ChatGPT prompt for writing a decision memo

This is the prompt we run on any decision that has been dragging for more than three days. Its core rule: it does not just argue for one option, it then separately builds the strongest case against it. You decide having seen both sides, not one.

The prompt to copy

Help me decide: {decision}. Context: {paste}. Options I see: {list}. Produce a one-page memo: framing of what's actually being decided, options with honest pros/cons and what each optimizes for, what information would change the answer, then argue for the strongest option — and separately steelman the strongest case against it.

Replace the fields in curly braces (the decision, the context, your options), then paste whatever you have, even in rough form. You do not need to have made up your mind: that is exactly the prompt’s job.

When to use it

Whenever a decision keeps circling back for several days without moving. A hire, a tool switch, a pricing table, taking or turning down a big project: the one-page format forces clarity where rumination just loops. It is the core of Playbook P9.

What you get back

1. The framing of what is actually being decided, often different from the question you started with.
2. The options laid flat, with what each one optimizes for and what it costs.
3. The list of information that would change the answer, so you know what is left to check.
4. An argument for the strongest option, then the strongest case against it.

Real example: a decision you keep mulling → a one-page memo

Input: Maya, who runs Brightside Studio, an 8-person brand agency, has spent two weeks weighing whether to create a ninth role. What she pasted, as-is:

decision: hire a full-time designer, or keep using freelancers per project
we're turning down projects for lack of design bandwidth. a full-time hire runs ~55k/yr. freelancers cost ~3k/mo right now but climb fast when we're busy
options: (a) full-time, (b) stay all-freelance, (c) a recurring freelancer 3 days/week

Output, the two parts that matter:

What’s actually being decided: not “designer or no designer,” but “at what level of steady load do we fix a structural cost to gain delivery reliability.”

The strongest case against the full-time hire: your design load is uneven. In a quiet month a full-timer costs 55k for unbilled time, where a freelancer drops to zero. Until you can show three straight months of projects turned down for lack of design, option (c) captures most of the benefit without the fixed risk.

Notice what the AI did: it argued for one option, then turned the argument against it, instead of handing back a one-sided recommendation. And it named the deciding fact, three straight months of turn-downs, that Maya can go verify before she signs anything.

This prompt is part of the library that ships inside the kit, with Playbook P9 turning it into a decision reflex.