You're three meetings deep before 11 AM. There's a PRD draft that's been 60% done for two weeks. Your roadmap deck needs updating before the exec review on Thursday. And the engineering team is waiting on six user stories to start the next sprint.
Product managers don't lack ideas. They lack time to write them down in the format every stakeholder actually needs.
These 35 prompts handle the five workflows that eat most PMs' writing time. Each one is built for copy-paste use — add your specific context, run it, edit the output. Most return a working draft in under 90 seconds.
How to Use These Prompts
Fill in every bracket before you run the prompt. A generic prompt gives a generic output. The model needs your actual context — feature name, user segment, metric target — before it can write something you'd actually send.
Use GPT-4 or Claude 3.5 Sonnet. Both handle structured PM documents well. For anything going to executives, always read and tighten the output before sending.
1. PRD Writing
A PRD that sits in draft for two weeks isn't a PRD problem. It's a starting problem. These prompts create the first draft fast so you can spend time revising instead of staring at an empty doc.
Prompt 1 — Full PRD Draft
Write a Product Requirements Document for [feature name].
Context: [describe the product and what the feature will do in 2-3 sentences].
Problem: [what user problem this solves — be specific].
Target users: [user segment].
Goals: [2-3 measurable outcomes — include specific metrics].
Out of scope: [explicitly exclude anything that might cause scope creep].
Key user flows: [describe 2-3 main flows at a high level].
Dependencies: [list any teams or systems this depends on].
Format as a clean PRD with sections: Overview, Goals, Requirements, Out of Scope,
Success Metrics, Open Questions.
Prompt 2 — Problem Statement Section
Write the problem statement section for a PRD. The feature is [feature name].
User research finding: [paste key insight or data point — e.g., "42% of users who
reached the checkout screen abandoned before completing payment"].
Current solution and its failure: [describe what users do now and why it's inadequate].
Business impact of the problem: [revenue, retention, or adoption metric affected].
Write 3-4 sentences that a skeptical engineer or designer would agree are accurate
and worth solving. No hype, no exaggeration.
Prompt 3 — Success Metrics Section
Define the success metrics for this feature: [feature name].
Primary goal: [e.g., "increase trial-to-paid conversion"].
Current baseline: [current metric value].
Target: [target value and timeline — e.g., "20% lift within 90 days of launch"].
Also suggest 2 guardrail metrics — metrics that should not decline — and 1 learning
metric we'd track to understand user behavior regardless of success or failure.
Format as a metrics table with: Metric Name | Current | Target | Owner.
Prompt 4 — User Research Synthesis for PRD
Synthesize these user research notes into 3 insight bullets I can use in a PRD
problem statement:
[paste raw interview notes or survey results].
Each insight should: (1) name a specific user pain point, (2) include a quote or
data point that supports it, (3) explain the downstream PM implication.
Avoid generic observations like "users want it to be easier."
Prompt 5 — Edge Cases Section
Generate edge cases and error states for this feature: [feature name].
Core flow: [describe the main happy path in 2-3 steps].
For each edge case include: scenario description, expected system behavior,
and any open design/eng questions it raises.
Think like a QA engineer and a power user who will immediately try to break the
feature. Give me at least 8 edge cases across: empty states, permission edge cases,
concurrent users, API failure scenarios, and mobile/offline behavior.
Prompt 6 — PRD Executive Summary
Write a 4-sentence executive summary for this PRD. The feature is [feature name].
Include: (1) the user problem in one sentence, (2) what we're building in one sentence,
(3) the business case in one sentence, (4) the launch timeline and key dependency.
This will be read by the CPO and CEO. Assume they will read only this section.
Prompt 7 — PRD Open Questions Log
Generate the Open Questions section for this PRD. Feature: [feature name].
Here are the unresolved decisions we know about: [list what you know].
Add 5 additional open questions a thorough PM should ask before this PRD is
approved — covering engineering feasibility, legal/privacy, pricing impact,
analytics instrumentation, and post-launch support requirements.
Format each with: Question | Owner | Must-resolve-by date.
2. Roadmap Communication
A roadmap deck isn't a list of features. It's a narrative about where the product is going and why. These prompts write the narrative layer so the features speak for themselves.
Prompt 8 — Quarterly Roadmap Narrative
Write a roadmap narrative for Q[X] [year].
Strategic theme: [1-2 sentence description of the quarter's focus — e.g.,
"Q3 is our retention quarter: every initiative reduces 30-day churn"].
Initiatives: [list 3-5 roadmap items with brief descriptions].
What we're NOT doing this quarter and why: [list 2-3 things explicitly deprioritized].
Audience: [engineering leads / entire company / board].
Write as a 3-paragraph narrative — context, this quarter's bets, and what success
looks like by quarter-end. Cut any language that reads like a feature catalog.
Prompt 9 — Now / Next / Later Roadmap Descriptions
Write Now/Next/Later descriptions for each item in this roadmap:
NOW (in flight, 0–4 weeks): [paste items].
NEXT (planned, 4–12 weeks): [paste items].
LATER (future, 12+ weeks): [paste items].
For each item write: 1 sentence on what it is, 1 sentence on why it's in this
time horizon. Avoid "TBD" for LATER items — state the condition that would move
them into NEXT instead.
Prompt 10 — Roadmap Trade-off Explanation
Write a roadmap trade-off explanation for a stakeholder who asked why [Feature A]
is not on the roadmap when [Feature B] is.
Context: [explain what Feature A and Feature B do].
Reason for the prioritization decision: [explain the actual reason — data, strategy,
or resource constraint].
Write a 3-sentence response that is honest about the trade-off, doesn't overpromise
a future date, and maintains trust with the stakeholder. No corporate hedging.
Prompt 11 — Roadmap Slide Talking Points
Write talking points for a 10-minute roadmap presentation to [audience: leadership /
engineering / sales / customers].
Roadmap items: [list 4-6 items with one-line descriptions].
For each item, write: the business reason we're building it (not just what it does),
and the specific outcome we expect.
Also write: 1 opening statement (not "Hi everyone, let me walk you through...")
and 1 closing ask from the audience.
Prompt 12 — Product Strategy Document
Write a 1-page product strategy document for [product name or team].
Current state: [where the product is today — users, revenue, key metrics].
6-month vision: [where it should be — be specific with numbers].
Strategic pillars: [2-3 areas of focus — e.g., "activation", "expansion revenue",
"platform reliability"].
For each pillar write: what it means, what success looks like in 6 months,
and the top bet we're making.
Format as a document I can share with my engineering and design leads.
Prompt 13 — Deprioritization Memo
Write a deprioritization memo for [feature or initiative name].
Why it was considered: [original business case].
Why we're deprioritizing it now: [honest reason — data, strategy shift, or capacity].
Impact of deprioritizing: [what we lose by not building it, honestly].
What would change this decision: [what data or change in context would bring it back].
This will be shared with the team and the requesting stakeholder. Be direct.
Avoid "we'll revisit in a future quarter" as a catch-all.
Prompt 14 — Competitive Feature Response
[Competitor name] just launched [feature description]. Write a 5-sentence internal
memo addressing:
1. What they launched and what it actually does (strip the marketing).
2. How our current product compares.
3. Whether this changes our roadmap priority (yes/no and why).
4. What we tell sales/CS who will get questions about this.
5. Whether a response feature is worth scoping in the next 90 days.
Keep it factual. Avoid panic and avoid dismissiveness.
3. Stakeholder Updates
The update that lands well isn't the most detailed one. It's the one that answers the stakeholder's actual question before they ask it.
Prompt 15 — Weekly Product Update
Write a weekly product update email for [date range].
Audience: [cross-functional team / leadership / board].
Shipped this week: [list 2-3 items with brief impact statements].
In progress: [list 2-3 items with expected completion].
Blocked or at risk: [list any blockers — be specific about what's needed].
Key metrics this week: [paste relevant data].
One decision needed from the team or leadership: [describe the decision].
Keep it under 250 words. One section, no filler intro.
Prompt 16 — Executive Briefing for a Feature Launch
Write a 5-bullet executive briefing for the upcoming launch of [feature name].
Include: what it does (for a non-PM audience), why we built it (business case),
who it affects, expected impact on [key metric], and launch date + rollout plan.
This will be read by the CEO before the all-hands. No jargon. No passive voice.
Each bullet should stand on its own.
Prompt 17 — Delay Notification
Write a stakeholder notification that [feature/project name] is delayed.
Original launch date: [date].
New expected date: [date].
Root cause: [be specific — not "unforeseen complexity"].
What changed to cause the delay: [what was learned or what blocked progress].
What we're doing to prevent further delay: [3 specific actions].
What stakeholders need to do (if anything): [describe or say "no action needed"].
Do not over-apologize. Be direct and solution-focused.
Prompt 18 — Beta Program Update
Write a beta program update for [feature name] after [X weeks] of beta.
Beta cohort: [size and description — e.g., "42 enterprise customers"].
Key usage data: [paste metrics or describe findings].
Top 3 user feedback themes: [describe what beta users said].
What changes we're making before general availability: [list].
GA launch timeline: [date or condition].
Format for an internal stakeholder audience who approved the beta.
Prompt 19 — Data Review Presentation Narrative
Write the narrative for a product data review covering [metric or product area].
Key metrics to present: [list metrics with current values and trends].
What the data tells us: [your interpretation].
What it doesn't tell us (and why that matters): [identify gaps].
Proposed next action based on the data: [1-2 concrete next steps].
Write this as a spoken narrative I can read from when presenting to [leadership /
cross-functional leads]. One clear insight per paragraph.
Prompt 20 — Sprint Review Summary
Write a sprint review summary for sprint [number/name].
What was planned: [list of stories or features planned].
What was shipped: [what actually got done].
What was not completed and why: [honest explanation].
Demo highlights: [what will be demoed].
Next sprint preview: [what's up next].
Audience: [stakeholders / leadership / full company].
Keep it factual. Highlight any patterns — if we've missed estimates two sprints
in a row, name it.
Prompt 21 — Stakeholder Ask Email
Write an email to [stakeholder name/role] asking for [specific thing needed —
e.g., engineering headcount, budget approval, legal review].
Context: [what you're building and why you need their help].
Specific ask: [exactly what you need, by when].
Impact of not getting it: [what slips if this isn't resolved — be honest].
Make the ask in the first two sentences. Don't bury it.
4. User Story Writing
A user story that needs three rounds of clarification before engineering can estimate it isn't a user story — it's a concept. These prompts write stories that stand on their own.
Prompt 22 — User Story with Acceptance Criteria
Write a user story for [feature or behavior name].
User type: [persona or user role].
Goal: [what they want to accomplish].
Context: [why they need this — what are they trying to do in the product].
Write the story in format: "As a [user type], I want to [action], so that [outcome]."
Then write 4-6 acceptance criteria in Given/When/Then format.
Include at least one AC covering an error or edge case state.
Prompt 23 — Story Decomposition
This epic is too large to ship in one sprint: [describe epic in 2-3 sentences].
Break it into user stories small enough for a 2-week sprint, where each story
delivers standalone value to the user.
For each story include: story statement, sizing estimate (S/M/L),
and which story it depends on (if any).
Flag any story that is a "skeleton" story — one that enables future stories but
doesn't deliver user-visible value on its own.
Prompt 24 — Definition of Done
Write a Definition of Done for this user story: [paste story].
Include technical and product criteria. Consider: unit tests written,
analytics events instrumented, error states handled, mobile-responsive
(if applicable), accessibility requirements met, product manager sign-off,
design QA completed.
Add 2 criteria specific to our context: [describe any team-specific
quality requirements].
Prompt 25 — Story for Non-Functional Requirement
Write a user story for this non-functional requirement: [describe it —
e.g., "the dashboard must load within 2 seconds for 95% of users"].
Frame it from the user's perspective, not the engineering team's.
Include: business reason this matters, specific performance threshold,
and how we'll measure whether the story is done.
Prompt 26 — Story Map Narrative
Create a story map narrative for [feature or flow name].
User goal: [what the user is trying to achieve end-to-end].
Activities (top of map): [list the high-level activities in sequence].
For each activity, write 3-4 user tasks the user performs.
Separate the first release (minimum path) from nice-to-have enhancements.
Format as a table: Activity | MVP Stories | Enhancement Stories.
Prompt 27 — Technical Story
Write a technical story (not user-facing) for this engineering task:
[describe the technical work — e.g., "migrate authentication service from
legacy OAuth1 to OAuth2"].
Include: why this is needed (impact on product or team),
specific technical outcomes that define done,
risks if not addressed,
and which product capabilities this unblocks.
Frame it so a non-technical stakeholder understands why this is on the roadmap.
Prompt 28 — Story Refinement Questions
I'm preparing to run a story refinement session for [feature name].
Here are the draft stories: [paste stories].
Generate the 8 most important questions the engineering team is likely to ask
during refinement, and my best answers to each based on this PRD context:
[paste relevant PRD sections].
If any question reveals an actual gap in the requirements, flag it explicitly
with "GAP: [description]".
5. OKR and Metrics Reporting
OKRs that sit in a spreadsheet and get reviewed once a quarter aren't OKRs. They're wishes. These prompts make the metrics layer visible and communicable.
Prompt 29 — OKR Draft for Product Team
Draft a set of OKRs for the [product team / product area] for [quarter and year].
Strategic priority: [company or product strategy this supports].
Propose 1 Objective and 3 Key Results.
Each Key Result must be: measurable with a specific number, owned by a named
role or team, and achievable but not trivial.
Also flag: which KR is most at risk and why, and which KR will take the most
cross-team coordination.
Prompt 30 — OKR Check-in Update
Write a Q[X] week [X] OKR check-in update for this objective: [paste objective
and key results with current status].
Current progress on each KR: [paste numbers].
What's driving progress or blocking it: [describe].
Confidence score for end-of-quarter completion (0-100% for each KR):
[your estimate].
Format for an internal team update. Flag any KR below 60% confidence with
"AT RISK" and name the specific action needed to recover it.
Prompt 31 — Metric Deep-Dive Analysis
Write a metric analysis memo for [metric name].
Current value: [X].
Target: [Y].
Trend over the last [X weeks/months]: [describe or paste data].
Hypothesis for why it's [above / below / on target]: [your theory].
What data would confirm or disprove this hypothesis: [list 2-3 data points].
Proposed experiment or action to move the metric: [describe].
Format for a data review meeting with [PM + Data + Engineering].
Prompt 32 — North Star Metric Presentation
Write talking points for a 5-minute north star metric presentation to [audience].
North star metric: [name and current value].
Why this is the north star (what user behavior it measures): [explain].
How it's trended over the last [quarter / year]: [describe].
What product initiatives are designed to move it this quarter: [list 2-3].
What would threaten it (risks to watch): [list 1-2].
Prompt 33 — Post-Launch Metrics Report
Write a post-launch metrics report for [feature name], [X weeks] after launch.
Metrics we said we'd track: [list from PRD or launch brief].
Actual results: [paste data].
What worked as expected: [describe].
What didn't match our hypothesis: [describe honestly — this is the most valuable part].
What we're changing based on what we learned: [next action].
Format for a learning-focused retrospective, not a success celebration.
Prompt 34 — Funnel Analysis Narrative
Write a funnel analysis narrative for [funnel name — e.g., "trial to paid",
"signup to first key action"].
Funnel stages: [list stages with conversion rates — e.g., "Signup: 100%,
Email verified: 68%, Onboarding completed: 31%, Key action: 19%, Converted: 11%"].
Biggest drop-off: [identify the stage with the worst rate].
Hypothesis for the drop-off: [your theory].
Two experiments worth running to improve the conversion at that stage: [describe].
Write this as a 200-word narrative I can paste into a product review doc.
Prompt 35 — Board-Level Metrics Summary
Write a board-level metrics summary for [product / company] for [quarter].
Key metrics to report: [list metrics with current values, trends, and targets].
Format: 3 sections — What went well (2-3 wins), What didn't (1-2 honest misses),
What we're focused on next quarter (1-2 priorities).
This will be read by board members who are not deep in the product.
Use plain language. Translate every metric into a business sentence.
Do not bury bad news. Surface it clearly and pair it with a plan.
Making These Work in Practice
The prompts above are starting points, not final drafts. Here's how to get 10× more out of them:
Add specificity before every run. Replace every bracket with your actual data. "User segment: enterprise buyers in financial services who manage $500M+ portfolios" produces a dramatically better output than "User segment: enterprise users."
Paste your actual documents. Many prompts work better when you paste a rough draft, raw notes, or existing data directly into the context field. The model improves existing text faster than it creates from nothing.
Iterate in the same thread. After you get a first draft, ask the model to: "Make it shorter," "Change the tone to be more direct," or "Rewrite bullet 3 — it's vague." Don't start a new chat for each revision.
Combine prompts. For a launch week, run Prompt 16 (exec briefing) + Prompt 15 (weekly update) + Prompt 17 (delay notification template to have ready) in the same session.
The best-ranking product managers I know treat ChatGPT like a first-draft engine and their own judgment as the editor. The prompts above cover 80% of the PM writing workflow. You bring the domain knowledge.
Want the Complete PM Prompt Library?
These 35 prompts cover the core PM writing workflows. The Product Manager AI Toolkit — $14.99 includes 100+ prompts organized by full PM lifecycle — discovery interviews, competitive analysis, launch plans, pricing pages, and job description writing — plus 15 fill-in-the-blank document templates.
Use LAUNCH30 for 30% off — limited uses remaining.
35 prompts. 5 PM workflows. Built for the PMs who write a PRD, two Slack messages, one stakeholder email, and a sprint planning doc before lunch.
Top comments (0)