A volunteer who disagrees with you does not argue. They agree pleasantly in the room, and then nothing happens. No email, no draft, no excuse — just a quiet gap where the work was supposed to be. The first few times this happened to me I treated it as a people problem. It took longer than it should have to see it as a feedback mechanism, and a far more honest one than anything I was getting at work.
I founded and chair Berkeley Florida, the Cal alumni community here, and almost everything that gets done there gets done by people who owe me nothing. That constraint turns out to be an unusually good teacher. In a company, bad delegation still produces output. Someone does the thing badly, or late, or resentfully, and the result lands on your desk and you file it and never learn what went wrong. Strip out the authority and the same bad delegation produces silence. Silence is information. You find out immediately which of your requests were actually requests and which were just instructions wearing a polite tone.
Authority is a painkiller, not a cure
Most delegation advice for leaders assumes the hard part is emotional — letting go, trusting people, resisting the urge to take the keyboard back. That is real, but it is not the binding constraint. The binding constraint is that most delegated work is badly specified, and in a paid environment nobody tells you, because they can be compelled to absorb the ambiguity themselves.
I have handed people assignments that were, on inspection, three tasks pretending to be one, with no definition of done and no decision rights attached. In a business those come back half-finished and I conclude the person needs development. In a volunteer organisation they come back not at all, and I am forced to conclude that I wrote a bad brief. The second conclusion is usually the correct one in both settings.
So the diagnostic I now use is simple. Take anything you currently delegate and ask whether it would still happen if you lost the power to require it. If the honest answer is no, the task is being carried by your authority rather than by its own design. That is fine occasionally. It is not fine as a general operating model, because authority is a painkiller — it masks the symptom and the underlying fracture keeps spreading.
Scope to something a person can finish
Volunteers will not accept an area. They will accept a thing. "Own our communications" gets nodded at and abandoned; "write the note that goes out the week before the event" gets done. The difference is that the second has an end, and a person can see themselves standing at that end.
I think open-ended ownership is one of the most overrated ideas in management. We tell ourselves we are empowering someone when we hand over a permanent domain, but what we have often handed over is an unbounded obligation with no completion state. The capable ones self-scope and survive. Everyone else slowly drowns and then looks unreliable.
The habit I brought back into business is to delegate the first finishable unit, not the territory. Give someone one shippable piece with a visible edge. When they deliver it, give them the next one, and after three or four you can hand over the domain — except now you are handing it to someone who has already demonstrated they can land work in it, and who knows what the standard is. Founder delegation fails most often at the first jump, when the founder goes from doing everything to assigning a department.
Hand over the decision, not the chore
The fastest way to lose a volunteer is to give them the work and keep the judgment. They produce something, you adjust it to match the version in your head, and they correctly conclude that they were a pair of hands. Nobody donates their evenings to be a pair of hands.
Paid people reach the same conclusion. They just stay, which is worse, because you now have a salaried person who has learned to stop thinking. I have watched this pattern in operating companies more than once: a competent team that routes every real decision upward, not because they lack ability, but because they were trained out of using it. Leadership trust is not a feeling you have about someone. It is a set of decisions you have actually let go of, and the evidence is that outcomes you would have chosen differently are allowed to stand.
Which means you have to decide, in advance, what "different from how I would have done it" costs. Some decisions are cheap to get wrong and easy to reverse — those are exactly the ones to hand over wholesale, and I have written elsewhere about reversibility as a way of sorting risk. Others are not. Being explicit about which is which is far more useful than a generic instruction to use good judgment.
Deadlines have to come from the world
I cannot give a volunteer a due date. I can give them an event. The event happens whether or not we are ready, and the whole group can see it coming, so the deadline has a reality that nothing I say could manufacture.
That taught me something about internal dates in businesses. Most of them are soft because everyone knows they were invented by a manager and can be uninvented by the same manager. The dates that hold are the ones tied to something external — a board meeting, a customer commitment, a release that has already been announced, a conference where the thing has to exist. Before I set a deadline now, I ask what makes this date true. If the only answer is that I said so, I either find a real forcing function or I stop pretending the date matters.
Taking it back costs more than you think
Pulling a task back from a volunteer usually ends the relationship. They may be gracious about it, but they do not come back for the next thing, and they are right not to. You asked, they committed time they did not owe you, and then you signalled it was not good enough without having told them what good enough was.
In a company the same act costs you something quieter and slower. The person stays, their standard quietly drops to "the boss will redo it anyway," and you have manufactured the dependency you complain about. Letting go as a leader is partly about tolerating a first version that is worse than yours — and partly about having defined the standard early enough that you do not need to rescue anything. If you did not say what the bar was before the work started, you have not earned the right to a rewrite.
When something genuinely has to be reclaimed, I try to do it as a change of scope rather than a verdict on the person. The task moves; the relationship does not get a wound. That is not softness. It is the difference between losing one deliverable and losing a contributor.
You recruit for motive, then for skill
In a volunteer organisation you cannot hire. You can only notice who is already leaning in and give them something real to do. That forces you to read motive first — why does this person want to be near this work — and treat capability as the second filter.
Businesses almost always invert that order. We screen hard for skill and assume motive will be supplied by the salary. It will not. The most reliable predictor of whether delegated work comes back finished is whether the person wanted the thing to exist in the first place, which is one reason I keep arguing for hiring for judgment and intent over raw production ability. Capability without motive produces compliance. Motive with adequate capability produces ownership, and ownership is the only thing that survives your absence.
The audit worth running on yourself
Every so often I take a plain list of what I have delegated — in the business, on boards, in the alumni work — and mark each item with the reason it gets done. Authority, habit, relationship, or genuine ownership. The honest tally is never as flattering as I expect. Anything living in the first two columns is work I have not actually delegated; I have merely assigned it, and it will stop the day I stop pushing.
Chairing something where nobody has to listen to you is the cheapest leadership education available, and I would recommend it to any operator who suspects their team has gone quiet on them. It is also why I think chairing is a different job from running and should be learned separately. The skill is not persuasion. It is building requests that stand up without you behind them — and if we can do that where we have no leverage at all, we have no excuse for doing it worse where we do.