What Kills Coordination Systems
I've watched six coordination networks die. Not fade — die. Full collapse, board goes quiet, people stop showing up, everyone pretends it was a nice idea that "just didn't work out."
It always worked. Something killed it.
If you're going to build one of these, you need to know what kills them. Not because you should be scared — because every failure I've seen was preventable, and every one of them came from the same short list. Fix these five things, or watch your network end up as a cautionary anecdote at someone else's kitchen table.
Killer #1: The Charismatic Founder Problem
This is the most common cause of death, and it's the most predictable.
One person starts the network. They're energetic, persuasive, organized. They build the board, recruit the members, facilitate the first exchanges, resolve the first disputes. The network is that person for the first six months.
Then that person gets a new job, or burns out, or moves, or has a kid, or just gets tired. And the network collapses. Not because the idea stopped working — because the idea was never transferred from one person's head into a system that could survive their absence.
I did this myself. Twice. Built something around my own energy, didn't distribute the load, watched it fall apart when I got pulled in another direction.
The fix is structural, not motivational:
Distribute coordination from day one. The person who manages the board shouldn't also be the person who recruits members, resolves disputes, and tracks exchanges. Split those jobs across at least three people immediately. Not because it's more efficient — because it means no single departure kills the system.
Rotate the roles. The board manager changes every three months. The person who organizes the monthly gathering rotates. Nobody becomes irreplaceable, because replaceable people build durable systems.
Document everything. The skills inventory, the exchange rules, the points ledger — all of it lives in the shared sheet, not in someone's head. If the founder disappeared tomorrow and a new person read the sheet, could they keep running the network? If no, fix that today.
Killer #2: Free Riders
Every network of more than eight people will attract someone who takes more than they give. Not maliciously — most free riders don't think of themselves as free riders. They just always need help this month and will definitely contribute next month and oh, something came up, but definitely the month after that.
In a small network, this is fatal. If one person in a ten-person group stops contributing, the other nine feel it. Resentment builds fast. The people who contribute the most start contributing less because they feel exploited. The network's best members leave first, because they have the most options.
The fix is visibility:
The board is the accountability system. When every exchange is tracked — who gave, who received, hours contributed, hours consumed — free riding becomes mathematically obvious. You don't need a confrontation. You need a spreadsheet.
Set a minimum. Not a punishment — a floor. "Every member contributes at least five hours per month." Below that for two consecutive months and someone from the group has a conversation with you. Not a tribunal. A conversation. "Hey, everything okay? We noticed you've been quiet on the board."
Let people leave gracefully. Some people join and realize it's not for them. That's fine. The worst thing you can do is guilt someone into staying when they don't want to contribute. A ten-person network with ten engaged members beats a fifteen-person network with five passengers every time.
Killer #3: Scope Creep
The network starts as a skill exchange. Simple. Tangible. Post a need, get it handled, move on.
Six months in, someone suggests using the network for collective purchasing. Then someone wants to start a community garden. Then someone proposes a shared childcare rotation. Then someone brings up mutual health insurance. Then someone wants to hold a meeting about the network's "vision and values."
None of these ideas are bad. But each one adds complexity, adds governance, adds the potential for disagreement, and pulls energy away from the core function: connecting needs to skills.
The network that tries to do everything ends up doing nothing well.
The fix:
Protect the core. The task board is the network. Everything else is optional and secondary. If the board is running smoothly — needs getting posted, exchanges happening, DONE column growing — the network is healthy. If the board goes quiet because everyone's arguing about the community garden bylaws, the network is dying.
Spin off, don't absorb. If collective purchasing is a good idea (it is), let the people who are excited about it start a purchasing group. Connected to the network, but not part of the board. Same for the community garden. Same for childcare rotation. Each initiative gets its own structure, its own leadership, its own channel. The task board stays clean.
Say no to meetings about meetings. The moment someone proposes a "planning session" to discuss the network's future direction, you've gone wrong. A coordination network doesn't have a direction. It has a board. The board reflects what people need right now. If the board is full, the network is working. If the board is empty, recruit better.
Killer #4: Imbalanced Exchange
This is different from free riding. Imbalanced exchange happens when the system structurally favors some members over others — not through bad intent, but through bad design.
The plumber's skills are in high demand. Everyone needs plumbing eventually. The graphic designer's skills are needed rarely. The plumber gives twenty hours a month and gets twenty hours back easily. The designer gives five hours a month and can't spend her points because nobody needs a logo.
Over time, the plumber feels valued and the designer feels useless. The designer leaves. Then the tutor leaves for the same reason. Then the bookkeeper. The network hollows out into a trades-only mutual aid group that's useful but limited.
The fix:
Broaden the definition of "skills." The graphic designer can't help if the only posts are plumbing and carpentry requests. But she can redesign menus for the guy who caters. She can build flyers for the woman who tutors. She can help the small business owner with branding. The problem isn't that her skills aren't valuable — it's that the network isn't recognizing the full range of needs.
Actively match underutilized members. When you see someone's CAN DO column filling up but their DONE column staying empty, that's a signal. Check in. "What do you need? Let's find it in the network." Sometimes people don't post needs because they don't want to "bother" anyone. Normalize asking.
Create internal projects. The network itself has needs — someone to manage the website, design the flyer that recruits new members, organize the monthly gathering. These are real contributions that use skills the board might not otherwise call for. The designer who can't find work through the board might be the person who designs the network's identity.
Killer #5: Letting the System Back In
This one is subtle, and it's the one that matters most.
Your network is humming. Exchanges are happening. People are connected. Value is being created and retained locally. And then someone suggests "professionalizing" the operation.
"We should incorporate." "We should get liability insurance." "We should build an app." "We should apply for a grant." "We should become a 501(c)(3)."
Every single one of these suggestions pulls the network back into the system you built it to route around. Incorporating means lawyers and filing fees. Insurance means a monthly premium flowing out to Hartford. An app means a tech dependency that will break, cost money, and eventually get acquired or enshittified. A grant means writing proposals, tracking compliance, and answering to a funder instead of to your members. A 501(c)(3) means a board of directors, annual filings, and a structure designed for a different century.
I'm not saying never do any of these things. Some networks, at scale, eventually need some of them. But the instinct to "professionalize" is almost always premature, and it almost always comes from someone who's more comfortable inside the system than outside it.
The fix:
Ask: who benefits from this? If the answer is the network's members, consider it carefully. If the answer is "it'll look more legitimate" or "it'll protect us legally" — from what? Protect you from what? You're a group of friends trading favors. You don't need a legal structure to fix your neighbor's fence.
Stay informal as long as possible. Formalization is a one-way door. Once you incorporate, you can't un-incorporate without effort. Once you accept a grant, you're on someone's spreadsheet. Every formal structure adds overhead and invites the extraction systems back in. Resist until the network is large enough and mature enough that formalization solves a real problem — not an imagined one.
Remember what you're building. You're building a parallel system. A system that works because it's outside the corporate and institutional framework. The moment you pull it inside that framework — with lawyers, insurance, compliance, and tax filings — it starts playing by their rules. Their rules are designed to extract from you. Don't volunteer for it.
The One That Doesn't Kill You
There's a sixth thing that people think kills coordination networks but actually doesn't: conflict.
Two members disagree about an exchange. Someone feels shortchanged. Somebody's work quality wasn't great. A personality clash makes two people avoid each other.
This isn't death. This is Tuesday.
Conflict in a small network gets resolved because the stakes are personal. Gene isn't going to screw over Rick because Gene has to see Rick at the hardware store. The social pressure of a tight network is the most effective conflict resolution system ever invented — far more effective than HR departments, legal contracts, or app-based rating systems.
When conflict happens (not if — when), handle it simply:
- The two people talk.
- If they can't resolve it, they pick a third person to listen.
- The third person doesn't decide — they facilitate.
- If someone needs to leave the network, they leave without drama.
That's it. No policy. No procedure. Just people who know each other working it out like adults. If your network can't handle interpersonal conflict without a rule book, the network is too big or the members don't know each other well enough. Both are fixable.
Start This Week
24 hours: If you're running a network, answer honestly: is one person doing most of the coordination work? If yes, pick two other members and split the load today. Not next month. Today.
7 days: Check your DONE column for balance. Who's contributing a lot and receiving little? Who's receiving a lot and contributing little? Have one honest conversation with someone on either end.
90 days: Audit for scope creep. Is the network still focused on the board, or has it accumulated side projects, committees, and initiatives that nobody asked for? If the board is active, the network is healthy. If the board is quiet but the meeting calendar is full, you've drifted. Strip it back. The board is the network. Everything else is extra.
The networks that die are the ones that stop being simple. Don't let that be yours.