What is the average Skool community size?

By Christopher Ward, Co-founder, SkoolgradesPublished

Short answer

There is no meaningful average. Across 18,412 mapped Skool communities the median is about 44 members once you correct for the small rooms that are hard to find from outside, and roughly 75% have fewer than 100 members. Community size follows a power law with an exponent near 0.92, and below 1.0 a distribution has no stable mean at all — so an average is not just misleading here, it is undefined. The median, read against communities your own size, is the only honest comparison.

See the State of Skool report18,412 communities measured. Benchmark your own against every chart.

Almost every owner asks this question in the same shape: am I behind? They have 80 members, they see a community with 9,000, and they quietly conclude something has gone wrong. The honest answer is that the comparison they are making is not a comparison at all.

What does a typical Skool community actually look like?

Across 18,412 Skool communities mapped from the outside, the median community has about 44 members after correcting for discovery bias, and roughly 75% have fewer than 100 members.

Both of those figures need a word of explanation, because the raw counts say something different.

Small communities are genuinely hard to find from outside. They are linked from fewer places, they surface in fewer directories, and they turn up in search far less often. So a straight count of what you can see over-represents big rooms. Measured raw, the median looks like 110 members and 48.1% look like they sit under 100. Corrected for how much of each size band is actually findable, the median drops to 44 and the under-100 share rises to 75.1%.

How much correction each band needs is itself measured, not assumed:

How much of each size band is findable from outside
Under 100
17.9%
100–999
51.4%
1,000–10,000
86.8%
10,000+
97.7%

Very large communities are essentially all visible. Small ones are mostly invisible. That single gradient is what separates "Skool is full of big communities" from "big communities are the ones you can see".

The limit worth stating: coverage is estimated with a capture-recapture model, so the corrected shares carry that model's assumptions. The median is 44 corrected against 110 observed — the direction is not in doubt, the exact figure is a best estimate.

Why is there no average Skool community?

This is not rhetoric. It is a property of the distribution.

Community size on Skool follows a power law with an exponent of about 0.92 after correction (0.81 on the raw counts). When that exponent sits below 1.0, the distribution has no stable mean: as you collect more data the "average" keeps climbing rather than settling, because each new sample can contain a community large enough to move it on its own.

So "the average Skool community" is not merely a bad summary. It is a quantity with no stable value. One 100,000-member community dropped into a sample of a thousand small rooms rewrites the average and tells you nothing about any of them.

The median does not have that problem, which is why every figure in this library is a median.

Where are all the members?

Here is the shape that surprises people most. Communities and memberships are distributed almost inversely:

Where the communities are vs where the members are
Share of communitiesShare of memberships
Under 10
16%
0.1%
10–49
37.2%
1.5%
50–99
21.9%
5.5%
100–499
15.5%
13.9%
500–999
4.7%
5%
1,000–5,000
3.6%
24.5%
5,000–10,000
0.6%
8.4%
10,000+
0.5%
41.1%

Half a percent of communities hold 41% of every membership on the platform. The smallest three quarters of communities hold about 7% between them.

That is not a Skool pathology — it is what every social platform, marketplace and content library looks like once you measure it. But it has a very practical consequence, and it is the reason this page exists.

Why comparing yourself to what you see is a rigged comparison

Owners meet Skool through the communities they have joined. Which communities are those? Overwhelmingly the large ones, because large ones are what get found, recommended, advertised and talked about.

So the owner of a 90-member room is measuring themselves against a sample drawn almost entirely from the top half of one percent — and concluding they are failing. Almost nobody is looking at a fair comparison.

If you take one thing from this page: compare yourself against the band you are in, not against the communities that reached you through discovery. A 90-member community is at the 60th percentile or so of everything on the platform. It is not small. It is normal, and most of what you have read about "what good looks like" was written by and about the outliers.

Does age explain size?

Partly, and less than you would think. Skool is a young platform and most of what exists on it was created recently:

CreatedShare of communitiesMedian members
202629%41
202547%101
202420%
Earlier4%444

Older communities are bigger, as you would expect. But note the second column: 76% of all Skool communities were created in 2025 or 2026. The platform's median community is young as well as small, and a two-year-old community with a few hundred members is comfortably ahead of the field rather than behind it.

One caveat on that table: dead communities are invisible from outside. These shares describe what survives today, not what was created. The survival rate is not something an outside measurement can see.

How should you actually read your own number?

Three rules, in order of how much they will change your thinking.

Use the median for your band, never the platform median. The right question is not "how do I compare with Skool" but "how do I compare with communities of roughly my size and door type".

Stop treating size as the headline metric. Size tells you how many people arrived. It tells you nothing about how many are still there, how many have ever contributed, or how much of what gets written comes from a handful of people. Those are separate measurements — creator rate for contribution, and the one-and-done rate for its shape — and they are far better predictors of whether a community is working.

Compare yourself to yourself a month ago. It is the only comparison you fully control and the only one that tells you whether what you are doing is working.

A 200-member community at a healthy participation rate is in better shape than a 5,000-member community where nobody writes anything. The second one is easier to screenshot. That is the only sense in which it is ahead.


These figures come from The State of Skool 2026, an independent measurement of 18,412 Skool communities. See the full report for the complete size distribution, the coverage model behind the correction, and 21 other findings.

See the State of Skool report18,412 communities measured. Benchmark your own against every chart.

Frequently asked

What is the median Skool community size?
About 44 members once the count is corrected for discovery bias, against 110 members on raw observed counts. The gap exists because small communities are much harder to find from outside: only about 18% of communities under 100 members are discoverable, compared with 98% of those over 10,000. Roughly 75% of Skool communities have fewer than 100 members.
Why can you not give an average Skool community size?
Because community size follows a power law with an exponent of about 0.92, and below 1.0 such a distribution has no stable mean. The more communities you measure, the higher the average climbs, because a single very large community can move it on its own. That makes the average undefined rather than merely unhelpful, which is why every figure here is a median.
Is a 100-member Skool community small?
No. Roughly 75% of Skool communities have fewer than 100 members, so a 100-member community sits comfortably above the middle of the platform. It feels small because owners meet Skool through the large communities they joined, which are drawn almost entirely from the top half of one percent. That is a biased sample, not a benchmark.
How many members do the biggest Skool communities hold?
Communities of 10,000 or more members make up about 0.5% of all Skool communities but hold roughly 41% of every membership on the platform. The smallest three quarters of communities hold about 7% between them. This concentration is normal for online platforms, but it means platform-wide averages describe almost nobody.
How old is the typical Skool community?
Young. About 29% of communities were created in 2026 and 47% in 2025, so roughly three quarters of the platform is less than two years old. Communities created before 2024 make up about 4% of the total and run a median of 444 members. Dead communities are invisible from outside, so these shares describe what survives today rather than what was created.
Should I compare my community to the platform median?
Only as a rough orientation. The more useful comparison is against communities in your own size band and with the same kind of door, because participation figures shift by band. The single best comparison is against your own numbers a month ago, which is the only one you fully control and the only one that tells you whether your current approach is working.
What owners say

Better than sliced bread.

As someone building my own Skool community, this has completely changed how I engage on Skool. It's simple, easy on the eyes, and helps me quickly find conversations where I can genuinely connect and add value. Instead of wasting time searching for posts, I spend more time building real relationships. It's one of the first tools I open each day.

Lorri Lynn Roche — Founder, The Inner Shift Collective

See this in your own community

Skoolgrades reads your Skool community through your own browser and turns it into a report you can act on: participation measured on real interactions, who is drifting, where your members go when they leave your room, and a ranked list of what to do next — with the draft already written.

Start with Flow — free

Keep reading

Last updated 2026-09-18. Skoolgrades is an independent tool and is not affiliated with or endorsed by Skool.