3. How do I act?
3.2.8 · 35–55 minutes · reflect, test, discuss, review
3. How do I act?
What you will learn
- Create a dated decision journal and premortem.
- Translate a concern into an indicator and review action.
- Review process and outcome separately without hindsight rewriting.
3.2 · The Human Side
The Human Side
The Human Side · the work you will do
- 1What do I know?
- 2What influences me?
- 3How do I act?
- 4Record and review
Before you begin
Your first decision
Your club is about to back DailyCup in a fictional investment committee. Imagine it is a year later and the decision has disappointed. Before you vote, write three plausible reasons why. Which could you investigate now?
Record your first answer before continuing.
Video coming soon · about 4 minutes
Build a Better Decision Process
Use the narration below while the recording is being made, then try the decisions in this lesson.
Read the full narration
Your club is about to back DailyCup in a fictional investment committee. Imagine it is a year later and the decision has disappointed. Before you vote, write three plausible reasons why. Which could you investigate now?
In this lesson, notice what affects your own judgement. You do not need to invest real money or share personal financial information.
Write the decision, objective, price and alternatives. Add the evidence, key assumptions, strongest opposing case and what would change your view. Date it. When you revise the recommendation, add a new entry with the reason rather than rewriting the old one. Your journal should let you see your learning, including the awkward parts.
A premortem asks you to imagine that the plan has gone wrong and work backwards to plausible causes. Start independently so the first confident speaker does not set every answer. For DailyCup, consider customers, cash, delivery and the committee’s own process. Use the exercise to find something worth checking, rather than compete to invent the most dramatic disaster.
“Sales might disappoint” gives nobody a job. Define the measure, source, checking date and response. For example, investigate a shortfall in paid repeat visits against the case plan, using a consistent customer group and reconciled records. A review trigger tells you to examine the decision; it need not mean an automatic sale. Explain any later change to the trigger.
At review, compare the original reasoning, what was done, the new evidence and the result. Choose one improvement you can use next time: an independent first estimate, a credible challenge, a clearer forecast or a dated review. A useful checklist supports judgement. It cannot replace it. Bring your process into the final company pitch and keep improving it afterwards.
Pause for the worked example and the choices. Use the feedback to improve the explanation, then apply it to the fictional case. Keep your first view as well as any revision.
Choose the one safeguard that would help you most. Explain when you will use it, what it could change and how you will tell whether it helped.
Leave a record your future self can use
Write the decision, objective, price and alternatives. Add the evidence, key assumptions, strongest opposing case and what would change your view.
Date it. When you revise the recommendation, add a new entry with the reason rather than rewriting the old one. Your journal should let you see your learning, including the awkward parts.
Research and nuance
A useful journal records the decision, objective, price, alternatives, evidence, assumptions, uncertainties, contrary case and review date. Include what would change the recommendation. Preserve the original entry; later revisions should be appended with dates and reasons. This makes it possible to compare what was believed with what became known rather than rewrite a smoother story afterward.
Imagine a failure before committing
A premortem asks you to imagine that the plan has gone wrong and work backwards to plausible causes. Start independently so the first confident speaker does not set every answer.
For DailyCup, consider customers, cash, delivery and the committee’s own process. Use the exercise to find something worth checking, rather than compete to invent the most dramatic disaster.
Research and nuance
A premortem asks the group to imagine failure and identify plausible causes before committing. Generate ideas independently, then group them into business, financial, operational and process concerns. The technique is associated with Gary Klein’s practical work and is also discussed in decision-making teaching. It does not forecast failure or remove uncertainty. Use the exercise to uncover overlooked mechanisms rather than invent dramatic stories.
Turn a worry into a check
“Sales might disappoint” gives nobody a job. Define the measure, source, checking date and response.
For example, investigate a shortfall in paid repeat visits against the case plan, using a consistent customer group and reconciled records. A review trigger tells you to examine the decision; it need not mean an automatic sale. Explain any later change to the trigger.
Research and nuance
Turn “demand may disappoint” into a defined indicator, source, frequency and proportionate response. Specify which condition requires a review, which requires immediate verification and what evidence would permit continuation. A trigger is not always an automatic sale rule. Some thresholds deserve reconsideration as the business changes, but record the reason for changing them rather than moving them to protect the thesis.
Build a process you would actually repeat
At review, compare the original reasoning, what was done, the new evidence and the result. Choose one improvement you can use next time: an independent first estimate, a credible challenge, a clearer forecast or a dated review.
A useful checklist supports judgement. It cannot replace it. Bring your process into the final company pitch and keep improving it afterwards.
Research and nuance
At review, compare original reasoning, execution, new evidence and results separately. Ask which error was avoidable, which assumption needs updating and what process adjustment is worth trying. A checklist can support consistent work, but cannot guarantee good outcomes. The final deliverable is a repeatable decision practice that pupils can use in their company pitch and in other uncertain business choices.
From vague worry to a review trigger
A trigger set before the result
At which observation is a review triggered? What still needs verification?
For a fictional monitoring example, the committee records a post-promotion paid repeat-visit threshold of 55% for a defined monthly loyalty cohort. It requires review after two consecutive completed cohorts below 55%, with at least 200 observed customers per cohort and reconciled payment data. Observations are 58%,53%,52%.
- The first cohort is above the threshold.
- The second is below; verify the measurement and continue the stated monitoring rule.
- The third produces two consecutive eligible cohorts below 55%, so the specified review is triggered.
- A review re-examines demand, promotion/mix and cash assumptions; it is not automatically an instruction to trade.
The threshold is an invented governance choice for this exercise, not a scientifically established safe loyalty rate. Population changes or data errors require explanation.
Put it into practice
1. Complete the journal
Write a dated DailyCup entry with objective, action, price, thesis, contrary case, evidence gaps, uncertainty and two review conditions.
Write your response in your workbook before opening feedback.
Compare with a model response
Keep the initial 250–400k range and 300k asking price conditional, note the realised-price question, and state what would justify revision. Do not claim a probability distribution from the range.
2. Run the premortem
Imagine the hypothetical investment disappointed. Independently list three plausible causes, then choose one material overlooked cause for evidence gathering.
Write your response in your workbook before opening feedback.
Compare with a model response
Possible mechanisms include promotion-led demand, underestimated staffing or capex, and a committee that suppressed a warning. Link each to evidence rather than simply predict a price fall.
3. Write the later review
Use a supplied outcome card to separate what was known then, what changed and what you would improve.
Write your response in your workbook before opening feedback.
Compare with a model response
Retain the original memo. Distinguish an avoidable missed source from an acknowledged adverse risk or new event. Recommend a specific process change without claiming it guarantees future returns.
Notice your own judgement
Choose the one safeguard that would help you most. Explain when you will use it, what it could change and how you will tell whether it helped.
Original On Judgement teaching. Research and further reading: Likierman; Yale teaching.
Jargon buster
Decision journal
A dated record of reasoning, uncertainty and review conditions.
Premortem
Imagining failure before action to investigate plausible causes.
Indicator
An observable measure linked to a risk mechanism.
Review trigger
A defined condition requiring reassessment or verification.
Implementation
How a decision is actually carried out.
Learning loop
Recording, acting, reviewing evidence and improving the process.
Knowledge check
Make each choice on the page. Feedback explains the reasoning. Use Try again after revisiting an idea.
AQA connection and further reading
AQA 7138 applied links: AO 2–AO 4 evidence, analysis and evaluation (4.2); uncertainty and change (3.3.4), with quantitative interpretation in Annex 6 where used. Named psychological effects and calibration methods are investing enrichment, not specified AQA content. Classroom exercises illustrate processes; they do not diagnose individual traits.
AQA 7138 specification
Primary curriculum reference; links and original classroom rubrics are interpretive, not endorsement.
Likierman · six elements explained by London Business School
Public LBS explanation checked on 1 November 2026. Six elements are attributed to Sir Andrew Likierman; the three-area grouping and investing exercises are On Judgement adaptations, not his published model or an endorsement.
Andrew Likierman · Judgement at Work: Making Better Choices
Publisher book page. The full book has not been reviewed; teaching uses the publicly explained six-element framework with original examples.
Premortem tool · AHRQ
Institutional method reference credits Gary Klein. The DailyCup journal and business-risk workshop are original.
Annie Duke · process and resulting interview
Author-hosted discussion informs the short attribution of resulting. Analyst cases and process tools here are original.
Understand yourself as an investor
Choose one safeguard you will actually use. Specify the situation, action and later evidence that would show whether it helped.
Keep this reflection private unless you choose to share. Use evidence from the fictional exercise; no personal finances or fixed investor-personality label is required.
Take forward
Create a dated decision journal and premortem. Translate a concern into an indicator and review action. Review process and outcome separately without hindsight rewriting.