You will set up your decision journal and complete a full review of one decision using the whole course.
You have spent this course collecting tools one at a time: argument maps, evidence rankings, chart checks, causal questions, a bias checklist, door types, pre-mortems, probabilities and expected value. Used separately, each one catches a particular kind of mistake. This project puts them together on your own decisions, and leaves you with a journal you will keep using after the course ends. Allow about an hour, split across two sittings if that suits you.
You will produce three things. The first is a decision journal, set up in a tool you actually open every week, with at least five live entries. The second is a full review of one past decision, worked through with the tools from every module. The third is a one-page summary of what that review taught you and the one thing you will change.
Choose where the journal will live. A spreadsheet works well because you can sort entries by review date and count predictions by probability later. A notes app or a paper notebook is fine if that is what you will open. The best tool is the one you already check every week. A beautiful system you never open is worse than a plain list you see every morning.
Set it up with the fields from lesson 8.2, What goes in a decision journal: date and decision, options, choice and reasons, expectations with probabilities, state of mind, and review date. Add a column for the outcome and a column for your process grade, both left empty for now.
Then write five entries for decisions you are making now or will make in the next few weeks. They do not all have to be large. A mix is good: perhaps one career decision, one work project, one purchase, one people decision and one smaller choice. Each needs at least one prediction with a probability and a time frame, as in lesson 7.1, Decisions are bets. Put the review dates in your calendar alongside the quarterly reviews from lesson 8.3.
Choose a decision from the past six to eighteen months whose outcome you now know. It should be important enough to deserve an hour and recent enough that you can reconstruct what you knew. Then work through it in order, writing brief notes for each step:
Argument map: what was the main claim behind the decision, what evidence supported it, what was the unstated warrant, and where was the weakest link (Module 1) Fallacies: were any of the arguments for or against it ad hominem, a false choice, an appeal to popularity, or a sunk cost (Module 2) Numbers: did any figure or chart influence the decision, and would it survive the checks on axes, averages, base rates and missing data (Module 3) Causes: did the decision rest on a belief that one thing causes another, and were the alternative explanations considered (Module 4) Biases: run your checklist from lesson 5.4 against the decision as it was made Door type and process: was it a one-way or two-way door, how many options were considered and was there anything like a pre-mortem (Module 6) Probabilities: what were the realistic outcomes and their rough probabilities at the time, what was the expected value and what was the worst case (Module 7) Process grade: using lesson 8.1, Good decisions can have bad outcomes, grade the process separately from the result
Daniel, from Module 4, reviews the app download campaign his team ran last year. The campaign raised downloads but spending per customer barely changed. His argument map shows the weakest link was the warrant that downloading causes spending. The numbers step finds the "twice as much" figure compared all app users with all web users, with no before-and-after comparison. The causes step names loyalty as the confounder nobody checked. His bias checklist catches confirmation: the team wanted the app to matter and searched only for supporting data. The decision was a two-way door treated as a one-way door, with the whole budget spent at once rather than tested on a random sample. Daniel grades the process poorly, and notes that the outcome would not have changed his grade, because even if spending had risen, the team would not have known whether the app caused it.
Finish with a single page that someone else could read in five minutes. It should cover the decision, the outcome, your process grade and the two or three steps of the review that taught you most. End with the one change you will make, written as specifically as the changes in lesson 8.3, Review your decisions every quarter.
Daniel's change: "Before any campaign over a set budget, we run a randomised test on a small sample first, as in lesson 4.2, and decide on the full campaign from the test result."
A finished project has a journal with at least five entries, each with a probability, a time frame and a review date in your calendar. It has a full review of one past decision with notes for every step, and a one-page summary that ends in one specific change. Nothing more is required, and nothing less will give you the habit.
The final project for this course takes the same tools and applies them to a decision still in front of you, so the journal you build here is where that decision will be recorded too. Set up the journal first, then the activity below takes you through the rest.
Set up the journal with five entries, complete the full review of one past decision, and write the one-page summary.
Junxiong-WFG Organisation is an authorised representative of AIA Financial Advisers Private Limited (Reg. No. 201715016G).