The after-action review: twenty minutes that turn one success into a repeatable method.

Imagine a natural history museum that has just opened a temporary night exhibit. The opening went well. Families moved through on time, the fragile specimens stayed protected, and the final visitors left smiling. By the next morning the staff is already talking about the next program. That is exactly when a useful lesson can disappear.

A failed event demands attention. A successful event lets everyone move on. But if nobody stops to ask why it worked, the good parts remain accidents. The next team has to discover them again.

The after-action review is built for this moment. The United States Army formalized the practice during training, and the structure is four questions asked in order. What was supposed to happen? What actually happened? Why was there a difference? What should we keep, and what should we change next time?

The order matters. A normal debrief often begins with “What went wrong?” That question makes people protect themselves before the learning starts. An after-action review begins by reconstructing the plan. Nobody is blamed for stating what the group intended. Then the group lays out what actually happened, using observable facts. Only after the plan and reality are visible together does anyone explain the gap.

Researchers including Marilyn Darling studied how military units used these conversations. The effective reviews were short, happened while memory was fresh, involved the people who did the work, and treated rank as less important than accurate observation. The review was not a polished report created weeks later. It was a practical conversation close to the action.

Run the four questions on the museum opening.

What was supposed to happen? The staff planned to admit one group every twelve minutes. A volunteer would greet each group, a guide would lead the gallery portion, and the conservation room would open only after the previous group had cleared it. The final tour was meant to end before the cleaning crew arrived.

What actually happened? The first two groups arrived early and crowded the lobby. A volunteer moved the welcome talk outside, which cleared the entrance. Midway through the evening, a display light overheated and the conservator switched to a backup case. Tours ran seven minutes behind, but the guide shortened the final demonstration and the schedule recovered. The last family left before the service doors opened.

Why was there a difference? The reservation message gave an arrival time but did not explain that doors would remain closed until the assigned slot. That created the lobby crowd. The lighting problem happened because the case had been tested during the day, when the room was cooler. The team recovered because the volunteer understood the goal, not just the script, and because the conservator had prepared a usable backup.

What should the museum keep? Keep the backup case ready. Keep giving volunteers authority to move the welcome talk when the entrance becomes crowded. Keep the shorter version of the final demonstration, since visitors still understood the main point.

What should change? The reservation message should explain when doors open. The equipment test should run under the temperature and lighting conditions expected during the real event. The schedule should include a small recovery gap before the last tour.

Notice what the review did. It did not turn a good night into a hunt for faults. It converted improvisation into a method. Without the review, “move the welcome outside” belongs to one quick-thinking volunteer. With the review, it becomes an option the next event team knows it has. Without the review, the backup case is just something a cautious conservator happened to prepare. With the review, it becomes part of the operating plan.

Amy Edmondson's work on psychological safety helps explain why the factual sequence matters. Teams learn only when people can name what differed from the plan. Putting the intended sequence and the observed sequence side by side makes the gap a shared object to examine. It becomes less about who failed and more about what the system asked people to do.

The review does not need a special meeting room or a long document. It needs the people who saw the work, twenty focused minutes, and a place where the next team will find the answers. For the museum, that might be a note attached to the event checklist. The format can stay simple: planned, actual, why, keep, change.

Do this after good work as well as bad work. If you only review failures, you learn what to avoid. If you review successes, you also learn what to repeat. Repetition is where speed and confidence come from. A team gets better when its best recovery stops depending on the same person being present.

There are two common mistakes. The first is turning the review into an evaluation of people. Save performance conversations for another setting. The second is producing so much documentation that nobody uses it. Capture only what will change a future decision or preserve a useful move.

Try this after the next event, shipment, presentation, or deadline your group completes. Within a day, gather the people who did the work. Give each of the four questions equal time. Start with the plan. Build the actual timeline from facts. Explain gaps without hunting for a villain. End with no more than a few specific things to keep and change, and put them where the next plan begins.

That is lesson twenty. A finished event gives you a result. A short after-action review gives you a method you can use again.