Semestree
Sign inStart free

Spacing out your review, without building a whole system

The same total hours produce very different results depending on when you spend them. Here's the college version, without the apps and the algorithms.

·4 min read

Two people spend four hours on the same material.

One spends all four the night before. The other spends one hour a week for four weeks, then walks in having done nothing that day.

Same hours. Not the same outcome, and it isn't close.

All at once, the night before 4 hours exam Everything arrives at once and starts leaving immediately. It works often enough to keep people doing it. Same four hours, spread out exam Each session starts with partial forgetting, which is uncomfortable and is exactly the part that makes the next one stick.
Same four hours either way. The gaps are doing the work, not the hours.

Why the gaps do the work

The intuition most of us have is that memory is a bucket you fill, and that the goal is to pour in as much as possible before the exam.

It's closer to a muscle. What strengthens it isn't exposure, it's effortful retrieval — and you can only retrieve something effortfully if you've partly forgotten it first.

Which means the forgetting between sessions isn't damage. It's what creates the opportunity. When you come back three days later and struggle to remember the four assumptions, that struggle is the whole mechanism. Cram everything into one night and you never let anything slip far enough to have to pull it back.

This has an unpleasant corollary. Spaced review feels worse than cramming, every single session, because you keep starting from a position of not-quite-remembering. Cramming feels great, because everything's fresh, and it stays fresh for approximately as long as it takes to hand in the exam.

The version that fits an actual semester

You don't need software for this. You need a habit that costs about ten minutes.

lecture Tue 10 min same day 8 min 2-3 days 6 min a week 5 min before exam One week of material, revisited four times Under thirty minutes total, spread across three weeks. Each return is shorter because less has leaked out. The exact intervals do not matter much. What matters is that there is a gap, and that you try to recall before you look.
Four short returns beat one long one, and they add up to less total time.

Same day, ten minutes. After a lecture, close the notes and write down what it was about. Not a summary of the notes — a recall of the lecture. This is the highest-value ten minutes in the whole schedule, because it's when the difference between "I was in the room" and "I have it" gets settled.

Two or three days later, eight minutes. Look at the cue words or the headings, try to reproduce the content, then check.

About a week later, six.

Once more before the exam, five.

Under thirty minutes per week of material, spread across three weeks. Compare that to what you actually spend re-reading the same chapter the weekend before.

The intervals are not sacred. What matters is that there's a gap, and that each visit starts with an attempt to recall before you look.

The rolling version, which is what I actually do

Scheduling per-lecture reviews is more bookkeeping than most people will sustain. Here's the low-effort form.

Once a week — same slot, mine is Sunday evening — do this:

  1. This week's material. Try to recall it, cold. Check. Note what didn't come back.
  2. One older week. Rotate through them. Week 1 this Sunday, week 2 next Sunday, and so on. When you run out, start over.
  3. Anything on the missed list, regardless of age.

That's it. Forty minutes a week per course, and it means nothing goes untouched for more than a few weeks. By the time finals arrive, you're reviewing rather than learning, and that difference is roughly the entire experience of finals week.

What to do with the flashcard apps

Anki does real scheduling — every card gets its own interval based on how you did last time. It's free on desktop and Android and it's the serious option. It also has a learning curve and it rewards people who like tinkering with settings, which is a genuine risk if you are the kind of person who will spend an evening on the settings.

Quizlet is the one everyone already has. Worth knowing: it removed its true spaced-repetition mode in 2020, and Learn and Test moved behind a subscription in 2022. It's still a fine tool for drilling a set of terms. It is not, in its current form, a spacing system, and people assume it is.

Honestly, for most college courses, a piece of paper works. The mechanism is the gap and the retrieval attempt, not the software. The software matters when you're carrying thousands of cards for a year — the reason medical students live in Anki — which is not most undergraduate courses.

Where spacing doesn't help much

Two honest limits.

Very close to the exam, spacing has nothing left to work with. If it's Thursday and the exam is Saturday, you don't have room for intervals. That's a different problem with a different answer.

Spacing is about retention, not understanding. If you never understood the derivation, reviewing it four times spreads out the confusion. Fix comprehension first — that's what office hours and explaining-it-out-loud are for — then space the retention.

The part of this that fails in practice is never the studying. It's remembering which week you're supposed to be rotating back to, three weeks after you stopped caring about the system. That bookkeeping is what Semestree does with material you've uploaded — it keeps the review list and surfaces what's due rather than making you track it. The ten minutes of actual recall is still yours, but the "what do I review today" question stops costing anything.

If you take one thing

Do the ten minutes on the same day as the lecture.

If you never do anything else on this list, do that. It's the cheapest, it's the one with the largest effect, and it's the one you can do on a bus. Everything else here is refinement.

Keep reading

All guides