What Is an After-Action Review?

A small team runs a big weekend sale in its online shop. On Monday, everyone is tired, the orders are shipped, and the next project is waiting. Nobody talks about what went wrong. Six months later, at the next sale, the same problems appear again. How can a team make sure it learns from what it has just done?

What is an after-action review?

An after-action review, often shortened to AAR, is a short, structured discussion held soon after a task, an event or a project. The people who did the work meet and compare what they planned with what really happened. They look for the reasons for the difference and decide what to keep and what to change. The method comes from the US Army, which uses it after training exercises and real operations.

Most versions are built on four questions.

1. What was supposed to happen?
2. What actually happened?
3. Why was there a difference?
4. What will we do the same or differently next time?

The order matters. The first two questions are about facts that everyone can check. The last two are about causes and opinions. Starting with facts gives the group a shared picture before anyone argues about reasons.

The Army’s official guide defines the review like this.

[An AAR] is a professional discussion of an event focused on performance standards, that enables soldiers to discover for themselves what happened, why it happened, and how to sustain strengths and improve on weaknesses.

The key words are “for themselves”. Nobody arrives with a verdict. The team finds the lessons together.

How does a review work after a real project?

Let us go back to the weekend sale. The team of four meets on Monday morning for 30 minutes. One person, who did not run the sale, guides the discussion and writes the answers on a board.

What was supposed to happen? The plan was 200 orders, all shipped within two days. The email to customers would go out on Friday at 6 p.m. The team expected fewer than five complaints.

What actually happened? The team agrees on the facts. There were 260 orders. A post on social media brought in 90 of them. The email went out three hours late. Two people packed all the orders, and 70 orders were shipped late. There were 15 complaints, almost all about late delivery.

Why was there a difference? Now the group looks for causes. The social media post worked better than expected, so demand was higher than the plan. Nobody had planned what to do if orders rose. The email was late because no single person was responsible for sending it. Each person thought someone else would do it. Notice the kind of question the guide asks here: “What in our process allowed this?” and not “Who forgot?”

What will we do next time? The team agrees on three points.

• Keep the social media post, because it brought in a third of the orders.
• Give every task one named owner, including the email.
• Agree in advance that if orders pass 150 by Saturday noon, two extra people come in to pack.

Each point has a name and a date next to it. The whole review took half an hour. At the next sale, the team does not start from zero.

The review made the gap between plan and result visible. It found the causes in how the team works, not in one person. Then it turned those causes into a few concrete changes.

What the after-action review teaches you

After anything that matters, ask yourself four questions: what did I plan, what happened, why was it different, and what will I do next time? Learning does not happen just because you have had an experience. It happens when you look back at it on purpose.

• Studies. After a test, compare the grade you expected with the grade you got. Then look at which kind of question cost you points. That tells you what to practise, which is more useful than feeling good or bad about the result.
• Work and projects. Teams in software, health care and aid work hold similar reviews. After the Indian Ocean tsunami of 2004, several aid organisations held a joint review to improve how they would respond to the next disaster.
• Personal life. A family can hold a ten-minute review after a holiday or a move. What did we plan, what really happened, and what will we do differently next year?

The method also has good evidence behind it. In 2013 the researchers Scott Tannenbaum and Christopher Cerasoli combined the results of 46 samples with 2,136 people. They found that debriefs, the wider family to which the after-action review belongs, improved performance by about 25 percent on average. The effect was similar for teams and individuals, in simulations and in real work, and in medical and other settings. Debriefs led by a facilitator worked about three times as well as those without one.

Is the review a hunt for the person at fault?

Many people think a review after a project is a meeting to find out who made the mistake. In fact, the Army designed it to be the opposite. The early reviews at its training centres were critiques. An observer told the unit what it had done wrong. Leaders became defensive, and soldiers learned what to do without understanding why. Over the years the observer’s role changed from critic to facilitator, who asks questions and lets the unit find the answers.

The reason is practical. When people fear punishment, they hide the facts the team needs. A review in which nobody admits the late email cannot fix the late email.

A second misunderstanding is that reviews are only for failures. A team should also review its successes. Otherwise it does not know which of its choices caused the good result, and it cannot repeat them.

How strong is the evidence, and where does the method fail?

The evidence is encouraging, but it has two limits. First, many of the studies in the 2013 review did not assign people to groups at random. So some of the effect may come from other differences between the groups. The authors say this themselves. Second, there were few studies without a facilitator or with little structure, so those comparisons are uncertain.

The method also fails in practice in three common ways.

1. It becomes a ritual. The team answers the questions, files a report, and changes nothing. A review without a concrete change has no effect.
2. People do not speak openly. If the boss is in the room and people fear for their jobs, the facts stay hidden.
3. The lessons go to the wrong place. Some companies collect lessons in a database for other teams. They forget that the main users of a review are the people who will do the next project.

Learn more

The after-action review has no single inventor. It grew out of US Army training research in the 1970s. The Army Research Institute developed it together with training exercises in which equipment recorded who “hit” whom in a mock battle. These records gave the unit objective facts to discuss afterwards.

Army researchers trace its roots further back, to the interviews that the historian S. L. A. Marshall held with soldiers right after combat in the Second World War. The method became standard at the Army’s National Training Center, which opened in California in 1981. The Army published a leader’s guide to it in 1993.

To use the method, start with the practical guide from ODI in the references below. It has the questions, tips for running the meeting and a real example. The review looks back after an event, and its partner looks ahead before one. That method is explained in What Is Inversion? Thinking Backward to Avoid Mistakes. Why open reviews need trust in a team is one of the themes of “The Culture Code” by Daniel Coyle: Summary and Review.