Home›AI for Project Managers›Day 5
Day 5 of 5
⏱ ~60 minutes
AI for Project Managers — Day 5

Project Retrospectives and Lessons Learned

Conduct better retrospectives with AI-structured prompts and build a lessons-learned library that survives the project.

Project Retrospectives with AI

Retrospectives are the highest-leverage meeting in project management — and the most commonly done poorly. Teams revisit the same issues project after project because the insights never get captured in a useful form. AI helps you run better retros and actually learn from them.

Structured Retrospective Prompts

Prompt — Pre-retro analysis
Analyze this project data and prepare a retrospective analysis.

Project timeline: [planned vs actual]
Budget: [planned vs actual]
Scope changes: [list of changes and their triggers]
Risks that materialized: [from your risk register]
Key stakeholder feedback: [paste any feedback received]
Team feedback survey results: [if available]

Identify:
1. Top 3 things that worked well (with specific examples)
2. Top 3 things that didn't work (with root causes, not symptoms)
3. What the team is likely to avoid discussing but should
4. Recommended "Start / Stop / Continue" items
5. One systemic issue this project shares with typical projects of this type

Facilitation Prompts

Prompt — Retrospective facilitation guide
Generate a 60-minute retrospective facilitation guide.

Team size: [N people]
Project type: [agile sprint / waterfall phase / full project]
Known tension points: [any issues you're aware of]

Create a facilitation plan with:
- Opening icebreaker question (not "how did the project go")
- Timed segments for each retro format (What went well / Delta / Actions)
- 3 probing questions for each topic area
- How to handle when the team blames a specific person
- How to close with energy and commitment

Lessons Learned Library

Prompt — Extract lessons learned
Based on this retrospective summary, extract structured lessons learned for our organizational knowledge base.

Retrospective notes:
[paste retro output]

For each lesson learned:
- ID: LL-[year]-[number]
- Category: [Planning/Execution/Stakeholder/Technical/Process]
- Lesson: [one clear sentence]
- Context: [when this applies]
- Action: [what to do differently next time]
- Applicability: [what types of projects this applies to]

Output as a table. Sort by category.
💡
Store lessons learned in Claude Projects. At the start of the next project, paste your lessons library into the system prompt so Claude can flag when you're repeating past mistakes.
Day 5 Exercise — Capstone
Run a Complete AI-Powered Retrospective
  1. Run the pre-retro analysis prompt on a completed project
  2. Generate the facilitation guide for a 60-minute session
  3. Conduct the retrospective (live or solo practice)
  4. Extract structured lessons learned from the output
  5. Add the lessons to a Claude Project for future reference

Day 5 Summary — AI for Project Managers Course Complete

  • Pre-retro analysis surfaces what teams avoid discussing — that's where the real lessons are
  • Use AI to extract structured lessons learned from raw retro notes
  • Store lessons in Claude Projects so they inform future planning automatically
  • Retrospective facilitation prompts help you run the meeting, not just document it
  • A lessons-learned library that gets used is worth more than ten retros that don't

Want to go deeper in 3 days?

Our in-person AI bootcamp covers advanced AI development, agentic systems, and production deployment. Five cities. $1,490.

Reserve Your Seat →
←
Previous
Day 4: Meeting Management — Agendas, Notes, and Action Items
Course Complete
Back to Course Overview
→
Finished this lesson?