Ben Silbermann shipped Pinterest the way almost every solo builder ships: he emailed all his friends and hoped. His own account of how that went, from Lenny Rachitsky's roundup of how the biggest consumer apps found their first users, is blunt: "no one really got it."
That's the default launch. You build the thing, you post it, you tell the people who already like you, and you wait for a curve that never bends.
If you built your product alone, the default is even more tempting, because you're tired, you've been heads down for weeks, and the part of launching that involves strangers feels like a different job. It is a different job. Here's what the public record says about doing it well, and where I think it's heading now that building has stopped being the hard part.
What the record actually says
Two sources are worth reading in full before you launch anything.
The first is Paul Graham's Do Things that Don't Scale, from July 2013. His core observation is one sentence: "The most common unscalable thing founders have to do at the start is to recruit users manually." He names the fantasy it replaces too, the one where you build a better mousetrap and people beat a path to your door. And he gives the examples everyone quotes: the Stripe founders taking a prospect's laptop and setting them up on the spot, Wufoo mailing each new user a handwritten thank you note, and Airbnb, which he describes as so fragile early on that "about 30 days of going out and engaging in person with users made the difference between success and failure."
The second is Lenny's How the biggest consumer apps got their first 1,000 users, from May 2020. He went through first-hand accounts from founders of Tinder, Uber, Slack, Airbnb, Pinterest, Etsy, DoorDash, Spotify and others, and sorted what they did into seven strategies: go where your users are offline, go where they are online, invite your friends, create FOMO, use influencers, get press, and build a community before launch.
The finding that matters most is the count. In his words, most of those companies found their early users from a single strategy, a few used a handful, and "No one found success from more than three."
Put the two together and you get something you can act on today. Getting the first users is manual work, and it's focused manual work. Not everything at once. One or two channels, done by hand, until they catch.
5 moves for your first 100 users
1. Pick one channel and write down why
Before you post anywhere, finish this sentence: "100 of the people I built this for already spend time in ____."
If you can't fill the blank with a specific place, you don't have a launch problem yet. You have a positioning problem, and posting to six platforms will hide it from you for another month.
Graham's essay records how Silbermann eventually found his: a lot of early Pinterest users cared about design, so he went to a design bloggers conference to recruit. One room, one kind of person, one reason for them to care.
2. Recruit by name, not by post
A launch post asks strangers to come to you. Manual recruiting goes to them.
The practical version for a solo builder is a list. Write down 100 real names or handles of people who have the problem, in the channel you picked. Then contact them one at a time, with a message about their problem, not your product. A reply rate of one in five still gets you twenty conversations, which is more signal than most launch days produce.
3. Do the setup for them
The Collison installation is the most copied move in the essay because it works on two levels. The obvious one is conversion: nobody bounces off a signup flow that someone else is filling in for them.
The less obvious one is what you learn. Every time you sit with a user and do the first run yourself, you watch exactly where a normal person would have given up. That's your onboarding backlog, written by reality instead of by you. For developer tools this is a screen share and a first install. It doesn't scale, and that's the point.
4. Start with a fire small enough to catch
Graham points at Facebook starting inside a single university for a reason. A product that's for everyone is for nobody on day one, because nobody feels like it was built for them.
Narrow the first audience until your 100 users could plausibly know each other. Framework users in one ecosystem. Freelancers who bill a particular way. Founders launching in the same month. Small, dense groups talk, and talk is the only growth loop you can afford when you're one person.
5. Help first, then ask
This is the move the older playbooks underrate, and the one I think matters most right now.
The people most reachable for a solo founder are other solo founders. They're launching too, this week, and they need exactly what you need: a real review, a testimonial, honest feedback on a landing page, a share to an audience that isn't their own. If you do that for them first, the ask that comes later isn't a cold ask. It's a return.
Where this is heading
Here's the part that changes the math.
Building a working product used to take long enough that the number of launches in any week was naturally limited. It isn't anymore. A solo developer with good agents can ship something real in days, and so can everyone else, so the number of products competing for a first look keeps climbing while the hours people have to look at them stay the same.
When generation gets cheap, the scarce input moves somewhere else. For launches, it moves to the unscalable part: human attention, given on purpose, by someone who actually tried the thing. Graham's advice from 2013 isn't getting old. It's getting more valuable, because it's the one part of shipping an agent can't do for you.
My bet is that distribution for small builders shifts toward reciprocity. Not audiences you rent, but rooms of founders who trade real help with each other, where the help is verified and the favor gets returned. The builders who figure that out early get their first 100 users from people who understand exactly what it costs to launch alone.
Where I'd start
That's why I built Favors.dev. It's a marketing co-op for founders who ship, built on one idea: other founders are your new growth channel.
It works in three steps. You do favors for other founders, like reviews, testimonials, shares and feedback. You earn verified points for the work you actually complete. Then you spend those points on a rally for your own launch, so the support shows up when you need it. Points are earned, never bought, which keeps the help honest.
To be clear about the bridge: it won't replace moves one through four. You still have to pick your channel and talk to your users by hand. What it does is make move five something you can do on purpose instead of by luck. It's free to join, and you start with 250 points, so the next step is to go help one founder whose launch is live today.
If you want a room full of people building the same way, the Vibe Coding is Life group has 335,000+ builders in it: join us on Facebook.
Top comments (0)