Almost nobody decides how big their community should be. They decide to grow it, which is a different thing and rarely a decision at all.
Size is the one variable everybody optimises and nobody chooses, mostly because the alternative — naming a number and stopping — sounds like giving up.
Growth is the default because it is the easiest thing to count
Every other question about a community is hard to answer and member count is a number that goes up.
Whether people trust each other, whether the good threads still happen, whether anyone would notice if you stopped — none of those fit on a chart, so they lose to the one that does. The goal was never really growth; it was legibility, and growth is what legibility looks like when you have nothing better to measure.
That matters because the cost of getting this wrong is invisible in exactly the number you are watching. Everything can improve on the dashboard while the room quietly stops being worth visiting — and the standard advice, including our own guide to growing one, assumes you have already decided that bigger is better.
What actually breaks is recognition
A community works when people are recognised, and recognition has a ceiling that has nothing to do with your platform.
Somebody remembers what you posted last month. Somebody notices you were away. Somebody answers you because they know you, not because answering is the rule. That is the entire product, and it is made of one person's ability to hold other people in mind — which does not scale, has never scaled, and will not start scaling because you added a category.
Past that ceiling, every thread happens among strangers, and rooms of strangers behave differently: less candour, more performance, more rule-lawyering, and far fewer honest questions. Nothing has broken that you can point at. The place has simply become somewhere people are careful.
Four sizes and what changes
Roughly, and with soft edges — these are characters, not thresholds, and they count active people rather than accounts.
| Active people | What it feels like | What it needs |
|---|---|---|
| Under twenty | Every absence is noticed; one person leaving is a hole | You, present, most days |
| Twenty to a couple of hundred | People recognise each other and reputation is real | A rhythm, and a reason to come back |
| Several hundred | Recognition thins; most threads are among strangers | Smaller rooms, and people who are not you |
| Thousands | A publication with a comment section | Deliberate small rooms inside it, or accept the broadcast |
The second row is where most paid communities should want to live, and a great many owners pass straight through it on the way to somewhere worse.
The number is active people, not members
None of the above works if you measure it with the figure on your dashboard's front page.
A community with four thousand accounts and ninety people who show up is a community of ninety, and it should be run like one. Reading it as four thousand produces every mistake in this post at once: too many spaces, announcements written for an audience that is not there, and a nagging sense that something is wrong with a room that is actually fine.
Count the people who did something this month. That is the number this whole question is about — the rest of the list is an archive of people who once joined, and treating it as a population makes the room feel emptier than it is.
Growing past your capacity is a downgrade you cannot see
There is a size at which you stop being able to read everything, and most owners cross it without noticing and then keep trying anyway.
That is where the exhaustion comes from — not from the work, but from doing a job that stopped being possible some months ago while believing you are behind on it. The room does not need you to read everything. You need to stop, and stopping requires admitting the community got bigger than the way you were running it.
The alternative is worse than it sounds. You keep the growth, lose the thing you liked, and the chart shows an improvement throughout — which is why this particular mistake tends to run for a year before anybody names it, and why it usually gets diagnosed as the community going quiet instead.
The money argument, honestly
There is a real reason to want more members, and it deserves a straight answer rather than a lecture about intimacy.
A paid community needs enough people to be worth running, and below some floor it is a hobby with invoicing. That floor is real. But the usual response — more members at the same price — is the harder of the two available moves, and often the worse one, because each new member costs you some of the attention that made the place worth paying for.
Two hundred people at a serious price is a business. Two thousand at a token price is a job. If the numbers do not work, the first thing to examine is what you charge, not how many more you can recruit — one of those levers preserves the room and the other spends it.
Splitting is the only growth that keeps the thing
If you do want to be bigger, the way to do it without losing what you have is more rooms, not a larger one.
Twelve rooms of forty is not the same object as one room of five hundred, even though the totals match. In the first, people are recognised; in the second, they are attendees. This is why big communities that still feel alive are almost always federations of small ones, and why the ones that feel dead are a single feed with impressive numbers.
The cost is honest: every room needs a reason to exist and somebody who cares whether it works, and an empty room is worse than no room because it proves nobody is there. Split when a group is already talking, never to create one — the rest of that judgement is in structuring a community.
What a dashboard can and cannot tell you
Some of this is measurable and the measurable part is worth setting up properly.
Mateflow's member analytics show active members as a headline figure alongside the total, and the ratio between them — which is the number this post is really about, and the one most owners have never looked at. You can also sort your spaces by active members, which answers the federation question directly: it shows you which rooms are genuinely alive and which ones you are keeping open out of politeness.
The limit is the timescale, and it happens to matter here more than anywhere else. Those views cover the last seven, thirty or ninety days. A community changing character as it grows does that over a year or two, which means the exact shift this post is about is the one your dashboard cannot draw — so if you want to see it, write down today what the place feels like, and read it back next year.
Three things not to do
Do not aim at a round number. A thousand members is not a goal, it is a milestone borrowed from somebody else's business, and it will shape a year of decisions that nothing in your community asked for.
Do not open the doors because it went quiet. Quiet is almost never a supply problem, and new arrivals land in a silent room and leave, which teaches you the wrong lesson twice.
Do not apologise for being small. "We are still quite small, but…" tells every visitor that smallness is a defect you are working on, when it is the reason the place works — and members repeat the framing you give them.
How to decide your number
The useful version of this question is not how many people, but how many you can run the community you described.
Write down what you want a member's ordinary week to contain: what they post, who answers, how long that takes, and who does the answering. Then count how many people that can hold — not aspirationally, at your real capacity, with the time you actually have. That is your number, and it is usually smaller than the one you had in mind and larger than you fear.
Then treat it as a decision rather than a ceiling you failed to break. It can change; it should be revisited when your capacity does; and it is the only version of this question that produces an answer you can act on — which is the same discipline as getting members to answer each other, since that is what raises the number most.
The bottom line
Communities do not have an optimal size. They have a size at which they stop being the thing you built, and it is different for each one.
Count active people rather than accounts. Notice the point where you stopped being able to read everything and stop pretending otherwise. Fix the money with price before headcount, grow by adding rooms rather than enlarging one, and write down what the place feels like now so you can tell later. Then pick a number on purpose, because a community that chose its size is a different place from one that simply never got any bigger — and the members can feel the difference even if they could never name it.