Name the thing in your community that you are still doing out of momentum.
You know the one. It made complete sense when you started it. Somebody asked for it, or you saw another creator doing it, or it was your best idea in a burst of energy fourteen months ago. Three people engaged with it enthusiastically for about six weeks. Now it takes you ninety minutes every Thursday and produces one reaction emoji, and you keep going because stopping feels like admitting something.
Everybody has one. Some of us have four. Let us build you a way to end things properly, because the ability to retire your own ideas cleanly is one of the least discussed and most valuable operational skills in this work.
Why we keep running dead programs
Worth understanding the pull, because knowing the mechanism makes it much easier to resist.
Sunk cost, obviously. You built it. You made a graphic. You said the word "weekly" out loud in public.
Fear of the signal. Ending something feels like announcing decline. You imagine members reading "we are stopping the Thursday thread" as "this place is dying," when what they will actually think is "fine, I never read that."
Loyalty to the three people who liked it. This one is genuine and it deserves respect rather than dismissal. We will handle it specifically.
And identity. If you have described yourself for a year as the person who runs a weekly critique session, stopping feels like losing a piece of who you are.
None of those are reasons to keep going. All of them are reasons it feels hard, which is different, and naming them takes most of the sting out.
The review, which takes about an hour
Do this twice a year. Put it in your calendar now.
Step one: inventory. Write down everything you currently run on a repeating basis. Every recurring thread, call, email, series, channel, program, ritual, and obligation. Include the things you do not think of as programs, like the monthly newsletter or the fortnightly check in with quiet members.
Most people find between eight and fifteen items and are slightly appalled. That reaction is the point.
Step two: cost. Next to each one, write the honest time it takes per month, including preparation, the thing itself, and the mental overhead of remembering it exists. Be generous. A weekly thing you prepare for thirty minutes is more like three hours a month once you count the low grade background awareness of it.
Step three: what it is for. One sentence per item. What job does this do? Not what it is, what it does. "Gets newcomers talking in their first week." "Gives me a reason to be present on a Monday." "Produces the material I reuse in my newsletter."
If you cannot write that sentence in under twenty seconds, you have found something to retire.
Step four: evidence. For each item, one honest note about whether it is doing that job. Not a metric necessarily. Just: is it working? Who uses it? Would anybody notice if it vanished for a month?
Step five: decide. Four verdicts, and you must give every item one of them.
Keep as is. It works.
Fix. It has a clear job, it is not doing it, and you can name the change. Give it a deadline: reviewed again in two months.
Pause. Might be seasonal, might just need a rest, might be right for a version of your community that does not exist yet. Pausing is underused and enormously useful.
End. It has run its course.
The rule that makes this work: you are not allowed to keep something in "fix" twice in a row. If it failed a fix cycle, it gets paused or ended. Otherwise "fix" becomes a comfortable place to store things you cannot bear to bury.
The one thing you cannot end
Before you get enthusiastic with the axe, protect one item.
Every community needs at least one absolutely predictable thing. One event or thread or moment that happens on a known schedule and does not move. That is the heartbeat, and if you retire everything at once you will discover that the community felt alive because of the rhythm rather than because of any particular program.
So identify your heartbeat first and mark it untouchable. Then retire whatever you like around it.
Handling the three people who loved it
This is the part that stops most people, and it has a clean solution.
Go to those three people individually, before you announce anything, and tell them first. "I am going to stop the Thursday critique thread. You have been the heart of it and I did not want you to read about it in an announcement."
Then offer them something. Not a consolation prize. A genuine question: "Is there a version of this that you would want to run yourself, or something else that would serve you better?"
Roughly a third of the time, one of them says yes and takes it over, and the thing lives on without your ninety minutes. Roughly a third of the time they say they had actually been finding it hard to keep up with too, which is a delightful and common outcome. The rest of the time they are briefly disappointed and completely fine, because you told them personally and that mattered more than the program.
How to announce an ending
Short, warm, factual, unapologetic. Four elements.
What is ending, named clearly. What is not changing, which is the reassurance people actually need. When the last one is. And thanks to whoever made it good.
For example: "The Thursday critique thread is finishing at the end of this month. Everything else stays exactly as it is, including the monthly call and the open help channel. The last one is on the twenty sixth, so bring something good. Big thanks to the handful of you who kept it alive, you know who you are."
Notice what is absent. No apology. No explanation of your workload. No "I have been thinking a lot about this." No implication that anybody failed. Endings announced confidently are read as competence. Endings announced anxiously are read as trouble, and members take their cue almost entirely from your tone rather than from the content.
And give the thing a proper last one. A final session, a final thread, a small acknowledgement that it happened and it mattered. Things that dribble out without ending leave everybody vaguely uncertain about what else might be quietly disappearing.
What to do with the archive
Practical housekeeping that people forget.
If there is a useful back catalogue, keep it and label it clearly as archived, with the date range. "This ran from March 2025 to September 2026 and there is good material in here" is a perfectly respectable state for a resource to be in.
If a channel is closing, set it to read only rather than deleting it, so that nobody loses work they put in. Members who wrote thoughtful posts there will be quietly grateful, and it costs you nothing.
Do not delete other people's contributions in order to tidy up. Their words are theirs. If you must remove a space entirely, tell people well in advance so they can save anything they want, and do not export anybody's material into something new without asking them first.
Do not replace it immediately
Strong temptation: end the Thursday thread and announce a new Wednesday format in the same breath, so it looks like an upgrade rather than a subtraction.
Resist for at least a month. Two reasons. You need to find out what the community actually feels like with the space empty, and you cannot learn that if you fill it instantly. And you need the practice of subtraction being allowed to be just subtraction, because a founder who can only stop things by simultaneously starting things will never actually reduce their load.
Enjoy the empty Thursday for a few weeks. It is a genuinely lovely feeling and it is information: if nobody misses it, you were right, and if three people ask about it, you have learned something too.
The mindset shift worth making
Here is the reframe that helps most. A program is an experiment, not a promise.
If you think of everything you launch as a commitment, then every launch is terrifying and every ending is a broken vow. If you think of it as an experiment with a review date, launching gets easier and stopping gets much easier, because stopping was always one of the possible results.
You can even say this out loud when you start things. "I am going to try this for three months and then we will see." That single sentence buys you enormous freedom, and members respect it. Nobody has ever thought less of a founder who said the words "we tried it and it did not work, so we are stopping."
The communities that stay good for years are not the ones that accumulate. They are the ones that are willing to prune.
Do this this week
Set aside one hour and write the inventory. Everything you run repeatedly, with honest monthly hours next to it.
Write the one sentence purpose for each. Notice which ones you cannot write.
Mark your heartbeat as untouchable.
Give every remaining item a verdict: keep, fix with a deadline, pause, or end.
Then pick the single most obviously dead one, message the two or three people who cared about it, and write the four line announcement.
You will get about three hours a month back and lose almost nothing, which is the best trade available to you right now. And there is something genuinely energising about a shorter list. It leaves room for the next idea, which is going to be much better than the one you have been dragging around since last spring.