Build a pre-launch checklist for a product feature

intermediateClaude SonnetProductProduct Managementproduct-launchproduct-managementrelease-managementchecklistsgo-to-market

Use case

Use this prompt 2–3 weeks before a planned feature launch. A thorough launch checklist prevents the most common launch failures: things nobody remembered to do, handoffs that didn't happen, and dependencies that were assumed to be done.

The prompt

You are an experienced product manager who has shipped dozens of features. Create a comprehensive pre-launch checklist for the feature described below.

**Feature name:** {{feature_name}}
**What it does:** {{feature_description}}
**Launch type:** {{launch_type}} (full release / beta / phased rollout / internal only)
**Target launch date:** {{launch_date}}
**Teams involved:** {{teams}}
**Customer segments affected:** {{customer_segments}}
**Key risks:** {{key_risks}}
**Rollout plan:** {{rollout_plan}} (e.g., 10% → 50% → 100%, or all-at-once, or feature flag)

Generate a launch checklist organized by phase and team. For each item:
- **Item:** Specific task or verification
- **Owner:** Role responsible
- **Due:** Relative timing (e.g., "2 weeks before launch," "Launch day")
- **Dependency:** What must be completed first
- **Status:** [ ] Checkbox

## Phase 1: 3–4 Weeks Before Launch (Foundation)

### Product
- PRD finalized and approved
- Success metrics defined and baseline captured
- Rollout strategy documented (phased, flag-based, etc.)
- Rollback plan documented

### Engineering
- Feature complete in staging
- Feature flag implemented (if phased rollout)
- Performance testing completed
- Load testing completed (if applicable)
- Security review completed (if applicable)
- Error monitoring configured

### Design
- Final designs approved and handed off
- Edge cases and error states designed
- Empty states designed
- Mobile responsive (if applicable)

### Legal / Compliance
- Privacy review (if collecting new data)
- Terms of service updates (if applicable)
- Accessibility compliance verified

## Phase 2: 1–2 Weeks Before Launch (Readiness)

### Engineering
- QA testing complete (all test cases passed)
- Regression testing complete
- Automated tests written and passing
- Feature deployed to production behind flag
- Monitoring dashboards configured

### Marketing / Comms
- Launch messaging approved
- Blog post or changelog written (not published)
- Email announcement drafted and approved
- Social posts drafted
- Sales talk track updated

### Customer Success / Support
- Product FAQ documented
- Support team trained
- In-app help content updated
- Escalation path for launch issues documented

### Data / Analytics
- Tracking events implemented and verified
- Dashboard configured for launch metrics
- Baseline metrics captured

## Phase 3: Launch Week

### Day Before Launch
- Final staging verification
- Feature flag ready for activation
- All communications approved and scheduled
- Monitoring war room scheduled (if needed)
- Rollback plan reviewed with engineering lead

### Launch Day
- Feature flag enabled (at planned %)
- Monitoring active
- Error rates checked immediately after activation
- Launch communications sent (if applicable)
- Team notified (Slack announcement)

## Phase 4: Post-Launch (Week 1 and Week 4)

### Week 1
- Day 1 metrics reviewed
- Any launch bugs triaged
- Customer support volume reviewed
- Phased rollout next step (if applicable)
- Learnings documented

### Week 4
- 30-day metrics review against targets
- Post-launch retro scheduled
- Feature flagging removed (if clean launch)
- Documentation updated

After generating the checklist, add:
- **Go/No-Go criteria:** Specific conditions that must be met to proceed with launch
- **Rollback plan:** How to disable the feature if something goes wrong
- **Escalation contacts:** Who to call for engineering, product, and comms emergencies

Variables

{{{{feature_name}}}}Replace with your {{feature name}}
{{{{feature_description}}}}Replace with your {{feature description}}
{{{{launch_type}}}}Replace with your {{launch type}}
{{{{launch_date}}}}Replace with your {{launch date}}
{{{{teams}}}}Replace with your {{teams}}
{{{{customer_segments}}}}Replace with your {{customer segments}}
{{{{key_risks}}}}Replace with your {{key risks}}
{{{{rollout_plan}}}}Replace with your {{rollout plan}}

Example

Input

feature_name: AI Insights — weekly engagement summary
feature_description: AI analyzes survey results weekly and generates a plain-English summary of top engagement insights with recommended actions
launch_type: Phased rollout (20% → 50% → 100% over 3 weeks)
launch_date: March 15, 2025
teams: Product, Engineering, Design, Marketing, Customer Success, Legal
customer_segments: All 800 paying accounts (all tiers)
key_risks: AI outputs could be inaccurate or misleading; customer data privacy concerns; potential for high token costs if usage spikes; some customers may be skeptical of AI
rollout_plan: 20% of accounts at launch (March 15), expand to 50% March 22 if metrics healthy, 100% March 29

Output

# Pre-Launch Checklist: AI Insights Feature
**Launch date:** March 15, 2025 | **Rollout:** Phased (20% → 50% → 100%) | **DRI:** VP Product

## Phase 1: 3–4 Weeks Before (by Feb 22)

**Product**
- [ ] AI Insights PRD approved by VP Product and Head of Engineering | PM | Feb 22
- [ ] Success metrics defined (Week 1 open rate, 30-day adoption, accuracy satisfaction) | PM | Feb 22
- [ ] Token cost model reviewed with CFO (cost per account per week at 100% rollout) | PM + Finance | Feb 22
- [ ] Rollback plan documented: feature flag disable process documented | PM + Engineering | Feb 22

**Engineering**
- [ ] AI output quality tested on 50 real anonymized accounts | Engineering | Feb 22
- [ ] Token cost cap per account implemented (circuit breaker) | Engineering | Feb 22
- [ ] Feature flag implemented — enables per-account activation | Engineering | Feb 22
- [ ] Rate limiting implemented on AI API calls | Engineering | Feb 22
- [ ] Error monitoring configured for AI generation failures | Engineering | Feb 22

**Legal**
- [ ] Privacy review: confirm no customer data is used for AI model training | Legal | Feb 22
- [ ] AI disclosure language approved ("Powered by AI — verify with your team") | Legal | Feb 22
- [ ] Terms of service updated to include AI-generated content disclosure | Legal | Mar 1

## Phase 2: 1–2 Weeks Before (by Mar 8)

**Engineering**
- [ ] QA: AI outputs reviewed for 20 test accounts — no hallucinations or inappropriate content | QA + PM | Mar 8
- [ ] Load test: simulate 200 accounts generating insights simultaneously | Engineering | Mar 8
- [ ] Monitoring dashboard live: tracks token cost, error rate, generation success rate | Engineering | Mar 8
- [ ] 20% cohort selected and feature flag configured for March 15 | Engineering | Mar 12

**Customer Success**
- [ ] CS team trained on AI Insights feature and common questions | CS Manager | Mar 8
- [ ] FAQ documented: "Is my data safe?" "How accurate is this?" "Can I turn it off?" | CS + PM | Mar 8
- [ ] Proactive email to beta customers (who gave early feedback) | CS | Mar 12
- [ ] Escalation path for AI accuracy complaints documented | CS Manager | Mar 8

**Marketing**
- [ ] Launch email drafted and approved ("AI Insights is here") | Marketing | Mar 8
- [ ] In-app announcement copy approved (legal cleared AI language) | Marketing + Legal | Mar 8
- [ ] Blog post drafted and legal-reviewed | Marketing | Mar 8
- [ ] Social posts drafted (no publication until launch day) | Marketing | Mar 8

## Launch Day (March 15)

- [ ] Feature flag enabled for 20% cohort at 9:00 AM ET | Engineering Lead
- [ ] Error rate and token cost monitored for first 2 hours | Engineering on call
- [ ] Launch email sent at 10:00 AM ET | Marketing
- [ ] In-app announcement enabled | Engineering / CS Ops
- [ ] Team notification in #product-announcements | PM
- [ ] CS team briefed — "launch is live, watch for inbound" | CS Manager
- [ ] Token cost checked against projections by noon ET | PM

## Go / No-Go Criteria (March 14 check)
**Launch proceeds if:**
- Zero P0/P1 bugs open in staging
- AI output quality review passed (QA sign-off)
- Legal has approved all customer-facing AI copy
- Token cost projection is within approved budget
- CS team confirms they're trained and ready

**Launch blocked if:**
- Any known AI hallucination or misleading output in test accounts
- Token cost projection exceeds $8K/month at 20% rollout (escalate to CFO)
- Legal has not approved AI disclosure language

## Rollback Plan
1. Engineering disables feature flag for all accounts — immediate (< 5 minutes)
2. PM sends internal Slack notification: "AI Insights rolled back — [reason]"
3. CS sends brief customer email if feature was already announced: "We've temporarily paused AI Insights while we improve quality. [ETA for return]"
4. Marketing holds all pending social posts

## Escalation Contacts
| Area | Primary | Secondary |
|------|---------|-----------|
| Engineering emergency | Marcus W. (on-call) | Dev A. |
| Product decisions | Jamie C. (VP Product) | PM on rotation |
| Customer communications | CS Manager | VP CS |
| AI accuracy concern | PM + Head of Eng | CTO |

Tips for best results

1The Go/No-Go criteria are the most important section. Define them before launch week — not on launch morning when pressure is high and judgment is compromised.
2Legal review of AI-related copy often takes longer than expected. Start it 4 weeks before launch, not 1 week.
3A feature flag with phased rollout is not optional for AI features — it's essential. Never launch AI to 100% of users on day 1.
4After every launch, run a quick retro: What was almost missed? What took longer than expected? What can be added to the standard checklist for next time?

Related prompts

Write a GTM brief for a new feature or product launch

intermediate

Generate a comprehensive go-to-market brief for a feature or product launch that aligns product, marketing, sales, and customer success on positioning, messaging, and launch execution.

Productgo-to-marketproduct-launchproduct-management

Write a post-launch review for a product feature

intermediate

Generate a structured post-launch review that evaluates what went well, what didn't, and what the team learned — to improve future launches and document impact.

Productpost-launchproduct-managementretrospective

Write a one-page product requirements document for a feature

intermediate

Generate a concise, structured one-page PRD that captures problem, scope, requirements, and success metrics — enough for engineering to begin planning.

Productproduct-managementrequirementsprd

Need help implementing this prompt in your workflow?

Book a call

You are an experienced product manager who has shipped dozens of features. Create a comprehensive pre-launch checklist for the feature described below. **Feature name:** {{feature_name}} **What it does:** {{feature_description}} **Launch type:** {{launch_type}} (full release / beta / phased rollout / internal only) **Target launch date:** {{launch_date}} **Teams involved:** {{teams}} **Customer segments affected:** {{customer_segments}} **Key risks:** {{key_risks}} **Rollout plan:** {{rollout_plan}} (e.g., 10% → 50% → 100%, or all-at-once, or feature flag) Generate a launch checklist organized by phase and team. For each item: - **Item:** Specific task or verification - **Owner:** Role responsible - **Due:** Relative timing (e.g., "2 weeks before launch," "Launch day") - **Dependency:** What must be completed first - **Status:** [ ] Checkbox ## Phase 1: 3–4 Weeks Before Launch (Foundation) ### Product - PRD finalized and approved - Success metrics defined and baseline captured - Rollout strategy documented (phased, flag-based, etc.) - Rollback plan documented ### Engineering - Feature complete in staging - Feature flag implemented (if phased rollout) - Performance testing completed - Load testing completed (if applicable) - Security review completed (if applicable) - Error monitoring configured ### Design - Final designs approved and handed off - Edge cases and error states designed - Empty states designed - Mobile responsive (if applicable) ### Legal / Compliance - Privacy review (if collecting new data) - Terms of service updates (if applicable) - Accessibility compliance verified ## Phase 2: 1–2 Weeks Before Launch (Readiness) ### Engineering - QA testing complete (all test cases passed) - Regression testing complete - Automated tests written and passing - Feature deployed to production behind flag - Monitoring dashboards configured ### Marketing / Comms - Launch messaging approved - Blog post or changelog written (not published) - Email announcement drafted and approved - Social posts drafted - Sales talk track updated ### Customer Success / Support - Product FAQ documented - Support team trained - In-app help content updated - Escalation path for launch issues documented ### Data / Analytics - Tracking events implemented and verified - Dashboard configured for launch metrics - Baseline metrics captured ## Phase 3: Launch Week ### Day Before Launch - Final staging verification - Feature flag ready for activation - All communications approved and scheduled - Monitoring war room scheduled (if needed) - Rollback plan reviewed with engineering lead ### Launch Day - Feature flag enabled (at planned %) - Monitoring active - Error rates checked immediately after activation - Launch communications sent (if applicable) - Team notified (Slack announcement) ## Phase 4: Post-Launch (Week 1 and Week 4) ### Week 1 - Day 1 metrics reviewed - Any launch bugs triaged - Customer support volume reviewed - Phased rollout next step (if applicable) - Learnings documented ### Week 4 - 30-day metrics review against targets - Post-launch retro scheduled - Feature flagging removed (if clean launch) - Documentation updated After generating the checklist, add: - **Go/No-Go criteria:** Specific conditions that must be met to proceed with launch - **Rollback plan:** How to disable the feature if something goes wrong - **Escalation contacts:** Who to call for engineering, product, and comms emergencies

Variables

{{{{feature_name}}}}Replace with your {{feature name}}

{{{{feature_description}}}}Replace with your {{feature description}}

{{{{launch_type}}}}Replace with your {{launch type}}

{{{{launch_date}}}}Replace with your {{launch date}}

{{{{teams}}}}Replace with your {{teams}}

{{{{customer_segments}}}}Replace with your {{customer segments}}

{{{{key_risks}}}}Replace with your {{key risks}}

{{{{rollout_plan}}}}Replace with your {{rollout plan}}

Example

Input

feature_name: AI Insights — weekly engagement summary
feature_description: AI analyzes survey results weekly and generates a plain-English summary of top engagement insights with recommended actions
launch_type: Phased rollout (20% → 50% → 100% over 3 weeks)
launch_date: March 15, 2025
teams: Product, Engineering, Design, Marketing, Customer Success, Legal
customer_segments: All 800 paying accounts (all tiers)
key_risks: AI outputs could be inaccurate or misleading; customer data privacy concerns; potential for high token costs if usage spikes; some customers may be skeptical of AI
rollout_plan: 20% of accounts at launch (March 15), expand to 50% March 22 if metrics healthy, 100% March 29

Output

# Pre-Launch Checklist: AI Insights Feature
**Launch date:** March 15, 2025 | **Rollout:** Phased (20% → 50% → 100%) | **DRI:** VP Product

## Phase 1: 3–4 Weeks Before (by Feb 22)

**Product**
- [ ] AI Insights PRD approved by VP Product and Head of Engineering | PM | Feb 22
- [ ] Success metrics defined (Week 1 open rate, 30-day adoption, accuracy satisfaction) | PM | Feb 22
- [ ] Token cost model reviewed with CFO (cost per account per week at 100% rollout) | PM + Finance | Feb 22
- [ ] Rollback plan documented: feature flag disable process documented | PM + Engineering | Feb 22

**Engineering**
- [ ] AI output quality tested on 50 real anonymized accounts | Engineering | Feb 22
- [ ] Token cost cap per account implemented (circuit breaker) | Engineering | Feb 22
- [ ] Feature flag implemented — enables per-account activation | Engineering | Feb 22
- [ ] Rate limiting implemented on AI API calls | Engineering | Feb 22
- [ ] Error monitoring configured for AI generation failures | Engineering | Feb 22

**Legal**
- [ ] Privacy review: confirm no customer data is used for AI model training | Legal | Feb 22
- [ ] AI disclosure language approved ("Powered by AI — verify with your team") | Legal | Feb 22
- [ ] Terms of service updated to include AI-generated content disclosure | Legal | Mar 1

## Phase 2: 1–2 Weeks Before (by Mar 8)

**Engineering**
- [ ] QA: AI outputs reviewed for 20 test accounts — no hallucinations or inappropriate content | QA + PM | Mar 8
- [ ] Load test: simulate 200 accounts generating insights simultaneously | Engineering | Mar 8
- [ ] Monitoring dashboard live: tracks token cost, error rate, generation success rate | Engineering | Mar 8
- [ ] 20% cohort selected and feature flag configured for March 15 | Engineering | Mar 12

**Customer Success**
- [ ] CS team trained on AI Insights feature and common questions | CS Manager | Mar 8
- [ ] FAQ documented: "Is my data safe?" "How accurate is this?" "Can I turn it off?" | CS + PM | Mar 8
- [ ] Proactive email to beta customers (who gave early feedback) | CS | Mar 12
- [ ] Escalation path for AI accuracy complaints documented | CS Manager | Mar 8

**Marketing**
- [ ] Launch email drafted and approved ("AI Insights is here") | Marketing | Mar 8
- [ ] In-app announcement copy approved (legal cleared AI language) | Marketing + Legal | Mar 8
- [ ] Blog post drafted and legal-reviewed | Marketing | Mar 8
- [ ] Social posts drafted (no publication until launch day) | Marketing | Mar 8

## Launch Day (March 15)

- [ ] Feature flag enabled for 20% cohort at 9:00 AM ET | Engineering Lead
- [ ] Error rate and token cost monitored for first 2 hours | Engineering on call
- [ ] Launch email sent at 10:00 AM ET | Marketing
- [ ] In-app announcement enabled | Engineering / CS Ops
- [ ] Team notification in #product-announcements | PM
- [ ] CS team briefed — "launch is live, watch for inbound" | CS Manager
- [ ] Token cost checked against projections by noon ET | PM

## Go / No-Go Criteria (March 14 check)
**Launch proceeds if:**
- Zero P0/P1 bugs open in staging
- AI output quality review passed (QA sign-off)
- Legal has approved all customer-facing AI copy
- Token cost projection is within approved budget
- CS team confirms they're trained and ready

**Launch blocked if:**
- Any known AI hallucination or misleading output in test accounts
- Token cost projection exceeds $8K/month at 20% rollout (escalate to CFO)
- Legal has not approved AI disclosure language

## Rollback Plan
1. Engineering disables feature flag for all accounts — immediate (< 5 minutes)
2. PM sends internal Slack notification: "AI Insights rolled back — [reason]"
3. CS sends brief customer email if feature was already announced: "We've temporarily paused AI Insights while we improve quality. [ETA for return]"
4. Marketing holds all pending social posts

## Escalation Contacts
| Area | Primary | Secondary |
|------|---------|-----------|
| Engineering emergency | Marcus W. (on-call) | Dev A. |
| Product decisions | Jamie C. (VP Product) | PM on rotation |
| Customer communications | CS Manager | VP CS |
| AI accuracy concern | PM + Head of Eng | CTO |

Tips for best results

1The Go/No-Go criteria are the most important section. Define them before launch week — not on launch morning when pressure is high and judgment is compromised.

2Legal review of AI-related copy often takes longer than expected. Start it 4 weeks before launch, not 1 week.

3A feature flag with phased rollout is not optional for AI features — it's essential. Never launch AI to 100% of users on day 1.

4After every launch, run a quick retro: What was almost missed? What took longer than expected? What can be added to the standard checklist for next time?