Where to launch to get your first users: it's not an event, it's showing up before and after

I used to think a launch was a day. Pick the date, hit publish on Show HN and Product Hunt, refresh, see what happens. I have now watched enough launches, including a few of my own that went nowhere, to believe the day is the least important part. The launch is everything around the day. The day is just when you ring the bell.

The mistake nobody warns you about

The failed launch pattern is always the same. A founder with no audience shows up on launch day to a community they have never participated in, drops a link, and expects the room to care. It does not. Why would it. You are a stranger who appeared to extract attention. These communities have long memories and good instincts for exactly that move.

So the first real rule: you cannot launch into a community you are not already a member of. On Reddit, in the subreddits where your users actually live, the approach that works is being a known regular, not a one-day guest. That means weeks, sometimes longer, of just being there. Answering things. Being useful with no link attached. By the time you post your thing, you are not interrupting, you are a member sharing what you have been working on, which is a completely different social act and gets a completely different reception.

Show HN, specifically

Show HN rewards a very particular thing: something real that a technical person can poke at, described plainly, with you present in the comments. No marketing voice, it gets sniffed out instantly and punished. The title says what it is in plain words. The first comment is you explaining honestly why you built it, what is rough, what you are unsure about. Then you sit in the thread for hours and answer every single comment like a human, including the harsh ones, especially the harsh ones. The engagement is the launch. A Show HN where the maker is genuinely present beats a slicker one where they are not. And you do not get a do-over, so it should be pointed at something actually buildable-on, not a waitlist page.

Product Hunt, the honest version

Product Hunt in 2026 is less about raw upvotes than it used to be. The algorithm now leans on engagement signals, comments, maker replies, time people spend on the page, and especially whether you bring new people to the platform rather than rallying the same crowd. So the gamed launch matters less and the genuine one matters more, which is good news if you are a nobody with a real product.

The timing mechanics still hold: it resets at midnight pacific, and mid-week, Tuesday through Thursday, is when the room is fullest. But the thing that actually determines the outcome happens in the four to six weeks before, building a small group of people who already know it is coming and will show up in the first hour to comment, not just upvote. You do not need hundreds. You need enough genuine early engagement that the page looks alive when the algorithm checks. A narrative-led launch with a small audience works by letting the story and the maker comments do the persuading, then seeding it into the niche corners where that story lands.

The part I believe most

Contribution before extraction. That is the whole thing, across every channel. Every place worth launching in can tell the difference between someone who has been giving and someone who has only ever shown up to take, and they treat you accordingly. So the launch strategy that works is mostly not a launch strategy. It is being a real participant in three or four specific communities for months, building actual relationships, being known as useful, and then, when you have something, telling the people who already trust you. The "launch" is just the moment that accumulated goodwill becomes visible.

Directories, briefly: submit to the relevant ones, the niche directories where your specific kind of buyer browses on purpose. They are a slow trickle, not a spike, and the trickle is fine, it is real intent. But do not mistake submitting to fifty directories for distribution. That is busywork that feels like progress. One community where you are genuinely known is worth more than fifty listings nobody scrolls to.

What I would tell myself a year ago

Pick your three places now, before you have anything to launch. Start showing up this week. Be useful with no agenda for a month. By the time you are ready to ring the bell, you will have a small room of people who actually want to hear it, and that room, small as it is, converts and refers in a way no cold launch ever will. The launch is not the day. The launch is the relationships you built before anyone needed to care, made visible all at once. Start the before now. There is no shortcut to it, and that is genuinely the good news, because most people will not do it.