← Back to Prompt Library

📊 Business & Productivity

Decision Memo Writer

Turn a fuzzy decision into a one-page memo: the question, the options, the trade-offs, what you would need to believe for each, and a recommendation.

The Prompt — replace [BRACKETS] with your details

Act as a chief of staff who writes decision memos for busy executives. Structure my decision.

The decision:
- What we are deciding: [STATE IT AS A QUESTION]
- Why now, and what happens if we do nothing: [CONTEXT AND THE DEADLINE, REAL OR SELF-IMPOSED]
- Who decides and who is affected: [PEOPLE AND TEAMS]
- The options as I see them: [LIST — say "not sure" if you want help generating them]
- What I know: [FACTS, NUMBERS, CONSTRAINTS]
- What I do not know: [OPEN QUESTIONS]
- Budget, time, and people available: [CONSTRAINTS]

Write a memo of one page with:
1. The decision, stated as a question with a date
2. Recommendation up front, in two sentences
3. Context: only what a reader needs to evaluate the recommendation, no history for its own sake
4. Options, including the one I did not list — doing nothing — each with: what it costs, what it gets us, what it forecloses
5. For each option, what we would have to believe for it to be right. This is the section that reveals the real disagreement
6. The trade-off I am consciously accepting with the recommended option
7. What would change my mind, stated as an observable thing
8. What we do in the next two weeks if this is approved, with owners
9. The two or three questions I expect to be asked, with honest answers

Rules: no adjectives doing the work of evidence. Where I have given you an assumption rather than a fact, label it. If my option set is missing something obvious, add it and say so.

How to use this prompt

  • Send the memo before the meeting and use the meeting to discuss section 5, where disagreements actually live.
  • Keep "do nothing" as a real option with real costs; it is often the honest baseline.
  • Write down what would change your mind before the decision, not after it goes wrong.

Why this prompt works

Most decision discussions are people arguing conclusions without surfacing the beliefs underneath. The "what would you have to believe" framing converts a values argument into a factual one you can sometimes resolve, and putting the recommendation first respects the reader who will only skim.

Variations to try

  • Ask for a pre-mortem: assume the recommendation failed a year from now, and list the reasons.
  • Add "the decision is reversible and cheap" and ask whether a memo is warranted at all.
  • Ask it to argue the strongest case against your recommendation before you circulate it.

Common mistakes to avoid

  • Presenting a single option with a supporting case, which is a proposal wearing a memo's clothes.
  • Burying the recommendation on page two, so busy readers form their own before reaching yours.
  • Letting assumptions read as facts, which is how a memo becomes a source of institutional error.

Works well with

Claude
ChatGPT
Gemini

Need a custom version of this prompt?

The free prompt generator builds a prompt tailored to your exact goal, framework, and target AI model — or paste this template into the optimizer to refine it.