There is no clever trick to spaced repetition. It is the most-studied finding in the psychology of learning, with results replicated for over a hundred years, and almost every student who tries it properly says the same thing afterwards: “I can’t believe I spent so long re-reading notes.”
This guide will not sell you anything. It walks through the boring science of why spaced repetition works, how to set it up in twenty minutes a day, and the small mistakes that quietly waste people’s time.
The forgetting curve, in one paragraph
In 1885 a German psychologist called Hermann Ebbinghaus memorised long lists of nonsense syllables, then tested himself at intervals to see how much he could remember. His results — replicated thousands of times since — describe a steep early drop in memory followed by a long tail. Most of what we forget is forgotten quickly.
How fast you forget without review
| Time since learning | Approximate retention |
|---|---|
| 20 minutes | 60% |
| 1 hour | 45% |
| 1 day | 33% |
| 1 week | 23% |
| 1 month | 15% |
Those are rough averages, not personal predictions. Some material decays more slowly because it connects to existing knowledge. Some — disconnected facts, vocabulary, dates — decays faster. The point is that all of it decays unless you do something about it.
Why cramming fails
Cramming feels productive because it produces strong short-term performance. You read a chapter the night before an exam, repeat it twice, and walk in feeling confident. The next morning, much of it is gone. A week later, almost all of it is gone.
There are three problems specific to cramming.
Problem 1: high encoding cost, low durability
When you read the same paragraph five times in one evening, the brain treats the second to fifth readings as low-priority information. It is already familiar — you saw it ten minutes ago. The encoding system stops marking it as important. Strong sense of familiarity, weak long-term trace.
Problem 2: no retrieval
Reading is recognition, not recall. You see the words, you nod, you move on. The skill an exam asks for is the opposite: a blank page, a question, and you producing the answer from nothing. Cramming gives you almost no practice at the actual skill.
Problem 3: sleep is doing the heavy lifting
Memory consolidation happens during sleep, especially in the first two cycles after learning. Cramming until 1 a.m. on five hours’ sleep is one of the worst things you can do for retention. The information goes in but is never properly filed.
How spaced repetition fixes the curve
The trick is to test yourself just as you are about to forget — not before, not long after. Each successful retrieval flattens the curve a little more. Five well-spaced reviews of one card, total time around four minutes, will outperform an hour of re-reading the same content.
The spacing intervals that have produced the most consistent research results look roughly like this:
- First review: same day or next day
- Second review: 2–3 days later
- Third review: a week later
- Fourth review: 2–3 weeks later
- Fifth review: 6 weeks later
After five successful retrievals, most material is essentially permanent at GCSE or A-Level intensity.
Three ways to actually do it
You do not need an app. You need a system that prompts you to review the right cards on the right day. Here are the three that work best in practice.
Anki (the digital path)
Anki is the most-used spaced repetition program in the world. It is free on desktop and Android, paid on iPhone (the developer subsidises everything else with the iPhone fee). The algorithm calculates each card’s next review date based on how easily you recall it. If a card is hard, it comes back tomorrow; if easy, in two months.
Strengths: zero scheduling overhead, syncs between devices, huge community decks for languages and medical school. Weaknesses: easy to make bad cards (long, vague, two answers in one), can become a wall of reviews if you miss a few days.
Leitner boxes (the paper path)
You write each fact on an index card and start it in Box 1. Each day you review Box 1; correct cards graduate to Box 2. Box 2 is reviewed every two days. Box 3 every four. Box 4 every week. Wrong answers, at any stage, drop back to Box 1.
Strengths: no screen time, very visible progress, hard to fake. Weaknesses: physical clutter, no syncing, you carry index cards around.
A timetable on paper (the lo-fi path)
Make a simple grid for the next eight weeks. For each topic you cover, write its name in today’s column, then again in three days, a week, two weeks, six weeks. Each evening, look at today’s column and rehearse the topics by closed-book recall.
Strengths: no setup beyond a pen. Weaknesses: you have to be honest about closed-book testing, easy to skip.
The 20-minute daily habit
Most students who succeed with spaced repetition do twenty minutes a day, every day, no exceptions. Twenty minutes is short enough to be defensible after a long school day. Daily is non-negotiable — if you skip, the queue grows, and the queue growing is the single biggest reason people quit.
A reasonable session structure:
- 2 minutes: review the cards or topics due today
- 12 minutes: closed-book retrieval (no notes, just attempt the cards)
- 4 minutes: mark each as easy / OK / hard
- 2 minutes: schedule any new cards you added today
Once a week, spend an extra ten minutes reviewing the cards you flagged hard. Those are the ones that need to migrate from short-term familiarity to genuine recall.
A study schedule template
If you have eight weeks until an exam and ten topics to cover, here is a workable schedule.
Weeks 1–4: cover one new topic per three days. After each new topic, add 8–12 cards or facts to your review system. Daily 20-minute review of accumulated cards.
Weeks 5–6: stop adding new cards. Continue daily review. Begin one past paper a week, mark it honestly, and add cards for any errors.
Week 7: two past papers, both timed. Review only.
Week 8: one final past paper at the start. Daily review tapering off. Sleep.
The point of stopping new cards two weeks out is not to look busy in the final stretch — it is to let the existing material consolidate.
Common mistakes
Cards that are too long. A card with three sub-questions is three cards in disguise. Split them.
Recognising instead of recalling. If you read the question and immediately flip to check, you have not retrieved anything. Force a guess every time, even if it is wrong.
Marking everything “easy”. This pushes intervals too long, and the card drops out of view at exactly the wrong time. Be honest.
Skipping a day, then skipping a week. The queue grows, you feel buried, you quit. Do five minutes on the bad days. Anything is better than nothing.
Adding too many new cards on one day. Twenty new cards added on a Sunday will all return on Tuesday and Wednesday. Spread the load.
Frequently asked questions
How long until I notice it working?
Two to three weeks. The first week feels like a lot of admin for no reward; the second week, you realise you remember things you would have forgotten by now; by week three, the schedule does the work and you stop thinking about whether it is worth it.
Is it worth using Anki for essay subjects?
Partly. Use cards for facts, dates, quotations, formulas — anything with a single right answer. Do not try to put essay arguments on cards; use a different technique (planning past questions, building a paragraph bank) for that. Spaced repetition is one tool in a larger study kit.
Should I make my own cards or download a deck?
Make your own for at least the first month. The act of creating a card forces you to identify what the actual question is, which is half the learning. Once you have a working habit, you can pull in shared decks for vocabulary or anatomy where the standard set is well-known.
What if I miss two weeks?
Do not try to clear the backlog in one evening. Set a smaller daily goal — say twenty cards a day from the overdue pile — until the queue is back to normal size. Better still, lower the daily new-card limit so the system feels light again.
One-on-one tutoring sits well alongside spaced repetition — the cards keep the facts in, a tutor closes the method gaps. Find a tutor on TheTutorLink when you want both.