Two rates, same data
Ask what share of a Skool course gets completed and the honest answer is a question back: out of whom?
Measured against every recorded enrollment, 1.3% stand at 100% complete.
Measured against only those enrollments showing any recorded progress at all, 32.1% stand at 100%.
Both are computed from the same 183,675 enrollments. Neither is wrong. The gap between them is why published course-completion figures range from the dismal to the respectable without anyone lying — the denominator is doing nearly all the work, and it is rarely stated.
Here is the distribution underneath both numbers.
Enrollments with any recorded progress
Where progress currently stands
Two features stand out. A single exact value — 100% — accounts for 32.1% of everyone with any progress. Below it, spread across intervals, the fullest is 10–19% at 14.3%, and the distribution thins toward the middle to 3.3% at 40–49%.
The shares themselves are directly comparable — every row is a share of the same total, so 32.1% really is larger than 14.3%. What differs is how much of the progress scale each row covers: 100% is a single value, while 10–19% spans ten of them. That makes the 100% figure more concentrated than the bands, not less — roughly a third of progressing enrollments sitting on one exact value, while the rest are spread across the ninety-nine below it. It is the shape the two headline rates come from.
What this chart does not show is when or whether anybody stopped. It is one observation per enrollment. A member at 30% today may be at 100% next month, and 100% is absorbing — once reached, nobody leaves it — so a snapshot accumulates people at the top over time regardless of the path they took. The shape is a fact about the present distribution; it is not evidence about abandonment, and this data cannot be made to answer that question.
The zero column, and what it does and doesn’t mean
The 1.3% figure exists because 95.9% of recorded enrollments show no progress at all.
That is a large number, and it needs reading carefully. Skool computes progress from lessons marked complete. An enrollment therefore records zero when:
- the member never opened the course;
- the member read or watched material but never ticked a lesson;
- the course is gated by level, tier or purchase and the member cannot reach it at all.
These are indistinguishable here. “No recorded progress” is what the data supports; “never opened” is a stronger claim that it does not. Since courses in Skool are frequently gated, and enrollment can be automatic for everyone in a community, the third case is likely to be a substantial share of that 95.9%.
It is not a handful of courses
An aggregate this lopsided invites the suspicion that a few large courses produce it. Computing the rate separately for each of the 190 courses with at least ten progressing members:
| Across 190 courses | Share of progressing members at 100% |
|---|---|
| Lower quartile course | 8.2% |
| Median course | 24.0% |
| Upper quartile course | 55.5% |
The median course has just under a quarter of its progressing members at 100%, and 55.3% of courses are above one in five. The spread is wide — many courses sit far below — but the pattern is distributed across courses rather than concentrated in a few.
The same shape from a different starting universe
The enrollment list is itself a judgement about who “had” a course. Running the identical calculation against every member-course pair observed in these communities — a starting universe nearly twenty-seven times larger, assuming nothing about enrollment — gives:
| Progress | Enrollments (183,675) | All pairs (4,900,892) | Difference |
|---|---|---|---|
| Share with any progress | 4.1% | 0.7% | −3.4 |
| Of those, where progress stands: | |||
| 1–9% | 11.4% | 11.9% | +0.5 |
| 10–19% | 14.3% | 12.9% | −1.4 |
| 20–29% | 9.0% | 9.1% | +0.1 |
| 30–39% | 6.6% | 6.3% | −0.3 |
| 40–49% | 3.3% | 3.6% | +0.3 |
| 50–59% | 6.1% | 7.2% | +1.1 |
| 60–69% | 5.7% | 4.8% | −0.9 |
| 70–79% | 4.2% | 3.5% | −0.7 |
| 80–89% | 5.3% | 5.2% | −0.1 |
| 90–99% | 2.1% | 2.3% | +0.2 |
| 100% | 32.1% | 33.4% | +1.3 |
The share with any progress moves a great deal, as it must — the wider denominator sweeps in every member against every course in their community.
The distribution behind it barely moves. Every band sits within 1.5 percentage points of its counterpart, and the four largest categories appear in the same order: 100%, then 10–19%, 1–9%, 20–29%.
Note what that comparison actually spans. The two starting universes differ by a factor of 27, but the sets being compared are the progressing enrollments each yields — 7,489 against 35,700, about 4.8 times apart. That is the honest size of the check.
Below the top four the middle bands do reshuffle: 50–59% sits under 30–39% on the enrollment side and above it on the other, and 40–49% and 70–79% swap likewise. So what survives is the terminal mass dominating, the fullest interval being 10–19%, and the overall magnitudes — not a rank-for-rank match through the middle, which the small counts there would not support anyway.
What would be needed to say more
The interesting questions here are behavioural — where members stall, whether a stall is permanent, whether shorter courses finish better — and none of them can be answered by a distribution observed once.
Answering them needs the same members followed across time, which requires progress history retained over months rather than days. That is a data-collection change, not an analysis one. Until it exists, the honest output is the census above: two defensible completion rates drawn from the same enrollments, a point mass at 100% with a thin spread below it, and no claim about how anyone arrived where they are.
What the census does settle is narrower and still worth having. Any single course-completion figure quoted without its denominator is close to meaningless, and on this platform the plausible range spans a factor of twenty-five.
Caveats
Every figure here is a snapshot. One observation per enrollment, with no history. Nothing in this article should be read as describing movement, dropout, or the eventual fate of anyone currently partway through.
100% is absorbing; the other bands are not. Members who reach 100% stay there and accumulate in the data over time, while members in intermediate bands may leave them in either direction. That alone can produce a large terminal band, so the size of the 100% group is not evidence about how people move through a course.
“No recorded progress” is not “never opened.” Skool’s measure is lessons marked complete, so reading without ticking registers as zero, as does having no access at all to a gated course. This is the largest weakness in the 4.1% figure and the reason the 95.9% is not described here as non-openers.
Completion is the platform’s own measure. Lessons marked complete, not material watched or understood. Clicking through a course counts as finishing it.
The sample is not random. These are courses inside communities already in view, which skew toward owners running the community as a business. There is no reason to assume courses elsewhere behave the same way.