CPA PM Elective common mistakes
CPA PM Elective · Understand the traps and how to avoid them.
When a scenario asks whether a division should be evaluated on ROI or residual income, the candidate computes both numbers and then recommends whichever is larger, instead of reasoning about the behavioural consequence of the metric choice (for example, a high-ROI division rejecting a project that earns more than the cost of capital but less than its current ROI).
Candidates treat the calculation as the answer and stop there. The elective case-style questions reward the interpretation and the behavioural implication, but under time pressure candidates default to the arithmetic they have practised most, and the metric-choice judgement is the step they skip.
What to do instead
Before computing, state the decision the metric is meant to drive. Then test the candidate project against both metrics and explicitly note the goal-congruence difference: ROI can induce under-investment when the division's ROI exceeds the firm's required rate, while residual income and EVA align with firm value. Practise writing one sentence of recommendation that names the behavioural risk, not just the number.
The candidate treats a balanced scorecard as a checklist of four perspectives and invents one or two generic measures per perspective, without linking measures to the organisation's specific strategy or showing cause-and-effect chains between perspectives.
Textbook familiarity with the four perspectives creates a false sense of completeness. Candidates do not build the strategy map logic (for example, training hours to customer satisfaction to revenue growth), so the measures they propose are plausible but not strategy-specific, which is what the marking guide rewards.
What to do instead
For every scorecard question, first write the strategic objective in one line, then build a short causal chain across perspectives and derive measures from the chain's links. Reject any measure you cannot connect to the objective. Check that each perspective has at least one leading and one lagging measure where the scenario supports it.
In a make-or-buy or special-order question, the candidate includes allocated fixed overhead in the relevant cost, or excludes an avoidable fixed cost, and then reaches a conclusion based on the distorted unit cost.
Candidates carry forward the absorption cost per unit from earlier workings because it is the number on the page. They do not re-examine which costs are avoidable, incremental, or opportunity costs specific to the decision, so sunk allocations contaminate the comparison.
What to do instead
Build a relevance table before calculating: list each cost, mark it relevant or irrelevant, and give the reason (future, differential, cash). Explicitly remove allocated common fixed costs and add opportunity costs such as lost contribution from displaced sales. Only then compare the relevant totals and state the qualitative factors that could override the number.
The candidate fits a cost function using the high-low method on the highest and lowest activity levels without checking whether those points are outliers or whether the relationship is non-linear, and reports the resulting fixed and variable components as if they were reliable.
High-low is quick and appears objective, so it is used as a default. Candidates do not inspect the data set for unusual periods, step costs, or mixed cost behaviour, and they do not question whether the chosen points represent normal operations.
What to do instead
Before applying high-low, scan the data for outliers, capacity step changes, and one-off events, and say which points you are excluding and why. Where the scenario provides multiple periods, note that regression uses all observations and is generally preferred, and comment on the likely direction of bias if an outlier remains in the high-low pair.
When a variance is unfavourable, the candidate lists generic causes such as 'higher prices' or 'poor cost control' rather than separating price and quantity effects and tracing them to the responsible manager and the operating cause.
Variance analysis is often learned as a calculation drill. Candidates compute the numbers correctly but have not practised the interpretive step, so under time pressure they write the first plausible cause instead of decomposing the variance into its operational drivers.
What to do instead
For each material variance, write the arithmetic decomposition first, then a two-column note: operational cause (for example, a supplier price rise, a change in mix, or inefficient use of material) and responsibility (purchasing, production, or sales). Flag variances that are interdependent, such as a favourable purchase price variance caused by buying lower-grade material that then drives an unfavourable usage variance.
The candidate prepares a cash budget by taking budgeted profit and adjusting only for depreciation, ignoring the timing of receipts and payments, capital expenditure, tax instalments, and financing flows.
Profit and cash are conflated because both start from the same sales and cost figures. Candidates do not build a receipts schedule based on credit terms and collection patterns, so the closing cash balance is materially wrong even when the profit figure is right.
What to do instead
Always start the cash budget from a collections schedule driven by the stated credit terms and bad-debt assumptions, and a payments schedule driven by supplier terms. Add separately the items that never appear in profit: capital purchases, loan drawdowns and repayments, interest, tax, and dividends. Reconcile the closing cash balance to the opening balance plus net cash flow as a check.
In a pricing question the candidate applies cost-plus markup mechanically, or applies the profit-maximising rule without regard to the demand curve, competitor response, or capacity constraints described in the scenario.
Cost-plus is the most familiar technique and produces a definite number, which feels safe. Candidates do not interrogate the scenario for price elasticity, spare capacity, or strategic pricing objectives such as penetration or skimming, so they miss the pricing logic the question is testing.
What to do instead
Classify the pricing situation first: is capacity constrained, is the product new, is there a competitive reference price, is this a one-off order? Then choose the technique that fits (target costing and value-based pricing for new products, relevant cost as a floor for special orders, full cost plus for regulated or long-term contracts). State the price range and the constraint that sets the floor and ceiling.
The candidate computes NPV and IRR, then recommends the project with the highest IRR even when the projects differ in scale or duration, or when NPV and IRR rank them differently.
IRR is expressed as a percentage and feels easier to compare across projects. Candidates do not recall that IRR assumes reinvestment at the IRR itself and can mislead with non-conventional cash flows or mutually exclusive projects of different sizes.
What to do instead
Use NPV as the primary decision rule for value maximisation, and present IRR as supplementary. Where rankings conflict, explain the cause (scale, timing, or reinvestment assumption) and show the incremental cash flow analysis. Also check for multiple IRRs when cash flow signs change more than once, and note the modified IRR or NPV profile as the resolution.
The candidate identifies risks in a scenario but does not map them to a specific control, owner, or residual risk assessment, and does not distinguish preventive from detective controls.
Risk questions read like comprehension tasks, so candidates list risks and stop. The professional-level answer requires the control response and an evaluation of whether the control is proportionate, and candidates have not practised structuring the answer that way.
What to do instead
Use a fixed structure for every risk answer: risk event, likelihood and impact, existing control and whether it is preventive or detective, residual exposure, and recommended additional control with an owner. Where segregation of duties is relevant, state explicitly which duties must be separated and why. Close by noting the cost of the control against the risk it mitigates.
The candidate presents a data analytics output, such as a regression or a dashboard, as a decision without discussing data quality, correlation versus causation, or the limitations of the model, and does not address how the result would be communicated to non-financial stakeholders.
Technical output feels authoritative and candidates want to demonstrate the analytical skill. The professional judgement layer, including the caveats and the communication plan, is treated as optional commentary rather than part of the required answer.
What to do instead
For every analytical recommendation, add three short paragraphs: data limitations (completeness, timeliness, bias), interpretation limits (correlation is not causation, model assumptions), and communication (audience, the one message, the visual that supports it, and the decision requested). This structure also protects marks in ethics and professional judgement questions where the candidate must recommend a course of action.