Recipes for AI
Strategy & Executive

The "Decision Memo Machine" (Prompt)

Turn a messy Slack thread into a crisp, stakeholder-ready decision document.

CONTEXT

No context provided.

THE PROMPT (Copy & Paste)

Act as a chief of staff helping an executive team make a structured, well-documented decision.

I'll give you raw context about a decision we need to make — this could be meeting notes,
a Slack thread, bullet points, or scattered thoughts.

Produce a decision memo with these sections:
1. Decision Statement — one sentence: "We are deciding whether to [X]."
2. Context — 2–3 sentences of relevant background
3. Options Considered — 2–4 options, each with a one-line pro and one-line con
4. Recommendation — your recommended option with a 2-sentence rationale
5. Risks if we're wrong — the top 2 things that could go sideways
6. Decision Owner & Due Date — placeholder fields

Keep it under 350 words. No fluff. Use plain section headers.

[PASTE DECISION CONTEXT HERE]

EXAMPLES

Raw Data:

"We've been debating whether to build our own collections dialer UI or buy Latitude/YAKK. Engineering says 6 months to build. Finance says the vendor costs $180K/year. Collections team says the current tool is broken and they lose 40 agent-hours/week to manual workarounds. Legal flagged that any new dialer needs TCPA compliance review."

Result:

Decision Statement: We are deciding whether to build an in-house collections dialer or purchase a third-party solution.

Options Considered:

  • Build in-house — Pro: Full control, integrates cleanly with our stack. Con: 6-month timeline means 6 more months of lost productivity.
  • Buy Latitude — Pro: Deploy in 4–6 weeks, TCPA compliance built in. Con: $180K/year recurring cost, vendor dependency.
  • Defer 90 days — Pro: Buy time for a proper RFP. Con: Collections team continues losing 40 hrs/week now.

Recommendation: Purchase Latitude. The $180K annual cost is recovered in under 6 months by eliminating 40 hours/week of manual agent work at fully loaded cost. Build capacity is better spent on core underwriting.

Risks if we're wrong: (1) Vendor lock-in limits flexibility if we scale collections significantly. (2) Integration with our servicing stack could surface hidden complexity post-contract.

Decision Owner: [Name] | Due Date: [Date]