Purpose

Every group that coordinates over time must answer one question: how do we decide?

The answer shapes everything, who feels safe, who burns out, who stays, who leaves, how fast the group moves, and whether it survives at all. Most groups inherit a decision model without choosing it. This document compares the four main models on the factors that actually determine whether a group thrives or falls apart, including hard numbers from research where they exist.

Efficiency vs. Effectiveness, The Core Distinction

Before the comparison, this distinction matters:

Efficiency = how fast you move. Effectiveness = how well you reach the actual goal.

A group can be highly efficient, fast decisions, clear authority, and deeply ineffective if those decisions don't serve people's actual purpose, erode trust, or produce animosity that slows everything down compoundingly. Running fast in the wrong direction is not effectiveness.

Coercion is the extreme case: a coerced person is efficient in serving someone else's goal and completely ineffective in serving their own. In a community setting, where no paycheck compensates, a coerced person simply leaves, or stays and corrodes the group from within.

The Four Models

Real Numbers from Research

Community failure rates (Diana Leafe Christian, Creating a Life Together, 2003)

• 90% of intentional communities fail before getting off the ground or in the first years
• The successful 10% all did the same five or six things right, shared vision documents, explicit agreements, clear governance, conflict repair processes, and decision models that don't override people
• Majority voting in communities: produces persistent minority resentment and is a primary driver of dissolution
• Hierarchy in communities: essentially non-functional, no paycheck means no compliance mechanism

Psychological safety (Google Project Aristotle, 2012–2016, 180+ teams)

Psychological safety, the degree to which people feel safe to speak up, disagree, take risks, and be honest, is the #1 predictor of team performance:

• <cite index="28-1">Psychological safety was correlated with 43% of the variance in team performance. Teams with high psychological safety showed 19% higher productivity, 31% more innovation, 27% lower turnover rates, and 3.6x more engagement.
• <cite index="29-1">High-performing teams were those where members felt comfortable expressing thoughts and ideas openly, leading to more productive discussions and innovative solutions.
• Key additional finding: high-performing teams had equal conversational turn-taking, all members had equal opportunity to speak. This directly maps to what ACT requires structurally.

Majority vote vs. consensus decision quality

• <cite index="10-1">A laboratory study found that social context and decision strategy affected decision acceptance, understanding, decision time, and affective reactions, though decision quality alone was not the differentiator, implementation and buy-in were.
• <cite index="7-1">Voting silences voices. Listening deeply to people in your group leads to more robust and better decisions. Formal consensus decision-making leads to broader engagement.
• <cite index="15-1">High quality results reside in a minority of responses and are often not identified by majority voting, the minority that knows the most gets overridden.
• <cite index="3-1">When group members view the status quo as unjustly difficult to change, they may begin to delegate decision-making to smaller committees or act unilaterally because they are frustrated with a stagnated group process. This is consensus degrading into informal hierarchy.

20-Factor Decision Model Matrix

Scores: 0–100. Higher = better (except Decision Load and Rework Rate, where lower = better). Sources: cited research above + OmniOne canon + Diana Leafe Christian + Psychological Matrix CIK-LIB-01 + structured design estimates.

Why Each Model Fails in Communities

Hierarchy

Nobody signs up to be controlled in their own home without a paycheck. Economic coercion, needing money to survive, makes people tolerate hierarchy at work. Remove the economic compulsion and the model collapses immediately. Even where it holds, it produces the lowest purpose-alignment of any model. The person at the top fulfills their goals; everyone else is efficient in service of those goals while their own remain unmet.

Psychological matrix risk (CIK-LIB-01): High authority shortcut (Pattern 5), dependency framing (Pattern 17), implied threat (Pattern 19). The entire decision environment is coercive by design.

Real number: Not a viable model for voluntary communities. Christian's research found zero successful long-running communities using pure founder/owner hierarchy. The model requires coercive compliance mechanisms to function, which communities don't have.

Majority Vote

Voting looks democratic but creates chronic losers. In a long-running community, the same minority can be overridden across years and decades. Someone may invest ten years building something and then have it changed by a vote they disagreed with. <cite index="19-1">Majority rule may need only 51% of votes to pass a measure and this can lead to a divided society, promoting competition and dominance instead of cooperation and equality.

The animosity this generates is not temporary, it compounds. Voting also fails because high-quality insight is often concentrated in the minority: the person who knows the most about a problem may be on the losing side of the vote.

Not Teal Paradigm aligned: Voting imposes on individual purpose, the minority's direction is overridden regardless of their genuine concerns. If a group member cannot be fulfilled in the direction chosen, they are being coerced by the majority.

Real number: <cite index="21-1">90% of communities fail, and majority voting is one of the primary identified failure modes in Christian's research across dozens of communities.

Consensus

Consensus is the closest conventional alternative and the only other model that sometimes works in communities. It fully protects the minority and tends to produce high trust, high implementation rates, and high buy-in when it works.

Its fundamental flaw: it can stall indefinitely. When full agreement cannot be reached, the system freezes. In practice, consensus communities often degrade, <cite index="3-1">group members begin to delegate decision-making to smaller committees or act unilaterally because they are frustrated with a stagnated group process. This means the model quietly reverts to hierarchy or minority rule by whoever has most energy.

Consensus also produces extremely high decision load, every new member, every changed circumstance, potentially reopens settled questions. There is no architecture for "already decided."

Real data: Diana Leafe Christian, after years of advocating consensus, moved away from it, her documented reason: communities using pure consensus-with-unanimity frequently stalled or allowed single members to hold the group hostage indefinitely. She subsequently advocated for sociocracy, which shares key principles with ACT.

ACT, Solutionary Consent

ACT solves both core failures of consensus:

1. Cannot stall, Round 2 and the Value-Based Decision Model provide resolution pathways without majority override
2. Disagreement is required to be generative, you must bring an alternative solution, not just an objection

ACT also dramatically reduces ongoing decision load because decisions connect to GenPlans. Once the foundational plan is established, most subsequent choices are already bounded. Teams add to plans rather than re-deciding everything from scratch.

Not Teal Paradigm aligned, wait, it is: ACT is the only Teal-aligned decision model. It is the only model where no one can unilaterally override anyone else, disagreement is made productive, and individual purpose is structurally protected.

The Decision Load Advantage

This is ACT's most underappreciated strength. The upfront cost, building GenPlans and foundational agreements, is real. But once in place:

Estimated decision elimination rate in mature ACT system: 90–99% of decisions that would otherwise require group deliberation are pre-resolved by the GenPlan and agreement architecture. Most "decisions" are really just role-level execution of what is already agreed.

Psychological Safety Comparison

Psychological safety, the belief that one can speak up, disagree, take risks, and be honest without fear of negative consequences, is <cite index="28-1">correlated with 43% of the variance in team performance (Google Project Aristotle).

Key insight: In hierarchy and voting, disagreement has social and sometimes material costs. This suppresses honest input, which degrades decision quality over time, even when single-decision speed appears high. ACT makes disagreement not just safe but required to be constructive. This is the rarest and most valuable property of any decision model.

How ACT Maps to Psychological Matrix Safeguards

ACT is structurally designed to counter the most dangerous influence patterns identified in the OmniOne Psychological Matrix (CIK-LIB-01):

Teal Paradigm Alignment

The Teal Paradigm (Frederic Laloux, Reinventing Organizations) is built on three properties: self-management, wholeness, and evolutionary purpose. All three require that individual purpose is not overridden by external authority.

ETHOS is the only organizational framework built to support Whole Living Systems alongside the Teal Paradigm, and ACT is the decision model that makes that possible. Without ACT, the framework cannot hold Teal alignment under pressure.

Summary Scorecard

Reference Sources

• Diana Leafe Christian, Creating a Life Together (2003), 90% community failure rate; failure mode analysis across dozens of communities; consensus-with-unanimity limitations
• Google Project Aristotle (2012–2016), psychological safety as #1 predictor of team performance; 43% variance correlation; 19% productivity, 31% innovation, 27% retention, 3.6x engagement
• Amy Edmondson, psychological safety research foundation
• Academy of Management Journal, consensus vs. majority vote: decision acceptance and implementation rates
• OmniOne Psychological Matrix CIK-LIB-01 (Notion: 357d14367d5f814680f3ccda79b887c8)
• OmniOne Full OS Comparison (vault: /system/comparisons/full-os-comparison.md)
• Frederic Laloux, Reinventing Organizations, Teal Paradigm framework

The numbers here are estimates, not measurements. They come from Decision Models's file projects/decision-models/act-decision-model-comparison.md, as it stood on 2026-09-16. Deeper analysis could change these numbers.