8-Week Board Exam Revision Plan (CBSE Class 10 & 12, GCSE, WAEC)

Quick answer: With eight weeks left, stop trying to cover everything. Use week 1 to rank topics by how many marks an hour of revision can realistically win back, weeks 2 to 6 for timed past-paper questions plus a written error log, week 7 for full-length mocks, and the final week for tapering with sleep protected.

I got this wrong the first time. Before my first IGCSE mock season I built a colour-coded timetable, then spent nine hours re-reading history notes because history is my favourite subject and re-reading felt calm. I scored almost exactly what I had before. What finally moved my grades was boring: sit a paper, mark it honestly, write the reason next to every mark I lost. Here is the eight-week version of that, for CBSE, GCSE (AQA, Edexcel, OCR), AP, Cambridge and WAEC/WASSCE students.

Week 1: rank topics by marks per hour, not by fear

Week 1 is not for revising. It is for deciding what you will deliberately ignore. Most students revise the topic that scares them most, which is usually the hardest one worth the fewest marks: the worst trade available.

For each topic get two numbers: how many marks it is typically worth (count it across your last three past papers and divide by three), and where four focused hours could realistically move your score.

My real comparison from Cambridge IGCSE chemistry:

• Electrolysis averaged about 11 marks a paper. I was scoring roughly 4. Four hours got me to 9. Gain = 5 marks over 4 hours = 1.25 marks per hour.
• Organic naming averaged about 6 marks. I was already scoring 5. Four hours might get me to 6. Gain = 1 mark over 4 hours = 0.25 marks per hour.

Same four hours, five times the return. Rank every topic that way and the top eight or so are your eight weeks. Two adjustments: push up anything other topics depend on, because mole calculations, rearranging equations and essay structure leak marks across whole papers; push down anything that has appeared once in five years.

Weeks 2 to 4: papers before notes, with an error log that says why

Do the paper first, then revise what it exposed. Notes-then-test is backwards, because it lets you re-revise what you already knew. Pulling an answer out of an empty head is much harder than recognising it on a page, and the hard version is the one that lasts.

The loop, once or twice a week per subject:

1. Print a real past paper from your board and level.
2. Set a timer for the actual exam length. No notes, no phone.
3. Mark it meanly against the mark scheme. If the examiner would not give it, you do not give it.
4. Next to every lost mark write a reason: did not know the content, knew it but wrote it vaguely, ran out of time, or careless slip.
5. Total the marks in each category. That total is your revision plan.

A real breakdown from one of my papers, 52 out of 80:

• 11 marks: genuinely did not know the content
• 7 marks: knew it, wrote it in my own woolly words, no keyword
• 6 marks: ran out of time, left the last question half-written
• 4 marks: unit errors and one arithmetic slip

Only 11 of those 28 marks needed learning. Seventeen were technique, and technique moves in days rather than weeks. Almost nobody reaches step 4, which is why students revise the same way three years running and get the same grade.

Weeks 5 and 6: drill in the mark scheme's words, not yours

Examiners award marks for specific wording, so learn the wording. It is the fastest grade jump available in eight weeks and needs almost no new knowledge.

In biology, "there is more oxygen outside the cell" often scores nothing where "down the concentration gradient" scores the mark. In physics and WASSCE maths, method marks need the formula stated and the substitution shown, so a bare correct answer can score less than a wrong answer with working. A CBSE 3-mark question wants three separate scoring points, not one paragraph with three ideas buried in it. GCSE level-marked responses want linked reasoning and a judgement, not volume. AP free-response points are earned by naming the concept, not implying it.

So build your cards from mark-scheme phrases, not your own paraphrase, and run them through flashcards daily for the rest of the eight weeks. When a definition comes out in the examiner's words automatically, those marks are banked.

Week 7: one full mock per subject, sat properly

Sit each mock at full length, in one sitting, handwritten if the real paper is handwritten. It only works if conditions are honest: splitting a two-hour paper in half tells you nothing about whether you can still think at minute 95.

Work out your pacing number first, by dividing minutes by marks. A 75-mark paper in 90 minutes gives 1.2 minutes per mark, so a 5-mark question gets about 6 minutes and a 12-mark essay about 14. Write that at the top of the page. Past double the allowance, leave a gap and move on: an unattempted 6-mark question at the end costs far more than an imperfect one in the middle.

Mark it, log the reasons again, and compare category totals with week 2. If "wrote it vaguely" has not shrunk, your wording drills are not working. Out of real papers? the mock exam generator generates another from your syllabus.

The last 10 days: taper instead of cram

Cut volume by about a third in the final ten days and stop learning new topics in the last five. What you gain from cramming a new topic on day 3 is smaller than what you lose by arriving tired, and tired students misread questions.

Those ten days hold short daily passes over mark-scheme phrases and formulas, one or two questions from your weakest topic, and full nights of sleep. Rehearse the practical parts too: what time you leave, what is in your bag, which pens, which calculator. The night before, do one question you know you can do, so the last thing your brain remembers is succeeding.

What AI actually helps with in eight weeks

AI is good for the parts of revision that need a fast second opinion, useless for the parts that need your own nervous system.

• Mark my answer for the "knew it but wrote it vaguely" pile: give it your answer and the question, and it shows you where the answer is thin.
• the quiz generator for extra questions on the topic that keeps costing you marks, once the past-paper questions on it are used up.

What it cannot do: sit still for 90 minutes for you, build hand stamina, replace real past papers, or predict what is on this year's paper. Anyone claiming the last one is guessing.

Test yourself

1. Topic A is worth 12 marks and you score 5. Topic B is worth 12 and you score 11. Which gets your next four hours, and roughly what is the marks-per-hour gap?
2. Why does splitting lost marks into "did not know" and "knew but wrote badly" change what you do next week?
3. Your paper is 75 marks in 90 minutes. Roughly how long should a 5-mark question take?

FAQ

Is eight weeks enough to fix a subject I have barely studied?

Enough to move it a lot, not enough to master it. Pick the six or seven highest-frequency topics, learn those properly, and accept partial marks elsewhere. Spread eight weeks thinly across an untouched syllabus and you half-answer everything.

How many past papers should I actually do?

Fewer papers marked properly beat more papers skimmed. One paper fully marked, with a reason next to every lost mark, is worth about three you skim and file away. Four to six analysed papers per subject is plenty.

What if I fall behind in week 3?

Do not rebuild the timetable. Re-run the week 1 ranking with the time you have left and cut the bottom of the list. A plan amended twice is still a plan; a plan abandoned because it slipped is how people end up revising by vibes.

Should I keep revising subjects I am already strong in?

Lightly, and mostly on technique. English is my strongest subject, and the marks I still dropped were structural, not knowledge, so one timed essay a week with feedback beat any re-reading.

In short

Eight weeks is a resource problem, not a willpower problem. Rank topics by marks per hour, do timed papers before notes, write the reason behind every lost mark, learn the examiner's wording instead of your own, sit honest mocks in week 7, then taper and sleep. Cutting hard is not giving up. It is the only version that fits the time you have.