Sociocracy vs Holacracy — and why what matters is not the model

Sociocracy (S3): a living toolbox

Sociocracy 3.0 — or S3 — grew out of a simple observation: decision-making in organisations is often captured by a few, while the useful intelligence is distributed throughout the system. S3 offers a set of modular practices to remedy that, organised around a few founding principles.

At the heart of S3 are circles : semi-autonomous units with a purpose, authority over their domain, and linked to each other through delegation. Each circle can decide how it organises itself — within the limits of what the parent circle has entrusted to it. That fractal structure lets the organisation hold both global coherence and local autonomy.

S3 introduces the notion of roles rather than positions: a person can hold several roles in different circles, and a role is defined by its purpose and its domain — not by a static job description. Roles evolve with the circle's needs.

Decision-making in S3 rests on consent : a decision is adopted not when everyone approves it, but when nobody has areasoned objection — that is, an argument showing the proposal would cause harm to the circle or the organisation. That is a fundamental difference from consensus, which seeks everyone's agreement.

Finally, S3 is organised around tensions : any perceived gap between what is and what could be constitutes a tension — and every circle member is invited to process their tensions by proposing improvements. It is an organisation that learns continuously from the ground.

« S3 is not a system to install. It is a grammar — a set of patterns you learn to use according to context. Nobody asks you to adopt them all at once. »

Holacracy: a constitution, not a guide

Holacracy, codified by Brian Robertson in the 2000s and popularised by Zappos, takes a different route. Where S3 offers an open toolbox, holacracy offers a constitution — a legal, precise document defining exactly how the organisation functions, how decisions are made, how roles are created and changed.

Holacracy draws a radical distinction between roles and people : you are not your job, you are a person who temporarily holds one or more roles. Each role has a purpose, domains of authority, and explicit accountabilities. When you act in a role, you carry its authority — and nothing else.

Holacracy organises its meetings into two distinct types. The Governance Meeting deals with the structure of the organisation: creating, changing or removing roles and policies. The Tactical Meeting deals with day-to-day operations: project progress, synchronisation, operational tension. Those two streams — governance and operations — are strictly separated.

As in S3, holacracy relies on tensions as the engine of change. But the process for handling them is more formalised, more procedural. There is a meeting facilitator, a secretary, defined roles in every circle, and precise steps to follow.

The differences that actually matter

On paper the two models look close. In practice they imply very different organisational cultures. Here is an honest look at the dimensions that separate them.

One difference the table does not capture well: the relationship to leadership. In holacracy, the manager's authority is formally redistributed to roles — which can be experienced as a liberation or as a loss of bearings, depending on the person and the organisation's culture. In S3 the transition is less head-on: you introduce practices, observe what changes, adjust.

Why both can fail

I have worked with organisations that adopted one or the other model with all the goodwill in the world — and exhausted themselves doing it. The reasons for failure look alike, whichever model was chosen.

• Ideological adoption. « We have decided to do holacracy », says the director, as if ordering software. The model becomes an end in itself. You measure conformity to the model rather than the organisation's real effectiveness. People learn to speak the language without living its spirit.
• Jumping without cultural preparation. Participatory governance assumes people are able to speak up, to formulate objections, to hold their authority in a role. If the organisational culture is strongly hierarchical, or if people have learned to stay quiet, the tools do not work — they reproduce the same dynamic under a different veneer.
• Process overload. Some organisations end up spending more time governing than working. Governance meetings pile up. Roles proliferate. The holacratic constitution runs to more than twenty pages — and every sentence counts. It is not that the model is bad: it is that it was applied without discernment.
• Abandoning human leadership. Neither sociocracy nor holacracy removes the need for clear, embodied leadership able to hold a vision. Confusing participatory governance with the absence of leadership is a fatal error. Both models redistribute authority — they do not make it disappear.
• Forgetting the human factor. Tensions between people, things left unsaid, informal power dynamics, relational wounds — no governance model resolves those dimensions. Applying a rigorous model on unworked human ground is like planting seeds in concrete.

« Imposing a governance model like a straitjacket is as damaging as having no governance at all. In both cases the organisation cannot breathe. »

The real question: what does this organisation need?

Before choosing a model, I always start with a diagnostic phase. Not to tick boxes — but to understand the living system in front of me. These are the dimensions I explore.

This diagnosis is not a questionnaire. It is a conversation — often long, sometimes uncomfortable — with the people who live inside the organisation. It is from that understanding that you can choose what will genuinely serve.

Take what serves, adapt the rest, create what is missing

My approach is not to sell a model. It is to build, with the organisation, the governance that looks like it — at this precise moment in its history.

Here are concrete examples of what that looks like in practice:

• Keep consent, drop the weekly tactical format. A 30-person company adopted decision-making by consent for all structural decisions — but declined formal tactical meetings in favour of an asynchronous tool. The practice lives because it was grafted onto an existing reality, not imposed on a blank page.
• Use circles without the role-appointment process. A non-profit structured its teams into semi-autonomous circles with clear purposes — but kept a human, informal negotiation for assigning roles. The skeleton is sociocratic; the flesh is human.
• Adopt the governance / operations split without the holacratic constitution. Several organisations found immense value in the simple distinction « we don't talk about two different things at the same time » — without adopting the full holacratic framework. That alone massively reduces confusion in meetings.
• Create an ad-hoc tension-facilitator role. When an organisation is not ready for a formalised governance process, you can simply name someone responsible for making sure tensions surface and find a space to be processed. Simple, powerful, appropriate.
• Introduce reasoned objections in the leadership committee only. Sometimes change starts at the top — and moves down gradually as teams observe that it works. There is no obligation to start with everyone at once.

« The real competence is not knowing S3 or holacracy inside out. It is knowing what this organisation, at this moment, can receive, digest and bring to life. »

A framework for choosing — and for knowing that every choice is provisional

If you are thinking about your governance, here are some markers to orient your reflection. They are not rules — they are questions to hold.

• What problem are you actually trying to solve? Decisions too slow? Team engagement low? Meetings unproductive? The answer should point towards specific practices — not towards a global model.
• Is your organisation ready to learn new behaviours? A governance model demands competences: formulating tensions, distinguishing a valid objection from a preference, holding a role with authority. If those competences do not yet exist, the first investment is in training — not in the model.
• Is your leadership ready to transform itself — not only to transform others? That is the hardest question. And the most decisive.
• What level of formalisation suits you? If you need clarity and rigour, holacracy offers a precise frame. If you need flexibility and experimentation, S3 fits better. Both are legitimate — for different organisations, or for the same organisation at different moments.
• Are you ready to revisit this in six months? Governance is not a destination. It is a living process. An organisation that adopts it as a fixed conviction condemns itself to ossification. One that adopts it as a hypothesis to test stays able to learn.

What I have learned working with organisations

After years working with European organisations on their governance questions, a few convictions have settled.

The first: models are maps, not territories. S3 and holacracy describe what is possible. They do not describe your organisation. There is always a gap between the map and the ground — and the real work happens in that gap.

The second: governance does not replace relationship. I have seen organisations adopt impeccable models while maintaining cultures of fear, silence and internal competition. The tools worked — but nothing really changed, because the relational quality had not moved. Governance is a container. What matters is also what you put in it.

The third: changing governance takes courage. Not just method. The courage to name what is not working. The courage, for a leader, to distribute an authority they held alone. The courage to cross the uncomfortable period where the old habits have gone but the new ones are not yet natural.

And the fourth, perhaps the most important: there is no good governance outside of context. Sociocracy can be the perfect tool for this organisation right now — and counterproductive in a similar organisation with a different history. Humility before the singularity of each situation is the condition of any useful support.