After action review template

This free after action review template walks a team through four questions after a piece of work: What was supposed to happen? What actually happened? Why was there a difference? What will we keep doing, and what will we do differently? It ends with an owner and a date for each change. Fill it in online or print it.

Fill in the template

AFTER ACTION REVIEW

What was supposed to happen?
   The goal or plan, as the team understood it at the start.
   ______________________________________________
   ______________________________________________

What actually happened?
   The facts, in order. Use data or records where you have them.
   ______________________________________________
   ______________________________________________

Why was there a difference?
   Causes in the work and the process, not who is to blame.
   ______________________________________________
   ______________________________________________

What will we keep doing?
   What worked and should happen again.
   ______________________________________________
   ______________________________________________

What will we do differently?
   One to three changes for next time.
   ______________________________________________
   ______________________________________________

Owner and date for each change
   Who will make each change happen, and by when.
   ______________________________________________
   ______________________________________________

Get the editable version: this template as a PowerPoint file with a worked example, in the Change Toolkit. Or learn to use it on your own change in the course Leading Change with Evidence.

How to use it well

These steps are practice suggestions, drawn from the research below where it applies.

Hold it soon, and keep it short. Aim for thirty minutes while memories are fresh.

Start with what was supposed to happen. If people had different goals in mind, that gap is a finding in itself.

Use records where you have them. The meta-analysis of after action reviews below found that objective performance review media were linked to larger effects (Keiser & Arthur, 2021, abstract read).

Look at the team's work as a team. The meta-analysis of after action reviews found that alignment to the individual or the team was linked to larger effects (Keiser & Arthur, 2021, abstract read). For a team review, look at team results.

Ask why, not who. Look for causes in the plan, the process and the tools. If the review becomes a search for someone to blame, people stop saying what happened. When one difference needs a deeper look, or the same problem keeps coming back, the root cause analysis template traces its causes and ends with fixes, owners and a check that they worked.

End with one to three changes, each with an owner. Keep the list short enough to get done. Put the changes on the agenda of the next review and check them first.

Run it on a rhythm. A regular review, held after each cycle of work, lets the team see whether last time's changes worked.

What the research says

Debriefs, also called after action reviews, have been tested in controlled comparisons.

About 25% better on average. A meta-analysis of 46 samples (2,136 people) linked debriefs to about 25% better performance on average than comparison groups (d = .67), falling to d = .54 with three outliers removed (Tannenbaum & Cerasoli, 2013, rated High in the Evidence Library after a full-text read). Average effects were similar for teams and individuals and across real and simulated settings (d = .56 in real settings, d = .70 in simulations). Many of the pooled studies were quasi-experimental rather than randomized.

Alignment and records. A later meta-analysis of after action reviews found them linked to an overall gain (d = 0.79 across 61 studies), and found two features linked to larger effects: alignment to the individual or the team, and objective performance review media (Keiser & Arthur, 2021, rated Medium from its abstract). The effects of the other features were often interactive; the combinations linked to the largest effects paired a self-led review with a team-aligned focus or with objective media.

Teams can run their own. In a cluster-randomized study with 174 business students in 35 teams, a guided team-led debrief, built on lessons from earlier debriefs, produced better team processes than an unguided one, and better processes were related to higher team performance (Eddy et al., 2013, rated Medium from its abstract). They were students, not employees, and only nine class sections were randomized.

Distance did not spoil it. In an experiment with 123 four-person teams in a training task, teams that held reviews had significantly higher performance and team efficacy than teams that did not, and reviews held at a distance did not show the expected losses (Jarrett et al., 2016, rated Medium from its abstract). It was a training task, not real work.

Not the same as trauma debriefing. A Cochrane systematic review of 11 randomized trials found that single-session individual psychological debriefing after trauma did not reduce psychological distress or prevent post-traumatic stress disorder (Rose et al., 2002, rated Medium from its abstract). That is a different practice with a different purpose. The evidence for reviewing the work does not carry over to it.

How it maps to the Unfrozen Loop

1. Week 0
   Before you start
2. 1
   Surface both ways
3. 2
   Make sense together, top included
4. 3
   Start it close to the work
5. 4
   Review on a rhythm, then hold the pace

   The debrief meta-analysis on this page is the High-rated evidence the Loop page lists for Part 4. The Loop's first move is a 30-minute review every two weeks, run by the managers, that ends with one named fix and one owner.

6. 5
   The middle’s own weather
7. Ring
   Safe to speak, and speaking changes something

   A review only surfaces what people are willing to say in the room. Keep it about the work, not about who is to blame.

Highlighted parts are the ones this idea speaks to. The mapping is my reading, not a tested finding. See the whole Loop and the evidence for each part.