strategyplaybook.aiSqwyz
Sign in
← All 124 tools
Monitor · Review

After action review

Four questions asked shortly after something finishes, to capture what actually happened before everyone reconstructs it. Only works where blame is genuinely off the table.

The printable canvas and the handoff into Claude Code are part of a paid plan. See what a plan includes. It needs the Claude desktop app on this machine, and your team and the prompt library already installed in that project — we cannot see your disk, so open it there.

Stage
05 Review
Works at
Business, Department
Maturity
Early startup → Large organisation
Time
An hour, within two weeks of the event

What it is

A structured review conducted shortly after an event, built on four questions and originating in military practice. Its discipline is that it establishes intent before outcome, which prevents the retrospective editing that makes most post-project reviews useless.

What we intended
The original objective and plan, stated before outcomes are discussed. Fixing this first is the method's key move.
What actually happened
The sequence of events as agreed by the participants. Disagreement about the facts is itself informative.
Why the difference
The analysis. What caused the gap between intent and outcome, examined without attributing blame.
What we change
Two or three specific, owned changes. Generic lessons are a way of appearing to learn.
Psychological safety
A precondition rather than a component. Without it the review produces a sanitised account and the actual lessons stay in the corridor.
The mistake people makeIf blame is still live, this becomes a hearing and everyone will present a defensible version. Either wait, or have it run by someone with no stake — and say explicitly at the start that it is not an accountability exercise.
What it’s forSomething has finished, well or badly, and you want the learning while it is fresh.
What it’s not forBlame is still live. Wait, or it becomes a trial.

How you run it

  1. Run it soon, but not immediatelyWithin a week or two. Too soon and emotions dominate; too late and everyone has settled on a version of events.
  2. Ask what we intended to happenEstablish the original intent before discussing outcomes, or it gets retrospectively edited to match what occurred.
  3. Ask what actually happenedFacts and sequence, agreed by the room. Disagreement here is itself a finding about visibility.
  4. Ask why there was a differenceThe analytical step. Requires the room to be safe, which is a condition of the exercise rather than a nice-to-have.
  5. Ask what we do differently, specificallyTwo or three changes with owners. "Communicate better" is not a lesson; it is a way of avoiding one.

The prompt

Two ways to run it

Run this tool in your own Claude

The short prompt starts your partner against the library on your disk. The long one carries everything with it and needs nothing installed.

🔒 Prompt locked

Everything above is free — what this tool is, what it is for, what it is not for, and how to run it as a workshop. Three tool prompts a month are free with an account; beyond three, and for the printable canvas, it is the paid part.

See plans — from £19 a month

Your playbook

It lands in the earliest stage this tool suits. Move it on the Playbook page.

You’ll need

  • Something that has finished, well or badly
  • The people who were actually involved
  • A facilitator with no stake and enough authority to hold the room

You’ll end up with

  • What was intended versus what happened
  • Why the difference, honestly
  • Two or three things that change next time

Runs well with

Assumption mappingEvidence logExecutive summaryOKR design