Most training tracking is manual somewhere, even when there's a real system involved. There's usually a spreadsheet, because the system reports the wrong thing.
/progress is built around what people actually ask for.
What it returns
Live completion by team, exportable on demand.
Two properties matter more than the format. It's live, because completion is recorded at the moment a module is finished rather than synced later — the module and the record are the same event, so there's nothing to reconcile. And it's per-person and timestamped, not an aggregate.
Why per-person matters
A dashboard showing 94% answers a management question: how are we doing. It doesn't answer either of the questions people actually turn up with.
Compliance asks: did this named person complete this specific policy, and when? That needs a name, a course, a timestamp, and an export — because the request usually comes from someone without access to your tools.
A manager asks: who on my team is behind, and what do I say to them? That needs names. "3 of 8 incomplete" is a prompt to go and find out who, which is friction that stops most managers acting.
An aggregate percentage satisfies neither, which is why the spreadsheet exists in most organisations.
The test worth running
Pick a person and a policy at random. Time how long it takes to produce a record showing they completed it and when.
If it takes more than a minute, or involves asking someone, the tracking isn't doing its job. Much better to fail that test on a Tuesday than during an audit.
Check the population too
The most dangerous number in training reporting is 100% completion of an incomplete population.
If enrolment is manual, someone who joined during a busy month was never assigned, so they're not in the denominator. Completion reads perfect while a real gap exists — and nothing looks wrong, which is what makes it dangerous.
/assign targeting a channel or everyone who joins from a date forward keeps the population correct without maintenance. When you check /progress, confirm the people you expected are actually in it.
Where people stopped
Binary completion — done or not done — hides the useful diagnostic.
Someone who started and stalled at module four is in a different situation from someone who never opened it. The first suggests module four is too long, unclear, or assumes something people don't have. The second suggests the assignment or delivery didn't reach them.
Those need opposite responses. If four of five people stalled at the same module, that's a content problem to escalate rather than a discipline problem to chase.
Push it to managers
Central chasing is slow and unpleasant, and the person doing it usually has no relationship with the people they're chasing.
/progress runs in Slack, so a team lead can get their own team's picture in a few seconds without an account on anything else and without leaving the tool they already work in. A manager who can see three of their eight people are behind will usually close it with one message — faster than any escalation path, and with far less friction.
That only happens if the number is somewhere they already are, which is the argument for it being a Slack command rather than a web dashboard.
Retention
Where training recurs annually, keep the historical records rather than overwriting them. An auditor asking about coverage for a specific past period needs the record for that period, not the current one.
Add Poplearn to Slack — free under 20 seats, installs in about a minute.
