What the examiners said about AQA A Level Maths — and exactly what to do about it
We read every AQA examiner report for 7357 from June 2024 to June 2025, mapped each question to a topic, and turned the comments into a checklist of the mistakes that cost the most marks.
Summaries are in our own words with links to the originals. Predictions are informed guesses, not a guarantee.
Ranked for June 2027
What to focus on for June 2027
June 2026 papers have been sat, but their examiner reports were not publicly readable when we compiled this page. We will add them when they are. Ranked by how often each area appeared, how many marks it carried and how weakly it was answered.
Informed prediction, not a guarantee: any topic on the specification can be examined.
1
Functions, domain and range in set notation
The most-examined area across the three papers in both years we read, and set notation for domain/range was answered fully by under half in 2024 and half in 2025.
The 10 mistakes that cost the most marks (and 5 more)
Ordered by how often and how widely they were reported. The first 5 are free in full; Pro unlocks the fix and a worked example for every other one.
#1"Explain" and "interpret" answers without context
What examiners saw: Explanation marks are lost more than any other type: answers restate a number ("negative correlation"), give generic facts, or give several reasons where one is asked for and one is wrong.
Fix: Answer in the context's words and units, give exactly what is asked (one reason means one), and link it to the numbers.
Worked example (our own): A student says "the sample of 5 runners shows coach B is better because the sd is larger". Comment.
Loses marks
"Coach B has a bigger standard deviation."
Earns the marks
Coach B's times are more spread out (larger sd), so coach B is more likely to produce an extreme time, including the fastest one.
With only 5 runners, the evidence is weak: a larger sample would be needed.
Why: "Explain" and "interpret" marks need a statement about the context (runners, times), not a restatement of the number.
#2"Show that": skipped steps, working back from the answer, no final line
What examiners saw: The most repeated comment across papers. When the answer is printed, the marks are for the route. Candidates jump steps, work backwards from the given result, or finish without writing the given line (e.g. the "= 0" or "f′(x) =").
Fix: Start from the question, write every line, and end with exactly the printed result. If the answer is an equation, include "= 0".
Worked example (our own): The line y = 3x − 1 meets the curve y = 2x² + x − 5. Show that the x-coordinates of the points of intersection satisfy x² − x − 2 = 0.
Loses marks
3x − 1 = 2x² + x − 5
so x² − x − 2 = 0
Earns the marks
3x − 1 = 2x² + x − 5
0 = 2x² − 2x − 4
0 = 2(x² − x − 2)
x² − x − 2 = 0, as required
Why: In a "show that" the answer is given, so the marks are for every step in between, and the final line must match the printed answer exactly (including "= 0").
What examiners saw: Most can find the boundary value, but set notation is wrong or missing, strict and inclusive inequalities are mixed up, and the domain is written in terms of the wrong variable.
Fix: Use braces, a variable and the right inequality: {x : x ≥ 3}. Check whether the end value is attained.
Worked example (our own): f(x) = (x − 2)² + 3 for x ∈ ℝ. State the range of f in set notation.
Loses marks
range: f ≥ 3 (or {y > 3})
Earns the marks
The minimum value is 3 (when x = 2) and f can be any value above it
#4Mechanics: too many significant figures after using g = 9.8
What examiners saw: Final answers given to 4 or more significant figures (or as fractions) after substituting g = 9.8 lose the accuracy mark, and units are often missing.
Fix: After using g = 9.8, give answers to 2 or 3 s.f. with units. Exact multiples of g are usually fine.
Worked example (our own): A stone is dropped from rest and falls for 2.5 s. Taking g = 9.8 m s⁻², find the distance fallen.
Loses marks
s = ½ × 9.8 × 2.5² = 30.625 m
Earns the marks
s = ut + ½at² = 0 + ½ × 9.8 × 2.5²
s = 30.625…
s = 31 m (or 30.6 m)
Why: g = 9.8 is only given to 2 significant figures, so an answer to 5 significant figures claims accuracy you do not have. Give 2 or 3 s.f.
What examiners saw: Values are usually correct, but the conclusion misses continuity, misstates it ("x is continuous"), or never says a root lies in the interval. Some forget to rearrange to the form = 0 first.
Fix: Rearrange to f(x) = 0, evaluate both ends, then state: sign change + f continuous on the interval ⇒ root in the interval.
Worked example (our own): f(x) = x³ − 2x − 5. Show that f(x) = 0 has a root between x = 2 and x = 2.1.
Loses marks
f(2) = −1, f(2.1) = 0.061
so there is a root.
Earns the marks
f(2) = −1 < 0 and f(2.1) = 0.061 > 0
There is a change of sign and f is continuous on [2, 2.1] (it is a polynomial)
so f(x) = 0 has a root α with 2 < α < 2.1
Why: The conclusion needs three things: the sign change, continuity, and a statement that a root lies in the interval.
#6Logarithms: different logs on each side and too many steps at once
What examiners saw: Taking logs of each term (not each side), using two different bases in one equation, or jumping several log laws in one line, so a single slip loses every mark.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
#8Moments: wrong distances and forces missing from the diagram
What examiners saw: Force diagrams with a reaction drawn vertically instead of perpendicular to the rod, extra or missing forces (including the rod's own weight), and distances measured from the wrong point.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
#9Variable acceleration: suvat used when acceleration is not constant
What examiners saw: When acceleration or velocity is a function of t, some still use suvat, forget the constant of integration or the initial condition.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
What examiners saw: Setting the second derivative equal to zero is common; checking that it changes sign (or classifying properly) is not, so the conclusion mark is lost.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
#11Conditional probability: dividing by the wrong probability
What examiners saw: P(A | B) is set up correctly in many scripts, but values are read from the wrong region of a Venn diagram and the denominator is P(A) instead of the given event.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
What examiners saw: "Exceeded by 95%" and similar phrases lead to the wrong tail; z-values like ±2 are used instead of 1.6449; signs are lost when setting up simultaneous equations.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
#13Proof by contradiction: the assumption is not the negation
What examiners saw: Candidates know the shape of a contradiction proof but assume the wrong thing (for "a and b are irrational" they assume "at least one is rational"), and do not state the contradiction.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
#14Trapezium rule: wrong strip width and missing brackets
What examiners saw: h is found using the number of ordinates, the outer bracket is missing, and "over or underestimate" answers talk about the curve increasing instead of its shape (concave/convex).
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
#15Binomial expansion (rational n): factor not raised to the power, validity not stated
What examiners saw: Taking out a factor is common, but the factor is not raised to the power n, and the range of validity is left out or not compared with the value asked about.
Pro: the fix and a worked example (wrong vs right method) for this mistake are in Pro. See plans
How often each topic appeared, and for how many marks
Every question in the papers we read, matched to a topic. Columns marked * count questions (q) because the marks could not be verified from the mark scheme. Hover or tap a cell to see the questions.
Confidence: high for the series shown — every question was read and mapped by hand. "Series" = how many of the 2 series examined the topic.
Paper by paper
Every paper we analysed, question by question
Topics are ours (matched from each report's question-by-question comments). We don't reproduce the questions: open the official paper from our past-paper page.
Top 10 mistakes examiners see in AQA A Level Maths
A one-page PDF of the ten most common mark-losing mistakes and how to avoid them. Print it and stick it inside your revision folder.
We'll also send occasional revision tips. Unsubscribe any time; we never share your address.
Sources
Where this comes from
We read the AQA 7357 Reports on the Examination for Papers 1–3 in June 2024 and June 2025 in full. AQA's mark schemes could not be opened, so the trend table counts questions rather than marks; multiple-choice questions are not mapped to topics.
What do examiners say students get wrong in AQA A Level Maths?
The most repeated points are: "Explain" and "interpret" answers without context; "Show that"; Domain and range in set notation. Each one on this page has the fix and, where free, a worked example showing the wrong and right method.
Which AQA A Level Maths topics come up most?
In the material we analysed, the biggest areas were Algebra and functions, Integration and differential equations, Differentiation.
Is this a prediction of the June 2027 paper?
No one outside the exam board knows what will be on the next paper. The "focus" list is an informed prediction from how often topics appeared and how weakly they were answered; it is not a guarantee, so revise the whole specification.
Where does this information come from?
We read 6 AQA examiner reports (June 2024, June 2025) in full, summarised them in our own words and linked each point to the original.