An end-to-end guide for planning and executing a product launch. Covers everything from early planning through post-launch review. Use this alongside the checklists and templates in this toolkit.
| Week | Focus | Key Activities |
|---|---|---|
| Week 1-2 | Planning | Determine launch tier, identify stakeholders, create master checklist, assign owners, set target date |
| Week 3 | Preparation | Engineering: feature development wrapping up. Marketing: draft content. Support: start documentation. |
| Week 4 | Testing & Content | QA: full test cycle. Marketing: finalize content. Support: training session. Legal: review. |
| Week 5 | Staging & Rehearsal | Full staging deployment. Rollback drill. Marketing content scheduled. Go/no-go prep. |
| Week 6 | Launch | Go/no-go meeting (Mon/Tue). Launch (Wed/Thu preferred). Monitor. Post-launch review (following week). |
| Week | Focus | Key Activities |
|---|---|---|
| Week 1 | Planning & Prep | Create checklist, assign owners, begin documentation and content |
| Week 2 | Testing & Content | QA cycle, staging deployment, content finalized, support briefed |
| Week 3 | Launch | Go/no-go (async), launch, monitor, day-7 review |
Hold this meeting at the start of the planning period. 30 minutes.
1. Feature overview (PM, 5 min) — What are we launching and why?
2. Tier determination (Launch Owner, 5 min) — What level of process do we need?
3. Timeline review (Launch Owner, 5 min) — Target date, key milestones
4. Owner assignment (All, 10 min) — Who owns each checklist area?
5. Risk identification (All, 5 min) — What could go wrong?
| Mistake | Impact | Prevention |
|---|---|---|
| Launching on a Friday | If something breaks, you're debugging over the weekend | Launch Tuesday-Thursday. Never Friday, never before a holiday. |
| No one monitoring after deployment | Issues discovered by users, not by you | Assign on-call for the launch window. Set alerts. Watch dashboards. |
| Marketing goes out before deployment completes | Users click through to features that aren't live yet | Marketing waits for deployment confirmation before publishing |
| No rollback plan | When things break, you're improvising under pressure | Document and test rollback BEFORE launch. See rollback checklist. |
| Launching too much at once | Can't isolate which change caused a problem | One major feature per launch. Bundle small changes separately. |
| Skipping the go/no-go meeting | Silent disagreements surface after launch | Always hold the meeting, even if it's 10 minutes and everyone says GO |
| Not briefing support | Support team gets tickets they can't answer, users get bad experience | Demo + Q&A session with support team before every Tier 1/2 launch |
| Not defining success metrics beforehand | After launch, nobody agrees on whether it was successful | Define metrics and targets BEFORE launch. Write them down. |
Conduct the full post-launch review using templates/post-launch-review.md. The most important output is the action items — things to improve for next time.
The best product teams treat launches as a repeatable process, not a heroic event.
Signs of a healthy launch culture:
Signs of an unhealthy launch culture:
Different audiences need different information at different times. This matrix prevents the two most common communication failures: telling people too late and telling them the wrong level of detail.
| Audience | When to Tell | What They Need | Format |
|---|---|---|---|
| Engineering team | Week 1 (planning) | Technical scope, timeline, on-call expectations | Sprint planning, Slack |
| QA team | Week 1-2 | Acceptance criteria, test plan scope, staging timeline | Meeting + written spec |
| Support team | Week 3-4 (before launch) | Feature walkthrough, FAQ, canned responses, escalation path | Demo session + documentation |
| Marketing team | Week 1 (planning) | Value prop, target audience, key dates, asset needs | Kickoff meeting |
| Sales team | Week 3-4 | Talking points, demo access, competitive positioning, pricing changes | Enablement session |
| Customer success | Week 3-4 | Customer-facing FAQ, proactive outreach plan, upsell opportunities | Meeting + playbook |
| Executive/leadership | Week 1 + launch day | Business case, success metrics, launch status, results | Email update (brief) |
| Legal/compliance | Week 2-3 | Data privacy review, terms changes, accessibility compliance | Review request (async) |
| Customers (beta) | Week 3-4 | Early access, feedback requests | Email + in-app |
| Customers (all) | Launch day | What's new, how it helps them, how to use it | Email + changelog + blog |
| Anti-Pattern | What Happens | How to Fix |
|---|---|---|
| "Surprise launch" | Support gets tickets they can't answer, sales finds out from customers | Follow the communication matrix. Brief support and sales 1+ week before launch. |
| "Asymmetric information" | Engineering knows about a risk that PM doesn't, leading to launch-day surprises | All risks go in the go/no-go template. No side conversations that skip the launch owner. |
| "Over-communicating noise" | Daily status emails that nobody reads, leading people to ignore real updates | Limit updates to weekly during planning, daily during launch week only. Use clear subject lines. |
| "Announcement before ready" | Marketing publishes before deployment completes; users click to a feature that isn't live | Marketing waits for explicit "all clear" signal from engineering before publishing anything. |
| "Radio silence on bad news" | A blocker exists but nobody escalates; discovered too late to adjust | Daily standups during launch week surface blockers. Any P0/P1 goes to Slack immediately, not "next standup." |
Not every launch needs 20 metrics. Pick 3-5 that answer the question: "Was this launch worth it?"
Framework: Input → Output → Outcome
| Level | What It Measures | Example |
|---|---|---|
| Input (effort) | What you put in | "We launched feature X on time" |
| Output (adoption) | Whether users found and used it | "1,200 users tried the feature in week 1" |
| Outcome (impact) | Whether it changed the business metric you care about | "Activation rate improved from 68% to 74%" |
Most teams stop at Output metrics. The real measure of success is Outcome metrics.
| Launch Type | Recommended Metrics |
|---|---|
| New feature | Feature adoption rate (day 1, 7, 30), impact on North Star metric, support ticket volume |
| UX improvement | Task completion rate, time-on-task, user satisfaction score, bounce rate change |
| Performance improvement | Load time reduction, error rate reduction, user-reported issues |
| Pricing change | Conversion rate change, ARPU change, churn rate change, revenue per visitor |
| Platform/infra migration | Error rate, latency, support tickets, user complaints (goal: no change) |
| Target Type | Description | Example |
|---|---|---|
| Baseline | What the metric is today, before the launch | "Current activation rate: 68%" |
| Minimum viable | The smallest improvement that makes this launch worth it | "Activation rate ≥ 70% (+2pp)" |
| Target | What you actually expect based on analysis | "Activation rate = 72-74%" |
| Stretch | Ambitious but not unrealistic | "Activation rate ≥ 76%" |
Setting all four levels prevents two problems:
1. Declaring failure when you hit 71% (below stretch but above minimum viable)
2. Declaring success when you hit 69% (barely moved the needle)
You likely don't need separate Engineering, QA, Marketing, and Support checklists — one person may cover multiple areas. Simplifications:
This toolkit is designed for your size. Use it as-is, with these tips:
You may need to extend the system:
By PM Toolkit Pro | $29
A comprehensive launch management system for product teams. Covers every phase of a product launch — from engineering readiness through marketing, support, legal, and rollback planning. Includes a master checklist, area-specific checklists, runbook templates, and post-launch review frameworks.
Get the full Launch Checklist System and unlock everything.
Get the complete guide with every chapter unlocked, including code samples, diagrams, and best practices.
Access all interactive tools with complete data, all workload profiles, and the full scenario library.
Downloadable source code, configuration files, and working examples from every chapter.
Free updates for life. Every new chapter, tool, and improvement included.