Quick answer
A good TMUA question bank covers both papers, goes beyond the finite past papers with original questions, and gives a worked solution for every one. CrackTMUA is exactly that: 400+ questions including 100+ original trap-based ones, filterable by topic and difficulty, with spaced repetition and a predicted band. Where to find TMUA questions.
Most TMUA prep advice stops at "do the past papers". That is correct for about three weeks, and then you run out. There are only a handful of official papers, they are older-format, and a serious candidate works through them well before the test. After that you need an ongoing supply of genuine, well-calibrated questions, which is what a question bank is for. This guide explains why the past papers run dry, what separates a good bank from a bad one, and how to practise with a bank so it actually moves your score.
Key fact
A question bank is not "more questions". The thing that improves your score is the loop around each question: a full worked solution that names the trap, the ability to filter to your weak topics, and a system that brings your mistakes back. Volume without that loop just teaches you to be wrong faster.
Why the past papers run out
The official past papers are the best TMUA practice in existence, and they are free from UAT-UK. Use them first, every one of them, properly. But there are only a handful of complete papers, and that supply has two problems for anyone preparing seriously.
The first is simply that it is finite. Each sitting is two papers of 20 questions, so the whole back catalogue is a few hundred questions. A committed applicant working a topic at a time, then sitting timed mocks, gets through that in weeks, not months. Once a paper has been seen, its value as a timed mock is gone, because you remember the answers.
The second is that the papers are older-format. They were written before the 2024 move to UAT-UK and Pearson VUE, when the test reported three grades rather than one and the on-screen interface did not exist. The mathematics is still completely valid, and you should absolutely use them, but the supply was frozen at that point. It does not grow, it does not reflect the current single-score on-screen experience, and it cannot be refreshed.
So past papers are the foundation, not the whole building. For exactly which papers exist and where to download the originals, our practice questions guide is the place to start. An ongoing question bank is what you reach for once that foundation is laid.
What a good TMUA question bank actually needs
"Question bank" is a label anyone can slap on a pile of PDFs or an AI-generated problem set. The ones worth your time share a specific set of properties. Here is what to look for, and why each one matters.
| What to look for | Why it matters |
|---|---|
| Both papers, including Paper 2 | Past papers under-serve reasoning and proof; Paper 2 is the most trainable part of the test |
| Original questions beyond the official set | The official supply is finite; you need fresh material once it runs out |
| Full worked solution to every question | A bare answer key tells you nothing; the method and the trap are the value |
| Filter by topic and by difficulty | Lets you practise by weakness, not by random shuffle |
| Spaced repetition | Resurfaces the questions you got wrong before you forget the fix |
| Accurate difficulty calibration | Off-calibration questions teach false confidence or pointless panic |
It covers both papers, including Paper 2. This is the single biggest gap in most banks. Paper 1, applications of mathematical knowledge, is easy to write for because it is familiar pure maths. Paper 2, mathematical reasoning, tests logic, necessary and sufficient conditions, negation, and spotting the flawed step in a proof. It is genuinely unfamiliar to most applicants, it is where the test feels hardest, and it is the most improvable part of your score. A bank that quietly skips it, or pads it with weak imitations, leaves your biggest opportunity untouched.
It goes beyond the official questions. Because the official set is finite, a real bank adds original questions written to the same specification, difficulty and style. These have to be authored carefully, not generated and hoped for, which brings us to calibration below.
Every question has a full worked solution. Not an answer letter, a method. The whole value of a TMUA question is locked inside its solution: the clever observation that turns a slow grind into a ten-second move, the counterexample that kills a Paper 2 claim, the specific slip that one of the wrong options was designed to catch. A bank that gives you the right letter and nothing else is barely better than the official answer key.
Tip
When you read a worked solution, do not just check that you got the right answer. Read it even when you were right, because the official route is often faster than yours, and speed is exactly what a no-calculator, four-minutes-a-question test rewards. The fastest method, not the correct answer, is the thing worth learning.
Filter by topic and by difficulty
A pile of questions in fixed order is a paper. A bank lets you slice it, and that slicing is most of what makes a bank more useful than a PDF.
By topic. Early in your prep you want to drill one area until the standard moves feel automatic: surds and indices, logarithms, sequences and series, coordinate geometry, calculus, and the logic and proof that dominates Paper 2. Topic practice surfaces your specific weak spots, which a full mixed paper hides behind an aggregate score. If your syllabus coverage is patchy, practising TMUA questions by topic is exactly how you find out where.
By difficulty. The TMUA spans gentle opening questions to genuinely hard ones, and the marks that separate competitive candidates live at the top of that range. Once your basics are secure, you want to deliberately seek out the harder questions rather than coast on comfortable ones. A bank that tags difficulty honestly lets you do that; one that does not just gives you a shuffle.
Slicing by weakness, in practice. This is where the two filters combine into something a PDF cannot do. Say logarithms are your weak spot. Instead of grinding random questions and hoping logs come up, build a single ten-question logarithms set that climbs in difficulty, then do the whole thing in one sitting, mark it, and note the exact step where it broke. A good climb looks like this:
- Questions 1 to 3, difficulty around 3: clean law-of-logs manipulation, combining and splitting log a + log b and the like, just to warm up the rules.
- Questions 4 to 6, difficulty around 4 to 5: changing base, solving a straightforward log equation, spotting an extraneous root.
- Questions 7 to 9, difficulty around 5 to 6: logs tangled into inequalities and indices, where the slip is usually a dropped domain restriction.
- Question 10, difficulty 6: a disguised log equation hidden inside another topic, such as a sequence or a coordinate-geometry question that only yields once you take logs.
Because you sat the whole set in one go, the mark tells you precisely where logs stop being automatic for you, and the one question that broke is the one worth logging and re-surfacing. That is the mechanics of using a bank: not "I did some questions", but "I aimed ten questions at one weakness and found the exact rung I fall off".
Here is a genuine official Paper 1 question, the kind a good bank would tag as coordinate geometry and differentiation at mid difficulty. Give it a proper attempt before you reveal the solution, and notice how the worked answer names the exact trap that one wrong option is built around:
Spaced repetition: getting weak questions back
Here is the part almost no PDF can do. You get a question wrong, you read the solution, you understand it, and then two weeks later you make the same mistake because you never saw it again. The fix is spaced repetition: the questions you got wrong come back on a schedule, with the gaps widening each time you get them right, so a fragile bit of knowledge gets reinforced just before you would have forgotten it.
This is the difference between practising and actually fixing weaknesses. Without it, a bank is a one-pass experience: you attempt each question once, and your specific mistakes quietly survive. With it, the bank becomes a system that hunts down your gaps and keeps closing them. It is the single feature that turns "I did a lot of questions" into "I stopped getting these wrong".
Key fact
The honest test of any prep tool is whether your mistakes come back. If you can attempt 200 questions and never see your wrong ones again, you have a worksheet, not a study system. Spaced repetition is what makes the difference, and it is worth weighting heavily when you choose a bank.
How to practise with a bank
Having a good bank is not the same as using it well. Three habits separate the applicants who improve from the ones who just rack up attempts.
Little and often beats marathon sessions. Twenty focused minutes most days will move your score more than one three-hour binge a week, because spacing is how memory consolidates and because tired practice teaches sloppy habits. Short, regular, deliberate sessions are the whole game.
Practise by weakness, not by comfort. It feels good to answer questions you already find easy, and it does almost nothing. Use the topic and difficulty filters to point yourself at the areas you are worst at, and let spaced repetition keep bringing those back. The goal is to spend your time where the marks you are currently losing actually are.
Add the clock, but not too early. Timed practice is a separate skill from learning the content, and trying to do both at once early on just teaches you to panic. Start untimed and topic-focused, switch to mixed harder sets in the middle of your prep, and add time pressure once the standard moves feel automatic. Save a few full mocks under genuine exam conditions for the final weeks. There is more on sequencing all of this in how to prepare for the TMUA.
How CrackTMUA's question bank is built
We built CrackTMUA around exactly the loop above. It starts with every official past paper turned into interactive questions, then adds 100+ original, trap-based questions written to mirror the exam's specification, difficulty and style, for 400+ questions in total. Both papers are covered properly, including the Paper 2 reasoning and proof that most banks skimp on.
Every single question carries a full step-by-step worked solution that names the trap and shows the fastest method, not just the correct letter. You filter by paper, topic and difficulty, so you can drill TMUA questions by topic and push into the harder end deliberately. An SM-2 spaced-repetition engine reschedules the questions you got wrong so they come back before you forget the fix. On top of the bank sit 18+ full mock exams in a replica of the real Pearson VUE on-screen interface, plus a predicted TMUA band with topic-by-topic analytics so you can see exactly where you stand, and a community to ask when a question is genuinely confusing.
The free tier is real: 10 questions a day plus the specimen mock and every worked solution, at no cost. Premium unlocks the whole bank, the mocks, spaced repetition and the predicted band for £37 one-time, with 12 months of access (no subscription, no auto-renewal). That is below the established paid platforms, which run from around £59 up past £200, though there are cheaper newer options too, so weigh features, not just price. For the wider landscape and the free official sources, the practice questions guide lays it all out.
The headline is simple. Past papers are the foundation, but they run out, so an ongoing bank is what carries you the rest of the way. Choose one that covers both papers, explains every answer, lets you practise by weakness, and brings your mistakes back. Then put in short, regular, deliberate sessions, and the score follows.
See how you measure up, free
Create a free account to start practising the question bank, official papers plus original trap-based questions, 10 a day free, and track your accuracy and coverage as you go. Your full predicted band and weak-spot map come with Premium.