# Project retrospective prompt: free ChatGPT template > Turn your notes on a finished project into a lessons-learned review: what worked, what slipped, why, and three changes to try on the next one. Source: https://learn.ynteractive.com/content/prompts/project-retrospective-prompt · Updated 2026-10-02 · Free from Stepthrough (https://learn.ynteractive.com) [Reports and planning](https://learn.ynteractive.com/content/prompts#job-reports-and-planning) · [Operations](https://learn.ynteractive.com/content/prompts#industry-operations) · Works in ChatGPT, Claude, Gemini and Copilot Updated October 2, 2026 · By [Robert Breen](https://learn.ynteractive.com/content/about) Use this the week a project wraps up, while the details are still fresh. Most teams skip the retrospective because writing it feels like homework. This prompt does the sorting: you dump rough notes, complaints and dates, and it organizes them into what worked, what didn't, the likely cause of each problem and a short list of changes. The causes are suggestions to discuss, not verdicts, so the team still has the conversation. ## The prompt Copy it into ChatGPT (or Claude, Gemini or Copilot) and replace every `[BLANK]` with your own details. It uses the four parts from [Prompt Writing 101](https://learn.ynteractive.com/content/prompt-writing-101-role-context-task-format): Role, Context, Task and Format. Copy prompt ```text Role: You are a neutral facilitator writing the lessons-learned summary for a finished project. Context: The project was [PROJECT]. It ran from [START] to [END]. The plan said: [PLAN]. What actually happened, in rough notes: [NOTES]. People involved: [PEOPLE]. Task: Organize the notes into: what went well, what didn't, and for each problem a "likely cause" marked as a guess for the team to confirm. Compare the plan with what happened on dates and scope. Suggest three specific changes for the next project, each with an owner role, not a name. Do not blame individuals and do not add events that are not in the notes. Format: Four short sections with headings: Went well, Didn't go well, Plan versus reality, Three changes. Bullets, under 350 words in total. End with two questions to discuss at the review meeting. ``` ## Fill in the blanks **`[PROJECT]`** The project name and a one-line description. **`[START]`** The date work began. **`[END]`** The date it actually finished. **`[PLAN]`** What the original plan promised in scope, dates and budget. **`[NOTES]`** Everything you remember, messy is fine: delays, surprises, praise, complaints. Facts only. **`[PEOPLE]`** Roles that took part, so the changes can be assigned to a role. ## Example, filled in **A made-up example** from the free lesson [Build an AI Maintenance and IT Request Form](https://learn.ynteractive.com/content/ai-maintenance-it-request-form-lovable-n8n). Ridgeline Supply Co. is the made-up warehouse business from the operations lessons. Its made-up operations team has just finished rolling out the internal maintenance and IT request form and wants to write down what they learned. Copy example ```text Role: You are a neutral facilitator writing the lessons-learned summary for a finished project. Context: The project was the internal maintenance and IT request form at Ridgeline Supply Co. It ran from September 3 to October 11. The plan said: launch by September 27, with the form, priority and team assignment, and a confirmation email. What actually happened, in rough notes: form was ready on time but nobody agreed on the priority labels until week three; two teams kept emailing requests instead of using the form; confirmation emails went to spam for the first week; warehouse staff liked the form once they tried it; testing with ten made-up requests caught two wrong assignments. People involved: operations manager, office manager, IT contact, warehouse leads. Task: Organize the notes into: what went well, what didn't, and for each problem a "likely cause" marked as a guess for the team to confirm. Compare the plan with what happened on dates and scope. Suggest three specific changes for the next project, each with an owner role, not a name. Do not blame individuals and do not add events that are not in the notes. Format: Four short sections with headings: Went well, Didn't go well, Plan versus reality, Three changes. Bullets, under 350 words in total. End with two questions to discuss at the review meeting. ``` ## What a good answer looks like - "Plan versus reality" states the gap plainly: planned September 27, finished October 11. - Each problem has a cause labeled as a guess, such as "labels weren't agreed early". - The changes name roles (operations manager, IT contact), not people. - Nothing appears that your notes didn't mention. ## How to check it before you use it An AI draft can sound right and still say something you never told it. Use the habit from the [Reply Faster lesson](https://learn.ynteractive.com/content/chatgpt-reply-to-customer-emails): check every promise, name, date and number against what you gave it before anything goes out. - Check the dates and the gap. If the summary says two weeks late, count it yourself. - Strike any cause the team knows is wrong before the meeting, so the guess doesn't harden into a fact. - Make sure positives survive. Retrospectives drift negative, and the team should hear what to keep doing. - Share it as a draft, and ask the people involved what is missing. ## Why this prompt works A retrospective is only useful if it changes something. Asking for three changes with an owner role forces the notes to end in action, in the same spirit as an [action items](https://learn.ynteractive.com/content/glossary/action-items) list. ## Practice it in the free lesson **[Prompt Writing 101: Role, Context, Task, Format](https://learn.ynteractive.com/content/prompt-writing-101-role-context-task-format)**: Ask ChatGPT for the same customer email twice, once vaguely and once with a four-part prompt, and see exactly why structure gets a better answer. You do every step yourself in a practice copy of ChatGPT, and nothing touches your real accounts. [Start the free lesson](https://learn.ynteractive.com/modules/prompt-writing-101) Also useful: [Build an AI Maintenance and IT Request Form](https://learn.ynteractive.com/content/ai-maintenance-it-request-form-lovable-n8n) ## Terms used here [Action items](https://learn.ynteractive.com/content/glossary/action-items) · [Iterative prompting](https://learn.ynteractive.com/content/glossary/iterative-prompting) · [Standard operating procedure (SOP)](https://learn.ynteractive.com/content/glossary/sop) ## Related prompts - [Project status update prompt](https://learn.ynteractive.com/content/prompts/project-status-update-prompt): Write an honest project status update: On track, At risk or Off track with the reason, what's done, what's next, blockers and… - [Process improvement ideas prompt](https://learn.ynteractive.com/content/prompts/process-improvement-ideas-prompt): Describe how a task runs today and where it breaks, and get five specific fixes tied to real steps, with effort, risks and a way… - [Incident report prompt](https://learn.ynteractive.com/content/prompts/incident-report-prompt): Draft an internal incident or near-miss report from your notes and witness statements: facts in order, who said what, gaps… ## More reports and planning prompts - [Project risk register prompt](https://learn.ynteractive.com/content/prompts/project-risk-register-prompt): List what could go wrong with a project from your notes: each risk with likelihood, impact, an early warning sign, a response… - [Purchase request justification prompt](https://learn.ynteractive.com/content/prompts/purchase-request-justification-prompt): Write a short, honest business case for a purchase: the problem, the cost, the options considered and what happens if you wait… - [Request for quote email prompt](https://learn.ynteractive.com/content/prompts/request-for-quote-email-prompt): Ask suppliers for prices with one clear email: the same specs, quantity and deadline for everyone, and a list of what each quote… [All free prompt templates](https://learn.ynteractive.com/content/prompts) · [AI glossary](https://learn.ynteractive.com/content/glossary)