Why can my student hold a conversation in class but freeze the moment we hang up? The hour on video isn't where fluency gets built; it's where fluency gets rehearsed. What a student can say next Tuesday is decided by what they did on Thursday, Saturday, and Sunday morning while nobody was watching.
Online teachers feel this pressure more sharply than in-person ones. There's no hallway chat, no textbook left open on a desk, no classmate to compare notes with. Between sessions, the student is alone with a phone. That's the terrain this piece is about, broken down by the kinds of homework problems that actually show up, and what each one calls for.
The Student Who Does the Homework and Still Can't Speak
This is the most common case, and the most misread. The student finishes every worksheet, hits every streak on their app, and shows up prepared. Then you ask an open question and they stall.
Effort isn't the issue; the practice they're doing is receptive rather than productive. Reading a dialogue, matching words to pictures, tapping the right translation, these all feel like study. They build recognition.
Speaking is a different muscle. Merrill Swain's output hypothesis has been arguing for decades that learners need to actually produce language, out loud, under a little pressure, before it becomes available to them in real time. If the homework never asks for output, the lesson is doing all the productive work in a single hour a week. That's not enough.
The fix is boring and it works. Assign speaking tasks between sessions, with a way to hear them back: a one-minute voice memo answering a specific prompt, a read-aloud of a paragraph they wrote, or a recorded reply to a question you sent Tuesday morning. Polish isn't the goal; reps are.
Teachers who want to sharpen the craft side of this can read a useful primer on how to get better at teaching a language remotely.
The Student Who Studies Hard the Night Before
Cramming is the second failure mode, and video lessons hide it well. The student blocks off Sunday evening, powers through the deck, arrives on Monday sharp, then forgets most of it by Wednesday. On a webcam you rarely see the collapse. You just notice, three months in, that nothing is sticking.
The research on this is old and unusually settled. Spaced practice, the same total study time broken into shorter sessions across several days, produces meaningfully better retention than a single long session. A recent study with adult EFL learners found gains in both vocabulary use and speaking fluency when retrieval was spaced across days rather than piled into one sitting.
For an online teacher, this changes what you assign, not how much. Three ten-minute touches beat one thirty-minute block. Send a short prompt Tuesday, a review card set Thursday, a recorded answer due Saturday. Small, frequent, unavoidable.
The Student Who Downloaded Five Apps and Uses None of Them
Tooling has a way of becoming the problem it was supposed to solve. Every new platform adds a login, a notification pattern, a place things live. By the third one, the student is spending more energy managing tools than using the language.
Pick a small stack and defend it. A rough working set for most online teachers:
- One place for live sessions. Whatever video tool you already use well. Don't switch for features you won't touch.
- One place for assignments. A shared doc, a simple LMS, or even a running email thread. The student should never wonder where homework lives.
- One place for spaced review. A flashcard app the student actually opens counts more than the best one they don't.
- One place for voice. A single channel where they send recordings and you send corrections back.
If a tool doesn't earn its place in that list, cut it.
Designing Homework That Fits a Real Week
Adult learners online carry the whole load of scheduling their own practice, and the ones who succeed are the ones who self-regulate well. Homework design either helps them do that or expects them to be exceptional. Assume they're not exceptional. Assume they're tired on Wednesday.
A between-session plan that survives contact with real life tends to share a few traits:
- Short and daily beats long and weekly. Ten minutes a day is a habit. Ninety minutes on Sunday is a chore that gets skipped.
- Every task has an output. Something the student says, writes, or records. Reading alone doesn't count.
- One prompt, not a menu. Choice fatigue kills follow-through. Tell them the exact thing to do.
- A visible finish line. They should know when they're done. Vague homework never gets done at all.
The lesson hour is the smaller part of the job. Design the six days around it and the hour tends to start working the way you meant it to.