Decision Methods

Ask a community how it decides things and you will usually get one word back. Consensus, or consent, or “we vote”. It is almost never true, and the gap is not dishonesty — it is that a group notices its formal method and stops noticing the four informal ones running alongside it.

The useful question is not which method is best. It is which decision gets which method, and whether anyone wrote that down.

Producing a picture, or carrying a motion

Most confusion about decision methods comes from putting them all on one shelf. They do two different jobs.

Some methods produce a picture. They tell a group where it stands, or which of several options it prefers. Nothing is adopted at the end, because there was nothing on the table to adopt — no proposal, no yes, no no. These are for the stage before a decision, and using them well is the reason some communities’ proposals pass easily: the disagreement was found and resolved before anything was drafted.

Others carry a motion. There is a written proposal, and at the end it is adopted or it is not. What separates them is what counts as a no — a shortfall against a threshold, an objection, or an objection that somebody has judged serious enough to stand.

A group that reaches for the wrong shelf produces a familiar mess: a poll on a question that needed a decision, so nothing is settled; or a binding vote on something the group had never discussed, so the losing half feels ambushed.

The three that produce a picture

A poll asks members to pick one option and reports how the group split. Which weekend for the work party. Which of three names. It is the most under-used method in community practice, because groups reach straight for a binding vote on questions where nothing needed to bind.

A multi-question poll puts several linked questions out at once and tallies each on its own — visitors should join work hours, there should be a maximum stay, hosts are responsible for their guests — before anybody drafts a visitor policy. What comes back is a map of where agreement already exists and where the real argument is. It is the single cheapest way to stop a proposal that was doomed from the first line. VoteCast calls this Common Ground; the practice has no settled name in the field.

A ranked vote Ranked-choice voting Members rank the options; last place is eliminated and its votes transfer until one option holds a majority. elects one option from several without splitting the vote. Five candidates for facilitator, and the winner is not someone twenty-two per cent wanted and everyone else could not stand.

The three that carry a motion

A threshold vote puts a written motion to the group and adopts it if approvals clear an agreed share — more than half, or a supermajority Supermajority A pass threshold above half, usually two-thirds, used where a bare majority is too thin a basis. for the weightier things. It is dismissed too quickly in community circles. It is fast, it is legible, and for reversible operational business it is usually correct. Its cost is structural: it manufactures a losing minority every time, and a minority that loses repeatedly stops attending.

Consensus Consensus A decision method requiring the agreement of all participants. Powerful for trust, slow under scale and fatigue. asks whether anyone objects, and expects the proposal to change until nobody does. It produces buy-in nothing else matches, and the amendment loop is a genuinely good teacher — a group that has consensed on something carries it out because it owns it, not because it passed. Past a dozen people it begins handing a veto to whoever is most tired or most stubborn. A stand-aside Stand aside Recording that you disagree with a decision without stopping it. The release valve that makes consensus survivable at any size. convention, letting someone disagree on the record without stopping the group, is what keeps it survivable.

Consent Consent A decision passes when no one has a reasoned, paramount objection. Not the same as everyone thinking it is the best idea. narrows the question again: not do you object? but do you have a paramount objection Paramount objection Under consent, an objection serious enough to stop a decision — a named harm to the group's aim, not a preference. — a named harm to the group or its aim, judged serious enough to stand. The bar moves from enthusiasm to tolerability, and the burden moves onto the objector. Its standard is that a proposal be good enough for now and safe enough to try, which assumes a review date and quietly breaks without one.

The pair people most often confuse has a page of its own: consensus vs consent. So does the one this section is really about: majority vote vs consensus.

The settings that quietly decide fairness

A method name settles less than groups expect. Four numbers settle more, and each is a trade with a real cost on both sides.

How many must take part. The quorum Quorum The minimum number of members who must take part before a result counts at all.. Set it high and one quiet month blocks everything; set it low and five people decide for forty.

What share passes it. More than half, or two-thirds. A supermajority protects a minority from a thin majority, and hands a determined third of the group a standing veto over any change.

When the tally becomes visible. Live results let people see where the group is heading. They also let people follow it rather than think — the reason a running count and a deliberative decision sit badly together.

What an absence means. Whether silence counts as agreement or is simply left out. Silence-as-consent lets a busy group decide anything, including things most of them never read.

None of these appear in the name of the method, and a community that has answered them differently from yours is running a different system under the same word.

No community needs only one

This is the part most groups arrive at the hard way, usually after an evening spent on tile colours.

Sort decisions by what they cost to get wrong. The RCOS Standard names three tiers Decision tiers Sorting decisions by what they cost to get wrong, and giving each class its own method and timeline., and they hold up:

• Operational — day-to-day business inside rules already agreed. The rota, the repair, the order. Fast, reversible, low ceremony. A threshold vote or a circle’s own consent, decided in a day.
• Strategic — direction and real money. A new building, a loan, a change in what the land is for. Careful: a supermajority, a proper deliberation window, a delay before it takes effect.
• Constitutional — the community’s purpose, its agreements, its membership rules, and how it decides. Slow on purpose: consensus, a long alignment phase, and a window in which a late objection still counts.

One rule makes the scheme work rather than leak: a decision that cannot be clearly classified goes to the higher tier. Without it, every borderline matter drifts downward, because the lower tier is always the easier meeting.

VoteCast ships this as presets — an operational preset at a two-thirds bar with a day of deliberation, a constitutional preset running consensus with a week of deliberation and a two-week objection window. The specific numbers are one community’s answer, not the answer. The sorting is the transferable part.

Which tier is this decision?

Operational decisions are day-to-day business inside rules already agreed. Strategic decisions commit real money or change direction. Constitutional decisions change the community’s purpose, its agreements, its membership rules, or how it decides. Five decisions — which tier does each belong to?

1. Replace the failed greenhouse pump — €340, from the maintenance budget agreed in March.

Operational correct answer Right. It spends money the group has already allocated, for a purpose already agreed, and it is entirely reversible. This is exactly what a maintenance circle’s domain is for — and if it still reaches a general meeting, that domain was never real. Strategic The amount feels significant, but the test is not size — it is whether the decision commits the community to something new. This one operates inside a budget the group already approved. Constitutional Nothing here touches the community’s purpose, agreements or membership. Treating routine spending this way is how a group ends up with four-hour meetings about pumps.

2. Borrow €120,000 to convert the barn into four more homes.

Operational No amount of routine framing makes a twenty-year debt operational. The test that catches this: how long would it take to undo, and what would undoing it cost? Strategic correct answer Right. It commits significant resource and changes the community’s direction, but it does not change the rules the community lives by. A supermajority, a real deliberation window, and a delay before it takes effect all earn their cost here. Constitutional Close, and worth arguing. If four more homes would take the community past a size its agreements were written for, it becomes constitutional too — which is a good illustration of why the escalation rule exists.

3. Lower the quorum for budget votes from two-thirds of members to half, because meetings keep failing to reach it.

Operational This is the trap. It arrives as housekeeping — a number that is not working, adjusted. But the number is part of how the community decides, which puts it in a different category entirely. Strategic It commits no money and changes no direction. What it changes is the machinery, and that is a category of its own. Constitutional correct answer Right. Any change to how the group decides is constitutional, however administrative it looks — and it should be decided under the *old* rule, not the new one. The underlying problem is real, and the honest fix is usually to lower the quorum for operational business while leaving it high where it matters.

4. Accept a neighbouring farmer’s offer to graze sheep on the lower field for three years. No money changes hands.

Operational The absence of money is doing a lot of work here. A three-year commitment of land is not reversible next season, and "it costs nothing" is not the same as "it commits nothing". Strategic correct answer Right. It commits a significant shared asset for years and forecloses other uses of that field — a direction decision wearing the clothes of a favour. Resource does not only mean cash. Constitutional Defensible if the community’s stated purpose says something specific about how its land is used — in which case this is testing that purpose and belongs higher. Otherwise it is a significant commitment inside existing aims.

5. Let a member’s adult son stay in their spare room for six months while he looks for work.

Operational It is operational *if* the group has a written guest policy that covers a six-month stay. Most groups do not, and answering it case by case is how a community decides its membership rules by precedent without ever noticing. Strategic No resource is being committed. What is at stake is who lives here and on what basis — a question about the rules rather than about the plan. Constitutional correct answer Right, in the absence of a written rule. Anything touching who lives here and for how long is membership territory, and an unclassifiable decision defaults to the higher tier. The real answer is to decide it once, properly, as a guest policy — and then every case after it genuinely is operational.

What breaks

The method is adopted in name only. A group says “consent” without defining grounds for objection, so it is still unanimity. It says “consensus” and runs something closer to consent. Both are common, and both mean years of arguing about a rule nobody wrote down.

One method for everything. Whichever one it is, it fails at one end. Consensus everywhere exhausts the people with jobs and children until only the retired are left deciding. Majority everywhere carries constitutional change on a margin of one and creates a permanent minority in decisions about the community’s own identity.

The numbers were set after the vote. A threshold or quorum decided once the result is visible is not a rule, and everyone present will remember that it was not.

Sources & further reading

1. 1 RCOS Core — Layer 2: Governance & Decision Logic — §4.1–4.2: the three decision types and the mechanisms available to each
2. 2 VoteCast — six ways to decide — where EcoHubs implements these methods, with its own names for two of them
3. 3 Buchanan, J. & Tullock, G. — The Calculus of Consent (1962) — why the best decision rule differs by class of decision
4. 4 Beyond the Vote: Why the Most Resilient Communities are Replacing Consensus with Consent — the longer EcoHubs argument
5. 5 Seeds for Change — free, practical consensus and facilitation guides

Written by EcoHubs members, with AI assistance for drafting and editing, and reviewed by a person before publication. Facts are checked against the sources listed; anything we could not verify is marked. How this is written