When the same mistake appears in several cases, pause the full mocks long enough to repair it. Another forty-minute attempt can easily give you the same feedback again. The useful work is to identify the moment where your reasoning went wrong and practise a better response.
I would start with one question: if we replayed that moment now, would you know what to do differently? If the answer is no, the feedback is still too vague.
Turn the feedback into an observable event
“Weak business judgement” could mean you chose the wrong metric, ignored a constraint or failed to explain a perfectly sensible conclusion. “Poor communication” could mean too much detail, missing logic or an answer to a different question. Those labels do not select an exercise.
Ask your partner to reconstruct the moment: what information you had, what you said and what consequence followed. “You recommended the promotion because revenue increased, without calculating its effect on profit” is something you can repair.
For every calculation or branch, ask what it helps the client decide. Make that connection explicit before doing more analysis.
Distinguish understanding from execution
Give yourself the same question again without time pressure. Explain the logic before doing the arithmetic.
If you still cannot explain why you are dividing one number by another, learn the concept. If the explanation is sound but you copy the wrong figure while speaking, change your note layout or execution routine. If you calculate correctly but give no recommendation, practise interpreting the result.
The exercise should target the cause:
| Observed mistake | First thing to investigate | Useful repair |
|---|---|---|
| Wrong percentage | Which base did you intend to use? | Label numerator and denominator before calculating |
| Generic structure | How would each branch change the decision? | Rebuild one prompt after changing its objective |
| Long, unclear answer | Which sentence actually answers the question? | State that sentence first, then justify it |
| Ignored new information | Did you understand what changed? | Update the calculation and recommendation aloud |
| Correct result, no implication | What comparison would make the result useful? | Compare with a target, threshold or alternative |
Quick Case Math Drill
A meal-kit company plans a price cut to win more orders. Marketing argues that the extra volume will easily pay for the lower price.
- Current price per box
- £50
- Variable cost per box
- £25
- Proposed price cut
- 10%
By what percentage must box orders rise to keep total contribution unchanged?
Work through a mistake that looks like good maths
A manufacturer has 200 hours available on its bottleneck machine each week. Product A contributes £40 per unit and takes two machine hours. Product B contributes £30 per unit and takes one hour. Demand is sufficient for either product, and these contributions already deduct variable costs.
A candidate chooses A because £40 exceeds £30. The subtraction is not the problem. The candidate has compared the wrong measure for the constrained resource.
A contributes £20 per bottleneck hour; B contributes £30. Using all 200 hours for A produces 100 units and £4,000 contribution. Using them for B produces 200 units and £6,000. Subject to the stated assumptions, B creates £2,000 more contribution from the same capacity.
The repair is to explain why the denominator matters: “Machine time limits output, so I need contribution per machine hour.” That sentence connects the economics to the constraint.
Now change demand. If the market can absorb only 120 units of B, allocate 120 hours to B and the remaining 80 to 40 units of A. Total contribution becomes £3,600 + £1,600 = £5,200. “Always choose B” was never the lesson. Allocate scarce capacity using the economics and the actual demand limits.
Practise the missing choice on a new problem
Do not immediately repeat the same factory numbers until the answer feels easy. Transfer the reasoning to a different setting: a repair shop with limited technician hours, a courier with limited van space or a clinic with limited appointment capacity.
Make yourself identify what is scarce before selecting the comparison. Sometimes the relevant denominator changes. If demand is scarce but capacity is abundant, maximising contribution per machine hour may no longer answer the decision.
Attempt the fresh problem before looking at the answer. Then use the explanation to locate and correct the mistake in your reasoning.
Prioritise the error that changes the answer
Repair a misunderstood objective, an invalid equation or an ignored constraint before polishing small wording choices. Those errors can undermine everything that follows.
Next consider repetition. A calculation slip you caught immediately may need less attention than consistently failing to choose what to investigate next. Keep one main target for the next session so your partner can watch for it carefully.
You can still note other problems. Put them in the log and return to them after the primary repair has improved. Trying to change your structure, speed, eye contact, creativity and conclusion simultaneously makes progress difficult to observe.
Quick Answer Repair Drill
A cinema chain asked a candidate how to raise its profit. The candidate recommended rolling out the pilot loyalty scheme across every site.
- Client objective
- Higher operating profit this year
- Pilot membership
- Up 30% across pilot sites
- Pilot site profit
- Unchanged versus the prior year
Which flaw should the candidate repair first in their next practice session?
Resolve conflicting feedback with the case in front of you
If one partner says “more detail” and another says “be concise,” both may be right about different moments. Ask where the missing detail was and which sentence could have been removed.
For a structure, compare the reasoning behind each recommendation. One reviewer may have noticed a capacity constraint the other missed. A disagreement about three branches versus four is less useful than a discussion of whether the decision is covered.
If you cannot diagnose the conflict, bring your prompt, notes and feedback to an experienced reviewer. The coach-selection guide explains how to make that session productive.
Record a result you can use next week
Write one short log entry: mistake, cause, repair and retest result. For the factory example: “Compared contribution per unit despite a machine-hour constraint. Practised identifying the scarce resource. Retested on a technician-hours problem without a prompt.”
Revisit the skill in a later full case. A successful replay shows understanding; a fresh, unprompted use shows the correction is becoming available when you need it. That is the progress worth tracking before you add more mocks to the calendar.
Frequently asked questions
Working out what actually went wrong
My feedback just says 'weak business judgement'. What do I do with that?
Ask your partner to reconstruct the moment it happened: what information you had, what you said and what followed. The label could mean the wrong metric, an ignored constraint or a sound conclusion you failed to explain. Each of those needs a different exercise, so the vague label alone cannot tell you what to practise.
How can I tell whether feedback is specific enough to act on?
Imagine replaying the moment where things went wrong. If you would know exactly what to do differently, the feedback is usable. If you would not, it is still too vague. 'Poor communication' fails this test, while a description of the sentence you said and what it led to usually passes.
How do I know whether I misunderstand a concept or just slipped under pressure?
Retry the same question without time pressure and explain the logic before doing any arithmetic. If you still cannot say why you are dividing one number by another, you need to learn the concept. If your explanation is sound, the problem lies in execution, and the repair is different.
I understand the logic but keep reading out the wrong figure. What should I change?
Change your note layout or the routine you follow when executing, rather than relearning the concept. Copying the wrong figure while speaking is an execution problem. Re-studying the theory will not help if the explanation was already correct.
My maths is right, but I never land a recommendation. Is that a separate problem?
Yes. Correct arithmetic followed by no recommendation means you need to practise interpreting the result. After each calculation, say what the number means for the decision. This is a different repair from fixing a concept or a note-taking routine.
One partner says add more detail and another says be concise. Who is right?
Both may be right about different moments. Ask the first partner where detail was missing and the second which sentence could have been cut. Tied to the actual case, the two comments usually point to separate fixes rather than contradicting each other.
Two reviewers disagree on whether my structure needed three branches or four. How do I settle it?
Compare the reasoning behind each suggestion, not the branch count. One reviewer may have spotted something the other missed, such as a capacity constraint. Knowing why a branch matters to this case is more useful than arguing about how many branches a structure should have.
Repairing the reasoning
Why is choosing the higher-contribution product wrong in the factory example?
Because machine time, not demand, is the limit. Product A earns £40 but uses two bottleneck hours, so it makes £20 per hour. Product B earns £30 for one hour. Across 200 hours, B produces more total contribution. The candidate's subtraction was fine, but they compared per unit when they should have compared per scarce hour.
Should I redo the same case until the answer feels easy?
No. Repeating the same numbers mainly tests your memory. Apply the reasoning to a different setting instead, such as a repair shop with limited technician hours, a courier with limited van space or a clinic with limited appointments. Each time, make yourself identify what is scarce before choosing the comparison.
Is contribution per bottleneck hour always the right measure?
No. The right denominator depends on what is scarce. If demand is the real limit and capacity is plentiful, comparing per machine hour may no longer give the right choice. The skill you are building is to identify the scarce resource first, and then choose the comparison.
Prioritising and tracking fixes
I have several mistakes. Which should I fix first?
Fix errors that change the answer first: a misunderstood objective, an invalid equation or an ignored constraint. These can undermine everything that follows. After that, look at frequency. A slip you caught straight away matters less than consistently failing to choose what to investigate next.
How many things should I work on in my next practice session?
Pick one main target so your partner can watch for it closely. You can still note other issues as they come up, but spreading attention across several fixes makes it harder for your partner to judge whether the one that matters most has improved.
How should I record a mistake so I can track it?
Write one short entry covering the mistake, its cause, the repair and the retest result. For the factory case: compared contribution per unit despite a machine-hour constraint; practised identifying the scarce resource; retested on a technician-hours problem without a prompt. The log tells you what to check in later sessions.
If I get the replay right, is the mistake fixed?
Not yet. A successful replay shows you understand the correction. Using it unprompted on a fresh problem shows it is becoming available when you need it. Return to the skill in a later full case to confirm it holds in a complete conversation.