Decision Card: one bet in writing

Goal. The card turns an important decision into one bet in writing: the question, the options, the choice with reasons, your confidence as a percentage, a stop signal and a review date. The decision stops hanging for months, and later you judge how you decided, not only the outcome, which always holds some luck.

In 15 minutes. (our estimate, not measured) The shortest path to one bet in writing. Take the decision that has waited longest and use only the fields in brackets.

1. (2 min) Write the question in one sentence: “We are deciding whether … or …” (field 2).
2. (3 min) List three options, including “we change nothing” (field 3).
3. (3 min) Choose one. Write two reasons and your confidence as a percentage (fields 4 and 5).
4. (4 min) Imagine the review day has come and it didn’t work. Write the most likely cause and a stop signal (fields 6 and 7).
5. (3 min) Put the review date in your calendar and write one line: “I decide that … Who … By when … I will check …” (fields 8 and 9).

Instructions (full version) (estimate: 15 to 20 minutes per decision; we have not measured this)

1. Check whether the decision goes on a card (the list at the bottom). Write small, easily reversed matters down in one sentence, or not at all.
2. Write the question in one sentence: “We are deciding whether … or …”. If you can’t write it, the decision isn’t ready yet.
3. List at least three options, including “we change nothing.” Choose one, give two or three reasons and your confidence as a percentage.
4. Imagine the review day has come and it didn’t work. Write down the most likely cause and what you do about it right now.
5. Set a stop signal and put the review date in your calendar. On review day, judge the process first (the four questions at the bottom), and only then the result.

Template fields

Does the decision go on a card? Yes, if it meets at least one condition: ☐ it is hard to reverse ☐ it exceeds [x] PLN or [x]% of monthly revenue ☐ it concerns a key customer, a hire or a price ☐ we have already come back to it once.

Review on the set date. Read the card only up to field 7 and answer before you check the result: ☐ were the options real ☐ were the reasons and the information sensible for that day ☐ was the confidence a number, not a word ☐ was the stop signal set and was someone watching it. Only then: what happened, how much of it was luck, and what we change for the next decision.

Fictional example (the firm and the numbers are fictional, for this template)

An accounting firm, 9 people.

1. November 4 · the owner decides · we ask the team manager.
2. We are deciding whether to hire a second accountant now, or first to move the 20 smallest clients to a fixed package with ready-made document templates.
3. Options: hire now; the package first, with a decision on the hire in February; we change nothing and work overtime.
4. Choice: the package first. Reasons: 40% of the team’s time goes to the 20 smallest clients; a new person takes 3 months to get up to speed, and the peak comes in January; the package can be reversed.
5. We are 70% sure this will work, and “work” means: at least 12 of the 20 clients in the package by December 15, and 10 fewer hours of work a week on their accounts.
6. Cause of failure: clients don’t want to change the way they send documents. Now: the team manager calls five of them before we send the offer.
7. We stop if fewer than 6 clients say “yes” by December 1. Then we go back to the first option.
8. Review: December 15.
9. The whole team hears about it at the Friday meeting. First step: a draft of the package, the team manager, by November 8.
10. Agreed: the owner and the team manager. Disagreed: the senior accountant, who would rather hire, but carries out the decision anyway.

From us. The ideas come from books: Annie Duke (Thinking in Bets) gives confidence as a percentage, imagining failure in advance, and judging a decision separately from its outcome; Andy Grove (High Output Management) gives the rule that people carry out a decision even if they disagreed with it; John Hammond, Ralph Keeney and Howard Raiffa (Smart Choices) give a well-framed question and options wider than “as usual.” What we added is one page per decision, the “does it go on a card” list and four questions for the review day.

When to call us.

• You can’t write field 5 as one measurable line: you don’t know how you will tell it worked.
• You are coming back to the same decision for the third time, or every option has equally good reasons.
• The decision concerns a key customer, a hire or a price, and people in the business see it differently.

The first call (30 minutes) is free. We will use it to see whether the card is enough or the decision needs more work.

Prompt for your AI. Paste it into your own ChatGPT, Claude or Gemini account; our own method, working from your company’s knowledge, is what we do in a Diagnosis.

You are a calm coach for the owner of a small business. Walk me through the
“Decision Card”: one question at a time, and wait for my answer after each.
At the start, remind me: “Do not paste passwords, account numbers or personal data.”
Order: 1) the question “We are deciding whether … or …”; 2) at least three
options, including “we change nothing”; 3) my choice, two reasons and my
confidence as a percentage; 4) the most likely cause of failure; 5) a stop
signal; 6) a review date.
If an answer is vague, ask for a number or an example from my business.
Never invent facts about my business; if you don’t know something, ask.
At the end, write one line: “I decide that … Who … By when … I will check …”.