Where Coordination Collapse Shows Up Every Day

When the system isn’t holding the work

Most nonprofit work looks coordinated from the outside. Programs are delivered. Grants are submitted. Teams stay aligned. There is very little visible friction, and nothing that immediately signals something is wrong.

But inside the organization, that coordination is often being held together by a person, not a system.

Last week, I wrote about coordination collapse, where work appears coordinated but the coordination itself lives in someone’s head. What’s changing now is not the problem itself, but how visible it has become. AI is starting to make that visible.

Where AI exposes the work

I was talking with a nonprofit leader recently about how she approaches grant writing. At one point she paused, almost mid-thought, and said, “AI helps a lot. But it feels like I’m cheating.”

She didn’t say it jokingly. It sounded more like she was trying to reconcile something that didn’t quite add up.

As she walked me through her process, it became clear what she meant. She has a folder filled with past proposals, program descriptions, and language she’s refined over years of doing the work. When a new opportunity comes up, she opens that folder and starts pulling pieces together.

Nothing in it is new. The thinking has already been done. But there isn’t a way to use it directly, so each time she rebuilds it into something that fits the moment.

We’ve all done it. You’re working across documents and tabs, pulling pieces together from things you’ve already written. Then you get interrupted, and when you come back, you don’t just pick up where you left off. You have to rebuild the context to even continue.

Most of what she is doing isn’t writing. It’s reconstruction.

What AI is doing is compressing hours of that reconstruction into minutes. And in doing that, it exposes something that’s easy to miss when you’re in it: how much of the effort was never about writing in the first place.

AI can generate from what exists, but it can’t organize what an organization can’t access.

Reconstruction is the hidden work

After watching that process, it becomes hard not to see it elsewhere.

Teams move between systems that don’t connect, pulling pieces of context together just to get back to a shared understanding. Decisions get revisited, not because they were wrong, but because the reasoning behind them isn’t easily accessible. Information exists, but using it requires effort.

The work continues to move, but only because someone is carrying the organization’s memory in their head, holding the relationships between all of those pieces and rebuilding them as needed.

That person isn’t just doing the work. They are where the work lives.

This is what coordination collapse looks like in practice. It doesn’t stop progress, but it ties progress to the people who can hold everything together.

AI works best when context is clear and connected. In most organizations, it isn’t. That’s why the friction becomes visible so quickly. AI doesn’t remove the need for coordination. It makes the absence of it obvious.

When the question changes

Once that becomes visible, the conversation starts to shift.

In some organizations, the effort is still accepted as part of the work. Someone has to hold everything together, and that responsibility simply becomes part of the role.

In others, a different question starts to emerge. If all of this already exists, why does it have to be rebuilt every time?

That question is hard to ignore once it’s asked, and it doesn’t stay contained to the work itself.

Where it shows up most clearly

It shows up in how decisions get made.

When the system can’t hold context, decisions don’t build on what’s already known. They restart. People don’t just revisit the work. They reconstruct the reasoning behind it, often under time pressure and with incomplete information.

What looks like careful decision-making is often just delayed access to what the organization already knows.

AI can generate options, but it can’t tell you which one makes sense if the reasoning behind past decisions isn’t accessible. Without that context, more information doesn’t lead to better decisions. It just adds more to sort through.

What changes when the system starts to hold

When the system starts to hold the work, the change is not abstract.

The organization stops relying on a few people to remember everything. New work builds on what already exists instead of starting over. Decisions don’t require reconstructing months of context just to move forward.

Progress is no longer tied to who happens to be in the room.

Until that happens, the work will keep moving. But it will move at the pace of the people holding it together.

A bridge forward

This is the problem we are focused on solving at BeMo.

Not how to help organizations do more work, but how to ensure the system can hold the work they have already done so they can move faster, decide with confidence, and stay aligned with their mission.