Choose something small enough to be real
Think local. The playground on your street that has been broken for two years. An elderly neighbor nobody checks on. A corner where the lighting makes people walk the long way around. Kids with nowhere to go between three and six. A stretch of road where the trash never gets picked up.
Specific enough to actually address. Not "fix inequality" — "the playground on Elm Street is broken and there is nowhere for the children on this street to go."
The scale matters more than it seems. National problems are where arguments happen; local ones are where things get built. And the point of this exercise is to end up with something you could start this week, not a position you could defend at a dinner party.
Design it
Five questions. Write the answers down rather than just thinking them.
Who else cares? Not who ought to care — who already does. Almost every local problem has a handful of people quietly annoyed about it who have never met each other. Finding them is usually the whole first phase.
How does it get funded? Donations, a crowdfunding page, a local business that would sponsor it for the goodwill, membership dues, a one-time collection, someone's spare weekend and a borrowed tool.
How does it get organized? An informal group, a nonprofit, a group chat for the street, two people and a spreadsheet. Choose the lightest structure that works, and resist making it more formal than it needs to be.
What sustains it? This is the question most plans skip and most efforts die of. What keeps people showing up in month four, when the initial enthusiasm has gone and the thing still needs doing?
And: what is the first step? Not the plan. The first action, the one you could take in the next seven days.
The part that decides it
You now have a design. The honest question is whether anything happens next, and for almost everyone who does this exercise the answer is no.
That is not a character flaw. It is what happens when a thing has no deadline, no boss, and no consequence for skipping it — which describes every voluntary undertaking, and is exactly the objection people raise against voluntary systems.
Which makes this the smallest honest test of the whole philosophy. Not whether voluntary cooperation could work in principle, but whether you will do one voluntary thing that nobody is making you do.
The first step is smaller than you think
You do not need permission. You do not need a law, a budget, a nonprofit, or anyone's approval.
You need one conversation with one other person who is annoyed about the same thing. That is the entire first step, and it is usually enough to make the second one obvious.
Most things that exist began roughly this way, with no plan to become what they became. The distance between a problem nobody addresses and a thing that quietly works is generally one person deciding to ask someone else about it.