How ACCA AAA is actually marked: what seven examiner reports reveal
Most candidates prepare for AAA by studying content. The examiners, in every report they publish, say the problem is somewhere else. We analysed seven consecutive ACCA AAA examiner reports — September 2022 through June 2025 — line by line, and the striking thing is how little changes between them. The same marking principles are restated, and the same candidate failures are described, sitting after sitting, in language that is sometimes nearly identical from one report to the next. That consistency is good news: it means the marking behaviour of this exam is knowable, and you can prepare for it deliberately.
Here is how the marks actually work, with every claim drawn from the published reports.
The structure: 80 technical, 20 professional
Every sitting, the exam carries 100 marks: 80 technical marks for applying knowledge to the requirements, and 20 professional skills marks spread across four skills — communication, analysis and evaluation, professional scepticism and judgement, and commercial acumen. One fifth of the exam is therefore not testing what you know at all; it is testing how you work. Candidates who ignore this are attempting a 100-mark exam with 80 marks available. (We cover the professional skills marks in depth in their own guide.)
Principle 1: a point unconnected to the scenario earns little or nothing
This is the single most repeated statement in the examiner reports. Across all seven sittings we reviewed, the examining team makes essentially the same point: vague answers that are not tailored to the scenario do not achieve high technical or professional skills marks, and generic responses that fail to build on the detail provided will gain little to no credit. The December 2023 report adds a sharper edge — generic responses with speculative risks, ones not evident from the information given, get the same treatment.
In practice this means the unit of marking is not the true statement but the applied point. "There is a risk that revenue is overstated" is a true statement about almost any company and earns accordingly. The point that scores is built from the scenario's own facts: this specific arrangement, under this specific accounting requirement, could misstate this specific balance in this direction — ideally with the scenario's numbers attached. The June 2025 report notes that stronger candidates used calculations from the exhibits to demonstrate the significance of the risks they raised, and were rewarded for it.
Principle 2: an identified issue is not an evaluated issue
The reports repeatedly distinguish identification from evaluation. In the June 2025 sitting, the examining team described the recurring restriction on marks as risks that are "not well explained": many candidates spotted that poor working conditions at a supplier posed a risk, but stopped at "reputational impact". Full credit required following the chain to its consequence — the association damages reputation, and therefore sales and profits, as customers buy elsewhere — with the aggravating facts from the scenario folded in. The examiners are explicit that completing this reasoning chain is also what earns the commercial acumen and analysis professional marks. Half-explained points are the difference between a marginal fail and a pass far more often than missing knowledge is.
Principle 3: materiality has its own marks, and they are lost the same ways every sitting
Since the September 2022 syllabus changes, Section A has carried dedicated marks for materiality — three technical marks for determining a threshold, per the reports from 2022 through 2025. The required steps are stable: use the benchmark the partner's email instructs (in June 2025, profit before tax with an expected range of 5–10%), calculate the monetary range, then exercise judgement to select a threshold within it and briefly justify the choice against the risk profile of the engagement.
The failures are equally stable, because the reports list the same ones each time: using a different benchmark than instructed; calculating correctly but then never applying the chosen threshold, reverting instead to fresh calculations against revenue or total assets for every item; and failing to justify the threshold at all. The examiners state that any reasonable justification earns the mark — but that "materiality is set low because the audit is high risk", without more, does not. They also note, with visible frustration across multiple reports, that some candidates still raise materiality when risk is high, inverting the relationship — assumed knowledge from Audit and Assurance at Applied Skills level.
Principle 4: the requirement's exact instruction is part of the mark scheme
When the June 2025 Section A question asked candidates to evaluate and prioritise the risks of material misstatement, the report records that a high proportion of answers gave no indication of prioritisation at all — discarding professional marks by ignoring an instruction printed in the partner's email. Merely calling the risks material was not enough, since only material risks belong in the answer anyway; the examiners wanted reasoning about which risks matter most and why. The same literalism applies in the other direction: the same report notes candidates wasting time prioritising business risks, which was never asked. Requirement verbs and scope words ("using only Exhibits 2 and 3", "you are not required to...") are marking instructions, and both reports from 2024 and 2025 record candidates losing marks on both sides — omitting what was asked and supplying what was not.
Principle 5: in ethics, naming the threat is the beginning, not the answer
The June 2025 report is unusually direct about this: identifying a self-interest threat scores only when the candidate explains why the threat arises in these circumstances and what it may lead to — for example, an audit team reluctant to challenge management for fear of losing the engagement. The briefing notes are addressed to the audit engagement partner, who — as the reports put it — does not need to be educated on what a self-interest threat is. Definitions written to the partner cost time and earn nothing; application earns both technical and professional marks. Conclusions matter too: many candidates in June 2025 discussed the guidance on gifts but only the better answers actually concluded that a 50% staff discount is not trivial and should be declined.
Principle 6: in reporting questions, critique what is in front of you
Section B completion-and-reporting questions regularly ask candidates to critically appraise an extract from a draft auditor's report. Both the December 2024 and June 2025 reports describe the same two failure modes: candidates critiquing what they believe is missing from an extract (when the requirement is explicit that it is an extract), and candidates writing out the circumstances in which each opinion type would apply — knowledge display rather than critique, which the examiners state fails to answer the question and scores nothing. Meanwhile straightforward marks go unclaimed: in June 2025 most candidates missed that the draft opinion paragraph named no financial statements, no reporting framework, and no accounting period.
The meta-finding: the exam is consistent, so preparation compounds
Read seven reports in sequence and the closing sections are almost interchangeable: a persistent gap between candidates who apply knowledge to the scenario and those who treat the exam as a factual exercise; recurring weakness in assumed knowledge for materiality, ethics, and auditor reporting; and a standing instruction to study the published model answers to understand the depth expected. An exam this consistent rewards preparation aimed at the marking behaviour itself — which is exactly the preparation most candidates never do, because until recently there was no way to have a full answer marked against these expectations and find out where the marks actually went.
That is the problem ExamMind was built to solve. Write a complete AAA answer, and the marking engine — calibrated against these published examiner expectations — returns line-by-line feedback and a mark estimate.