Skip to content

Run a Strategy-Review Prep

When to use this

A strategy review is coming up. The inputs are everywhere. Notes from three meetings, a half-finished deck, last quarter's numbers, two conflicting memos. Use this to pull all of it into one tight pre-read. The point isn't a prettier document. It's starting the review on the real questions, not the warm-up.

This isn't a one-shot prompt. It's a loop. You feed in the raw material, the AI drafts a pre-read, and then you push on it. You cut what doesn't matter, sharpen what does, and force the hard questions to the front. Treat the AI as a thinking partner that organizes, not one that decides.

A prompt to start

Open the loop with something like this, then keep going turn by turn:

You are helping me prep a pre-read for an upcoming strategy review. I'll paste in my scattered inputs below — notes, numbers, memos, half-formed thoughts.

First, organize all of it into a draft pre-read: where we are, what's changed since last time, and the open decisions on the table. Keep it tight. Flag anything that contradicts itself or looks thin.

Then list the 5–7 questions this review should actually start on — the real ones, not the easy ones. Rank them. Then wait.

I'll react, cut, and reorder. We'll go back and forth until the pre-read is sharp.

My inputs:
[PASTE YOUR NOTES, NUMBERS, AND MEMOS]
[WHAT THIS REVIEW NEEDS TO DECIDE]

Work through it step by step

  1. Dump everything in first. Don't tidy it up for the AI. Messy notes, conflicting figures, loose ends — paste it all. The organizing is its job, and the contradictions are exactly what you want surfaced.
  2. Read the draft for what's missing. It will arrange what you gave it. It can't know what you left out. Scan for the input that should be there and isn't, then add it and ask for another pass.
  3. Push on the question list. This is the heart of it. Ask why those questions and not others. Cut the ones that are really last quarter's debate. Add the one everybody's avoiding.
  4. Make it cut, not pad. A good pre-read is short. If a section is filler, tell it to drop it. Keep pressing: "What can come out and still leave the decision clear?"
  5. Pressure-test the framing. Ask it to argue the opposite of how you've framed a decision. If the reframing is stronger, you found that before the room did.
  6. End on the opening question. Settle what the review starts on. One sharp question beats a status recap. That single line is the real output of the session.

What stays your call

The questions are yours. So is what makes the cut. The AI is good at organizing scattered inputs and proposing what to ask. It has no idea which decision actually matters most to your business, or which question your team needs to face this quarter. Use it to pull the pre-read together and to surface candidates. Then you decide what the review opens on, and what gets left out. Don't let a tidy draft talk you past the question you know is the one that counts.

Watch out for

  • Tidy is not the same as right. A clean pre-read can read well and still bury the real issue. Organized output feels finished. Check that the hard decision is on the page, not smoothed over.
  • Keep sensitive specifics out of a tool you don't control. Unannounced deals, named personnel issues, legal exposure — don't paste those in. Prep those parts in general terms, or offline.
  • It can flatten a real disagreement into a neat summary. If two memos conflict, that tension might be the whole point. Make sure the pre-read shows the disagreement instead of papering over it.
  • The question list is a starting point, not a script. It will propose questions that don't fit your situation. You know the review. Drop what doesn't apply and add what it missed.