Gavel Gavel
Experts Guides Skills
Guides
Launch

A product launch checklist that moves the needle

A useful product launch checklist skips the press and the swag. It is a short sequence. Build an email list before the product exists. Recruit your first users by hand, the way Paul Graham describes. Ship the jankiest version that still delivers value, the way Garry Tan argues. Pick one channel instead of five. Set the number that tells you it worked. Below is the checklist, each step cited to an operator who ran it. They disagree on how big an audience you need first.

Why this matters. "How do I launch my product" is where most first-time founders freeze. The advice they find is a generic 40-item to-do list. It cannot tell the two steps that matter from the thirty-eight that do not.

Ask Gavel, free

Ask Gavel to turn this checklist into a dated plan for your specific launch.

5

hair-on-fire users who will tolerate a janky, embarrassing MVP is the whole early-validation game. That's per Garry Tan. Five who need it teach you more than five hundred who are curious.

Garry Tan YC Office Hours

The short answer

The founder launch checklist

Seven steps, ordered. Run them top to bottom and the launch stops being a one-day gamble and becomes a probe you can read.

  1. 1

    Build the list before you build the launch

    Stand up a pre-launch landing page with email capture weeks ahead, so launch day has an audience instead of silence. Colin sized his Sheets & Giggles list to a revenue goal before the product existed.

  2. 2

    Know what they would do without you

    Write down the real alternative a buyer would pick if you did not exist. Excel. An intern. Doing nothing. April Dunford's point: differentiation exists only against that alternative, so name it before you launch.

  3. 3

    Recruit your first users by hand

    Do not wait for a funnel. Email, DM and meet the first ten users one at a time. Do the unscalable work that delights them. Paul Graham: startups do not take off on their own.

  4. 4

    Ship the jankiest version that still works

    Cut scope until launching feels uncomfortable, and keep the core loop working end to end. Janky still works. Broken does not. If someone uses the janky version, the problem is real.

  5. 5

    Pick one channel, not five

    Choose one of Hormozi's Core Four and work it for a quarter. Warm outreach or cold outreach. Content or paid ads. Sampling all four gets nothing past escape velocity.

  6. 6

    Make one thing remarkable

    If you cannot describe what makes it remarkable in one sentence, it will not spread by word of mouth. Seth Godin: design the remarkable quality for the sneezers, not the masses.

  7. 7

    Set the number that tells you it worked

    Decide your success signal before launch. Sean Ellis draws the line at 40% of active users who say losing the product would leave them very disappointed.

The order is the point. List first. Hand-recruited users second. A janky ship third. One channel fourth. Read the signal last. Skip the sequencing and launch day becomes a coin flip you cannot learn from.

The cited playbook

  1. 1

    Build the launch list before the product is ready

    Weeks before you ship, put up a pre-launch landing page and capture emails. Use founder-market fit and brand personality to earn signups. Colin built the Sheets & Giggles email list before the product existed. He calculated the list size from his target conversion rate and revenue goal.

    Colin (Sheets & Giggles) · Pre-launch email capture
  2. 2

    Nail what customers would do without you

    Before launch day, define the real alternative a buyer would use if you did not exist. A spreadsheet. A manual process. Doing nothing at all. April Dunford's rule: differentiated value only exists against that alternative. Name it before you launch, or first-user feedback is impossible to read.

    April Dunford · Competitive Alternatives
  3. 3

    Recruit your first users by hand

    Do not launch into a void and hope. Go find the first ten users one by one. Email them. Message them. Meet them. Do the unscalable setup work that delights them. Paul Graham's argument is that this by-hand work is the only thing that produces real users and real feedback early on.

    Paul Graham · Do Things That Don't Scale
  4. 4

    Ship the jankiest version that still delivers value

    Cut the launch backlog to the few items that produce minimum value end to end and strike the rest. Garry Tan's filter: launch the jankiest MVP that still works. If someone uses a janky product, the problem is real. That shows how much they need it solved. Janky is fine, broken is not.

    Garry Tan · Launch-jankiest rule
  5. 5

    Pick one channel and go deep

    Do not launch across every channel at once. Choose one of the Core Four. Warm outreach or cold outreach. Content or paid ads. Commit to one for a quarter before adding a second. Alex Hormozi's point is that founders who rotate channels weekly never push any single one past the point where it compounds.

    Alex Hormozi · Core Four
  6. 6

    Make one thing remarkable enough to spread

    A launch with nothing remarkable in it needs paid promotion every single day. Seth Godin's test: can you describe what makes the product remarkable in one sentence? Does it target the sneezers who spread ideas rather than the passive masses?

    Seth Godin · Purple Cow
  7. 7

    Set the signal you are aiming for

    Decide before launch what result would mean it worked. Sean Ellis surveys active users on how they would feel if they could no longer use the product. He draws the line at 40% saying very disappointed. That number moves weeks before a retention curve can confirm the same thing.

    Sean Ellis · The Sean Ellis Test

Where experts disagree

Where operators disagree: how big an audience you need before launch

Gil (Subscribr)

builds the audience and distribution first, then launches to it. That way the audience proves demand before a line of code ships. Lower risk and more predictable, but slower to the first dollar.

Yaser (Chatbase)

had sixteen followers when a single tweet went viral and turned into a seven-figure business. His bet: you do not need an audience at all, one viral post or Reddit thread can launch you from zero.

ChatGPT averages this into 'build an audience.' Gavel shows you both the audience-first safety and the zero-audience upside. Bet the one that matches your runway and your appetite for variance.

FAQ

Common questions

What should a product launch checklist include?

These are the steps that produce users, not press. Build a pre-launch email list. Define your real competitive alternative. Recruit the first users by hand. Ship a janky-but-working MVP. Pick one acquisition channel. Make one thing remarkable. Set a success metric like Sean Ellis's 40% test. Order matters more than length.

Do I need an audience before I launch?

It helps. You do not need it. An audience or email list makes a launch faster and more predictable, which is why operators like Colin build one first. But founders like Yaser have launched to seven figures from sixteen followers on a single viral post. Both paths work; pick against your runway.

How janky can my launch be?

As janky as it can be while the core loop still works end to end. Garry Tan's line is that janky is not the same as broken: the product must still provide minimum value. If someone uses an embarrassing version, the problem is real. If they walk away because of the jank, it was a nice-to-have.

How do I know if my launch worked?

Set the signal before you launch. Sean Ellis's test asks active users how they would feel if they could no longer use the product. 40% saying very disappointed is a leading indicator of fit. Below that, you do not have fit yet. It doesn't matter how many signups you got.

Should I save everything for one big launch day?

No. Treat launch day as a Tuesday and the first launch as a probe. Garry Tan's advice: launch early and often. Diagnose week by week. Launch again every three months. The market does not remember most launches, so the pressure to make it perfect is a story you are telling yourself.

Bring your actual numbers. Get a cited answer you can defend.

Try Gavel free

Related guides

Launch

Product launch plan

A product launch plan is a timeline. Pre-launch: waitlist and pre-sell. Then the launch windo...

Launch

Product launch strategy

Product launch strategy is the bet before the checklist. Audience-first or product-first. Dis...

Distribution

Customer acquisition framework

There are four ways to get customers. Warm outreach. Cold outreach. Content. Paid ads. Hormoz...

Keep exploring

Experts & frameworks

Every operator Gavel cites, with timestamped sources.

Is ChatGPT good for…

The decision-by-decision take on where generic AI falls short.

Playbooks

Multi-expert guides for pricing, PMF, first customers.