Launch day mechanics: the week that decides the day
A launch is a logistics exercise dressed as a marketing event. Almost everything that decides the outcome is done in the week before you post — which is good news, because that part is controllable.
The short version
- Almost everything that decides a launch happens before you post — which is good news, because that part is controllable.
- Tell twenty people individually, by name, with a date. It is the highest-leverage hour of the whole launch.
- Prepare the assets and fix the first-run experience first: launch traffic is the least patient traffic you will ever get.
- Stay in the thread for the first six hours, answer the criticism publicly, and harvest testimonials and objections in the week after.
Launch outcomes look like luck from the outside. From the inside they are mostly preparation: the assets that existed before the day started, the people who already knew it was coming, and the founder being genuinely available for two days rather than heroically busy.
The week before
- Tell twenty people individually. Not a broadcast — twenty messages that mention them by name and say when. This is the single highest-leverage hour of the whole launch.
- Prepare the assets: a thirty-second recording, three screenshots, the one-sentence description, and the longer version for comment replies.
- Fix the first-run experience. Launch traffic is the least patient traffic you will ever receive; anything confusing converts to nothing.
- Test the signup flow on a phone, on mobile data, signed out. This is how most of your visitors will arrive.
The day itself
Post early in the day of your largest audience's timezone, then stay in the thread. Reply to everything for the first six hours, including the criticism — especially the criticism, which is read far more carefully than praise and where a good answer converts skeptics into advocates in public.
The week after
The spike decays within days; what persists is what you harvest. Email everyone who signed up, personally, while they still remember why. Turn the good comments into testimonials with permission. Write down the three objections you heard most — that list is your next month of work, handed to you by the market for free.
None of this contradicts the harder truth that a launch will not hand you a user base — the spike fades, and the durable channels are the slow ones. It means the day is worth preparing for anyway, because the artifacts it leaves behind outlive the traffic it brings. And then launch again. Nothing in the rules says you get one. A meaningful new capability, a new audience, a new angle on the same product — the founders who compound are the ones who treat launching as a repeatable operation rather than a birthday.
Put this article to work
Ask it a question, or turn it into a to-do list for your own project. Both answer strictly from this article — nothing invented.
Answers are generated from this article only and are a working draft, not advice.
Questions this answers
How do you prepare for a product launch?
In the week before, not on the day. Message twenty people individually with a date, prepare a short recording, screenshots and the one-line description, fix the first-run experience, and test the signup flow signed out on a phone over mobile data. Launch traffic is the least patient traffic you will ever receive.
What should you do on launch day itself?
Post early in your largest audience’s timezone and then stay in the thread. Reply to everything for the first six hours, especially the criticism — negative comments are read far more carefully than praise, and a good public answer converts skeptics into advocates.
Can you launch a product more than once?
Yes, and the founders who compound do. A meaningful new capability, a new audience or a new angle on the same product all justify launching again. Nothing in the rules grants you exactly one, and treating launches as a repeatable operation beats treating them as a birthday.
Building something?
Put it in front of founders who read this — free listing, community-voted, reviewed before it goes live.
Keep reading
Your first hundred users will not come from a launch
Launch days produce a spike and a lesson. The founders who get to a hundred real users almost always get there through channels that look too small to bother with.
Read the article →
Build a distribution radar: the agent that finds conversations
Answering the right thread is the highest-quality distribution a new product has. Finding it every morning is a chore, and chores get dropped — so hand the finding to an agent and keep the answering.
Read the article →
Nine red flags your users see before you do
Most early products do not fail from one dramatic mistake. They fail from a handful of small signals that quietly tell visitors this is not for them.
Read the article →