PCM common mistakes
PCM · Understand the traps and how to avoid them.
Treating a SWOT list as the strategy itself: the candidate writes four bullet lists of strengths, weaknesses, opportunities, and threats, then concludes the plan is 'to leverage strengths and address weaknesses' without choosing a position or allocating resources.
SWOT is taught early and feels like the safe, complete answer. Under time pressure, listing feels productive, and the candidate mistakes coverage of the framework's boxes for the analytical judgment the item is actually testing.
What to do instead
After drafting any SWOT, force a conversion step: write one sentence pairing an internal factor with an external factor (strength-opportunity, weakness-threat) and state the resulting action, owner, and resource shift. If a bullet does not change a decision, delete it. Practise items by answering with the chosen strategic option first, then using the framework only as supporting evidence.
Confusing marketing objectives with tactics or KPIs: the candidate writes 'run a social campaign' or 'increase engagement' where the question asks for a marketing objective, or writes an objective with no metric, baseline, or timeframe.
Objectives, strategies, tactics, and KPIs sit in adjacent boxes of the planning hierarchy and are described with overlapping vocabulary. Candidates who memorise the hierarchy without practising classification default to whichever term appears in the stem.
What to do instead
Apply a three-test filter to every candidate objective: does it name a target market, a direction of change, and a quantified level with a date? If it names a channel or activity, it is a tactic; if it names only a measure, it is a KPI. Drill by rewriting ten tactic statements into objective form and ten objectives into KPI form until the boundary is automatic.
Choosing a research design by habit rather than by question type: the candidate selects a focus group for a price-elasticity question, or a survey for a question about unconscious purchase triggers, and then justifies the choice with generic phrases like 'rich insights'.
Exploratory, descriptive, and causal questions look similar when written in prose, and candidates anchor on the method they know best instead of reading what the decision requires. The stem often signals the design only through the type of inference needed.
What to do instead
Before answering, classify the decision as 'what is happening', 'how many or how much', or 'what happens if we change X'. Map these to exploratory, descriptive, and causal designs respectively, then check whether the method can support the required inference (a focus group cannot estimate incidence; a cross-sectional survey cannot establish causation). Write the inference sentence the method must support and reject methods that cannot produce it.
Treating statistical significance as practical significance: the candidate recommends rolling out a campaign because a test result is significant at p < 0.05, without checking effect size, confidence interval width, or whether the lift exceeds the cost of the change.
Significance is the number most prominently reported and most heavily drilled. Candidates conflate 'unlikely under the null' with 'large enough to matter', especially when the item supplies a large sample size that makes trivial effects significant.
What to do instead
For every significance result, compute or read the effect size and the confidence interval, then compare the lower bound of the interval against the decision threshold (break-even lift, minimum detectable effect). State the recommendation in terms of the interval, not the p-value. If the interval straddles the threshold, the correct answer is usually 'gather more data' or 'test at larger scale', not 'roll out'.
Answering brand-equity questions with awareness alone: the candidate proposes a campaign to build brand equity and measures success only with aided or unaided awareness, ignoring associations, perceived quality, and loyalty.
Awareness is the easiest equity dimension to recall and to measure, so it becomes the default proxy. Items that describe a brand already known but not preferred are designed to catch exactly this substitution.
What to do instead
Memorise equity as a multi-component construct (awareness, associations, perceived quality, loyalty, and proprietary assets) and, for each item, identify which component is deficient from the described symptoms. If the brand is known but not chosen, awareness is not the gap. Pair each diagnosis with a measure from the same component so the metric and the problem match.
Misapplying the product life cycle or the growth-share matrix as a prescription: the candidate says 'invest because it is a star' or 'harvest because the product is mature' without checking the evidence for the stage or the quadrant, and without considering that the classification itself may be the disputed point.
Both models are memorable and give confident-sounding verbs. Candidates skip the diagnostic step that assigns the stage or quadrant and jump to the recommended action, which is often the opposite of what the described evidence supports.
What to do instead
Always write the evidence for the classification before the action: for life cycle, cite growth rate, competitor count, and margin trend; for the portfolio matrix, cite relative share and market growth with the actual numbers. Then state the action and one condition under which it would flip. If the stem gives no growth or share data, treat any quadrant label as an assumption to be flagged, not a fact.
Confusing reach, frequency, and impressions when evaluating a media plan: the candidate recommends the plan with the highest impressions or the lowest CPM without checking whether it delivers the required reach against the target segment or enough frequency to move the response curve.
Impressions and CPM are single numbers that are easy to compare, while reach and frequency require reasoning about overlap and duplication. Candidates optimise the easy metric and miss that a cheap, high-impression plan can under-deliver unique reach.
What to do instead
Restate the communication task as a reach-and-frequency requirement before comparing plans: how many target buyers must see it, and how many times, to achieve the desired effect? Then evaluate each plan against both, and compute effective reach rather than total impressions. When plans trade reach against frequency, justify the choice from the purchase cycle and decision complexity, not from CPM alone.
Judging digital channel performance on last-click attribution: the candidate credits the converting channel and cuts the upper-funnel channel that initiated demand, or recommends budget reallocation based on a last-click report.
Last-click data is what standard platform reports show by default, and it produces a clean ranking. Candidates accept the report's framing instead of questioning whether the attribution model matches the multi-touch purchase journey described in the stem.
What to do instead
Check the purchase cycle length and the number of touchpoints before trusting any attribution report. If the journey is multi-touch, reason with a position-based or data-driven model, or at minimum with assisted conversions and incrementality tests. When an item shows a channel with strong last-click performance and weak assisted performance, the correct move is usually to test incrementality before reallocating.
Treating customer satisfaction as customer loyalty: the candidate concludes that high CSAT or a high NPS means retained customers, or designs a retention programme around satisfaction scores when the described problem is switching driven by price or a competitor's offer.
Satisfaction and loyalty are measured with similar survey instruments and are often reported side by side, so the causal link is assumed. Items describing satisfied defectors are built to expose the assumption.
What to do instead
Separate attitudinal measures (satisfaction, NPS) from behavioural ones (retention rate, share of wallet, repeat purchase, churn). For any retention item, ask which behaviour is actually at risk and what drives it; if the stem shows satisfied customers leaving, look for a structural driver such as switching cost, relative price, or distribution access rather than a service-quality fix. Recommend a behavioural metric alongside the attitudinal one.
Designing channel structure without segment and coverage logic: the candidate recommends direct-to-consumer, intensive distribution, or a new intermediary purely because it is growing, without checking whether the target segment's buying behaviour, service requirements, and channel economics support it.
Channel questions invite trend-based answers, and channel conflict, margin stack, and coverage are harder to reason about than a headline growth rate. Candidates answer the question 'what is modern' instead of 'what fits this segment and this margin structure'.
What to do instead
Before choosing a structure, write three constraints: how the target segment wants to buy (service level, information needs, purchase size), the margin available at each tier after the channel's costs, and the coverage required to reach the segment. Evaluate each option against all three and name the conflict it creates with existing partners. An option that fails the margin or coverage test is wrong regardless of its growth rate.