CMA Final Paper 19 common mistakes
CMA Final Paper 19 · Understand the traps and how to avoid them.
In essay questions on strategic analysis, the candidate describes the company's situation generically (for example, 'competition is intense and costs are rising') without applying the named framework's categories to the specific facts given in the scenario, so the marker cannot award application marks.
Under time pressure candidates fall back on memorised framework headings and write about the industry in general rather than extracting the facts the case actually supplies. They treat the framework as a checklist to name rather than a lens to apply, and they have often practised recall of definitions but never practised mapping case facts onto categories.
What to do instead
Before writing, spend two minutes building a two-column table: framework element on the left, verbatim case fact on the right. Write only from that table, and for each element state the fact, then the implication, then the consequence for the decision the question asks about. If a framework element has no supporting fact in the case, say so explicitly and move on rather than inventing generic commentary.
In financial reporting questions, the candidate records a transaction using its legal form or its cash effect rather than the substance and recognition criteria of the relevant standard, for example recognising revenue on a bill-and-hold or consignment arrangement when control has not transferred, or treating a lease as a simple operating expense without assessing the right-of-use asset.
Candidates learn journal entries by pattern-matching to familiar transactions and rarely rehearse the recognition and derecognition criteria themselves. Exam scenarios deliberately add a twist (side agreement, repurchase option, variable consideration) that breaks the pattern, and the candidate answers the familiar version of the transaction rather than the one described.
What to do instead
For every transaction question, write one line answering: what is the performance obligation or asset, has control or the risks and rewards passed, and what is the measurement basis? Then post the entry. Practise by taking a standard entry you know and writing the three or four fact patterns that would change it, so you recognise the twist when it appears.
In cost and management accounting, the candidate computes a single cost figure (overhead absorbed, product cost, variance) without stating and checking the basis used, so they mix absorption and variable costing, use budgeted versus actual rates inconsistently, or report a variance without its sign convention and whether it is favourable or adverse.
Exam pressure pushes candidates straight to arithmetic. Cost accounting questions often contain both a budgeted and an actual driver quantity, and the candidate grabs whichever appears first. Sign conventions for variances are learned inconsistently across topics, so an adverse variance is reported as favourable or netted incorrectly.
What to do instead
Adopt a fixed layout for every cost computation: state the driver, the rate, the quantity used and the source of each figure before calculating. For variances, always write the formula, substitute, and label the result favourable or adverse by comparing actual to standard explicitly rather than by sign. Do three mixed absorption/variable reconciliation questions in one sitting until the layout is automatic.
In financial management calculations, the candidate produces a numeric answer for net present value, weighted average cost of capital or valuation without interpreting it or linking it to the decision, and without sanity-checking the magnitude (for example, a cost of capital of 2% or 60%, or a project NPV that ignores the initial outlay's timing).
Candidates treat these questions as maths problems with a single right number and stop at the figure. They also mis-key cash flows by one period, forget working capital recovery or tax timing, and never ask whether the answer is plausible, because they have not built the habit of a magnitude check.
What to do instead
After every calculation, write one sentence: what the number means and what the decision should be. Then run two checks: is the discount rate plausible against the risk described, and does the NPV change sign if the discount rate moves a few points? Build a standard timeline template (year 0 to year n, tax, working capital, scrap) and use it every time so no cash flow line is omitted.
In business law and ethics questions, the candidate answers with a moral opinion or a general statement that the conduct was 'unethical', instead of identifying the specific legal rule or ethical principle engaged, applying it to the facts, and reaching a conclusion on liability or appropriate action.
Ethics is often revised as a list of theories without practice in applying them, so candidates default to everyday moral language. In law questions, candidates recognise the topic (for example, contract formation or directors' duties) but cannot recall the precise element that the scenario is testing, so they write around the issue.
What to do instead
Use a fixed answer structure: identify the issue, state the rule with its elements, apply each element to the facts, conclude. For ethics, name the framework (for example, rights, justice, virtue, or a professional code principle), apply it, and state what the professional should do and to whom they owe the duty. Practise converting three theory notes into applied mini-answers each week.
In information technology and data analytics questions, the candidate answers at the level of tool names and buzzwords ('use a dashboard', 'apply data analytics', 'implement an ERP') without specifying the data required, the control objective, the test procedure, or how the output would change a decision or an audit conclusion.
These topics are often revised from vendor-style summaries and syllabus headings rather than from worked scenarios, so the candidate has vocabulary but no procedure. They also underestimate how concretely the exam expects analytics to be described, including population, fields, exception criteria and follow-up.
What to do instead
For any analytics question, answer in four parts: the question being asked, the data fields and population, the test or model and its exception criteria, and the action taken on the exceptions. For controls questions, state the control objective, the specific control, and one test of operating effectiveness. Rehearse this on realistic scenarios such as duplicate payments, journal entry testing and segregation-of-duties analysis.
The candidate allocates time uniformly across all questions or in proportion to how comfortable each topic feels, so a high-weight essay or a multi-part calculation is left half-finished while early short questions are answered at excessive length.
Candidates do not convert the marks available into minutes before starting, and they write until they run out of ideas rather than until they have earned the available marks. Anxiety also drives over-investment in the first question, which feels safe.
What to do instead
In the first five minutes, write the mark allocation beside each question and convert it to minutes using your target marks per minute. Cap each answer at its allocation and move on, leaving a visible gap to return to. Practise under a timer with the same conversion so the pacing habit is trained, not improvised on the day.
The candidate answers the question that was expected rather than the one printed, for example producing a full ratio analysis when asked to recommend a financing choice, or discussing all four framework elements when only two are requested, and so earns marks for irrelevant material while missing the required conclusion.
Pattern recognition from past practice triggers a rehearsed answer. Candidates also skim the requirement under time pressure, and verbs such as 'evaluate', 'recommend' and 'advise' are read as 'describe'.
What to do instead
Underline the requirement verb and the specified scope before reading the scenario, and restate the requirement in one sentence at the top of your answer. End every answer with an explicit recommendation or conclusion sentence that directly answers the verb. If you notice you have written material outside the scope, cut it rather than leaving it to dilute the answer.
In financial reporting and management accounting, the candidate presents figures without showing workings, or shows workings in a disorderly sequence with unlabelled numbers, so that a single arithmetic slip early in a multi-step computation destroys the final answer and forfeits method marks.
Candidates believe the final figure is what is marked and treat workings as scrap. They also compute in one continuous chain without intermediate labels, which makes an error impossible to locate or to be credited for method.
What to do instead
Present every computation as labelled lines: describe the item, show the formula or source, then the amount. Carry forward your own figure consistently even if you suspect it is wrong, and state any assumption. This preserves method marks and lets you spot the error quickly when you review.
The candidate treats the professional ethics and business law content as common sense and skips structured revision of it, then loses easy marks on conflicts of interest, confidentiality, independence, insider dealing or directors' duties because they cannot name the governing principle or the required course of action.
These topics feel intuitive, so they are deprioritised against calculation-heavy areas. But exam scenarios are built around specific, testable duties and thresholds, and intuition alone does not supply the named principle, the party owed the duty, or the escalation route.
What to do instead
Build a one-page table of the duties and principles in the syllabus: the duty, to whom it is owed, a triggering scenario, and the required action including escalation. Test yourself by writing the required action for three short scenarios a week, and check each against the code or legal rule rather than against your own judgement.