Networks Do Not Scale Like Programs: They Grow Like Mycelium

The real question is not how to fund a network, but what kind of thing a network is

Most organizations still try to finance networks as if they were projects, and that is the original mistake. A project has a beginning, a budget line, a manager, and a finish. A network has none of those things in the tidy sense. It is closer to a living ecosystem, where value emerges through relationships, repetition, trust, and shared sensemaking. The challenge is not just to pay for activities, but to sustain the conditions under which coordination becomes possible.

That is why so many well intentioned network efforts feel fragile. They launch with energy, convene impressive people, produce a few useful outputs, and then quietly fade. The failure is usually blamed on weak execution or insufficient funding. But the deeper issue is structural: we often ask networks to prove themselves using the logic of programs, while networks create value using the logic of connection.

A network is not a machine for delivering outputs. It is a medium for generating alignment, meaning, and collective action.

This distinction changes everything. If a network is a living system, then funding it is less like buying supplies for a construction site and more like maintaining the soil, water, and weather conditions that let a forest take root.

Why networks create value that budgets struggle to see

A network is useful precisely because it can do things no single organization can do alone. It can translate across sectors, circulate ideas faster than formal hierarchies, and make collaboration feel less like a transaction and more like a shared practice. Yet these benefits are subtle. They are often felt as reduced friction, faster trust, fewer duplicated efforts, and better decisions, all of which are real, but hard to capture in a simple output metric.

This is where funders get trapped. They look for direct attribution, clean causal chains, and measurable units of service. Those tools are appropriate when the thing being funded is discrete and bounded. They are less useful when the core product is relational infrastructure. If the network helps ten organizations coordinate instead of compete, the value may appear in avoided waste, not obvious revenue or headcount.

Imagine trying to evaluate a city by counting how many roads it built without noticing whether those roads connect people to jobs, schools, and each other. The asphalt matters, but only because of what it enables. In the same way, a healthy network is not mainly a bundle of meetings or a directory of members. It is connective tissue.

This is why a network can be both indispensable and underfunded. Its best outcomes are often distributed, indirect, and delayed. The paradox is that the more successful a network becomes, the harder it can be to prove its necessity in the short term, because its work gets absorbed into the normal functioning of the field.

The hidden economy of trust, attention, and coordination

If we want to finance networks well, we need a broader accounting system. Traditional budgets track visible expenses: staff, events, tools, communications. But network value also depends on three hidden currencies: trust, attention, and coordination capacity.

Trust lowers the cost of collaboration. When people trust each other, they share early, disagree honestly, and move faster. Attention determines whether a network stays coherent or fragments into noise. And coordination capacity is the ability to turn many independent actors toward a common goal without forcing them into a rigid hierarchy.

These are not soft extras. They are the actual infrastructure of collective action.

Think about a community of practice. On the surface, it may look like a recurring convening or a discussion forum. But its real value is that people return because they expect to learn, be challenged, and be understood by others facing similar constraints. Over time, the group becomes a shared memory bank and a laboratory. It does not merely transfer knowledge; it changes what members notice, how they interpret problems, and what they consider possible.

That is why networks often matter most in uncertain environments. When the environment is stable, you can optimize a process. When the environment is changing, you need a structure that can learn faster than the problem evolves. Networks are adaptive precisely because they are decentralized enough to absorb variation, yet connected enough to propagate learning.

The financing trap: treating the visible parts as the whole

The most common mistake in network funding is to pay only for the visible layer: the convenings, the platform, the facilitation, the reports. Those are necessary, but they are not sufficient. A network also requires the less glamorous work of curating relationships, onboarding newcomers, translating across institutions, maintaining shared norms, and handling conflict before it metastasizes.

If this sounds familiar, it should. It is the same mistake people make with software. They pay for the app and forget the maintenance, the updates, the integrations, and the user support that keep it useful. A network is more like a product ecosystem than a one time event. Without ongoing stewardship, the channels clog, the language drifts, and participation becomes performative.

There is also a deeper financing error: expecting the network itself to be the only deliverable. In reality, a network should be evaluated on both its direct outputs and its system change effects. Sometimes it will produce shared tools. Sometimes policy influence. Sometimes faster diffusion of innovations. But its most important effect may be the creation of a field that is more legible to itself.

This is a powerful but often overlooked shift. When people in a field can see one another, name common problems, and exchange practical methods, the field becomes more governable from within. Coordination becomes cheaper. Isolation becomes less normal. And ideas travel farther than they ever could through isolated grants or one off pilots.

A better model: fund the network like an organism, not a pipeline

The question, then, is not whether networks should be funded. It is how to fund them in ways that match their biology.

One useful model is to think of a network as an organism with four functions:

1. Metabolism: the basic operations that keep the network alive, such as facilitation, communications, administration, and platform upkeep.
2. Circulation: the movement of information, opportunities, and learning between members.
3. Immune function: the ability to resolve conflict, protect trust, and prevent capture by any one actor.
4. Growth and adaptation: experimentation, convening new subgroups, and evolving the network’s purpose as conditions change.

Most funding supports metabolism, sometimes circulation, and almost never immune function. Yet immune function may be what determines whether a network survives contact with real power. Without mechanisms for norms, governance, and conflict resolution, networks can become either chaotic or dominated by the loudest players.

A second useful model is to distinguish between seed funding, stewardship funding, and infrastructure funding. Seed funding helps a network form. Stewardship funding helps it mature through curation and adaptation. Infrastructure funding sustains the underlying tools and shared assets. Confusing these stages leads to disappointment. A young network should not be judged as if it were mature, and a mature network should not be starved of the resources needed to maintain coherence.

This is not just a technical distinction. It is a moral one. Underfunding stewardship creates a system where collaboration is endlessly demanded and rarely maintained. People are asked to coordinate across organizations, but no one pays for the labor of making that coordination possible.

Collaboration has a cost. If we pretend otherwise, we simply shift the burden onto the most relationally generous people in the system.

What changes when you fund for relationships instead of events

Consider two different ways of supporting the same field. In the first, a grantmaker funds an annual summit, a website, and a series of reports. The field gets visibility, a few inspiring keynote speeches, and some short term energy. In the second, the grantmaker also funds ongoing facilitation, peer learning groups, data sharing norms, conflict mediation, and lightweight coordination between convenings. The field gets something more durable: a shared nervous system.

The difference is not cosmetic. In the first case, people may leave inspired but isolated. In the second, they leave connected to a living network that can continue to metabolize what they learned. That matters because most system change does not happen at the moment of inspiration. It happens later, when people try to implement what they heard and discover that someone else has already solved the same problem.

This is why communities of practice are so effective when they are done well. They turn individual experience into collective intelligence. They make tacit knowledge shareable. They also create a place where people can admit uncertainty without losing status, which is essential for learning in complex systems.

The practical implication is simple: if a network is meant to change a field, then funding should prioritize the routines that keep the field learning. A network that only meets when there is a grant deadline is not a network. It is a pause between silos.

Key Takeaways

• Stop evaluating networks like projects. Look for changes in trust, coordination speed, learning quality, and field alignment, not just visible outputs.
• Fund the invisible labor. Facilitation, norm setting, conflict resolution, and member curation are not overhead, they are the operating system.
• Treat stewardship as essential infrastructure. Networks need ongoing care to remain coherent, especially as they grow and diversify.
• Measure circulation, not just production. Ask whether ideas, tools, and relationships are moving through the network and changing practice.
• Design for adaptation. The best networks do not just deliver a fixed agenda, they evolve as the field’s needs change.

The deepest return on network funding is not scale, it is collective intelligence

We have inherited a habit of thinking that bigger is always better, and that the purpose of funding is to produce visible scale as efficiently as possible. But networks reveal a different kind of ambition. Their highest value is not simply that they can reach more people. It is that they can help a field think together.

That is a more profound achievement than it first appears. When a network works, it changes the unit of intelligence from the organization to the ecosystem. Problems become less mysterious because more people can name them. Solutions become more portable because they are translated by peers, not imposed by outsiders. And action becomes more resilient because it is distributed across many committed actors rather than concentrated in one heroic institution.

So the next time someone asks how to fund a network, the better question is this: what would it mean to finance the conditions under which collective intelligence can emerge and persist? Once you ask that, the whole conversation changes. You stop buying events and start cultivating relationships. You stop funding appearances and start funding the invisible architecture of collaboration. And you begin to see that the future of system change may depend less on building bigger institutions than on learning how to keep living networks alive.