"Microlearning" became a buzzword faster than it became a discipline. Chopping a 45-minute course into nine 5-minute videos isn't microlearning; it's the same course with more scrollbars. Real microlearning is a design decision about how retrieval and spacing work, not just a runtime target.
What actually makes it stick
- One decision per module, not one topic per module. A module that teaches "how to de-escalate an angry customer call" beats a module that teaches "our de-escalation policy": the first ends in a decision, the second ends in a definition.
- Retrieval, not review. Asking learners to recall and apply something cold, even if they get it wrong the first time, builds retention far better than presenting the same fact a second time in a nicer format.
- Deliberate spacing. The forgetting curve is real. A single 5-minute module has almost no advantage over a 45-minute one if it's never revisited. The spacing between exposures, not the length of any single exposure, is what drives long-term retention.
- A reason to open it that isn't "it's due." Microlearning tied to a real, current task ("here's how to handle the objection you just got on a call") gets opened voluntarily. Microlearning tied only to a compliance deadline gets opened at 11:58pm on the due date.
Microlearning is a spacing strategy wearing a runtime constraint as a costume.
A pattern that works
The most durable microlearning programs I've built follow a simple shape: a short scenario-based module introducing a single decision point, followed by two or three spaced "retrieval nudges" over the following weeks: quick, low-stakes questions that ask the learner to apply the same decision again in a slightly different context. The initial module does less work than people expect. The spaced nudges do more.
If a microlearning library isn't producing measurable behavior change, the fix is rarely "make it shorter." It's usually "make it come back."