How do I track member activity on Skool?

Tracking member activity sounds like a data problem. It is really a triage problem. You do not need to know everything about all 400 members. You need a short list, every week, of the ten or so people where doing something today changes the outcome.

Everything below is aimed at producing that short list.

What counts as activity in the first place?

Activity means one of four things — contribution, poll votes, presence, or likes — and they are not equivalent, so getting the definition wrong wastes everything downstream of it:

• Contribution — posts and comments. What most people mean, and the strongest signal.
• Poll votes — low effort but deliberate. Worth counting alongside comments.
• Presence — showing up and reading without saying anything. A completely different state from absence, and the one owners most often collapse.
• Likes — nearly free to give. Weak signal.

Skoolgrades counts engagement as comments plus poll votes, and does not use likes at all. Per-post liker data was deliberately dropped because it added noise without changing a single decision. If you are building your own tracking, do the same.

That choice has a name. Skoolgrades tracks a participation rate — comments, comment replies and poll votes — rather than page loads, so it reads lower than a visit-based engagement rate (the reasoning behind that). Tracking activity is precisely the job that needs the stricter of the two definitions.

There is more on which signals earn their keep in what member behaviors matter.

The other decision, and the more important one: track members against their own baseline, not against each other. Ranking members against each other produces a leaderboard, which is an excellent member-facing game and a poor owner-facing diagnostic. Your steady, quiet, four-year member never appears on it, so when they leave, nothing changes on it.

Can I track member activity by hand?

Tracking member activity by hand costs nothing, starts immediately, and is genuinely fine below roughly 150 members.

1. Open your member list and sort by most recent activity.
2. Scroll to the bottom — the members who have not been around. That is your at-risk pool.
3. Cross-reference against join date. Someone who joined ten days ago and has not posted is an onboarding problem. Someone who joined a year ago and just went quiet is a retention problem.
4. Open the last two weeks of posts and note who is commenting. Compare with last month's names. Anyone who disappeared from that list is worth a message.
5. Keep the list somewhere. Memory is not tracking.

The failure mode is not accuracy, it is consistency. This works when you do it every week and produces nothing at all when you skip three weeks, which everyone does.

Is a spreadsheet good enough for tracking members?

A spreadsheet is good enough for most communities, and what it adds over doing it by hand is memory. One row per member, and columns for the things that do not exist anywhere natively:

The comparison between the last two columns is where the value is. A member who went from twelve comments to one has not "left" by any dashboard's definition, but they have made a decision, and this is the week to notice.

Downsides: it is manual, it decays the moment you stop updating it, and it still knows nothing about what your members do outside your community.

How does automatic member tracking actually work?

Automatic tracking reads your community for you on a schedule, which is what Skoolgrades does, and the mechanism is worth stating precisely because it determines what is possible.

Collection runs through a Chrome extension the owner installs, reading the community on the owner's own logged-in Skool session in their own browser. No Skool API, no partnership, no datacenter scraping — to Skool it looks like an admin browsing their own community, because that is what it is. Skoolgrades is not affiliated with or endorsed by Skool.

Eleven streams are collected: member roster, posts, comments and engagement, DM channel list, notifications, member profile enrichment, cross-community activity, peer data, leaderboards and analytics. Scheduled jobs then build per-member activity heatmaps, synthesize member themes, and materialize 26 cohorts that rebuild weekly.

What that adds over a spreadsheet

Three things a spreadsheet structurally cannot give you.

Baselines computed for you. Activity heatmaps mean rhythm changes are detected rather than eyeballed. You do not have to remember what normal looked like for each member.

Lifecycle states instead of raw numbers. The Cohort Blender sorts every member into star, rising, steady, fading, drifted, ghosted, untapped, or going dark. A state implies an action in a way a comment count never does. "Fading" tells you what to do. "7 comments" does not.

Outside context. Your spreadsheet knows what happens in your room. It cannot know that a silent member is posting daily in three other Skool communities. That distinction is covered properly in what other communities your members are in, and it is the single thing that most often flips a read from "lost cause" to "best opportunity in the list."

Is tracking my own members invasive?

Tracking your own members is not invasive when the tool reads only what is already public plus the community you administer, and two points are worth stating explicitly in case a member asks.

Skoolgrades reads only what is already publicly visible on Skool — public profiles and public communities — plus the community you already administer. It does not read your members' private messages. It does not access anything hidden.

Chat collection is list-only: the DM thread list and unread state, so you can see which conversations are waiting on a reply. No message content, no history backfill.

And the framing matters. This is not monitoring people. It is remembering them. Every owner does this manually for their first thirty members — checks a profile before replying, notices someone has been quiet, remembers what they are working on. Tracking is just that behavior surviving past the point where a human brain can hold it.

What can't activity tracking tell me?

Activity tracking has three honest limits, all worth knowing before you lean on a number.

Cross-community counts undercount deliberately. "Nineteen members active elsewhere" is a floor, not a total. The real number is higher. Certainty is preferred over coverage.

Large communities are sampled. Above roughly 3,000 active members, collection reads a fixed sample of around 900 members instead of a full census, so those figures are estimates rather than exact counts. Use them to rank and decide, not to report precise percentages.

History is shallow. Long-range trend lines and predictive forecasting are not offered, because the honest answer is that there is not enough accumulated history to support them. Withholding a trend line is better than drawing a confident one from three data points.

What should my weekly routine look like?

A weekly member-activity routine takes about twenty minutes and stays the same whichever tracking method you use.

1. Who broke rhythm? Members whose participation dropped against their own baseline. Message these first, and reference something specific they actually engaged with.
2. Who never started? Members past two weeks with no first contribution. They need an easy public on-ramp, not a nudge.
3. Who is newly active? Someone who was quiet and just commented is raising their hand. Reply within the hour if you can.
4. Who is prolific elsewhere and silent here? The untapped list — 25 or more posts elsewhere, two or more communities, seen in the last 14 days, joined 30 or more days ago, silent with you 30 or more days.
5. Send. Where the move is a message, build the list, draft it verbatim or personalized per member, approve every draft, and send from the extension panel one click at a time. Nothing goes out on its own — that is deliberate, because blast sending is how owners get flagged and how members learn to ignore their inbox.

Four lists, twenty minutes, once a week. That beats any amount of staring at a member count.

If you want the same routine framed around people rather than process, start with finding your most engaged members.