30 ChatGPT Prompts for Product Managers to Ship Faster

Product management is the art of making good decisions with incomplete information under constant pressure. The right ChatGPT prompts for product managers help you write sharper PRDs, prioritise more defensibly, communicate clearly with stakeholders and think through product decisions with greater rigour — without adding hours to your day.

Below are 30 prompts across the product lifecycle: discovery and research, strategy and prioritisation, specs and execution, stakeholders and launch, and metrics and learning. They work in ChatGPT or any other AI chatbot.

Product prompts are only as good as the context behind them. Include:

• Product and users: what the product does, who uses it and their main jobs to be done
• Goals and metrics: the company objective and the numbers that define success
• Constraints: team capacity, deadlines, tech debt and dependencies
• Real evidence: paste anonymised research notes, data and feedback rather than describing them

Model choice matters too. GPT-6 Astra handles detailed specs and trade-off analysis well, and Claude Sonnet 5 tends to write clear, nuanced stakeholder communication. Avoid pasting confidential data you are not permitted to share.

Great products start with understanding the problem. The AI summarizer condenses long interview notes, and Gemini 3 Pro can synthesise many transcripts in one go.

1. User research interview guide

Write a 30-minute interview guide to understand how [user type] currently handles [problem]. Focus on past behaviour rather than opinions about future features, include warm-up questions, probes for each topic and a closing question, and avoid leading wording.

2. Opportunity assessment builder

Build an opportunity assessment for [idea]. Cover the problem, who has it, how they solve it today, evidence it matters, the size of the opportunity, how it fits our strategy, risks and what we would need to believe for it to succeed.

3. Interview synthesis

Here are anonymised notes from [number] user interviews: [paste]. Identify recurring pain points, needs and workarounds, rank them by frequency and intensity, and pull out one supporting quote for each.

4. Jobs to be done

Based on this research — [paste] — write the key jobs to be done for [user type] in the format ‘When… I want to… so I can…’, and note the functional, emotional and social dimensions of each.

5. Competitor feature analysis

Compare how [competitors] handle [feature or workflow] based on these notes: [paste]. Identify strengths, gaps, patterns across the market and where we could differentiate.

6. Problem statement

Turn these observations into a crisp problem statement: [paste]. Include who is affected, what they are trying to do, what gets in the way, the impact and how we will know the problem is solved — without prescribing a solution.

Prioritisation is where product managers earn trust. GPT-5.6 Sol is good at structured frameworks and clear reasoning you can defend.

7. Roadmap prioritisation framework

Prioritise these initiatives: [list with any data on reach, impact, confidence and effort]. Score them with [RICE / ICE / value vs effort], show the ranking and reasoning, and flag where the scores depend on assumptions we should test.

8. Feature trade-off analyser

We can only build one of these this quarter: [option A] or [option B]. Compare them on user value, business impact, effort, risk, strategic fit and what we give up, and recommend one with the key assumption behind the choice.

9. Product strategy document

Draft a one-page product strategy for [product] covering the vision, target users, the problems we will focus on, how we will win, what we will not do and how we will measure progress.

10. Product OKRs

Write product OKRs for [team] for [quarter] linked to the company goal [goal]. Draft two objectives with three outcome-based key results each, and flag any key result that is really an output.

11. Now, next, later roadmap

Organise these initiatives into a now, next, later roadmap: [list]. Group them by the problem they solve rather than by feature, and write one sentence on why each sits where it does.

12. Saying no to a feature request

A stakeholder has asked for [feature]. It does not fit our current priorities because [reason]. Write a respectful reply that acknowledges the need, explains the decision and suggests what would change it.

Clear specs save engineering time. Claude Haiku 4.5 is quick for batches of user stories and acceptance criteria.

13. PRD writer

Write a PRD for [feature]. Our product is [describe], our users are [describe] and the problem is [describe]. Include the problem statement, goals and metrics, user stories, functional and non-functional requirements, out of scope, open questions and dependencies — specific enough that engineers can start scoping.

14. User story generator

Break [feature] into user stories in the format ‘As a… I want… so that…’, sized small enough for a single sprint, with acceptance criteria in Given/When/Then format.

15. Edge case finder

Review this feature spec — [paste] — and list the edge cases, error states, empty states, permission issues and accessibility considerations we have not covered.

16. Questions for engineering

Before our refinement session on [feature], list the technical questions I should ask engineering about feasibility, effort, risks, dependencies and simpler alternatives.

17. Release plan

Create a release plan for [feature]: phased rollout, feature flags, beta group, success criteria for each phase, rollback plan and who needs to be informed at each step.

18. Bug triage

Help me triage these bugs: [paste]. Rate each by severity and user impact, group related issues and suggest which to fix now, schedule or close.

Alignment is half the job. The AI email writer helps with stakeholder emails, and the social media post generator adapts launch messages for each channel.

19. Stakeholder update writer

Write a stakeholder update for [audience] from these notes: [paste]. Cover what shipped, progress on key metrics, what is next, risks and decisions needed, in under 250 words.

20. Decision memo

Write a one-page decision memo for leadership recommending [decision]. Include the context, options considered, recommendation, evidence, risks and what we need approved.

21. Launch announcement writer

Write launch announcements for [feature] for three audiences: an internal Slack post, an email to customers and a short social post. Lead with the user benefit, not the technical details.

22. Sales and support enablement

Create an enablement brief for sales and support on [feature]: what it does, who it is for, key talking points, common questions with answers, limitations and where to get help.

23. Release notes

Turn these technical changes into user-friendly release notes: [paste]. Group them into new, improved and fixed, and explain each change in terms of what users can now do.

24. Customer feedback reply

A customer sent this feedback: [paste]. Write a reply that thanks them, shows we understand the problem, is honest about whether and when we might address it, and asks one follow-up question.

Metrics tell you whether you built the right thing. DeepSeek V4 Pro is useful for working through funnels and experiment results step by step.

25. Metrics framework designer

Design a metrics framework for [product or feature]: a north star metric, input metrics that drive it, guardrail metrics to watch, how each is defined and how often to review them.

26. Experiment plan

Plan an experiment to test [hypothesis]. Define the variants, primary metric, guardrails, audience, how long it should run and what result would lead us to ship, iterate or stop.

27. Funnel analysis

Here are our funnel numbers for [flow]: [paste]. Identify the biggest drop-offs, possible reasons for each and three experiments to improve conversion.

28. Dashboard review

Here are this month’s product metrics compared with last month: [paste]. Summarise what changed, what might explain it, what needs investigating and what to share with leadership.

29. Product retrospective facilitator

Plan a retrospective for the [feature] launch: what went well, what did not, what we learned about users and our process, and how to turn the findings into owned actions.

30. Post-launch review

Thirty days after launching [feature], here are the results: [paste metrics and feedback]. Compare them with the goals in our PRD, decide whether it succeeded and recommend whether to invest more, iterate or sunset it.

Chat Smith gives you ChatGPT alongside Claude, Gemini, Grok and DeepSeek in one app, so you can draft a PRD with one model and stress-test it with another. Save your most useful prompts as templates and bring the same rigour to every feature.

Use Gemini 3.5 Flash for quick drafts between meetings. AI speeds up the writing — your judgement about users, trade-offs and timing is what ships great products.