Two structural decisions shape everything about how Poplearn courses behave: how the material gets divided, and what the questions are for.
Why modules are about six minutes
The threshold that matters isn't runtime, it's whether finishing requires a decision.
Above roughly ten minutes, a module competes with everything else in someone's day — and loses, reliably, not because it isn't important but because everything else is more immediate. It has to be scheduled, and anything scheduled gets deferred.
Below it, completion happens in the gap between two meetings without deliberation. Six minutes sits comfortably inside that.
The counterintuitive result is that splitting one twenty-four minute module into four six-minute ones raises total completion even though the total content is identical. The first version asks someone to find a twenty-four minute block; the second never asks them to find anything.
How the split is decided
The split follows the document's own structure. Sections in a well-written document are usually conceptually complete, so each becomes a module that ends somewhere that feels finished rather than mid-thought.
This has a direct consequence: heading quality in your source drives module quality in the output. If section titles are real heading styles, the split lands where you'd expect. If they're bold text at a larger size, they look like headings to a human and like ordinary paragraphs to anything automated, so the split has less to work with.
Checking the outline view of your document before converting is the highest-return preparation step available.
Modules should stand alone
Because modules are delivered one at a time and completed separately — sometimes days apart — each needs to make sense without the previous one fresh in mind.
After generating a course, scan for modules opening with "as described above" or "building on this". Someone who did the previous module last Tuesday won't have that context. Either add a sentence or merge the two.
This is a five-minute pass and it noticeably improves how a course reads in practice.
What the quizzes are for
A quiz converts "was sent" into "engaged with". That's the whole job — it's what makes a completion record mean something rather than recording that a message was delivered.
Poplearn writes questions for each module as part of /courses new. They're worth reviewing, because generated questions are the most variable part of any conversion.
Look for the obvious-answer problem. If one option is clearly the responsible-sounding corporate choice, someone who skipped the module still gets it right. Distractors need to be plausible — things a reasonable person might actually do.
Prefer decisions over wording. "According to section 3, what is the threshold?" tests document navigation. "A supplier asks for same-day payment above your limit — what do you do?" tests the behaviour you actually care about.
Keep it short. Two to four questions per module is right. Longer assessments push people into guessing and clicking through, which measures nothing.
Spacing comes for free
Because modules arrive one at a time over days rather than all at once, the material is naturally spaced rather than crammed.
Spacing is one of the more robust findings in how people retain things, and here it's a property of the delivery rather than something you have to design. A course completed across a week sticks better than the same course completed in one sitting.
Delivery
Modules arrive as Slack direct messages. The person reads the module there, answers the quiz by clicking inside the message, and completion is recorded — no portal, no password, no link out.
Because it's a DM, it's private: nobody sees what someone is being trained on or how many attempts a question took.
Add Poplearn to Slack — free under 20 seats, installs in about a minute.
