but a credible, grounded place to put it. The commonest thing we are told, in different words each time, is some version of I can see it, I have the means, and I do not know where to begin.
A baseline of trust with each other, a different openness to deep critique, direct access to capital in some families — and, in most, real influence over the generation that holds it. Experience of that generation ranges from heavy expectation to genuine release. The great wealth transfer is the single largest handover of capital in history, and it is already underway.
We are not trying to get people to move money into things we happen to like. We are building something together, which takes longer and holds better.
Ten years of pledge events have taught a generation of wealth holders to expect a hand out at every gathering. People leave those rooms tired. The test for any room we hold is whether the family received more than it was asked for.
Where entrepreneurs appear, it is in a short, bounded, well-vetted session — or remotely. The rest of the time the room belongs to the families.
The friendships are a by-product, not the product. We have watched convening become the deliverable elsewhere and are designing against it.
Radical clarity about what we know, what we do not, what we earn and what we want is the only approach that does not trigger the wariness this world has rightly built.
Families told us the four layers were beautiful and unusable. So here they are converted into the kind of specific, falsifiable priority we are actually willing to be judged on. In build These are our current candidates under active research, not a published thesis. Thesis development is the first workstream in Part V.
Does the biophysical capacity to renew measurably improve, at a named place, over a named period?
Does the rule, right or flow change — such that better behaviour becomes the easier behaviour?
Can a group that previously could not reach a durable decision now do so — and does it hold?
Do named people hold more real agency and responsibility a year later than they did before?
Five live referrals from members of this group are running as test cases through the route on page 19. They are described here without names. The companies have not consented to appear in a circulated document, and naming them would imply an endorsement we have not made. The named register is held privately, and is shared with a family on request only once the company has agreed.
A frontier-biology venture — tests whether we can assess a deep-science bet systemically, rather than on the strength of the story.
A waste-to-resource infrastructure business — AI and robotics turning a cost line into a revenue line. Infrastructure, not a product.
An agri-tech and food-supply case — where the margin to solve the problem may simply not exist inside the current system.
A biodegradable-materials case — a solution for the world we want, not the world we are in. Useful precisely because it exposes that distinction.
A monetary-design proposal — deepest leverage, longest horizon, least legible to conventional diligence.
None is a shortlist entry and none is an endorsement. No candidate becomes a priority without a causal account, source evidence, alternative explanations, expert review and a named decision owner.
Each of these is a constraint on our own behaviour, not a claim about the world. They are applied to our own decisions in writing, and a decision that fails one does not proceed on the strength of the others.
Sources attached. Uncertainty named. Alternative explanations considered. The uncomfortable version reported alongside the flattering one.
Including people who were not in the room, and people who come after. A benefit captured by the already-powerful is not the benefit we aimed at.
Or does it make people more dependent, more managed, more observed? Local agency is preserved, not optimized away.
The horizon is not the fund life or the news cycle. It is what we are handing forward, and to whom.
The map is a hypothesis, not a doctrine. We hold models lightly, use power carefully, and preserve the ability to stop.
Applied after the fact, not before. Intentions are the cheapest thing in this field. What actually happened is the only report that counts.
A compass governs decisions. These three govern the group that makes them. Where organisations in this field come apart, it is almost never the analysis — it is one of these three left incomplete.
The standard we set for each other: spine strong, heart open, soul on fire — and the role is closer to immune cells in a body than to an institution in a market.
We are participants in something larger than ourselves, not its masters — so we use power carefully, hold our models lightly, protect local agency, and keep the ability to stop.
The route a concern travels to become a thing that exists in the world, with a name on it. Seven stages, five gates, and a defined point at which each one can be stopped.
Each gate is a place where work stops unless something specific has been demonstrated. The named owner of each refusal is on page 24. Nothing on this diagram is optional, and no stage can be skipped because a funder is enthusiastic.
“A thousand right answers do not add up to one effective pathway.” The route exists to turn a field of good intentions into one sequence that can actually be followed.
“What does success at each of these stages look like, so that if I do something, I know I am doing what everyone wants?” Each gate answers that question in writing, with a named owner.
“Where is the trunk? What can we do today?” Stage 01 begins now, on live cases, inside existing organisations — not after the structure is settled.
Wait. Not every concern should be funded now. Deciding not to act, and recording why, is a legitimate outcome of this route — and historically the one most often skipped.
Stated as a condition in the world, in plain language, with a date. Not as a programme, a product, or a round.
The constraint, named. Usually it is not money. It is a missing owner, a rule, an incentive, a relationship, or a capability.
The depth of intervention is chosen deliberately, using the ladder on the facing page, not chosen by default.
The instrument is the last decision, not the first. A fund is not the strategy. Neither is a foundation, nor a piece of software. They are tools.
This sequence is fast, legible, and reliably produces activity without changing the condition that caused the problem. It is also how most well-intentioned capital behaves.
ThisSlower to start and harder to pitch. This is our working preference, not a proven rule. We believe it produces interventions shaped by the system rather than by whichever instrument happened to be available, and the first ninety days exist partly to test whether that belief survives a real case.
Depth is a choice, and each level is legitimate. The failure is not working shallow — it is working shallow while believing you are working deep. Every piece of work we take on is assigned a level in writing.
Treat the harm in front of you. Food parcels during a shortage. Necessary, immediate, non-compounding.
Change what moves and how fast it is sensed. Shorten a supply chain; measure soil annually.
Change who decides and who holds the asset. Community land trust; procurement rules; cooperative equity.
Change what the system is for. From yield per acre to nutrition per acre, and the accounting that follows.
Change what is thought normal. From land as extractable asset to land as inherited trust. Slowest, hardest to attribute, and in our view the most durable — a belief we hold, not a result we can show.
Defend what must not be lost. The cheapest form of progress, and the most frequently neglected.
Carry people and institutions through the transition without collapse. Most of the real work sits here.
Highest variance, longest horizon, needs the most patient capital of the three.
A method that cannot be corrected by evidence is a doctrine. Our operating model is reviewed against outcomes each quarter, in the same report that covers the projects. Where we were wrong, that is stated in the first paragraph, not the appendix.
Nearly all tangible evidence available today sits at levels 1–3. Levels 4 and 5 are where our thesis claims the leverage is, and are the hardest to evidence. We mark that distinction explicitly in Part IV rather than blurring it.
The same split runs through the portfolio: solutions for the present and solutions for the future. The first keep people whole through the transition. The second depower the current system — and because the market is substantially co-opted, that second category rarely wins on price alone.
Working at depth is not the same as working slowly. It means choosing the level at which the change can hold.
| Step | What happens | What you receive | Typical elapsed | Who leads | What we need from you |
|---|---|---|---|---|---|
| 01 See | We map the field — conditions, actors, causal account, and what the evidence does not support. | Field map with sources attached and uncertainty marked | 2–4 weeks | Intelligence layer | One conversation about what you are genuinely worried about |
| 02 Locate | We identify the points where a small, well-placed force changes the trajectory, and the depth at which it must act. | Leverage note, two pages, with the depth level named | 3–6 weeks | Intelligence layer | Your reaction — including where you think we are wrong |
| 03 Choose | You decide which of these you want to carry. This is a family decision and we do not make it for you. | Your written priorities, signed by you | 2–3 weeks | Your family | A decision, and clarity on who in the family holds it |
| 04 Explore | Diligence on the specific route, with a deliberate case against it, reviewed by people who disagree. | Diligence pack with a red-team section | 6–12 weeks | Diligence and partners | Access to your advisers; tolerance for bad news |
| 05 Assemble | The smallest system that can do the work: owner, capital, governance, capacity, baseline, stop condition. | Term sheet or grant memo — numbers, dates, owners | 4–10 weeks | Structuring | Confirmation of amount, instrument and co-funders |
| 06 Act | Capital moves and work begins. The project owner is accountable for delivery, not us. | Signed instrument and an operating plan | On signature | Project owner | Signature, funds, and patience with the operating phase |
| 07 Learn | We report against the baseline — failures, variance, and what we changed as a result. | Quarterly evidence report, published to participants | Continuous | Shared | Willingness to read the uncomfortable version |
Proposed service levels, not delivery history — we have never run this end to end. Added up sequentially these stages total 17–35 weeks, which is why the ninety-day pilot deliberately runs 01–04 on a narrow case and does not promise 05–07.
A family can exit after any step. The artefacts produced up to that point remain yours.
Steps 01–02 can run for a whole cohort and overlap; 03–06 are specific to one family and one project. The family's own time is roughly a half-day a month, plus a decision at step 03 and a signature at step 06.
Most well-funded failures are not failures of money. They are missing one of the six components below, and the money arrived anyway. Where a component is absent, our job is to build it before the capital moves — or to decline.
What must be true, for whom, by when — in one sentence a sceptic could verify.
Fails if: the goal is an activity, not a condition.
The one thing whose absence stops everything else. Often not capital.
Fails if: we assume the constraint is money because money is what we have.
The least structure that can hold the work and still be stopped cleanly.
Fails if: the vehicle becomes more interesting than the outcome.
One site, one cohort, one city. Evidence before replication.
Fails if: we scale the model before we have evidence it works.
Already working, under-capitalised. Cheapest and most under-used.
Technology built for something else, redirected to the plan.
Buy the infrastructure, then change what it is pointed at.
Nothing exists. Back a venture builder; accept the failure rate.
| Gate | What must be demonstrated | What disqualifies | Who can say no | If it fails |
|---|---|---|---|---|
| 1 · Truth after Stage 01 |
A causal account with attached sources, named uncertainty, and alternative explanations considered | Conviction without sources; a single expert's view presented as consensus | Research lead, in writing | Returns to Stage 01 or is published as an open question rather than a priority |
| 2 · Fit after Stage 03 |
The family's own written conviction that this is theirs to carry, and who in the family holds it | Enthusiasm from an adviser rather than a decision from a principal | The family | Stops. We do not substitute our conviction for yours, and no fee is earned |
| 3 · Quality during Stage 04 |
Passes all six compass questions in writing, including a deliberate case against | Strong returns with a failing dignity or stewardship answer | Any compass reviewer, unilaterally | Stops or is redesigned. A failure here cannot be offset by a strong financial case |
| 4 · Readiness during Stage 05 |
All six MVS components present: owner, capital, governance, capacity, baseline, stop condition | Capital ready and ownership unresolved — the most common failure mode in this field | Structuring lead | Capital is held. We fund the missing component first, or we decline |
| 5 · Authority before Stage 06 |
The decision sits with the party who bears the consequence, and they have made it | Us deciding on your behalf, or a project owner deciding on a community's behalf | The family, and the affected party | Stops until the right party has decided. Speed is not a reason to skip this |
Gates 1, 3 and 4 are ours to enforce and we can be audited on them. Gates 2 and 5 are explicitly not ours — they belong to the family and to the people affected. That division is the single most important structural commitment in this document.
The failure is rarely the amount. It is using return-seeking capital for work that cannot return, or using a grant for work that could have been self-sustaining and now never will be.
For work that should never have to pay for itself: public goods, research, relief, the slow repair of institutions.
Source · Flourishing Futures Foundation
Makes something investable that was not. Feasibility, legal structuring, the first hire, the pilot site, the baseline study.
Source · Catalyst Pool (proposed, p.26)
For work with a real business underneath it, where the mission can be protected in the capital structure rather than in the pitch.
Source · Family capital, regulated vehicles
Where nothing suitable exists and the gap is important enough to justify founding risk. Rare, deliberate, and reported as such.
Source · Venture building, co-funded
Convening, standards, procurement, narrative, relationships. Often the highest-leverage and lowest-cost instrument available to a family.
Source · Network and convening capacity
The constraint is not yet understood, the owner does not yet exist, or the moment is wrong. Recorded with a date to revisit, not quietly dropped.
Source · No capital deployed
A public-equity account. Roughly fifty listed companies selected for the good they do, held in a separately managed account — seeded with $500,000 this year, targeting $10–20m within one to two months of launch. Daily liquidity, full public reporting, open at $50,000 rather than to qualified investors only. The argument is structural repricing: make values-aligned behaviour visible in brand equity and share price.
A founder consortium. Thirty to forty founders of eight-figure businesses each roll 10–30% of their company into one aggregate vehicle and receive a share of every other. Sovereign businesses, united balance sheet, shared economies of scale — and a single entry point for impact capital that cannot currently diligence forty companies separately.
Allocation discipline. The instrument is selected after the constraint is named, never before. Each commitment records why the other five were rejected. Blended structures are allowed; disguised ones are not. Where philanthropy subsidises a return-seeking party, it is disclosed in the memo.
Neither instrument above is a Third Attractor product. Both would be run by their proposers. They appear here because they are the two most concrete capital proposals on the table.
Almost every project we have examined needed money before it was investable — and that money has no natural home. The Catalyst Pool exists to fund the work that makes a thing fundable, and to recycle whatever returns back into the same gap.
Proposed Discussed and not objected to on calls · no party has formally agreed, approved or committed capital
Pre-investment capital of this kind does exist elsewhere — in project-preparation facilities, recoverable grants and some catalytic philanthropy. We are not claiming to have invented it. We are claiming that the families we work with rarely have access to a version of it they control.
It is also among the clearest things a family could fund first: small, bounded, legible, and useful to every other family in the cohort. Nothing about it is agreed.
Size, ceiling and stop. A first pool small enough that losing it changes nothing material for any contributor, with a written ceiling and a date at which it is wound down if the evidence does not arrive.
Who decides, and who can say no. Allocation sits with a named committee that includes contributors; any single contributor can veto the use of their own capital without explaining why.
What recycling actually means. Returns go back to the gap, not to a management fee and not to our balance sheet. If we ever benefit from the pool's growth, the instrument has already failed.
Holds the purpose. Cannot be bought, diluted or redirected by a return requirement.
What it doesOperating Entity established; year-one strategic programme and end-of-year report in preparation.
Why this side exists. A mission that can be outvoted by a return requirement is not a mission. The foundation is the only place in this structure where the answer to is this still what we said we would do cannot be overruled by a financial argument.
Holds the operating capacity. Earns, invests, builds — within limits set on the other side of the membrane.
What it doesIn build Partner capabilities operating; the combined entity, ownership and governance are being formalised.
What is genuinely open. Whether this side is a partnership, a joint venture, a new company, a merger or an open-source commons is undecided. Five options are live, none is agreed, and the choice is a gating item — not a detail to be settled later. See page 34.
Everything in this part is a specific, checkable claim with a status and a verification flag. Where a figure comes from a partner and we have not yet audited it, it says so on the same line.
Source: the 6 November 2026 working package prepared from a two-hour recorded call, plus the Flourishing Futures Foundation planning session of 3 November 2026. Both are cited page by page. Neither has been independently audited, and we do not present them as if they had been.
On 6 November 2026 the platform was shown live: entity and subsidiary maps, ownership and equity tracking, data rooms, accounting views, and a network knowledge base. What has not yet happened is our own hands-on walkthrough with real access; that distinction is maintained below. This is one of four infrastructure types the work requires — financial, digital, physical and cultural — and the only one with running software behind it today.
| Asset | What it is, concretely | Status | What is still required to call it proven |
|---|---|---|---|
| Symviosis platform Coordination and systems architecture |
Company and subsidiary mapping, ownership and cap-table tracking, governance circles, data rooms, financials and systemic analysis | Operating Shown on screen |
Guided access for our own team and one complete case walked end to end, by us, in our own data |
| Network knowledge base “Second brain” across the network |
Intra- and inter-company knowledge, bottlenecks, inefficiencies and dependencies mapped across a live SME network | Demonstrated Reported |
A worked example of a decision that was materially changed by this layer, with the before and after |
| Volatility and horizon models Open-source inputs |
Forward timeline of destabilising and regenerative events, with the underlying indicators and formulas published openly | Operating Open source, unaudited by us |
Independent review of method and back-testing. An open model is inspectable, which is not the same as validated |
| Think tank / intelligence layer Democracy, systems thinking, governance |
Research and expert-convening function, carried on the network's distribution with philanthropic backing | In build Demo shown; launch intended |
A staffed team, a funded operating budget, a decision interface, and the first published brief |
| Live collaboration record Question → evidence → decision → follow-up |
Meeting intelligence linking a question asked on a call to the later answer, the decision and the named follow-up | In build Product extension, not certified |
One cycle completed on this group's own calls, including a source correction that reopened a conclusion |
| Analysis throughput Companies and funds, continuously |
Dozens of company and fund analyses per week against systemic criteria — the engine behind any claim to source at scale | Operating Reported, not sampled by us |
A sample of ten completed analyses released to us, including two the team considers weak, so we can judge the standard |
| Open intelligence MVP Bring your own model |
A forthcoming release letting a family connect the datasets and research corpus to its own GPT, Gemini or Claude — intended to be open source and collectively owned, not held by one entity | In build Announced; not yet released |
A release date, a licence, and one family using it on a question it actually has. Open-source intent also requires a named governance holder |
| Conditionals framework Capital with enforceable strings |
A standard book of processes and protocols intended to stop a company becoming extractive as it succeeds | Proposed In development |
Drafted terms, legal enforceability tested in one jurisdiction, and one company that has accepted them |
The honest summary of this page. The strongest part of our evidence base and the weakest part of our verification. The software exists and is in daily use by a real network. We have seen it; we have not yet used it. Our first obligation before any family commits capital on the strength of this page is to convert “shown on screen” into “operated by us, on a real case.”
The regulated perimeter is the difference between a thesis and a transaction. It also carries the highest penalty for overstatement, so every figure here is marked with its source and its verification state — and three of them are specifically scheduled for confirmation before any ask is made.
| Asset | What is claimed, and by whom | Status | What must be confirmed before we rely on it |
|---|---|---|---|
| UK asset manager | Acquired outright by Better World for a reported sum in the low single-digit millions; holds an FCA licence, back-office and reporting capability | Operating Reported in call |
The legal entity, the actual permissions on the FCA register, permitted jurisdictions, and who holds regulated decision authority |
| Stewardship capacity | Capacity to steward approximately five billion under that licence | Demonstrated Unverified, currency unstated |
Capacity is not the same as assets under management, and neither is the same as a mandate. We will state current billable assets or state that we do not know |
| PLC and SME coordination | A listing-ready PLC; roll-up of SMEs into shared and employee ownership. A reported pooled-equity performance figure has been withheld from this document pending substantiation and appropriate regulatory review | In build Experiment under way |
How many SMEs have actually joined, and the audited equity movement |
| Family service revenue | €180,000 per family per year, covering platform access, analysis, relationship management and workshops; revenue shared between partners | Operating Reported; split not disclosed |
The service specification, the number of paying families, retention, the revenue split, and the true cost to serve |
| Fee models for the vehicle | Two shapes discussed: 1% management with 10% performance; or 2% / 20% with half the management fee redirected into experiments | Proposed Neither agreed |
Nothing is agreed. The arithmetic in the source (1% of 100m = 1m) is an illustration, not a report of contracted assets |
| First candidate project | One named project put forward as a possible first case | Proposed Candidate, not selected |
A candidate, not the automatic choice. It has passed no gate in Part III |
A capacity is not a mandate. A licence is not a permission. A discussed fee is not a price. We would rather lose a page of impressiveness than one line of accuracy.
Software can be rebuilt and licences can be acquired. A working relationship between a family-service group operating for many years, a systems and software house, and the owner of a regulated vehicle cannot be assembled quickly. It is also the thing most easily broken by a badly designed deal.
Operating Reported; structure informal
In build No terms agreed
Proposed Outreach under way
The purpose of external stress-testing is to prevent the two failures we were founded to avoid: becoming an ordinary impact fund with better language, or reproducing the frameworks we set out to improve on.
The source transcript mislabels several speakers, and some remarks are attributed to the wrong participant. We have followed explicit conversational references and left unconfirmed ownership open rather than guessing. Where this document names a person as responsible for something, that person has either said so themselves or will be asked to confirm before the next version is issued.
The horizontal axis is how far something has progressed. The vertical axis is what kind of check stands behind it. Placements are our own judgement, not an independent assessment, and several rest on a single partner's account. The empty top row is the honest part of this page.
The mass sits bottom-right: capability that is reported to be substantial, thinly evidenced. Until the checks are done we cannot tell you whether that is a verification problem or a credibility problem. Assume the checks may move items down as well as up.
Move anything up this chart by assertion. An item moves only when a named checker, a date and a scope are attached to it. The chart is reissued each quarter with every change marked, including downgrades.
A small, bounded first commitment can be spent almost entirely on moving items up and to the right. That is the cheapest risk reduction available to any family here.
| What is missing | Why it matters to you | Status | How and when it gets closed |
|---|---|---|---|
| A track record of completed interventions | Nobody here can yet point to a closed project, measured against a baseline, with results published. This is the central gap and we will not disguise it | Gap | One case taken end to end in the first 90 days, with its baseline published before it starts, not after |
| An agreed legal structure across the partners | Today the collaboration runs on trust, a shared capital pool and revenue sharing. Admirable, and not something most family advisers will sign against | Gap | Decision 3 of the six on page 47. Requires a UK and a US specialist, and a decision about geography |
| Independent confirmation of the regulated perimeter | Until the FCA register is checked and authorised individuals are named, the licence is a reported fact and should be treated as one | Gap | Register check and a call with the asset manager's compliance function, before any ask is finalised |
| A costed, approved operating plan | The €994,400 envelope on page 41 is an author-built planning illustration, not quoted salaries, an agreed budget or secured funding | Proposed | Replace the illustrative loaded rate with real, location-specific offers for the first three roles |
| Named owners for most roles | Three of six workstreams have no confirmed lead. A plan with unassigned ownership is a wish list with a budget attached | Gap | Programme lead, intelligence lead and engineering capacity first; analysts and diligence only once the pilot shows the workload |
| A tested causal thesis | The four candidate priorities in Part II have falsifiable tests but have not yet been reviewed by people with deep domain expertise who disagree with us | In build | Thirty-day outreach to research partners, then written review of each candidate priority with dissent recorded |
| A capital-recovery mechanism for experiment money | Where a family expects its capital back, no instrument, downside, liquidity or governance terms have been defined. This is the question most likely to kill a commitment late | Gap | Decision 4 of six. Drafted terms for each separate pool before any capital is requested, not after |
| Supply of investable projects | The binding constraint named in the source call: for many priority areas, no company exists whose financial structure lets it address the problem without becoming part of it | Gap | Test four routes on a small number of cases — partner, acquire and repurpose, build, or defer — and report which actually works |
This is the page most documents like this do not contain. Its absence elsewhere is the reason this field is difficult to trust.
Take it to your adviser and ask them to add to it. If they find a ninth gap, we want it in the next version with their name against it.
The previous page lists what we cannot yet show you. This page lists what we have not yet resolved among ourselves. Each item was raised by a member of the group on a recorded call; the wording below is paraphrased and unattributed; no speaker will be quoted externally without consent. A coalition that cannot publish its own open questions is not a coalition; it is a launch.
| The constraint | As it was raised, paraphrased | Status | What would resolve it |
|---|---|---|---|
| Ownership and contribution | “Who has ownership? What am I sharing in terms of data? My relationship capital — what's mine, what's yours?” A full-body yes at the heart, and an unanswered question at the business layer | Gap | Written contribution-and-claim terms per participant before anything is pooled, including an explicit answer on data and relationships |
| A definition of success per component | “What does success at each of these stages look like, so that if I were to do something, I know I'm doing what everyone wants?” | In build | One page per component of the minimal viable system: what exists, what done means, who decides, and the metric |
| Governance without hierarchy | Building a new airport means coordinating some 800,000 tasks over a decade — and it only works because there is a hierarchy. Here there is none. Who decides the master plan? Who assigns the roles? | Gap | A named decision-rights model, tested on one real allocation decision before it is written into any entity |
| Intention is not impact | “Intention and impact are very widely different in our culture.” Two decades in extractives produced the renewables build-out — and with it cobalt demand and child labour in the supply chain | In build | A mandatory second-order-consequence section in every intervention memo, with a dissenting reviewer named |
| Pace: excitement without ground | A previous network of ours generated enormous energy and many simultaneous actions with no ground to hold them. “Patient enough, silent enough to understand the shape of its impact” | Proposed | Two parallel tracks, explicitly: a mission council acting now, and a slower strategy track building the foundation |
| Sequencing of people | Not everyone enters at the same depth or the same time, and pretending otherwise creates quiet resentment. Anyone can walk away at any point | Proposed | Clear agreements negotiated relationship by relationship, on a stated sequence, rather than one blanket membership |
| Minimum viable context | How much shared context is required before collaborative sense-making is even possible — and whose ontology do we use to decide what counts as a focus area? | Gap | One shared ontology drafted and stress-tested against three live candidate interventions, not in the abstract |
| The relational layer | “We have no map for togetherness.” Founders sit at the tip of the spear and carry sole decision-making weight — a faithful reproduction of exactly what went wrong at scale | In build | Rupture-and-repair practice built into the operating rhythm, not offered as an optional retreat module |
Several people here have been in rooms that looked exactly like this one and dissolved. The difference, if there is one, is whether the hard questions were named early or late.
Work continues inside existing organisations. But three items are prerequisites, not parallel tasks: contribution terms, decision rights, software ownership.
Read this page before the plan in Part V. If an item here would stop you committing, say which one — that is more useful to us than agreement.
Dashed lines mean contribution and coordination. They do not mean ownership, control, or an agreed commercial relationship. Solid lines mean a confirmed, documented arrangement — and there are currently only two of them.
Reported to us, not verified: that Better World owns the asset manager, and that the foundation holds a mission lock. We have seen neither the share register nor the governing instrument, and placement on the map above confers no right on anyone. Everything else is a working relationship without a signed basis. Collective ownership is the stated intent of all parties, and all parties also want it designed rather than defaulted into — informality is a current fact, not a governance philosophy.
What you should ask us: which of these lines will be solid in ninety days, and which will still be dashed in a year?
| Founder support at scale | An exited-founder network with diligence and operating support, offered as the first joint MVP to take to market |
| Narrative and distribution | A top-ten media and PR practice focused on conscious founders, plus a production company and a conscious broker-dealer licence in progress |
| The human layer | A coherence institute with a trained coach network, embodied-leadership and nervous-system capacity for families and operators |
| Incentive redesign | An eight-year protocol for redirecting currency flows that does not depend on altruism to work |
| Trusted family access | Long-standing committee and network relationships with family offices already pursuing transformational investment |
| Physical hubs | Two consciously built venues offered at no cost to aligned missions, plus a private ranch and a residence available for gatherings |
Six workstreams, four quarterly gates, a ninety-day pilot, eight roles, one costed envelope, and four funding routes kept deliberately separate.
The quarters are planning choices, not dates that have been agreed with all parties. The costs are author-built assumptions, not quoted salaries. Both are published in this form so that they can be argued with rather than accepted.
Each workstream shares context with the others but keeps distinct responsibilities and decision rights. Each has a stated evidence of progress — the observable thing that would show it is working — rather than a milestone that can be met by activity alone.
A documented service, clear access, a named relationship lead, a case walkthrough and a feedback loop.
Evidence of progressFamilies can understand the service and its constraints, and the team can deliver it without one person.
Owner · Relationship leads, to confirm
A lead plus analyst capacity. Map the selected interventions, find missing evidence, convene expert review, produce decision briefs.
Evidence of progressEvery proposed priority has a causal account, sources, alternatives and a named decision owner.
Owner · Intelligence lead to appoint; thesis contribution proposed
A coherent onboarding route. Link a question asked on a call to the later answer, decision and follow-up. Test on ourselves first.
Evidence of progressA user moves from a meeting question to evidence, a reviewed answer and an owned next action.
Owner · Product / engineering lead, to confirm
Confirm the asset manager's actual permissions and roles, review options with specialists, document costs and fee terms.
Evidence of progressA named entity and named authorised people own each decision. Reporting and client terms are written down.
Owner · Asset-manager leadership and compliance
A small number of cases. Examine incentives and conditionals. Compare four routes: partner, acquire, build, or defer.
Evidence of progressThe first case has evidence, delivery capacity, aligned terms and a reviewable decision — or is deliberately deferred.
Owner · Case leads and domain experts, to appoint
Track delivery effort, recurring costs, service revenue and case outcomes. Keep grants, fees and repayable experiment money separate.
Evidence of progressThe team can state cost to serve, cash needs, and which activities still require subsidy.
Owner · Programme lead and finance lead, to appoint
Note on ownership · Three of these six have no confirmed lead. We have left the gaps visible rather than filling them with names that have not agreed. Workstreams 01–03 are staffed first; 04–06 follow once the pilot demonstrates the actual workload.
Only Q1 is costed, and only Q1 is being asked for. Q2–Q4 are planning hypotheses Everything after the first gate is shown so you can see the shape of the thinking, not because it has been agreed, priced or committed. Treat Q2–Q4 as an appendix to the 90-day plan on page 39. Two tracks run at once, deliberately — a mission council for work that can start now under existing organisations, and a slower strategy track, externally facilitated, taking roughly six months to build a foundation able to hold the weight.
Appoint the core leads, agree scope and permissions, onboard the group, select the first case, and run one full question → evidence → decision → follow-up cycle.
Review gateDemonstrate an end-to-end case, identify who can decide, and show the actual effort and cost — not the estimate.
Run a small agreed set of cases, improve onboarding and diligence, test the research service, and decide whether the proposed US relationship model is actually useful.
Review gateDemand, delivery quality, cost to serve and required permissions justify the next hires. If they do not, we do not hire.
Extend proven workflows across participating families or SMEs. Test one bounded replication route and a reviewed conditionals framework.
Review gateReplicated value and operational capacity are evidenced. No automatic expansion from the size of the addressable market.
Compare realised costs against income, service outcomes, project consequences and failed assumptions. Refine governance. Build the next plan from that evidence.
Review gateContinue, change or close each experiment — with a documented reason, published to everyone who funded it.
This distinction decides what ninety days can honestly claim. Complete at day 90 means: one case taken from question to a documented, inspectable decision, with the real effort and cost recorded. It does not mean capital deployed, a project under way, or a measured outcome — those take years, and page 43 shows an eighteen-month stop condition for a reason. Work starts once scope, budget and authorised responsibilities are agreed, not before.
Named platform routes, a written service description, an entity and role map, one guided case, versioned sources and explicit permissions.
Closes the gap between “shown on screen” and “used by us”.
Selected calls produce linked questions, partial answers, decisions and follow-ups. A source correction can reopen a conclusion.
This is a proposed product extension, not a feature this document certifies.
One case whose causal claims, economics, incentives, affected stakeholders, risks and decisive missing evidence can all be inspected by anyone in the group.
The first real test of Gate 1 and Gate 3 from Part III.
Actual staff effort, actual costs, real service demand and any unresolved obligations — supporting the next decision.
No invented outcome or return target fills a data gap.
Four artefacts, one case carried to a documented decision, honest cost data, and the right to stop with no further obligation.
Proceed to a defined next scope · Redesign and re-run · Wait, with the reason recorded · Stop, keeping the artefacts. Acceptance is the quality of the decision record, judged by the family, not the attractiveness of the conclusion.
Roughly one half-day session a month plus two reviews, one named family contact, and access to whatever information the case actually requires. Where oversight of regulated activity is needed it comes from the existing authorised entity — no new hire is assumed or agreed.
Four roles are marked First and four After pilot; the ninety-day pilot assumes roughly four people, not eight. Existing staff, internal transfers, fractional work and location could all change this. Legal, compliance, finance, design and domain expertise are held deliberately outside this list as separately costed specialist support, so that the core stays small enough to stop.
| FTE | Role | What they own | Hiring phase | Why not later |
|---|---|---|---|---|
| 1 | Programme and service lead | Delivery, dependencies, service design and coordination across all six workstreams | First | Without this role, every question returns to one person — the stated bottleneck |
| 1 | Intelligence lead | The research process, expert review, and source-backed decision briefs | First | Gate 1 cannot be enforced by anyone who also owns the deal |
| 2 | Product and integration engineers | Connect existing workflows, improve reliability, implement the live intelligence features | First | The coordination record is the pilot's main deliverable |
| 2 | Systems and evidence analysts | Map selected cases, check claims, compare explanations, maintain the evidence base | After pilot | Added only when the pilot has demonstrated the actual workload |
| 1 | Diligence / capital-operations lead | Case analysis with authorised specialists, operating responsibilities, documentation | After pilot | Needed before the first transaction, not before the first case |
| 1 | Family relationship / onboarding lead | The agreed family journey, and testing the proposed North American route | After pilot | Depends on Decision 3: what the US scope actually is |
A twenty-person shared intelligence function was described in the source conversation as a longer-term possibility. It is not an agreed first-year headcount, and this plan deliberately starts at eight.
One senior operating role has a possible candidate in mind. No approach, offer or terms exist, and the person is not identified here. Any engagement would start as a bounded trial, not a full-time contract.
Several highly capable people around this work are better placed as fractional advisers and ecosystem builders than as full-time executives. Saying that clearly is kinder than discovering it after a hire.
A subscription, a management fee, a grant and an investment are four different things, and blending them is how mission drift begins. Every figure below is an author-built planning assumption — not a quoted price, an agreed budget, secured funding or a forecast.
| Line | Basis | Amount |
|---|---|---|
| People | 8 FTE × 12 months × €7,500 loaded | €720,000 |
| Specialist support | Legal, compliance, finance, domain | €90,000 |
| Infrastructure | Platform, servers, tooling | €48,000 |
| Pilots and convening | Case work, travel, family sessions | €46,000 |
| Subtotal | Before contingency | €904,000 |
| Contingency | 10% of subtotal, stated not buried | €90,400 |
| Steady-state annual cost | If all eight are in post all year | €994,400 |
At 3 → 5 → 7 → 8 people by quarter, people cost falls to €517,500 and the year lands near €772,000. The €994,400 figure is the run-rate, not the 2027 cash need.
Roughly €170,500: four people over three months, weighted legal set-up, one case, a quarter of infrastructure, 10% contingency. The only number tied to a defined deliverable.
What is excluded, and why that matters. Baseline payroll at partner organisations; contributed time from the current team; employer taxes beyond the assumed loading; location differentials; working capital; and all project investment capital. The €5,600 of headroom against a round million is arithmetic, not evidence that the exclusions are affordable. Figures are euro-denominated; a dollar or sterling version requires rebuilding the local cost base, not applying a rate.
| Request | To whom, for what | Instrument | Amount | Approval needed |
|---|---|---|---|---|
| 90-day pilot | Third Attractor, to run one case end to end | Service purchase or restricted grant — not decided | ~€170,500 illustrative | The family. Scope and stop conditions agreed first |
| 2027 operating envelope | Third Attractor, to staff the function for a year | Operating grant and/or service revenue | €772k phased / €994k run-rate | Multiple contributors. No single family is asked for this |
| Platform seed | A joint venture that does not yet exist, for shared software | Equity or convertible — structure undecided, five options live (p.34) | €1–3m discussed | Gated on business model and ownership. Nothing accepted until settled |
| Catalyst Pool | A pre-investment pool, funding work that makes projects fundable | Recoverable grant / catalytic capital (p.26) | No amount proposed | Allocation committee, contributor veto, ceiling, wind-down date |
The pilot is severable: buy it and stop. The envelope needs more than one family. The seed is blocked until the structure question is answered. The pool is independent. They are not stages of one ask, and should not be read as a ladder.
Client assets are not company revenue. A seed cheque is not committed capital. This page is not an offer of securities or a solicitation.
Five qualities determine whether capital moves well: credibility, credible relationships, integrity, character and competence. Four of those are about who we are. The fifth — competence — is the only one that can be bought, and it is the one that matters least if the others fail.
| Activity | How we are paid | The conflict this creates | How it is controlled |
|---|---|---|---|
| Field maps and leverage notes Stages 01–02 |
Cohort-funded or grant-funded. Shared across families at no marginal cost | Low. There is a mild incentive to make the field look more urgent than it is | Sources attached, uncertainty named, and the same map goes to every family including the ones who decline |
| Diligence and structuring Stages 04–05 |
Fee for work done, disclosed in advance, not contingent on the deal closing | Moderate. Fee-for-work still rewards activity over restraint | We are paid the same whether our recommendation is proceed, redesign, or stop. A declined case is a completed piece of work |
| Regulated asset management Via licensed partners |
Management fee on billable assets, via the authorised entity | High. Asset-based fees reward gathering assets, whether or not deployment is wise | Not our decision to take. Regulated authority sits with named authorised individuals, and the compass veto sits outside the fee line |
| Software and coordination | Subscription, reported at €180,000 per family per year | Moderate. Subscription income rewards retention, which can reward comfort over honesty | Published service specification, and the uncomfortable version of the quarterly report goes to subscribers first |
| Venture building | Equity in what we help create, alongside other founders | High. Equity rewards building rather than finding something that already works | Build is the last of three options, after find and strengthen. Every build decision records why the first two were rejected |
| Foundation grant-making | Not paid. No fee is earned on the foundation's own grants | No fee. Residual conflicts are reputational — a grant that reflects well on us, or on a family we also serve commercially | Structural, not discretionary. This is what the membrane on page 27 is for |
Which is why the conflicts are printed here, in the plan, not disclosed later.
This is a construction, not a case study, and it is marked as such on every row. It exists because the sequence in Part III is abstract until you see it carry one concern all the way through to a number, a name, and a date.
Illustrative No project named, no capital committed
| Stage | What happens in this example | Artefact | Gate, and who holds it |
|---|---|---|---|
| Concern | A family says: “if the supply chain into our region fails for three weeks, what actually happens to people here?” | A question, written down and dated | — |
| 01 See | We map the region's actual food system: where calories come from, days of buffer, load-bearing single points of failure, and what the evidence does not support | Field map, sources attached, three claims marked as unsupported | Gate 1 · Truth Research lead |
| 02 Locate | Not production or capital. No entity holds responsibility for regional buffer capacity, and procurement rewards the lowest-cost distant supplier | Leverage note. Depth assessed at Level 3 — rules and ownership | — |
| 03 Choose | The family decides this is theirs to carry, and names which family member holds it | Written priority, signed by the family | Gate 2 · Fit The family |
| 04 Explore | Diligence on two routes: a procurement-standard change, or a producer-owned buffer co-operative. A red team argues that both fail without a named operator, and that the co-operative risks capture by the largest producer | Diligence pack with the case against stated first | Gate 3 · Quality Any compass reviewer |
| 05 Assemble | Catalytic money funds the baseline study, the legal structure and the first operator. Procurement reform is pursued through convening, not capital. A stop condition is set: if buffer days do not improve within eighteen months, this ends | Grant memo and term sheet. Six MVS components confirmed in writing | Gate 4 · Readiness Structuring lead |
| 06 Act | The operator signs. The co-operative is incorporated with ownership held locally. The family's capital is one of three sources and not the largest | Signed instrument and operating plan | Gate 5 · Authority Family and affected parties |
| 07 Learn | Buffer days are measured against the published baseline every quarter. The procurement route stalls, and that is reported as a failure | Quarterly evidence report, published to participants | Feeds back into the field map |
The constraint was not money. The right depth was rules and ownership. Two instruments were used, one of which was not capital at all.
That we have done this. We have not. Page 33 is explicit that a completed, measured intervention is our central missing piece of evidence.
Because the first ninety days exist to replace this page with a real one — and you should be able to see, in advance, exactly what shape that real one will take.
What success looks like, the six things we must be willing to say out loud, the decisions that are actually required, and what seven families told us was wrong with the last version of this document.
A document that makes refusal difficult is not a plan, it is a pitch. The pages that follow are designed to make it easy to say no, and specific enough that a no tells us something useful.
Intentions are the cheapest thing in this field. So each horizon below is written as a condition someone outside this group could check, and each carries the result that would tell us we were wrong.
Ninety days produce documents but no case, or the cost data is unavailable.
Delivery still depends on one or two individuals, or no failure has been reported.
Our own institution has become the thing being preserved.
These are not values. Values are free. These are specific, uncomfortable sentences, and each one has a cost attached. The test of this document is whether you hear them from us in two years without having to ask.
In the first paragraph of the quarterly report, not the appendix. Cost: looking less competent than our peers who do not publish this.
Said at the point of recommendation, not afterwards. Cost: revenue we could have earned by staying quiet.
Any compass reviewer can say this unilaterally, and a strong financial case cannot offset it. Cost: declining deals that would have made the portfolio look better.
No invented outcome or return target fills a data gap. Cost: appearing less prepared than people who fill gaps with confidence.
Recorded as a decision with a review date, not quietly dropped. Cost: capital sitting undeployed while we are judged on deployment.
Including handing on work we started and would prefer to keep. Cost: our own institutional importance.
Gates 1, 3 and 4 are ours and can be audited. Gates 2 and 5 are not ours at all. No success fee or carry attaches to a family's first commitment, and the quarterly report goes to everyone who funded the work, in the same version. The compass veto is a design intention, not a documented legal right — no instrument exists yet. Treat it as a commitment we are making, not a protection you hold.
Three belong to the partner group and must be settled before a family is asked for anything; three belong to a family. Each is bounded and each has a holder — but decisions 2 and 3 require legal and compliance input and will not be settled in a single meeting. They can, however, be scoped in one.
| Decision | The actual question | Who holds it | Consequence of not deciding |
|---|---|---|---|
| 1 · First collaboration scope | Which one or two workflows demonstrate value first? The family platform, the intelligence function and the thesis all attracted interest; the focus was never chosen | Partner group | Effort spreads across three fronts and none reaches an evidenced result |
| 2 · Governance and authority | Who owns the operating plan, controls the budget, and makes each regulated or project decision? Shared analysis is not delegated investment authority | Partner group | Capital arrives before ownership is resolved — the most common failure mode in this field |
| 3 · Joint structure and geography | Five live options and no terms agreed on any of them: partner, joint venture, new company, merger, or open-source the stack and run it ourselves. A sixth option was raised on a call — a nucleus into which each organisation contributes 10–20% of itself and receives a share of every other. It is an idea, not a proposal anyone has worked up | Partner group | The collaboration continues on trust, which works until it does not, and cannot be signed against |
| 4 · Funding instrument and recovery | Is a given contribution a service purchase, a grant, an operating investment, or experiment capital? If repayment is expected, what realistic mechanism returns it? | The family | The question that most often kills a commitment late, after goodwill has been spent |
| 5 · People and costs | Which capacity can existing teams provide, which roles are genuinely incremental, and what are the real loaded rates? | Partner group, with the family | The €994,400 figure stays an illustration and cannot become an ask |
| 6 · Intervention thesis and safeguards | Which cases meet the causal and incentive tests? Which conditionals are actually enforceable, who bears the risk, and how can an affected party challenge a decision? | The family, with the research partners | We invest at the depth that is easy rather than the depth that holds |
The only ask tied to a defined deliverable: four artefacts, one case carried to a documented decision, honest cost data, and the right to stop. It completes a decision process, not an intervention. Severable from everything else on page 41.
Pre-investment capital of this kind exists elsewhere; what is missing is a version the families here control. Small, legible, useful to every other family in the cohort — and entirely unagreed. Terms on page 26.
Ask them to find a ninth gap. Send it to us. That costs you nothing, commits you to nothing, and is genuinely the most valuable thing you could do for this work this quarter.
This is a simulated editorial exercise, not feedback from actual families and not an endorsement by anyone. Seven composite perspectives were constructed, each from a distinct kind of owning family and a distinct professional reflex, and the draft was read against them. The challenges below are invented tests, not quotations. We publish them so a reader can see what we were testing for and judge whether we responded.
| Perspective | The challenge the perspective was built to raise | What changed in response |
|---|---|---|
| Third-generation industrial family Europe · operating company roots |
Eighteen pages and I could not find a single number, date or name. We have run factories for sixty years. Show me the plan, the cost and the person responsible. | Parts IV and V are now almost entirely figures, dates and owners: the readiness matrix (p.32), the 2027 sequence and gates (p.38), the ninety-day pipeline (p.39), eight named roles (p.40) and a costed envelope (p.41) |
| Next generation US · two siblings, thirties |
This is beautifully written and I could not explain it to my father in five minutes. Where is the one page I can hand across the table? | Page 3 is now the whole thesis on one page — vision, mission, why, how, evidence and ask with a single master diagram. Page 2 explains how to read the status conventions before anything is claimed |
| Single-family office CIO Fiduciary, professional |
No fee schedule. No conflicts disclosure. No regulated perimeter. I cannot take this to an investment committee however good the thinking is. | Page 30 sets out the regulated perimeter with every figure flagged for verification. Page 42 publishes how we are paid, the conflict each route creates, and how it is controlled — including the three rated high |
| Philanthropic foundation family Nordic · grant-making tradition |
You blur philanthropy and investment. Which is which, who decides, and how do you stop charitable money subsidising someone's margin? | Page 27 introduces the membrane with explicit lists of what crosses and what does not. Page 25 matches instrument to constraint. Page 41 keeps the four funding routes separate |
| Trading-house family Gulf · generational commodity base |
Your only example was hypothetical. Where is the proof? One case you have actually done, measured, with the result published. | We could not produce one, so we said so. Page 33 names it as the central gap; the example on page 43 is labelled illustrative on every row; the first ninety days exist to replace it |
| Technology-exit family US west coast · operator instincts |
I understand the worldview. I do not understand the mechanism, and I especially do not understand what you are refusing to do. | Part III is now a seven-stage route with five named gates, what disqualifies at each, and who can say no (p.19, p.24). Page 13 states what Third Attractor is not. “Wait” is a formal outcome at every gate, not a failure to pass one |
| Multigenerational stewardship family Asia · century horizon |
Who decides? And if we want to stop, what happens to the work, the people and our name? | Gates 2 and 5 are explicitly not ours — they belong to the family and to affected parties. A family can exit after any stage, keeping the artefacts. Every project carries a written stop condition |
The previous draft described intentions in the grammar of accomplishments — everything written as though it had already happened. The fix was not softer language but a convention: every claim now carries a status and a statement of who checked it, on separate axes, so that “real” and “proven” can never be quietly conflated. The next correction is to replace simulated readers with real ones, with their consent.
Two of the simulated perspectives argued for cutting the philosophy. We kept it: the reason this work can be trusted at depth is the reasoning underneath it, and a document of numbers alone would be indistinguishable from the funds we were founded to improve on. We moved it earlier, made it shorter, and put the numbers where they can be found without reading it. A second round of fifteen simulated perspectives produced the evidence and attribution corrections carried through this version.
This is a working document prepared for discussion with families and partners. It is not an offer, a solicitation, a prospectus, or financial advice. It does not constitute a commitment by any party named in it. Figures described as illustrative are author-built planning assumptions and are not quoted prices, agreed budgets, secured funding or forecasts.
Organisations and individuals are named on the basis that consent has been sought, and described with accuracy, including where that accuracy is unflattering. This is our undertaking, not evidence that consent was given — written confirmations are being collected and are not yet complete. Any party who believes they have been described inaccurately should tell us, and the correction will appear in the next version with their wording.
A ninth entry for the gap register on page 33, with your adviser's name against it.
A claim on pages 29–31 you think we have overstated. We will either evidence it or withdraw it.
One of the four candidate priorities on page 16 that you would want to carry yourself.
Entirely welcome, and more useful to us than a polite maybe — especially with the reason attached.
Mission into tangible action · v6 · November 2026 · thirdattractor.com · Flourishing Futures Foundation
Next review: end of the first ninety days, with the readiness matrix reissued and the changes marked.