ICAP CFAP common mistakes
ICAP CFAP · Understand the traps and how to avoid them.
Computing an accounting ratio or variance and then reporting it without interpreting what it signals about the entity's position, or applying the wrong benchmark (e.g., comparing a gross margin to an industry average when the question asks for trend analysis).
Candidates treat analytical questions as arithmetic drills and stop once a number is produced, because the calculation feels like the 'hard part'. Interpretation requires a further judgement step that feels subjective and is therefore skipped under time pressure.
What to do instead
For every ratio/variance question, force a two-line written conclusion: (1) direction and size of the movement, (2) the most plausible business cause and its implication. Practise by covering the answer options and writing the conclusion before looking at them.
Preparing or evaluating a statement of cash flows by classifying interest paid, dividends paid, and interest/dividends received using the classification the candidate personally prefers rather than the framework stated in the question (e.g., IAS 7 allows choices; a local GAAP or exam-stem convention may fix them).
Candidates memorise one 'correct' classification from a textbook or prior exam and apply it universally, missing that the standard permits policy choices and that the question supplies its own basis.
What to do instead
Before answering, underline the reporting framework named in the stem. Maintain a one-page comparison grid of the permitted classifications under each framework you are examined on, and rehearse classifying the same transaction under each.
In capital budgeting questions, discounting the wrong cash flows - including sunk costs, allocated fixed overhead, or financing costs in the project cash flows, or using the accounting profit instead of incremental after-tax cash flow.
Candidates default to the income-statement figures they can see in the stem because those are familiar, and they forget that NPV/IRR operate on incremental, financing-excluded cash flows.
What to do instead
Use a fixed three-step template every time: (1) list only incremental cash flows, (2) strip out sunk costs, allocations and interest, (3) apply tax and depreciation tax shield, then discount. Explicitly write 'excluded: sunk cost X' in your workings so the marker sees the reasoning.
Calculating WACC using book values, or using the pre-tax cost of debt, or weighting with the wrong capital structure (target vs current) when the question specifies one.
Book values are the numbers printed in the financial statements and are easier to find than market values; the tax adjustment on debt is a single step that is easily dropped when rushing.
What to do instead
Write the WACC formula at the top of every answer and tick each input as you substitute it: market value weights, after-tax Kd = Kd(1−t), and the stated weighting basis. If the stem gives a target gearing, use it and say so.
Applying the wrong tax residency or source rule to a cross-border payment - for example, taxing a dividend or royalty without checking the withholding rate, the double tax agreement relief, or whether the recipient has a permanent establishment.
Candidates know the domestic rate well and assume it applies to every transaction, treating treaty relief as an advanced topic to skip. The stem often buries the recipient's status in a single clause.
What to do instead
For every cross-border item, run a four-question checklist: who is the recipient, what is the character of the income, is there a treaty and what article applies, and is there a PE? Answer each explicitly before computing the tax.
Ignoring the interaction between capital allowances and accounting depreciation, or claiming a deduction twice, when computing taxable profit from accounting profit.
The reconciliation from accounting profit to taxable profit is done mechanically line by line, and candidates forget that the depreciation add-back and the capital allowance deduction are a pair that must both appear.
What to do instead
Build the reconciliation as a standard pro-forma with fixed rows (accounting profit, add back depreciation, deduct capital allowances, add back disallowable expenses, deduct exempt income) and reuse it in every practice question until it is automatic.
Recommending a strategy (e.g., diversification, a pricing change, a make-or-buy decision) without quantifying the financial consequence or acknowledging the qualitative constraints such as capacity, culture, or regulatory limits.
Business strategy questions read like essays, so candidates write general management theory and never anchor it to the numbers or constraints in the scenario.
What to do instead
Structure every strategy answer as: recommendation, quantified financial impact, two qualitative enablers/barriers, and the key risk. If the stem gives figures, no recommendation is complete without at least one computed comparison.
Designing audit procedures that are vague ('review the ledger', 'discuss with management') rather than specifying the assertion tested, the population, the sample selection, and the evidence expected.
Candidates confuse describing an area of risk with designing a procedure, and they have not internalised that a procedure must be capable of detecting a specific misstatement.
What to do instead
Write every procedure to a four-part formula: assertion → action (inspect/recompute/confirm/observe) → source document or third party → evidence conclusion. Test yourself by asking 'could this procedure fail to detect the stated risk?'
Concluding on audit findings without escalating to the correct level - recommending 'more testing' when the matter is a suspected fraud, a going-concern doubt, or a disagreement with management that requires communication with those charged with governance or a modified opinion.
Candidates treat all exceptions as sampling issues and reach for additional substantive work, because escalation and reporting thresholds are learned as lists rather than as decision rules.
What to do instead
Memorise a decision ladder: misstatement magnitude vs materiality → pervasiveness → management response → required communication (those charged with governance, regulators) → opinion modification. Apply the ladder in that order to every findings question.
Treating ethics scenarios as abstract philosophy and missing the concrete safeguard or escalation step - e.g., identifying a self-interest threat but not stating the specific action (decline the engagement, remove the individual, disclose to those charged with governance, resign).
Candidates can name the threat categories but have not practised converting a threat into a documented safeguard, so answers stop at diagnosis.
What to do instead
Use a fixed three-step answer: identify the threat and whose objectivity is impaired, evaluate its significance, then state the specific safeguard or escalation with the party responsible. Rehearse with short scenarios until the third step is automatic.