Post-Mortem Meeting Agenda
A post-mortem is a 45-minute meeting after a project ends or an incident is resolved, where the people involved rebuild what happened and look for causes, not culprits. This project post-mortem meeting agenda ends with a short list of fixes, each with an owner and a due date.
Agenda
45 min · 5 parts
- 0:00Ground rules5 minBlameless: we look at systems and decisions, not at people
- 5:00Timeline10 minWhat happened, in order, with dates and decisions
- 15:00What went well5 minWhat helped, and what we should repeat
- 20:00What went wrong15 minThe biggest problems and the causes behind them
- 35:00Action items10 minA few fixes, each with an owner and a due date
Free to start. Guests join from a link, no account needed.
- Length
- 45 min
- How often
- Within one or two weeks of a project ending, or a few days after an incident is resolved, while memories are fresh.
- Who attends
- Everyone who worked on the project or the incident, led by a facilitator who wasn't directly involved.
- You leave with
- A shared timeline, the causes behind what went wrong and a short list of fixes, each with an owner and a due date.
The agenda, part by part
How to run a project post-mortem meeting
1. Ground rules
5 minOpen by saying why you're here: to learn, not to judge. State the ground rule out loud: people made the best decisions they could with what they knew at the time, so the discussion is about systems and decisions, not about who is at fault. Name a note-taker.
Ask: What do we most want to understand by the end of this meeting?
2. Timeline
10 minThe facilitator drafts the timeline beforehand from tickets, chat history and calendars, and the group corrects it and fills the gaps. Stick to facts: what happened, when, and what people knew at each decision point. Save the analysis for later.
Ask: Where does this timeline differ from what you remember?
3. What went well
5 minSpend a few minutes on what worked before turning to problems. In an incident it might be how quickly the alert fired or how well people coordinated; in a project, a practice the next team should copy.
Ask: What would you insist on doing again next time?
4. What went wrong
15 minPick the two or three problems with the biggest impact and look at each in turn. Keep asking why until you reach something the team can change, such as a missing check, an unclear handoff or a deadline set without the people doing the work. Most problems have more than one cause, so don't stop at the first.
Ask: What made this easy to get wrong, and why didn't we catch it sooner?
5. Action items
10 minTurn the main causes into a few concrete fixes, each with one owner and a due date. Put them where the team tracks its other work, agree when you'll check them, and share a short written summary with everyone affected.
Ask: Which fix would have prevented the biggest problem?
Lessons learned questions to ask
- What did we set out to do, and what actually happened?
- Which decision would we make differently with what we know now?
- Where did we find out about a problem too late?
- What did we assume that turned out to be wrong?
- Which handoff or dependency caused the most delay?
- What would have helped us detect the issue sooner?
- What should the next team copy from this project?
- What should we stop doing in every project from now on?
Example: a post-mortem after a delayed launch
A website relaunch went live three weeks late, and checkout failed for two hours on launch day.
Notes
- Timeline: content freeze slipped twice; load testing was cut to keep the launch date.
- Went well: the rollback worked in 15 minutes; support had replies ready for customers.
- Went wrong: nobody owned the final go/no-go check.
- Cause: the launch date was set before the team had estimated the data migration.
Action items
- Kenji: make load testing a required step in the release checklist, by Friday.
- Amara: add a named go/no-go owner to the launch plan template, by the end of the month.
- Lucía: share the post-mortem summary with all product teams, this week.
Ways to adapt it
Incident post-mortem
After an outage or a security issue, build the timeline from logs and alerts, note how long it took to detect and to resolve, and meet within a few days.
Lessons learned workshop
For a long project, send a short survey first and allow 90 minutes, so you can review each phase in turn instead of only the final weeks.
Written draft first
Have the facilitator write the timeline and open questions into a project post-mortem template, and use the meeting to correct it and agree on actions. 30 minutes is often enough.
Common mistakes
Looking for someone to blame.
Instead: Ask what made the mistake possible. When people fear blame, they hide the details you need to fix the system.
Holding one only when things go badly.
Instead: Run post-mortems on successful projects too. Knowing why something worked is just as useful as knowing why it failed.
Arguing about causes before agreeing on the facts.
Instead: Settle the timeline first. Many arguments about causes are really disagreements about what happened.
Writing a lessons learned document that nobody reads.
Instead: Turn each lesson into an action with an owner, track it with other work and review it at the next project kickoff.
Post-mortem vs retrospective
A retrospective is a regular meeting, usually at the end of every sprint, about how the team works day to day. A post-mortem happens once, after a project ends or an incident is resolved, and looks at one outcome in depth. Both are blameless and end with actions, but a post-mortem needs more preparation, usually a written timeline, and often brings in people from other teams.
Run it in Edworking
Your agenda keeps the meeting on time
Start a video meeting from any template. Edworking turns the agenda into the structure of the call, so nobody has to watch the clock.
Live agenda
A timer for every part
Everyone sees the agenda and how long is left in each part, with a clear moment when the meeting moves on.
AI notes
Notes that follow the agenda
AI takes the notes part by part and writes the summary and action items when the call ends.
Tabs in the call
Your work on screen, no sharing
Open docs, tasks and files as tabs in the call. Everyone sees the same thing without anyone sharing a screen.
Guests
Anyone can join from a link
Free to start. Clients, candidates and guests join from a link, with no account needed.
Frequently asked questions
45 to 60 minutes works for most projects and incidents. For a large project with several phases, plan 90 minutes or split it into two sessions.
Ground rules, a timeline of what happened, what went well, what went wrong and why, and action items with owners. Give the most time to the causes, because that's where the lessons come from.
The meeting looks at systems, processes and decisions rather than individual fault. People explain what they knew and why their actions made sense at the time, which brings out the real causes.
Mostly, yes. "Lessons learned" is the usual term in project management, while "post-mortem" is common in software and for incidents. The agenda is the same: what happened, what worked, what didn't and what will change.
Someone who wasn't directly involved, such as a project manager from another team. They keep the discussion neutral, make sure quieter people are heard and stop it from turning into a debate about who was right.
More meeting agenda templates
Sprint Retrospective Template
Look back on the sprint together and agree on the one or two things to change.
5 parts
Project Kickoff Meeting Agenda
Start a project with a shared goal, a clear scope, named owners and a first deadline.
7 parts
Decision-Making Meeting Agenda
Lay out the options, hear every concern and leave with a decision written down.
4 parts