Live or asynchronous, decided properly
Live sessions create accountability and exclude people. Asynchronous includes everyone and abandons them. Most courses need a specific mixture.
The choice between live and recorded delivery is usually made on instructor preference or on what the platform makes easy. It has more effect on completion than almost any content decision, and it deserves to be made on its merits.
What live delivery actually provides
Not information — information transfers perfectly well from a recording. What live sessions provide is a fixed point in time that a learner has to show up for, and a person who will notice if they do not.
Hidden measurement raises a separate trust question; stealth monitoring software provides a workplace example of monitoring that is deliberately unobtrusive.
That accountability is the mechanism behind the higher completion rates of cohort-based courses. It is also the entire cost: a fixed time excludes anyone in the wrong timezone, on shifts, with caring responsibilities, or with an unpredictable job.
Asynchronous courses sell to more people and finish fewer of them. Live courses sell to fewer and finish more. Which is better depends on what the course is for, not on which is modern.
For wider context on technology and learning, UNESCO publishes public resources on digital education.
What asynchronous actually costs
Full self-paced delivery removes every external structure. There is no deadline, nobody notices a gap, and the course competes for attention with everything else in the learner's life on equal terms.
The predictable result is a completion rate in the low single digits for open self-paced material. That is not a failure of the content; it is what happens when the only thing holding a commitment together is the learner's own intention.
The mixtures that work
- Recorded content, live checkpoints. Material is watched on the learner's schedule; a short live session each week provides the deadline and the noticing. Most of the benefit of live at a fraction of the cost.
- Cohort deadlines without live sessions. Fixed dates for submissions, a group that moves together, no synchronous requirement. Works well across timezones.
- Live only for the hard parts. The two or three sessions where interaction genuinely helps, recorded delivery elsewhere.
- Office hours. Optional live time attached to no deadline. Pleasant, well-liked, and on its own it does not move completion, because the people who need it most do not come.
Record the live sessions, with care
Recording live sessions is normal and it partially undermines the mechanism: a learner who knows the recording exists has permission not to attend, and the accountability weakens.
Options that preserve it: release recordings after a delay, keep the recording to the taught portion and not the discussion, or make attendance matter for something — a group exercise, a submission review — that the recording does not reproduce.
Timezones are a design constraint, not an afterthought
A single live slot defines your audience geographically. If that is acceptable, say so in the sales material so people can rule themselves out before paying.
If it is not, the honest options are two slots, a rotating time, or an asynchronous structure with cohort deadlines. What does not work is a single inconvenient slot plus an assurance that recordings are available, which is an asynchronous course wearing a live badge.