A Primer on Coordination

How to view coordination before you coordinate

Photo by Stephane Gagnon on Unsplash

Most people I’ve spoken to think coordination begins once people agree. It’s an assumption that sits underneath almost every meeting, project, and collective effort, and it quietly shapes how people try to work together. This idea stems from decades of theory in the field of Management Studies, and is a simplified description of what management theorists call Planned Coordination. Planned coordination relies on a concept called the Shared Mental Model, in which everyone who is meant to work together is supposed to work better if they are all on the same page first. The idea is relatively simple: if you get the right people in the room, define the goal, align perspectives, and then move, everything should work out. It sounds reasonable because it describes what coordination looks like once it has already stabilized. It is a retroactive model that assumes the coordination framework is valid, purely because coordination happens. It never inspects the conditions that gave rise to the coordination, or the moment-by-moment dynamics and interplay of events, or the resulting forms of Power that emerge from the kinds of coordination at play.

But if you pay attention to how coordination actually unfolds in real time, agreement turns out to be the endpoint of a process that is usually invisible to the people inside it. By the time a group reaches alignment, something else has already happened. The field has already narrowed. Some directions have already been made easier to pursue, while others have already fallen away. Participation has already settled into a pattern that feels natural, even though it was shaped moment by moment.

Research on emergency medical teams has found that effective coordination happens through what they call “mutual adjustment” long before any formal plan is in place. An abstract description of how it works looks something like this: Team members read each other’s movements and the demands of the situation in real time, exemplified by things like a nurse moving to make room for a doctor without being asked, or someone handing over a piece of equipment before the request is made. The agreement on who does what never gets spoken out loud. Instead, it emerges from the action itself, resulting from attunement to the environment and the movement of everyone in it. And this kind of mutual adjustment is precisely where i think most emergent coordination happens.

Seven years ago, almost to the day, I joined a mutual aid group (I will refer to it as MAG for the remainder of this article, but the actual group has no name). We stay informal and under the radar because living in a theocratic monarchy that depicts itself as benevolent, means grassroots collective movements are treated as threats to an authoritarian rule that is supposed to have full control and prosperity for everyone.

When I first joined the MAG, every meeting started with the same assumption. We needed to agree on a plan before anyone could act. So we would spend two to three hours debating the wording of a call-to-action, or the exact process for intake, or whether dried chickpeas were better to distribute than canned chickpeas. But here is what I noticed. The meetings and the agreement never came first. The real coordination was already happening in WhatsApp groups and in the pre-meeting gathering as people trickled in. Two people would decide to share a ride to pick up supplies. Someone would text a neighbor about a food delivery. A small subgroup would start mapping out a route for distributing blankets. Agreement came at the end, not the beginning. By the time we voted on something, the actual work had already been coordinated in small ways that never made it into the minutes.

If you want to actually learn how to coordinate, you have to start earlier than agreement. You have to look at the level where things are still fluid, where nothing has settled, and direction is still being formed. That means going smaller than plans, roles, and most of the concepts people use to describe collaboration. It means paying attention to what is being put into motion.

I call that a thread.

What is a Thread?

A thread is the smallest unit of coordination. It is not necessarily an idea sitting in someone’s head, and it does not have to be a finished proposal. It is something that enters the coordination field and creates the possibility of movement. It could be a suggestion, a reframing, a question that shifts direction, or a next step that someone puts forward. Once it appears, something very simple happens. Either someone picks it up, or they don’t. That dynamic, repeated over and over again, is what produces everything people later call coordination.

Before I go any further, I want to say something important. Many groups already work with threads without calling them that. You might have been in meetings where someone throws out a half-formed idea, and someone else says, “That reminds me of something,” and runs with it. Or where a facilitator asks “what do we need right now?” instead of pushing a pre-set agenda. Or where a group explicitly decides to let two competing approaches run in parallel for a while before choosing.

These are all ways of working with threads. They show up in traditions like Agile retrospectives, Open Space technology, sociocracy, and consensus-based organizing. They show up in good teaching, in skilled parenting, and in any team that has learned to trust each other enough to leave space.

The thread framework is not a claim that nobody has ever done this before. It is an attempt to name what works, to understand why it works, and to make it easier to learn and teach. If your group already does some of these things, you are ahead of the curve. What this document offers is a language to talk about what you are already doing, and a way to notice the moments when you accidentally shut down the very movement you want to create.

If you start watching for threads, meetings look very different. Instead of a neutral exchange of ideas, you see multiple lines of action being proposed and either carried or abandoned. Instead of asking what the group decided, you start noticing how a particular direction became the one that held. You begin to see that coordination is less about choosing between options and more about which threads manage to sustain themselves long enough to shape what happens next.

That shift in perception is the first step, but it is not enough on its own. Seeing threads does not automatically mean you can work with them. Most people continue to coordinate in ways that unintentionally shut down the very movement they are trying to create.

Planned Coordination is mostly just Power Over

Closed Threads

The most common mistake (I call it a mistake because it eventually leads to Power Over) when humans attempt coordination starts, fittingly, at the point of initiation. People assume that to coordinate well, they need to be clear, complete, and correct. So they present fully formed ideas, tightly defined plans, or strongly framed positions. What that actually does is close the thread. It leaves very little space for others to enter, which means others respond to it as something to evaluate rather than something to carry. For closed threads like these, agreement becomes the only way forward, and if agreement does not happen quickly, the thread dies.

In those early MAG meetings, people would show up with fully formed proposals. A three-page document on how to handle donations. A detailed roster system for volunteers. A complete schedule for the next month. The person presenting had done all the thinking ahead of time. But instead of feeling grateful, the rest of us felt locked out. There was nothing to add. So we would pick at small flaws instead of engaging with the idea. The proposal would get voted down or tabled. The thread died. And the person who brought it would stop bringing new ideas. We confused completeness with clarity, and it killed our momentum.

If you want a thread to move, it has to be open enough to be taken up. It needs direction, but it cannot be fully resolved. It has to leave handles that others can grab onto and extend. This does not mean being vague or indecisive. It means understanding that coordination requires shared construction. A thread that cannot be shaped by others will not travel far, no matter how good it is.

Over-controlled Threads

The second mistake shows up once a thread has been picked up. People often try to control how it develops. They correct interpretations too quickly, narrow the acceptable ways it can be extended, or pull it back toward the original framing. That creates a form of friction that is easy to miss. Others may continue to engage, but their engagement becomes constrained. The thread stops evolving and starts repeating. Over time, participation drops, even if no one explicitly withdraws.

I saw this happen whenever someone tried to delegate tasks in our MAG. A member would ask for help with a task. Someone would volunteer. But then the first person would keep checking in. “Did you do it this way?”, or “That is not exactly what I meant.”, or worse, “Let me just do it myself.” The volunteer would keep showing up, but the energy drained out of them. They stopped offering ideas. They just waited for instructions. The thread was technically still moving, but it was no longer alive. It was just one person pulling it forward while everyone else held on without contributing. This left both the delegator and the volunteer burnt out.

Working with threads that are meant to generate engaged and collective coordination requires a different approach. Once something has been taken up, it is no longer fully yours. It will change. The question is not how to preserve it, but how to stay with it as it evolves. This requires us to pay attention to how others are extending it, noticing where it is gaining energy, and making small adjustments that keep it moving without locking it in place. Coordination here is less about directing and more about maintaining conditions for movement.

Divergence is Strength, not Weakness

The third mistake I’ve noticed is when threads begin to diverge. In most groups, divergence is treated as a problem to be resolved quickly. Competing threads are collapsed into a single direction as soon as possible, often in the name of efficiency or clarity. What this actually does is reduce the range of possibilities the group can explore. It creates early closure, which can feel like progress but often leads to weaker outcomes.

Our MAG had a recurring debate about whether to focus on food distribution or housing support. Both were urgent. Both had people passionate about them. But the group kept trying to settle the question. We would take a vote. Food would win one week. Housing would win the next. We kept forcing our coordination into a single direction. What we did not realize at the time, and in retrospect was blatantly obvious, was that the two threads needed to run in parallel. The people doing food knew things the housing people needed to know. The housing people had contacts that could help with food. When we finally stopped forcing a choice and let both continue, the group actually started moving faster, not slower.

There is another way to handle this. Instead of collapsing threads too quickly, you can allow multiple threads to run in parallel for a period of time. This approach, when done right, creates a wider field of exploration, where different approaches can develop far enough to reveal their strengths and limitations. The role of coordination here is not to force convergence, but to track how these threads are developing and to notice when alignment begins to emerge naturally. If convergence happens at that point, it is usually more stable because it has been shaped by multiple perspectives rather than imposed prematurely.

Simple Shifts, Big Results

These three shifts, opening threads at initiation, allowing them to evolve without over-control, and resisting premature convergence, change the feel of coordination quite significantly. The space becomes more participatory, movement becomes more continuous, and outcomes tend to be more resilient because they are not dependent on a single line of thinking.

This also explains why most coordination we see defaults to more controlling patterns. When threads are closed too early, tightly controlled as they develop, and forced into quick convergence, participation narrows, and the space becomes “easier to manage”. Fewer variables, fewer surprises, fewer deviations. It feels efficient, but it comes at a cost. Over time, fewer people initiate, fewer ideas are explored, and the systems that are built with these threads become brittle. What remains is a pattern where direction is set by a small number of threads that consistently dominate, whether intentionally or not.

If you shift the way threads are handled, a different pattern starts to appear. More people initiate. Threads build on each other instead of replacing each other. Movement becomes distributed rather than centralized. At a certain point, the coordination no longer depends on any single person holding it together. The pattern itself begins to carry forward because enough threads are aligned in a way that sustains momentum.

From the outside, that can look like a highly effective team, a strong community, or a movement that has found its rhythm. But what is actually happening is much simpler. The coordination has reached a point where threads are being initiated, taken up, and adapted in a way that keeps the system moving without constant intervention.

Let it Cook

There is one more element that matters here, and it sits at a more personal level. Most people drop threads too quickly. If something is not immediately picked up or if it meets resistance, the instinct is to withdraw and try something else. Over time, this reduces the number of threads that enter the space at all. Coordination becomes reactive rather than generative.

Learning to coordinate also means learning to stay with a thread a little longer. Not forcing it, but not abandoning it at the first sign of friction either. Sometimes a thread needs time to be understood, or it needs to be reframed slightly to become accessible. The ability to hold that line of action without immediate reinforcement is what allows new patterns of coordination to form, especially in spaces where existing patterns are deeply ingrained.

When we first introduced the thread framework (myself and a fellow Marxist in the group), it did not land. I mentioned it in a meeting. People nodded and seemed to agree with the abstract concept, but nothing changed. The instinct was to drop it. Assume it was not the right moment or the right group. But we stayed with it. My friend started naming threads when she saw them. “That suggestion is a thread. Let’s see who picks it up.” I started asking open questions instead of making closed proposals. “Here is a loose direction. What would it need to become actionable?” It took several weekly meetings to begin catching on. But slowly, people started using the language themselves. The thread of the framework itself had to be held long enough to be taken up.

The same fashion show research I mentioned above documented a specific pattern. Stage managers would introduce a new way of organizing a task. Sometimes it was picked up immediately. Sometimes it was met with resistance or silence. The managers who succeeded were those who did not withdraw the thread at the first sign of friction. They reframed it slightly. They found one person willing to try it. They let the thread sit and resurface later. The thread that died at first often came back to life simply because someone held it a little longer.

Stop Focusing on Agreement

When we put all of this together, coordination is less a problem of alignment and more a practice of working with movement. We are not trying to control outcomes directly. We are shaping the conditions under which movement happens. We are paying attention to what enters our coordination field, what gets carried, and how it evolves.

Research on multi-agent systems demonstrates that coordination can emerge without any explicit agreement or negotiation. In simulated environments, agents following simple cooperative rules (like putting away a tool that someone else will need) produced global coordination even though no agent ever agreed on a plan with another agent. The coordination happened through action, not through alignment.

Our Mutual Aid group looks different these days. Meetings are shorter. More people initiate. Someone says, “What if we tried X?” and someone else says, “I can take that and do Y with it,” and a third person says, “That connects to what Z is working on.” We rarely need votes or formal agreement. The coordination just moves. And the group has become more resilient because it does not depend on any single person holding it together. The pattern continuously carries itself forward.

Once people start working at that level, the question changes. It is no longer about how to get everyone on the same page. It becomes a question of what kinds of threads we are introducing, what kinds of threads we are supporting, and what kind of coordination those patterns are producing over time.

I am not arguing against plans or agreements. They have their place. But they are a special case. They work when conditions are stable. When conditions are not stable, you need something else. You need the ability to coordinate in real time without waiting for everyone to agree. And that ability starts with seeing threads.

That is the level where coordination becomes something we can actually learn. Next time you are in a meeting, try this: watch for the moment someone offers a half-formed possibility. Notice whether it gets picked up or dropped. And if you can, be the one who picks it up.

This is the first part of a series I want to write about How to Coordinate. In upcoming essays, I will go into more detail on the different aspects of coordination and how small differences in how we approach coordination can have dramatic effects on the resulting forms of power.