Quick answer
Most candidates who do well spend roughly 40 to 70 hours across 8 to 12 weeks, which works out at 4 to 6 hours a week. In question terms that is every official past paper worked through properly (about 360 questions) plus several hundred more for the topics you keep getting wrong. Volume matters far less than review: 20 questions you study the solutions to beats 60 you only mark.
"How much practice do I need" is the most reasonable question a TMUA candidate can ask and the one that gets the vaguest answers. This guide gives real numbers: how much material actually exists, what a realistic total looks like, how to spread it across a term, and the point at which more questions stop helping you.
One caveat before the numbers. There is no published data linking hours to TMUA scores, so anything you read framed as "X hours gets you a 7.0" is invented. What follows is built from what the test actually contains, how much genuine material exists, and what the people who do well have in common, which is not volume.
Key fact
The single most reliable predictor of improvement is not how many questions you do, it is what fraction of them you review properly. A question you got wrong and then studied until you can name the move that unlocks it is worth ten you skimmed.
First, how much material is there?
Useful to know before you plan, because it bounds the whole exercise.
The official supply is the specimen paper plus a Paper 1 and a Paper 2 for every year from 2016 to 2023. That is 18 papers of 20 questions, so roughly 360 genuine questions. It sounds like a lot. At an honest working pace, including reading the solution to everything you get wrong, it is somewhere between 25 and 40 hours, and it is finite: once a paper has been seen, it is spent as a timed mock because you remember it.
So the shape of TMUA preparation is fixed by arithmetic. Whatever your total, the official papers cannot fill more than about half of it, and you have to decide in advance what fills the rest.
The realistic totals
| Level of preparation | Hours | Roughly what it covers |
|---|---|---|
| Minimum that is not a gamble | 20 to 25 | The specimen plus three or four recent papers, timed, with solutions read |
| What most successful candidates do | 40 to 70 | Every official paper properly, plus targeted topic work on weaknesses |
| Heavy preparation | 80 to 120 | The above, plus sustained Paper 2 reasoning work and repeated timed mocks |
The middle row is the one to aim at. Below about 20 hours you are essentially sitting the test cold and hoping the paper suits you. Above about 80 the returns thin out sharply unless the extra time is going into a specific named weakness rather than general practice.
Spread over a term that is 4 to 6 hours a week for 8 to 12 weeks, which for most Year 13 applicants means two evening sessions and one longer weekend session. Starting earlier and going slower beats compressing it, because Paper 2 reasoning is a slow skill that improves with spacing rather than with cramming.
What the hours should actually contain
A common failure is spending 50 hours and improving very little, because all 50 went on the comfortable activity: doing fresh questions and marking them. Here is a split that works.
- Roughly 40 per cent on new questions. Attempt properly, under something close to time, no notes.
- Roughly 40 per cent on review. For everything you got wrong, and everything you got right by luck, read the worked solution until you can state the one move that unlocked it. Write that move down. This is the part people skip and it is the part that moves scores.
- Roughly 20 per cent on timed full papers. Not from the start. These belong in the last third, once individual questions no longer surprise you, so the mock measures your pacing rather than your unfamiliarity.
If you take one thing from this guide: the 40 per cent on review is not optional padding. It is where the improvement happens. Our guide on how to prepare for the TMUA sets this out as a full week-by-week plan.
Free TMUA Survival Kit — every formula and the top traps on one page. Grab it before you go.
A worked weekly plan
For a candidate starting 10 weeks out at 5 hours a week:
- Weeks 1 to 2. Diagnose. Sit one full paper untimed, mark it, and identify your three weakest topics. Do not start grinding yet.
- Weeks 3 to 6. Topic work. One weak topic per week, roughly 20 to 30 questions on it, every wrong one reviewed to the point where you can name the move. Start Paper 2 reasoning in week 3, not week 8, because it is the slowest skill to feel natural.
- Weeks 7 to 9. Mixed and timed. Full papers under exam conditions, one a week, each followed by a session at least as long as the paper spent on review.
- Week 10. Taper. Re-read your list of named moves, do light mixed practice, sit nothing new and hard in the final two days.
Pace yourself against the clock, not just the count
Both papers are 20 questions in 75 minutes, so the average is three minutes and forty-five seconds a question, with no calculator. That average is misleading in a useful way: the easier questions should take you 90 seconds, which is what buys you the six or seven minutes a hard one needs.
This is why untimed practice eventually stops being enough. Once you can do the mathematics, the remaining marks are in triage, and triage is only trainable against a clock. Practising with a per-question timer is the cheapest way to build it without burning a whole past paper.
When more practice stops helping
Three honest signals that additional volume is not the answer:
- You are getting questions right but slowly. More questions will not fix this; timed practice and deliberate triage will.
- You keep missing the same topic despite doing lots of questions in it. The gap is understanding, not exposure. Go back to the underlying content, then return.
- You are re-doing papers you remember. The score is now measuring recall, which tells you nothing. You need unseen questions, not another lap.
Where the questions come from once the papers run out
Full disclosure: this is our site, so here is the honest picture rather than a sales pitch. The arithmetic above has an obvious hole in it. If a serious total is 40 to 70 hours and the entire official supply is 25 to 40, then something has to fill the second half, and for most candidates it ends up being re-doing papers they already remember, which is the least useful hour available.
That is the specific gap CrackTMUA fills. Every official past paper is here as interactive practice, and alongside them are 330+ original questions we write and calibrate ourselves to match the exam's difficulty and its favourite traps, 690+ in total. Every one carries a full worked solution that names the trap and the fastest route, which is what makes the review 40 per cent above actually possible rather than a nice idea. You can filter to a single weak topic, set a per-question timer to train triage, and let spaced repetition bring your mistakes back on the schedule that makes them stick. There are 26+ full mocks in a replica of the real Pearson VUE screen for the timed phase, and a predicted band so "am I ready" stops being a guess. Free is 10 questions a day; Premium is a one-time £49 for 12 months.
To be clear, the official papers still come first and nothing replaces them, especially as your final timed mocks. Ration two or three untouched. Everything else is what fills the weeks in between.
Start practising free, or try a timed drill if pacing is the part you need to fix.
Practise the real TMUA, free
Drill 690+ questions, every official past paper plus 330+ original, trap-based ones, each with a full worked solution, then sit full mocks in a replica of the real exam screen. Spaced repetition and a predicted band included. No PDFs.
Get the free TMUA Survival Kit
Every must-know formula, the fastest method for each question type, and the 10 traps that catch people out, on one page. Plus the key 2026 dates as the exam nears. No spam, unsubscribe any time.