How do you plan a product launch week?
A successful product launch week is not a single moment—it's a structured sequence of actions that runs from about two weeks before you ship to several days after. The goal is not to generate a spike of attention but to find and lock in the first users who genuinely need what you built, then learn from them fast enough to improve before momentum fades.
Start with one user, not an audience
The instinct before a launch is to think about reach—press, Product Hunt, social posts, influencers. That instinct is mostly wrong, especially for an early product. Paul Graham's argument in 'Do Things That Don't Scale' is that your real job in the early days is to find a single person whose problem your product solves acutely, then treat them like you're a consultant building something exclusively for them. That constraint forces you to learn things you cannot learn any other way.
So before you write a launch tweet or draft a press pitch, answer this: who is the one person whose life gets measurably better if your product works? Name them. If you can name an actual human being—not a persona, a real person—you have a launch target. Your launch week goal is to get that person (and a handful of people like them) using the product and talking to you daily. Everything else is secondary.
This reframe changes your pre-launch to-do list. Instead of 'write a blog post,' you write a list of ten specific people to call the morning you launch. Instead of 'set up a waitlist,' you schedule 30-minute onboarding calls for the first five signups. Reach can come later; depth comes first.
The two weeks before: prepare the foundation, not the fanfare
Two weeks out, your job is to eliminate every reason a first user might bounce before experiencing the core value of your product. Work backward from the moment a new user first opens the product and ask: what is the single thing they need to succeed in the first five minutes? Build that path until it's airtight. Everything else on your roadmap can wait.
At the same time, draft a short outreach list organized into three tiers: people who already know and trust you (warm contacts), communities where your exact target user hangs out (forums, Slack groups, subreddits), and a short press or newsletter list if your category warrants it. Tier one is where your first paying or deeply engaged users will come from. Do not skip it in favor of the glamour of press.
Set up your feedback infrastructure before you launch, not after. This means a simple way to capture every support message in one place, a shared doc or Notion board for user feedback, and a standing block of time each morning of launch week to review what came in overnight. Founders who set this up after launch are always a day behind the signal they need.
Launch day and the days immediately after
Launch day is not a single event—treat it as the start of a five-day sprint. On day one, activate your tier-one list personally. Send individual messages, not a mass email. Reference something specific about why this person in particular would find the product useful. The reply rate and conversion rate on personal outreach is dramatically higher than broadcast, and the conversations you have will give you language you'll use in every piece of marketing you write afterward.
If you're posting on public channels—Product Hunt, Hacker News, a relevant subreddit—do it in the morning of day one and then be present in the comments for the entire day. Founders who post and disappear get far less out of public launches than those who answer every question, take every criticism seriously in real time, and treat the comment section as a live user interview.
Days two through five are for following up. Every person who signed up but hasn't activated: reach out individually and offer to walk them through it. Every person who activated: ask them what confused them and what they wished was there. This is not customer service—it's product research with urgency. The information you collect in the 72 hours after launch is among the most valuable data your company will ever produce.
What to measure and what to ignore
Launch week generates a lot of noise metrics: page views, social impressions, follower counts, upvotes. Mostly these don't matter, and tracking them will distort your priorities. The metrics that tell you something real during launch week are: number of people who completed the core action your product is built around, number of people who came back the next day without being prompted, and number of conversations you had directly with users.
Set a specific, small target for each of these before the week starts—not a big aspirational number, a number that would prove there is genuine demand. Ten people completing the core action matters more than a thousand signups who never opened the app. Three people who return on day two without prompting tells you more about retention than any survey.
At the end of the week, the question to answer is not 'did the launch go well?' but 'do we understand our user well enough to make the product meaningfully better in the next two weeks?' If yes, launch week succeeded. If you have traffic but no conversations, you have work to do.
The post-launch week: don't let momentum die
The mistake most founders make is treating launch week as a crescendo rather than a beginning. After the initial push, activity drops, and without a plan, the team loses energy and focus. Prevent this by scheduling, before launch week starts, what happens the week after: a synthesis session to review all feedback, a short list of the top three changes users asked for, and a clear owner for each.
If you ran the launch with the depth this playbook recommends—personal outreach, daily user conversations, real-time feedback collection—you will end the week with enough material to inform the next month of product work. The founders who extract the most from a launch are those who turn raw user feedback into specific, actionable changes within two weeks of launching, then tell their early users what they built in response. That loop is how you convert a launch into a retained user base.
“The initial user serves as the form for your mold; keep tweaking till you fit their needs perfectly.”
— Paul Graham, source
The one thing to do
Before you write a single launch post, name one real person whose problem your product solves acutely, then make getting them on a call your first launch-week action.
Frequently asked questions
How long should a product launch week actually be?
Plan for five active days, but the work starts two weeks before and continues for at least a week after. The 'week' is really a three-week arc with the launch as the midpoint, not the endpoint.
Should I launch on Product Hunt during my launch week?
Only if your target user is actually on Product Hunt—many B2B and niche categories are not well represented there. If you do launch publicly, treat it as one channel among several, not the main event, and be present in comments all day.
How many users should I aim to get in launch week?
Set a target based on depth, not breadth: aim for a number of users you can personally talk to during the week, typically 10 to 30 for an early product. Understanding those users completely beats having hundreds you've never spoken to.
What's the single biggest launch week mistake founders make?
Optimizing for attention metrics—views, upvotes, press mentions—instead of direct user conversations. Attention without learning is wasted momentum that won't repeat itself.
Sources
- Do Things that Don't Scale — Paul Graham
- The Refragmentation — Paul Graham
- gstack: AGENTS.md — Garry Tan