You’ve probably sat in this meeting more than once. A big deal went to a competitor, or a launch slipped badly, and someone put a review on the calendar. The right people came. The slides were thoughtful. Everyone was polite, and the meeting ended on time with a list of lessons learned.
A few months later, it happened again, in nearly the same way.
So what went wrong in the review? Did people not care? Were they hiding things? Was the meeting badly run? No. Most of the time, it went exactly the way meetings like it tend to go. The conclusion had been settled before anyone walked in. Every cause named was somewhere outside the room: the market, the vendor, the timing, another department. And no one said the four words that make a review useful: “I got that wrong.”
Holding the meeting is the easy part
Scheduling a debrief takes thirty seconds. What makes a debrief valuable takes much longer to build, and it can’t be put on a calendar. The people in the room have to be comfortable enough with each other to be honest. They have to be open about what they actually saw and did, including the parts that don’t flatter them. And they have to want to learn more than they want to be right.
Most organizations hold the meeting, but far fewer have the trust that good outcomes depend upon, and you can’t tell the two apart from the calendar. Both show up as a meeting entitled “Lessons learned.”
That’s why a review can be run carefully, with good intentions and a skilled facilitator, and still teach no one anything. You can schedule the meeting and hire someone good to run it, but the candor that makes it useful can’t be scheduled or hired.
Why senior leadership teams find this hardest
You might expect a senior leadership team to get the most out of this meeting. The people in it are experienced, confident, and used to hard conversations. Often, though, it’s the group that gets the least out of it.
The reason is structural, not personal. The cost of saying “I got that wrong” rises with what you have to lose. A new team member who admits a mistake loses an afternoon. A senior leader who admits one may feel they’re putting a reputation, a budget, or a board’s confidence on the table. Everyone below is also watching how the leadership team treats a mistake, and they’ll take their cue from it. A senior leader who owns an error is also going against the habits that helped them get there: projecting confidence, having the answer, and never looking unsure in front of the people who report to them.
That doesn’t mean senior leaders are dishonest. They’re being careful, in a setting where care and candor pull in opposite directions.
Here’s the part that tends to surprise people. The teams who report to you may already have a working version of this. Many of them hold short, regular reviews of their own work, and some are very good at it. The group with the most power is often the one with the least practice, not so much because the people in it are less capable as because they’ve had more reasons to avoid it.
An organization that set rank aside
The US Army ran into this problem in a setting where the cost of not learning is about as high as it gets. It developed the after-action review in the mid-1970s and made it standard practice at its National Training Center in the 1980s. The Army’s own guidance is plain about what the review is not: “An AAR is not a critique.” And it’s just as plain about why rank can’t be allowed to run it: “No one, regardless of rank, position, or strength of personality, has all of the information or answers.”
So for the length of the review, rank is set down on purpose. The hierarchy is still there before and after. For that hour, the organization has decided that learning what actually happened matters more than protecting whoever was in charge when it happened.
This came decades before agile software practices and owes them nothing. It came from people who needed the truth about what went wrong and understood that the hierarchy was the thing most likely to keep it out of the room.
What changes the conversation
In my experience, two things work, and neither can be handed to someone else.
Someone with power goes first
That means a specific admission from the most senior person present, about a specific call that didn’t hold up, said out loud before anyone else speaks, rather than a general invitation to be candid that everyone has heard before. It takes vulnerability, and it only works if the people listening prove trustworthy with it. Then the room watches what happens next. If that person is visibly wrong and comes through it fine, the cost of honesty drops for everyone else. If they aren’t willing, no amount of process will make up for it.
I was surprised the first time a senior leader began a postmortem by explaining that he was wrong about a recent decision. He went on to explain how his reasoning was faulty and even described what he thought was the fallout, both short term and longer term. He was new to the organization, a place where truth telling and taking responsibility were not typical in such meetings. Although it took several months, his leadership led to a fundamental change in the way individuals in his leadership team actively participated in problem solving rather than making excuses.
Change the format, and keep changing it
A review run the same way every time lets people decide in advance what they’ll say, and among senior leaders that preparation is often political. When the format changes from one review to the next, people arrive without a script. They also look at the same events from an angle they haven’t tried before, and that’s where real thinking tends to start. A new format takes away the rehearsal, but it doesn’t make honesty any less risky. It only helps where the people in the review bring enough trust, transparency, and courage to use the opening it creates.
Running these meetings well is real work, and it can be learned: choosing a format that fits the moment and deciding who speaks first. None of this takes hold in a single meeting, though. In the story above it took several months, and I wouldn’t expect it to take less.
When going first isn’t enough
Perhaps culture is the single most difficult hurdle to overcome in institutions. A school district, a state agency, or a university carries habits that were in place long before any current leader arrived, and that will likely still be there after they leave. Procurement rules, tenure, union agreements, and appointment cycles all reward caution, and none of them are yours to rewrite.
What you can usually shape is the meetings you run yourself. Going first there won’t change the institution on its own, but it can change what the people closest to you believe is safe to say, and in a place where culture moves slowly, that may be where change has to start.
The harder question
It’s natural to ask whether your organization holds enough reviews. Most do. The harder question is when someone in your most senior review last said “I got that wrong,” and what happened to them afterward.
There’s one more participant to think about. More and more of your organization’s work now passes through AI agents, and no one in any of these reviews has asked what those AI agents could share about where things went wrong. That’s a subject for another day.
The Army’s guidance quoted here is from TC 25-20, A Leader’s Guide to After-Action Reviews (1993). Amy Edmondson introduced the idea of team psychological safety, the shared sense that it’s safe to take interpersonal risks with the people you work with, in “Psychological Safety and Learning Behavior in Work Teams,” Administrative Science Quarterly (1999).
