The 'debrief my week' prompt
Most people use AI to plan the week ahead. Looking backward is where the pattern recognition lives. This prompt turns AI into a thinking partner for the conversation you should be having with yourself anyway.
Text Rosey · Schedule a call →The prompt
Copy and paste into ChatGPT, Claude, Copilot, Gemini — works anywhere.
When to use it
Use it Friday afternoon or Sunday evening, before you plan the week ahead. Paste 5–15 bullets of what you actually worked on (not what you intended to work on — what you actually did, including the meeting that went 90 minutes over and the email thread that ate a morning).
The output usually surfaces something you knew but hadn't said out loud — that you avoided the hard conversation, that the project you keep moving to next week is actually a "no," that the meeting load is creeping back up, that the highest-leverage work happened in the smallest time blocks. Useful in proportion to how honest the bullets are.
Worked examples
- End-of-week reflection — the standard use case. Paste real bullets; let the questions provoke the thinking you usually skip.
- Quarterly debrief — same prompt scaled to a quarter. Useful before performance reviews, board meetings, planning offsites.
- Project debrief — 'here's what I did on this project; what patterns do you notice?' Especially useful when a project finished but didn't quite land.
- Team-level debrief — paste your team's standup notes from the week. The pattern recognition jumps a level higher.
- Pre-1:1 prep — debrief your direct report's week before the conversation, with their input. Surfaces what's actually worth talking about vs. status reporting.
Why this prompt works
Most AI use in 2026 is forward-looking — drafting the next thing, planning the next week, generating the next idea. Backward-looking AI use is rarer and higher-leverage. The pattern recognition that's hardest to do for yourself becomes much easier when an outside (AI) perspective surfaces it back to you.
SHRM 2026 found 73% of directors and above report creativity improvements from AI vs. 65% of individual contributors. Some of that gradient is access; some of it is that senior leaders use AI for thinking work like this. Deloitte Q4 2024 also found 21% of C-suite respondents feel AI is already transforming their organization vs. 8% of non-C-suite — directors are getting more value because they're using AI to think, not just to type.
McKinsey 2025's high-performer differentiator (3x more likely to have fundamentally redesigned workflows) shows up here too. A weekly debrief is workflow redesign at the individual level — turning Friday afternoon from a time to "wrap up" into a time to actually learn from the week.
Common questions
- Will the AI actually find useful patterns, or just summarize what I wrote?
- Depends on what you paste. If your bullets are precise and a little bit honest about how time actually went, the patterns are real. If they're idealized ('worked on Q3 strategy'), the output is shallower. Honest in, honest out.
- How long should the bullet list be?
- 5–15 bullets is the sweet spot. Fewer than 5 and there's not enough material for pattern recognition; more than 15 and the AI starts losing track of what mattered.
- What about privacy — should I paste names of clients, deals, etc.?
- Use enterprise-tier AI tools (no-training data handling) and follow your AI use policy on prohibited data categories. For most knowledge work the bullets are general enough to be safe; for client-sensitive work, abstract names ('the renewal client', 'the new prospect') without losing the specificity that makes the patterns visible.
- Can I use this with a team — like for a team retro?
- Yes. Paste the team's collected standups or PR descriptions for the week. Same prompt, broader signal. Especially useful when the team's velocity dropped or jumped and you want to know why.
- Does this replace journaling?
- It augments. Journaling is private reflection; this prompt is reflection with an outside observer. Most people who do both find them serve different purposes — one for emotional processing, one for pattern recognition. Doing only one usually means you're missing one half of the work.
Sources
- AI high performers are nearly 3x as likely as others to say their organizations have fundamentally redesigned individual workflows — The state of AI in 2025: Agents, innovation, and transformation, McKinsey & Company (QuantumBlack, AI by McKinsey), 2025
- 73% of directors and above report creativity improvements from AI vs. 65% of individual contributors — The State of AI in HR 2026, SHRM (Society for Human Resource Management), 2026
- 21% of C-suite respondents feel GenAI is already transforming their organization vs. 8% of non-C-suite respondents — Now decides next: Generating a new future — State of Generative AI in the Enterprise Quarter four, Deloitte AI Institute, 2025
Related
Text Rosey to begin.
Rosey is our executive-assistant bot. Text the number below — she'll ask two questions, offer three calendar slots, and put a 30-minute call on Jim's calendar.
Text Rosey · Schedule a call →