Engagement rate is the number community owners reach for first and understand last. It sounds like a standard metric, the way a conversion rate or a churn rate is. It is not. There is no official Skool engagement rate, no published definition, and no shared formula that everyone in the ecosystem uses. Every owner quoting a percentage has made three private choices — what counts as engagement, over what window, divided by which set of people — and those choices move the answer more than the underlying community does.
That is not a reason to skip the metric. It is a reason to define yours precisely and then stop changing it.
What does engagement rate on Skool actually mean?
Engagement rate is the share of your members who actually did something in a defined window, rather than merely existing on the roster.
That is it. Everything else is implementation detail. The metric exists because member count lies. A community of 800 where 40 people talk is a smaller community than one of 150 where 60 do, and no headcount will ever tell you that. Engagement rate is the correction.
There is a second, related sense of the phrase that gets used interchangeably and should not be: engagement per post, meaning the average comments a post attracts. That measures your content, not your community. Both are useful. They answer different questions, and mixing them is where most confused conversations start.
| What you want to know | Metric | Rough shape |
|---|---|---|
| How alive is my community? | Member participation rate | Participants divided by members, per window |
| Is my content landing? | Engagement per post | Comments divided by posts |
| Is the community getting healthier? | Participation rate trend | Same metric, plotted week over week |
| Are new members activating? | Cohort participation rate | Participants divided by members, joiners only |
Why does my engagement rate change depending on what I divide by?
The denominator you divide by moves your engagement rate more than your community does, which is why two honest owners with identical communities report wildly different numbers.
Say you have 500 people on the roster. Last week, 45 of them commented on something.
- Divided by total members (500): 9 percent.
- Divided by members active on Skool at all in the last 30 days (say 220): 20 percent.
- Divided by paying members only (say 130): 35 percent.
Same week. Same 45 people. Three numbers, from 9 to 35. None of them is a lie and none of them is comparable to anyone else's.
Each denominator answers a different question, so pick the one that matches your actual worry:
- Total members answers "how much of what I built is dormant?" It is the harshest number and the most honest about accumulated dead weight. Free communities with an open door will look grim here, correctly.
- Active members — people who have been on Skool recently, whether or not they showed up in your room — answers "of the people who could reasonably have participated, how many did?" This is usually the most actionable, because it separates gone from the platform from present but silent with me. Those are completely different problems.
- Paying members answers "are the people funding this getting value?" For a paid community this is the number worth putting on the wall, because a quiet paying member is a cancellation with a delay on it.
Pick one denominator. Write it down. Never quietly switch it, because the moment you do, your trend line stops meaning anything and you will read a definition change as a result.
What should I count as engagement?
Engagement should count deliberate acts rather than passive ones, and that boundary is the second private choice. Comments obviously count. Posts count. Votes in a poll are a real signal — low effort, but a deliberate act. Course progress is engagement with your material but not with your community. Logging in and reading is engagement with a lowercase e, and you almost certainly cannot see it.
Likes are the awkward case. They are the cheapest possible signal and they inflate the number without telling you much, and depending on what you are counting with, per-post like data may simply not be available to you.
For what it is worth, Skoolgrades counts engagement as comments plus polls. Per-post like data is not captured at all, so it cannot flatter the figure. That is a deliberately conservative definition: it counts acts that took a sentence or a decision, not a reflex. Your number will look lower than a like-inclusive count and it will move more meaningfully when it moves.
Participation rate versus a visit-based engagement rate
This is the distinction worth internalising, because it explains most of the gap between two numbers describing the same room.
| What it counts | Typical size | |
|---|---|---|
| Participation rate (what Skoolgrades reports) | True interactions with posts: comments, comment replies, poll votes. A member counts only if they actually did something. | Smaller, harsher |
| Visit-based engagement rate | Anyone who loaded the page in the window, including a member who read nothing and said nothing. | Larger, more flattering |
As of July 2026, Skool's own dashboard reports engagement on a visit basis. Skool ships changes often, so open your own dashboard and check rather than take that on faith. The honest punchline is that the visit-based number is always the bigger one: a community can display a perfectly healthy engagement percentage while almost nobody is talking. Both are legitimate: a visit measure tells you about reach, a participation measure about the room. Participation rate is the one that moves when you do something right.
The rule that matters more than the choice: count the same things every week. A metric that quietly gains a category is a metric that shows fake growth.
What time window should I measure engagement over?
Seven days is the sensible default window for an engagement rate, because window length changes the metric's character.
- Seven days is the operational window. It answers "how is it going right now" and it is responsive enough that a change you make on Monday shows up by the following Monday. It is also noisy — one good thread or one holiday week swings it hard.
- Thirty days is the health window. It smooths the noise and catches the members who show up fortnightly rather than daily. Most communities have a real population of monthly-rhythm people who look dead at seven days.
- Ninety days is mostly a retention question wearing an engagement costume. By that horizon you are really asking who is still here, which is covered better in Skool member retention.
Run seven-day and thirty-day side by side. A big gap means a large casual tier that a weekly-only view writes off unfairly.
How do I read my engagement rate without fooling myself?
Reading an engagement rate honestly means watching for four traps, all of which I have fallen into.
The founder effect. If you comment on everything, you are in your own numerator, and so is every reply people make to you. A community can look engaged when what is actually happening is one very busy owner. Try computing the rate with yourself excluded. It is a bracing exercise.
The whale effect. Ten people leaving forty comments each looks identical, in a raw comment count, to eighty people leaving five. Participation rate — distinct humans, not events — is immune to this, which is exactly why it is the better default. Track both and the ratio between them tells you whether you have a community or a very active book club inside a community.
The size effect. Engagement rate almost always falls as a community grows, and that is arithmetic rather than failure. A room of thirty people where everyone knows everyone will beat a room of three thousand on percentage forever. Compare yourself to yourself at a similar size, or to genuinely comparable communities — which is the entire argument in what is a good Skool engagement rate. For scale, measured across 20 Skool communities in July 2026 the median 90-day participation rate was 27.4 percent, with the middle half between 15.3 and 34.4 percent (Skool participation benchmarks).
Freshness skew. If you measure a post's engagement the day it goes up, you are measuring how fast people saw it, not how well it did. Posts keep collecting comments for days. Skoolgrades computes its peer benchmark only on posts matured 72 hours, so fresh content does not make the number jump and then sag.
How do I turn my engagement rate into something I can act on?
An engagement rate becomes actionable only when you decompose it, because the rate on its own tells you whether to look rather than what to do. What you do next comes from splitting it apart.
- Split by tenure. Newly joined, 15 to 90 days, and long-tenured members behave completely differently. A falling overall rate driven by new joiners never activating is an onboarding problem. Driven by long-tenured members going quiet, it is a value or fatigue problem. Same number, opposite fixes.
- Split by direction. Some silent members were never active. Others used to be active and stopped. The second group is the one to message this week — see how to spot members about to leave Skool.
- Check whether they are silent everywhere or only with you. A member who is quiet in your room but posting daily in three other Skool communities has the habit and has not pointed it at you. That is a fixable, conversational problem, not a dead member. It is also invisible from a single-community view.
- Name the individuals. A percentage cannot be messaged. A list of eleven names can. Every engagement metric should terminate in a list of people, or it is decoration.
That last step is the whole point. Engagement rate is a smoke alarm — useful, cheap, and completely uninterested in putting out the fire.
So what is my Skool engagement rate really telling me?
Skool engagement rate is participants over members over a window, and it is only as meaningful as the three definitions behind it. Fix your denominator, fix what counts, fix the window, then watch the trend rather than the level. If you want the arithmetic step by step, how to calculate Skool engagement rate walks it through with worked numbers. If the number is low and you want to know why before you fix it, start with why is my Skool community so quiet.