Launch templateChecklists and process
Post-launch retrospective template
A retro turns one launch into lessons for the next. Run it a week or two after launch, when the numbers have settled, and write down what you'd do differently while it's fresh. Be specific about channels: which one brought users who stayed, not just visits.
Use it for
- One to two weeks after launch.
- After any campaign or big release.
- Before planning the next launch.
Limits and rules
- Compare channels on users who stayed or paid, not only visits; a launch spike can hide which channel actually worked.
The template
Copy it, then replace everything in [BRACKETS]. Delete lines that don't apply rather than leaving them vague.
POST-LAUNCH RETRO - [PRODUCT] - launched [DATE] GOALS vs RESULTS Goal: [WHAT YOU SET] -> Result: [WHAT HAPPENED] Goal: [...] -> Result: [...] BY CHANNEL | Channel | Visits | Signups | Active after 7 days | Notes | | [PRODUCT HUNT] | | | | | | [LAUNCHRANKED] | | | | | | [HACKER NEWS] | | | | | | [EMAIL] | | | | | | [COMMUNITY] | | | | | WHAT WORKED - [...] WHAT DIDN'T - [...] SURPRISES - [...] TOP QUESTIONS PEOPLE ASKED - [QUESTION] -> [PAGE / FAQ WE'LL WRITE] WHAT WE'D DO DIFFERENTLY - [...] NEXT 30 DAYS - [ACTION] - [OWNER] - [DATE]
How to fill it in
Active after 7 days
The column that tells you which channel brought real users.
Top questions
Each becomes a FAQ entry or a page.
Next 30 days
Owners and dates, or it won't happen.
Filled example
Fictional example. Paperlark is a fictional product we made up for these examples: an invoice-reminder tool for freelancers that sends polite follow-ups when a client pays late. Names, prices and details are invented to show the template filled in; they aren't a real company or real results.
POST-LAUNCH RETRO - Paperlark - launched [date] GOALS vs RESULTS Goal: 200 signups -> Result: [actual] Goal: 20 paying -> Result: [actual] BY CHANNEL (fill in from analytics; we don't invent numbers, even for a fictional product) WHAT WORKED - The freelancer subreddit post: people shared their own late-payment stories - Waitlist email: most replies had useful feedback WHAT DIDN'T - Show HN: skipped because the demo needed a login; should have built a no-login demo SURPRISES - People cared more about the reminder tone than the schedule TOP QUESTIONS - "Will clients know it's automated?" -> FAQ + sample reminder page - "Does it work with my invoicing tool?" -> integrations page WHAT WE'D DO DIFFERENTLY - Build the no-login demo before launch NEXT 30 DAYS - Write 3 pages from top questions - Sam - end of month (Fictional example.)
Do
- Measure users who stayed, by channel.
- Write down surprises.
- Turn questions into pages.
- Share the retro with anyone who helped.
Don't
- Don't judge channels on visits alone.
- Don't skip what didn't work.
- Don't end without owners and dates.
FAQ
When should I run a launch retro?
One to two weeks after launch, once the numbers have settled but details are fresh.
What's the most useful launch metric?
Users who were still active after a week, split by channel. Visits alone can mislead.
What did we learn from our own launches?
On our own product, publishing about 2,500 posts in one month got indexing throttled; a pace around 20 a day worked. Write lessons like that down so you don't repeat them.
Launch guides that use this template
Related templates
- Launch week checklistA day-by-day launch week checklist: final checks, launch-day hour plan, follow-ups and the SEO steps that keep traffic coming after the spike.
- Feedback surveyA short product feedback survey for new users: seven questions on fit, value, friction and alternatives, with a filled example and how to read the answers.
- Changelog entryA changelog entry template grouped by Added, Changed, Fixed and Removed, written for users rather than commit logs, with a filled example.