A term calendar with the review days circled, which is the entire scheduling system this method needs.
The finding behind spaced repetition is old and unusually robust: material reviewed at increasing intervals is retained far longer than material reviewed in a block, for the same total minutes. Reading a chapter four times on Sunday and reading it once on Sunday, Wednesday, the following Monday and three weeks later cost the same and produce very different results in June.
The finding is not the hard part. Applying it for eleven weeks is. Software exists to handle the scheduling, and it works, but it introduces a failure mode that paper does not have: the day you miss becomes a backlog, the backlog becomes four hundred due cards, and the app becomes a thing you avoid opening. A paper system degrades more gracefully, because a missed day just means the box is fuller on Thursday.
The three box system
Get three containers. Shoeboxes, envelopes, elastic bands round three piles, anything that keeps them separate and visible. Label them daily, weekly and monthly. Every new card starts in the daily box.
The rule has one clause. A card you answer correctly moves one box forward. A card you get wrong goes back to daily, wherever it was. That is the whole algorithm, and it is deliberately cruder than what software would compute, because a rule you can apply while tired is worth more than an optimal one you stop applying.
Then decide the days, once, and write them on the calendar. Daily box every weekday. Weekly box on Saturday morning. Monthly box on the first Sunday of the month. Two hundred cards in the daily box is fifteen to twenty minutes, which is a bus journey. The weekly session runs half an hour by mid term. The monthly session is the longest and the most useful, because it is where you find out what has genuinely stuck.
The daily box fullest, the monthly box thinnest, which is what the system looks like after a month of use.
What goes on a card
Card quality decides whether the schedule is worth running at all. A card carries one question with one answer. Two facts on a card means you can be half right, and half right is unscoreable, so the card cannot move anywhere. Cards built from a bullet list with six items produce the same problem: the answer either comes in full or the card fails, and it will fail every time.
Write the question so the answer has to be produced, not recognised. The full case for that phrasing, with examples of fronts rewritten from weak to strong, is in flashcards that ask harder questions. Cards made from your own confusion are the best ones you will own, which is why the marked gaps from the one page summary, built while the lecture is fresh are the natural source.
Answer out loud or on paper before turning the card. Looking at the front, thinking "yes, I know that" and flipping is not retrieval, and it is the single most common way this system quietly stops working while appearing to run normally.
A row of ticked squares, the only progress record the system keeps.
Keeping it alive past week four
Most paper systems die from volume. Thirty cards a week per module across five modules is a hundred and fifty new cards weekly, and by November the daily box takes an hour and stops fitting into a day. The fix is a cap: no more than fifteen new cards per module per week, chosen from what you actually got wrong rather than from what the syllabus contains. Fifteen cards you failed are worth more than sixty you copied.
The second killer is the missed week. It happens to everyone, usually around a coursework deadline. Do not attempt to catch up by clearing everything at once. Shuffle the daily box, take the top forty cards, run them, and carry on the normal schedule the next day. The material you did not reach was already spaced further than intended, which is a much smaller problem than an abandoned system.
Three boxes, two clauses, one calendar:
- Right moves the card forward one box. Wrong sends it back to daily, from any box.
- The answer is spoken or written before the card is turned, every time.
- New cards are capped per module per week and come from what you failed, not from the reading list.
The limits deserve to be stated plainly. Spacing is a retention tool, not a comprehension tool. A card about a mechanism you never understood will be memorised as a string of words and will be useless the moment the exam rephrases the question. Understanding has to come first, from the lecture, the reading and the worked example.
Spacing also does not train performance. Knowing every card in the box does not mean you can plan and write an essay in forty minutes, or work through a six part problem under a clock. Those are separate skills with separate practice, and the practice is described in past papers, timed and marked honestly. A term of perfect card review followed by a first attempt at a full paper on the morning of the exam is a familiar and avoidable mistake.
Finally, the boxes need somewhere to live and something to sit on. If your desk is shared or your room changes weekly, an envelope system in a bag will survive where shoeboxes will not. The physical form is negotiable. The two clause rule is not.