Tools / Startup launch checklist

Startup launch checklist

A working checklist for the fortnight around a launch: what to have ready before you post, where to submit, and what to do in the week after. It remembers what you have ticked off, in your browser, with no account.

0 / 30 done

Saved in this browser

Before you post

the fortnight before · 0/10

Write the one lineWhat it is, who it is for, in under 15 words, with no adjectives. Everything else — the tweet, the directory listing, the OG image — is a rewrite of this line.
Put the product above the foldA screenshot or a 20-second loop of the thing working. Visitors from a launch arrive cold and leave in seconds.
Publish pricing, even provisional pricing'Contact us' loses the visitor and gets your product left out of AI answers, which need a number to compare.
Set the OG image, title and descriptionEvery share for the next year renders from these three tags.
Ship a real faviconDirectories, browser tabs and link previews all pull it. A default globe reads as unfinished.
Install analytics before launch day, not during it
Sign up as a stranger, on a phone, on cellular dataFind the step that breaks when the session cookie is not already there.
Have somewhere to receive the first complaintAn inbox you actually watch. Launch-day silence is usually a broken contact form.
Publish an llms.txtSo assistants read a summary you wrote instead of guessing from your nav bar.Do it here →
Add Organization and SoftwareApplication structured data

Launch week

the week itself · 0/10

List on bunny.directoryFree, permanent, dofollow on every tier. Do it first: the listing is the URL every other submission can point at while your own site is still changing hourly.Do it here →
Product Hunt, on a Tuesday or Wednesday, at 00:01 PTOne loud day. Have the assets, the first comment and the maker reply written in advance.
Submit to 20–40 directories over two weeksNot in one afternoon. A burst of identical listings on the same day is the pattern spam filters were built for.
Show HN, with the title as a plain descriptionNo emoji, no 'revolutionary'. Answer every comment for the first six hours.
Two subreddits where you already postPosting a launch into a community you have never commented in is the fastest way to a ban.
Post the story, not the announcementWhat you built, what broke, what you learned. The link goes in the first comment.
Email everyone who ever gave you their addressIncluding the ones from six months ago. They opted in for exactly this.
Send 30 personal messagesNot a broadcast. Thirty people you know, each told why you thought of them specifically.
Answer three questions where your product is the honest answerReddit, Stack Overflow, a Slack or Discord you belong to. Answer fully; mention the product last.
Put your directory badges on your own siteIt is the link back that keeps a free listing alive, and it is social proof for the next visitor.

The week after

when everyone else stops · 0/10

Reply to every comment, review and DMThis is the whole compounding mechanism. Launch traffic is a spike; replies are what it converts into.
Get ten of the new signups on a callTwenty minutes each. You will learn more than from the entire launch day's analytics.
Find the step where new users stop, and fix that one
Write what happened, with the real numbersA launch retrospective with actual figures gets read, linked and quoted long after the launch does.
Cut the retrospective into five posts
Update your directory listings with what shipped sinceA listing you keep current keeps ranking. Most people never touch theirs again.
Line up a second launch moment 6–8 weeks outA major feature, an integration, a report. One launch is an event; a cadence is a channel.
Ask your ten happiest users for a review, each on a different site
Ask ChatGPT, Claude and Perplexity about your categorySee whether you are named, and what they get wrong. That answer is now part of your funnel.
Keep showing up daily somewhere that rewards itDirectories with streaks, communities with reputation — the compounding is not on launch day.Do it here →

Launch day is the smallest part of a launch

The spike is real and it is short. What decides whether a launch was worth doing is the fortnight of preparation in front of it and the fortnight of replies behind it — and almost every checklist you will find covers only the middle day.

That is why this one is ordered by time rather than by importance. If you are reading it the night before, the first section tells you what is already too late to do properly, which is more useful than a list that pretends everything is still possible.

The three things that decide the outcome

The one line. Fifteen words, no adjectives, saying what it is and who it is for. Every other asset is a rewrite of it: the tweet, the Show HN title, the directory tagline, the OG description. Founders who skip this write nine different descriptions of their product in one week and wonder why none of them landed.

Somewhere permanent to point at. Launch posts expire. A Product Hunt page falls off the front page in a day; a Reddit thread is unfindable in a month. A directory listing with your description and a link to your site stays indexed and keeps being read — which is also why it is worth doing first, while your own landing page is still changing hourly.

Replies. The founders who get compounding traffic out of a launch are the ones who answered every comment for a week. It is unglamorous and it is the entire mechanism.

About submitting to directories

Twenty to forty is a reasonable target, spread over about two weeks. Two things matter more than the number:

Do not do them all in one afternoon. A burst of identical listings created the same day, all pointing at a domain with no history, is exactly the pattern link spam filters were built to catch.
Check the link before you submit. A directory that marks outbound links nofollow, or redirects them through a tracker, is worth submitting to for the traffic but not for the link. Both are fine — just know which one you are buying. Every tier on this directory is dofollow, including the free one.

Why the ticks are saved in your browser

No account, no email, no row in a database. The list is long enough that you will come back to it across several days, and asking for an address before showing you a checklist would be a lead magnet wearing a checklist's clothes. The trade is that your progress lives in this browser only — clear your site data and it is gone.

Questions

How long before launch should I start?

About two weeks. The first section of the checklist is the part that cannot be done well the night before: the one-line description, the OG tags, a landing page that shows the product, and a signup flow you have tested as a stranger on a phone.

What day should I launch on?

For Product Hunt, Tuesday or Wednesday at 00:01 Pacific gives you a full day on the leaderboard against slightly lighter competition than Monday. For everything else the day matters far less than whether you are available to reply for the following six hours.

How many directories should I submit to?

Twenty to forty, spread across about two weeks rather than done in one sitting. Prioritise directories that publish a real page for your listing, that get crawled, and that give you a dofollow link. A directory that only lists your name in a table is worth very little.

Do I need to pay to be listed anywhere?

No. Paid tiers on most directories, this one included, sell speed and placement rather than the link itself. Pay when you want to be published today or featured on a busy page, not because you think a free listing is worth less as a link.

Is my progress saved if I come back later?

Yes, in this browser. Nothing is sent to us and there is no account. Using a different browser or clearing your site data starts the list over.

Launch on bunny.directory

Free tier, permanent page, dofollow link on every tier. It takes about four minutes.

Submit free →