The creator rate tells you roughly one member in fifteen has ever posted. It does not tell you that most of them posted once — which is the more useful fact, because that group is the easiest one in your community to do something about.
The definition
One-and-done rate is the share of a community's posters who have written exactly one post.
The denominator is posters, not members. This is a statement about the shape of contribution among people who already contribute, not about the community as a whole.
What the number is
Across 85,909 poster-community records in our public panel — 79,915 distinct people — 65.4% have exactly one post, and between them they account for 17.4% of all content. At the other end, the 0.9% of posters with fifty or more posts produce 37.2% of everything.
| Posts written | Share of posters |
|---|---|
| Exactly 1 | 65.4% |
| 2–4 | 24.1% |
| 5–9 | 5.8% |
| 10–49 | 3.7% |
| 50 or more | 0.9% |
The pattern holds at person level, not just per community: of the 79,915 distinct individuals appearing as authors anywhere in the panel, 63.8% have posted exactly once across every community they belong to. It is not an artifact of counting the same person in several rooms.
This is what a community looks like everywhere, at every size in the panel. If your own figure is somewhere near two thirds, nothing has gone wrong.
Why this result was checked three times before publishing
A truncated crawl — one that captured only recent history — would make long-standing contributors look like newcomers and produce exactly this finding spuriously. Three checks were run before it was written down.
Completeness. Every completed crawl reports itself complete, and posts retrieved divided by the post count Skool itself declares equals 1.000. The deepest crawl reached 144 pages against a 400-page ceiling, so nothing was stopped by a cap.
Invariance to history length. If truncation were operating, the one-post share would rise with how much history a community has. It runs 67.4% / 70.5% / 67.0% / 64.9% across communities with under 50, 50–199, 200–999 and 1,000+ posts — flat, and marginally lower in the deepest histories, which is the opposite of the truncation signature.
Person level. The finding survives when records are resolved to distinct individuals, as above.
One limitation cannot be checked and is stated rather than hidden: deleted posts are invisible to any outside observer. A member who posted five times and deleted four is indistinguishable from a genuine single-post author. The figure counts posts visible today and is therefore an upper bound.
Why this is the most encouraging number in the report
It is easy to treat everyone who is quiet as one population. They are at least two, and they are not equally reachable.
A member who has never posted has not shown you anything yet. A member who posted once has already done the hard part. They wrote something in front of people, published it, and waited to see what happened. That takes more from most people than any owner's prompt can supply.
In most communities nobody went back to them. Not out of neglect — a single post from six months ago does not announce itself in a feed, and there is no natural moment at which it resurfaces. But it means the largest opportunity in a typical community is a group that has already raised its hand, is far smaller than the silent majority, is nameable, and has a specific first post you can refer to when you speak to them.
You do not have to activate a silent majority. You have to go back to the people who already started.
The other end of the same distribution: the engine
The same concentration shows up at the top, and the useful version of it is a ratio rather than a headcount.
Across 399 public communities with a complete crawled history, about 2.15% of members write three quarters of everything posted. Half of all communities fall between 0.92% and 4.05%. In a typical community, 2.15% is roughly four people. The figure was 2.15% at n=158 as well: the crawl finished, the sample more than doubled, and the number did not move.
The share is the finding. The four is not — because the number of people required climbs steeply with size while the share stays where it is:
| Median members | People writing three quarters of the posts |
|---|---|
| 133 | 4 |
| 461 | 9 |
| 1,675 | 37.5 |
| 8,292 | 199.5 |
That is a 62-fold range in community size. If a community were really carried by a handful of individuals regardless of scale, the right-hand column would barely move. It goes from four people to almost 200. Fitted properly, engine size scales with membership at a log-log slope of 0.90 (95% CI 0.82–0.97) — near-proportional, and just short of a perfectly constant share.
The obvious confound was checked. A community with a longer history has more chances to accumulate posters, which could manufacture that scaling on its own. Holding post volume constant, the engine share is 2.12% in communities averaging 248 members and 1.73% in communities averaging 1,332 members — five times the size, and the share moves by four tenths of a point rather than by a factor of five. It tracks members, not history.
Two limits, stated plainly. In a community with only eight posters, the top ten is everyone; this figure is computed only where at least twenty distinct people have posted, and it describes the median community rather than yours. And the check that was outstanding has now been run: the top poster IS disproportionately the owner, in 77.4% of communities, with a median rank of 1. So the finding is restated rather than defended. In three of four communities the loudest voice is the owner, writing a median 25.8% of everything. Set the owner aside entirely and the concentration still holds: a median 3.58% of members carry three quarters of what members write.
Putting the two halves together
Your community is carried by a group you could write on one hand at 140 members, and by a group of a few hundred at eight thousand. Either way it is a countable set of people, not an anonymous mass — and one step below them sits a much larger group who contributed once.
That is the whole of the work, and it is smaller than it looks from inside a quiet feed.
What this figure cannot tell you
Post counts here are all-time and undated, so a member who posted once last week is counted the same as one who posted once two years ago. "Posted once and stopped" and "posted once so far" are different situations, and separating them needs dated post history rather than totals. If you are working through your own list, start with the recent ones.