Learning platforms

Completion is the product, and it is mostly a design problem rather than a content one. Learners abandon at predictable points: when the next step is unclear, when a module is long enough to require a decision to start it, when returning after a week means not knowing where they were, and when there is no reason to come back on a particular day. Progress that is visible and granular, short units, and a resumption that lands exactly where they stopped do more for completion than any amount of production value.

Cohorts outperform self-paced courses on completion by a wide margin, and the reason is social rather than pedagogical: a start date, other people at the same point, and something happening at a fixed time create obligation that an always-available library never does. That has real product consequences — scheduling, live sessions, discussion tied to a specific week — and it is worth deciding early, because a platform built for self-paced study does not become a cohort platform by adding a calendar.

Assessment is where credibility is won or lost. If a certificate is going to mean anything to an employer, the assessment behind it has to be harder to game than a multiple-choice quiz with a retake button — which means question banks, some form of identity assurance, and a considered position on what you do about the fact that a model can now answer most written questions. That last one is not a future concern; it is the current state, and a platform that has not thought about it is issuing certificates that quietly mean less each year.