AAR · After Action Review
After Action Review
What is an AAR?
- Four questions: expected, actual, gap, next action
- Thirty minutes right after the task beats two hours a month later
- Learning stops the moment it becomes blame
Making AAR a standard practice changes what is left behind when a task ends. Before, a single results report ending in 'here is what we did and the outcome' went up the chain and that was that. A team that runs AARs leaves one more page: a checklist for kicking off the next task. The report goes up, and the checklist goes sideways. Keeping the team next door from repeating the same mistake is what this process actually produces.
In practice, AARs come in two kinds: a quick AAR done standing at the work site within 15 minutes, and a formal AAR with a separate facilitator and note-taker that runs 60–90 minutes. Three conditions are common to both: within 48 hours of completion, no more than 8–10 attendees, and facts before interpretation. With more than 12 people, half never speak up, and after a week each person's memory is already overwritten by their own interpretation. Don't confuse it with RCA (root cause analysis): RCA is an investigation that identifies causes and documents them, while an AAR is a learning session that changes the judgment criteria of the people who were there.
Keep the write-up to one A4 page. Use just four boxes (expectation, facts, gap, next action) and designate that page as required reading at the kickoff of similar projects; that is all the spreading mechanism you need. Cap action items at four, each with an owner and a deadline. Two items that actually get done change an organization more than 20 that nobody reads.
The most common mistake in practice is the belief that you only hold one when something fails. If you skip the projects that went well, you never learn why you won, repeat the same approach next time, and lose. The second is when a senior executive speaks first. If a division head opens with “I think the technical review was late,” the next 30 minutes circle around testimony that supports that one sentence, and the real gap in the quote logic never reaches the table. Once an AAR turns into blame even one time, participants start arriving with safe answers prepared. It is safer to have the most junior people speak first.
Jihye ChoiExpert in design thinking, new-hire onboarding, and train-the-trainer programs
John KimAuthor of B2B Marketing Design and CEO of Nanum ConsultingIt began as a review procedure the U.S. Army introduced in the 1970s to make training more effective. It presupposes a culture of stating facts regardless of rank, and it later spread to corporate project management and knowledge management.
Consider a hypothetical case: Company A, an industrial automation equipment maker with 340 employees and KRW 90 billion (≈ US$67,335,000) in annual revenue. After eight months of work on a KRW 1.2 billion (≈ US$897,800) public tender, it lost, finishing second, and its win rate on public-sector bids over the past three years had stalled at 18%. Four days after the loss notice, nine people from sales, engineering, and planning gathered in a meeting room.
- The head of sales set a 40-minute timer on the whiteboard and made one thing clear first: 'Nothing said today goes into performance reviews.' Then he went around the room, starting with the most junior employee.
- A planning team manager put up the original proposal strategy from kickoff and read aloud the line: 'Secure 35 of 40 technical points to overturn the 8-point price disadvantage.'
- The client's actual score sheet showed a 32-point gap on technology and 14 on price. The team learned only six days before the proposal deadline that the competitor had thrown in three years of free maintenance.
- When the proposal PM said, 'I thought sales was watching competitor moves,' the team pinpointed a role gap: only one site inspection had been done after the pre-specification notice.
- To keep, they set 'technical review completed two weeks ahead' (2 cases). To change, they set 'draft three competitor price scenarios 30 days before the bid' and 'require two site inspections', each with an owner and a deadline.
- After making this one-pager required material for the next bid kickoff, the company won 2 of 5 public tenders over the following six months. Its win rate rose from 18% to 40%, and proposal rework time fell by an average of 22 hours per bid.