UKMLA Revision Timetable: The Arithmetic, Not a Template
A timetable is three inputs — your date, your honest hours, and each specialty's blueprint weight — converted into questions per week. The conversion we use (60 seconds a question, from a measured 46-second median), why the published weights need normalising before you use them, and two worked examples: 12 weeks at 10 hours buys 45% coverage, 8 weeks at 6 hours buys 17% and defers 13% of the blueprint.
The short version: a UKMLA revision timetable is arithmetic, not a template. Three inputs decide it — your exam date, the hours you'll genuinely do, and each specialty's weight in the GMC blueprint — and the output is a number of questions per specialty per week. The near-universal mistake is splitting time evenly across specialties, which hands Breast the same week as Heart & Vasculature despite the exam weighting them roughly 1–2% against 10–14%.
Below is the whole method, including the maths we use ourselves. If you'd rather not do it by hand, the generator at the end does exactly this and emails you the grid.
Step 1 — Work backwards from the date, not forwards from today
Count the weeks between today and your exam and build the grid backwards from the exam. Forwards-planning produces a timetable that runs out of road in the final fortnight, which is precisely when you need slack for mocks.
Two practical rules:
- Round up, not down. Someone 13 days out has two weeks of runway, not one. Flooring costs them six days.
- Reserve the last week. More on that below, but it isn't a study week.
If you don't know your date yet — common for UK students until the school publishes its assessment calendar — build to a placeholder and rebuild when it lands. Don't wait: the first weeks of any plan are broad coverage, which doesn't change with the date.
Step 2 — Budget honest hours, then convert them to questions
This is where most timetables die. People plan for their best week and then miss five days out of seven, so by week three the grid is fiction and gets abandoned.
Plan for a bad week. If you can reliably do six hours, plan six — a plan you beat is motivating; a plan you fail is abandoned.
Then convert hours into work, because "revise cardiology, Tuesday 6–8pm" is not a plan, it's a wish. Here's the conversion we use, and where each number comes from:
| Quantity | Value | Basis |
|---|---|---|
| Time per question | 60 seconds | Our users answer at a median of 46 seconds (1,628 in-session answers; 26s at the 25th percentile, 87s at the 75th). 60s budgets the median plus reading the explanation. |
| Time per flashcard | 20 seconds | Steady-state review pace, not first-encounter. |
| Share of time on questions | 60% | |
| Share on flashcards | 15% | |
| Remaining | 25% | Notes and reading around what you got wrong. |
| Share of question time on review | 20% | Re-attempting previous incorrects, not new questions. |
Ten hours a week therefore buys roughly 360 questions and 270 flashcards, of which about 72 questions are revisiting things you already got wrong. That last line is not optional. A timetable of only new questions is a timetable for forgetting.
Step 3 — Allocate by blueprint weight, not evenly
Every specialty in the content map carries a published weighting — Heart & Vasculature at 10–14%, five areas at 6–8%, a tail at 1–2%. Your timetable should mirror that.
One wrinkle worth knowing if you're doing this by hand: the published weights are ranges, and their midpoints sum to about 122%, not 100. You can't use them as percentages directly. Normalise them — divide each midpoint by the total — before turning them into hours or questions. Skip this and your grid will over-allocate everything and run out of weeks.
The practical consequence is stark. At a proportional allocation, the heaviest specialty earns several times the question volume of the lightest. If your current timetable gives each specialty a tidy equal block, it is systematically under-preparing you for the largest part of your paper.
Step 4 — Interleave presentations; don't block them
The content map has clinical areas and presentation groups — chest pain, breathlessness, the abdominal presentations. They aren't a specialty to schedule on a Thursday; they're how the exam frames its questions.
In our own bank the 14 presentation and skills groups hold 2,164 of 7,198 SBAs — about 30%. So roughly three in every ten questions you do each week should be presentation-framed, mixed in rather than saved for a block at the end. A timetable that leaves presentations until week ten is leaving 30% of the practice until week ten.
Step 5 — Take a taper week
The last week before the exam introduces nothing new. No new specialty, no new resource, no first-encounter cards. It's retrieval of what you have, light timed blocks to keep pacing warm, and sleep.
This is not a wellbeing platitude. Material met for the first time in the final week hasn't been through a single review cycle, so it's the least likely to be retrievable under pressure, and it displaces revision of things that would have been.
Two worked examples
Both generated by the same builder behind our free plan, so the numbers are real rather than illustrative.
Twelve weeks, ten hours a week. 120 hours total, which buys about 3,262 questions — 45% of a 7,198-question bank. No specialty gets dropped. A typical week is 252 condition questions, 108 presentation questions and 270 flashcards, with the week's focus weighted to the heaviest areas. Week twelve is the taper.
Eight weeks, six hours a week. 48 hours, about 1,240 questions — 17% coverage. At this budget the plan cannot cover everything, so it defers the seven lowest-weighted areas — Breast, Genetic & Congenital, Other Knowledge, Drugs & Drug Effects, Male Genitourinary, and the rest of the 1–3% tail — and tells you plainly that they represent 13% of the blueprint.
That second example is the honest one. A timetable that pretends 48 hours covers 694 topics isn't a plan, it's a comfort object. Knowing you've consciously set aside 13% of the paper is a decision; discovering it in the exam is an accident.
When you fall behind — and you will
Every timetable slips. The good ones slip gracefully, and that means having the triage rule decided in advance:
- Cut from the bottom of the weighting, never the top. Dropping a 1–2% area costs you a fraction of what dropping a 6–8% area does.
- Cut new content before you cut review. Reviewing 200 questions you've done beats meeting 400 you'll never see again.
- Never cut the mocks. They're the only instrument telling you whether the plan is working. How to use them.
- Rebuild the grid, don't try to catch up. A plan you're four weeks behind on has stopped being a plan. Recalculate from today's date and today's honest hours.
Or let it build itself
All of the above is deterministic, which is why we automated it. The free revision plan takes your exam date, your weekly hours and your pathway, and emails a week-by-week grid: specialty focus weighted by the blueprint, question and flashcard targets per week, mocks scheduled, a taper week, and an explicit list of anything it had to defer with the share of the exam that represents. No account, and it takes about thirty seconds.
Related, if what you're really asking is a different question: how long you need before you start, the fixed 12-week plan if you want one already laid out, or the four-week intensive if that's your runway.
Frequently asked questions
How many hours a week do I need for the UKMLA? There's no universal figure — it depends on your runway and starting point. What matters more is that the number you plan is one you'll actually hit. Ten hours a week over twelve weeks buys roughly 3,262 questions at a realistic pace; six over eight buys about 1,240.
Should I give each specialty equal time? No. Weight each by its share of the content map. Equal blocks systematically under-prepare you for the heaviest areas, which are several times the size of the lightest.
How many questions should I do per day? Derive it rather than guessing: 60% of your study time, at about a minute per question including the explanation. Roughly 50 questions per hour of question time — and about a fifth of them should be re-attempts of previous incorrects.
What if I don't have my exam date yet? Build to a placeholder and rebuild when the date is published. The early weeks are broad coverage and don't change.
Should the last week be revision or rest? Retrieval, light timed blocks and sleep — but nothing new. Material first met in the final week hasn't been reviewed once and displaces revision that would have stuck.