Day 0: two hours of triage before any revision
Do not open a textbook yet. Spend the first two hours building a map, because every hour after this one is spent according to it.
- 1
List every topic on the syllabus, one line each
From the module outline, not from your notes — your notes only contain the lectures you attended. Twenty to forty lines is normal.
- 2
Score each topic out of 5 for how likely it is to be examined
Evidence, not vibes: how many past papers featured it, how many lecture hours it got, whether the lecturer said "this is important". Weight past papers heaviest — see how to use past papers.
- 3
Score each topic out of 5 for how well you know it
Be brutal. "I attended the lecture" is a 1. If you cannot explain it out loud right now, it is at most a 2.
- 4
Rank by (likelihood × gap) and draw a line
Multiply likelihood by (5 minus your knowledge). Sort descending. Draw a hard line under whatever fits in 30 hours — realistically 10 to 14 topics. Everything below the line is abandoned, deliberately, in writing.
- 5
Write the abandoned list somewhere you'll see it
This sounds like a formality and is the load-bearing step. Unabandoned material comes back at 11pm on day 5 as guilt and eats the evening you'd planned for the topics that actually matter.
The seven days
| Day | Focus | The output that proves you did it |
|---|---|---|
| 1 | Top 3 ranked topics. First pass: read once, then self-test. | One blank-page brain dump per topic, plus a question list for each gap. |
| 2 | Next 3 topics, same method. Review day 1's gaps first (20 min). | Three more dumps. Day 1 gaps re-tested and mostly closed. |
| 3 | Next 3 topics. Review days 1–2 gaps first (30 min). | First full past paper attempt, open-book, untimed, in the evening. |
| 4 | Remaining topics above the line. Mark day 3's paper properly. | A marked paper with your losses categorised: knowledge, technique, or time. |
| 5 | Second past paper, closed book, strictly timed. Then fix what it exposed. | A score, and a list of the three topics it proved you still can't do. |
| 6 | Those three topics only. Plus one more timed paper if there's time. | Everything above the line has been tested at least twice. |
| 7 | Review only. No new material. Finish early, sleep normally. | One pass through every brain dump and every gap card. Stop by 8pm. |
Why past papers move from day 6 to day 3
The instinct is to revise first and test at the end, when you'll do better. That gets it exactly backwards. A past paper on day 3 is a diagnostic — it tells you which of your remaining four days is being spent on the wrong thing, while you still have four days. A past paper on day 6 is a score you can no longer act on.
Sit the first one open-book if closed-book would produce a blank page. The purpose is not the mark; it is discovering that question 4 requires a technique you have never practised.
How to actually spend a revision hour
Each topic gets roughly two hours. Split them like this, and note how little of it is reading:
- 15 minutes reading the notes or chapter, once, with a pen. Not three times.
- 20 minutes blank-page dump — write everything you know about the topic without looking. This is the actual revision.
- 10 minutes checking against the source, marking only the gaps.
- 45 minutes on questions — past-paper questions on that topic, or generated ones if you've run out.
- 10 minutes converting the gaps into questions you'll re-test tomorrow morning.
If you find yourself re-reading for an hour because it is comfortable, you have reverted to the illusion of fluency and the week is being spent. The discomfort of the blank page is the signal that it's working.
Sleep, and the day-7 temptation
Every hour you take from sleep this week costs you more than it gives, with one exception: if the exam is tomorrow and you have genuinely not covered a major topic, an extra two hours tonight is worth it. That exception does not apply on days 1 through 5, where lost sleep degrades tomorrow's encoding and quietly makes the whole week less productive.
Day 7 should end early. Reviewing until midnight the night before feels responsible and reliably costs more marks in exam-hall attention than it adds in recall — the arithmetic in all-nighters is the same one day out.
If you have three days, not seven
Same method, harsher line. Rank the topics, take the top five, and run days 1, 3 and 5 of the plan above compressed: triage, two topics a day with testing, one timed past paper on the middle day, review on the last. Do not attempt breadth. Five topics known well outscores fifteen topics recognised, on almost every mark scheme ever written.