Why notes fail as revision material
- They contain the answers. You can't test yourself on a document that shows you what you're trying to recall.
- They're ordered by delivery, not by importance. Lecture order reflects teaching sequence, not exam weighting.
- They're too long to re-read repeatedly, so re-reading becomes skimming, which is recognition, not recall.
- They reward the wrong activity. Re-reading them feels productive and produces very little — the rereading versus retrieval gap.
The four output formats
| Format | Best for | Made from |
|---|---|---|
| Question cards | Facts, definitions, discriminations | Any starred claim in your notes |
| Problem sets | Method-based subjects | Worked examples with the solutions removed |
| One-page topic maps | Seeing structure and connections | A whole topic compressed to a side of A4 |
| Essay plans | Argument-based subjects | Likely questions, answered in five bullet points |
Most subjects need two of these, not one. A pure card deck is inadequate for essay subjects and a pile of essay plans is inadequate for anatomy — pick by what the paper actually asks you to produce.
The ten-minute conversion
- 1
Star the five to eight examinable claims
Do this in the same week, when the lecturer's emphasis is still in memory. Revision-you cannot reconstruct which claims were stressed.
- 2
Turn each into a question with one answer
If the answer is a list, split it into several questions. List-shaped cards are the most common reason decks get abandoned.
- 3
Add one application question per lecture
"Given this scenario, what happens?" This is where the marks are and it's the question type nobody writes enough of.
- 4
Note what confused you as a question too
Your confusions are higher-value than the clean facts, because they're where you'll lose marks under pressure.
- 5
Put them where they'll be scheduled
A file of questions you must remember to open won't be opened. A queue that surfaces them will — see spaced repetition.
Compress before you convert
For each topic, produce one side of A4 that contains the whole thing — the structure, the key claims, the connections. The compression is itself a strong learning activity, because deciding what fits on one page requires knowing what matters.
That page also becomes your final-week revision artefact and, in an open-book exam, your index. Ten pages of A4 for a module is a manageable object; two hundred pages of notes is not — this is the same discipline as making a cheat sheet, used all term rather than the night before.
Don't convert everything
The instinct to make a card for every fact produces a deck of eight hundred items, half of them trivia, which becomes unmanageable by November and is then abandoned entirely — taking the useful half with it.
- Convert what's examinable, judged by past papers rather than by lecture emphasis alone.
- Convert what you got wrong, always. Error-derived questions are the highest-value items you will ever have.
- Skip anything you'd look up in practice. If it's a reference value, it doesn't need to be in your head.
- Skip what you already know cold, and be honest — testing it once to check costs thirty seconds.
Do it weekly, not in revision week
The economics are stark. Converted in the same week, a lecture costs ten minutes and produces questions shaped by what the lecturer stressed. Converted in April, the same lecture costs forty minutes, produces textbook-shaped questions, and is competing with nine other modules for the time.
Ten minutes per lecture across a term is roughly two hours a week for five modules — which is less than most people spend rewriting notes neatly, an activity that produces nothing testable at all. Put it in a fixed slot; the weekly review is the natural home for it.
Test the material before you rely on it
A converted set is a hypothesis about what matters, and it should be checked against a past paper early rather than trusted until March. Sit one paper, then look at how many questions your material would have prepared you for.
The usual finding is that the cards cover the recall marks and nothing covers the long application questions, which is worth knowing in November. How to use past papers covers the wider habit; here it's a calibration check on your own conversion.