Here is the hardest truth in this whole book. The best product does not win. The best distributed product wins. You can build something genuinely better than everything out there and watch it die in silence, while a worse product with a working channel prints money. If you take one idea from this chapter, take that one.
Peter Thiel put it plainly in Zero to One: poor sales rather than bad product is the most common cause of failure. Founders, especially technical ones, refuse to believe this. We think if we just build something great, people will find it. They will not. No product distributes itself. Growth comes from you pushing it into the world, every single day.
So split your time. However much you spend building, spend at least as much getting it in front of people. A founder who spends ninety percent of their time building and ten percent on distribution has it exactly backwards.
The channels, with the honest version
There are many ways to reach people, and each one is good for some businesses and useless for others. Here is the straight version of the main ones.
Content and SEO. You write things people search for, and over time you get free, compounding traffic. The upside is that it builds an asset that keeps working while you sleep. The downside is that it is slow, often six to eighteen months before it pays off, Google changes the rules, and AI answers are now eating clicks for simple questions. Great when your customers are actively searching for a solution.
Building in public and an audience. You share what you are building on X or LinkedIn, and you build trust and a warm list of people who will buy when you launch. Cheap and powerful. The downsides are that it is slow to compound, it depends on you personally, and it does not transfer if you ever sell the business. X skews toward developers and makers, LinkedIn toward business and B2B.
Product Hunt and launch days. A launch gets you a spike of attention, some early users, social proof, and a backlink. The mistake is treating it as a strategy. It is a one-day bump, the audience is mostly other makers, and ranking well takes weeks of preparation and a network you built beforehand. Launch, take the spike, but do not build your business on it.
Communities. Reddit, Discord, niche forums, the places your customers already gather. These are high-intent and free, and they are the best place to find your first users and real feedback. The catch is that self-promotion gets you banned fast. You have to show up, give value, and earn the right to mention what you built.
Cold outreach. Email and DMs to a clear target customer. Predictable, controllable, and it works from zero audience, which is why it is so good for B2B. It is also labor-intensive, and you have to manage deliverability so you do not end up in spam. The previous chapter on email is your playbook here.
Partnerships and integrations. You borrow someone else’s distribution by plugging into their platform or marketplace, or by working with a company that already has your customers. Powerful, but slow to land and risky to depend on, and it usually works better once you already have some traction to offer in return.
Referrals and virality. When it works, it is the cheapest growth there is, because your users bring you more users. But it only works if the product is naturally shareable or gets better with more people on it. Bolting a referral program onto a product that is not inherently viral almost never creates real virality.
Newsletters. You build an audience you actually own, with no algorithm in the middle, high engagement, and the ability to sell to them directly. The cost is that lists grow slowly and you are on a permanent content treadmill.
Paid ads. Fast, scalable, and measurable, which makes them tempting. But they only make sense once you know that a customer is worth clearly more than it costs to acquire one, and you can do that math with confidence. For most founders before they have real traction, paid ads just burn cash and hide the fact that you do not yet have a working free channel. Earn the right to pay for growth.
Pick one channel and go deep
The single biggest distribution mistake is spreading yourself across all of these at once. You end up mediocre everywhere and great nowhere. The fix comes from a book called Traction by Gabriel Weinberg, the founder of DuckDuckGo, and Justin Mares.
Their core idea is the Bullseye framework, and it is simple. Picture three rings. The outer ring is every channel that could possibly work, all nineteen of them. The middle ring is the three or four that seem most promising for you right now, which you test cheaply and quickly. The inner ring, the bullseye, is the single channel that is actually working, which you then pour everything into.
The insight underneath it is that at any given stage, one channel usually drives the bulk of your growth. Your job is not to do all of them well. It is to run a few cheap tests, find the one that works for your specific business, and go deep on it until it stops working. Then you find the next one.
So do not ask “which channels should I use.” Ask “which one channel can I make work right now,” and give it everything.
Your first thousand users
Before any channel scales, you get your earliest users the hard way, by hand. Paul Graham’s essay “Do Things That Don’t Scale” is the bible here, and the examples in it are famous for a reason. The founders of Stripe did not wait for signups. When someone agreed to try it, they would take the person’s laptop and set it up on the spot, which became known as the Collison installation. Airbnb’s founders went door to door in New York, recruiting hosts and taking better photos of their apartments themselves. Wufoo sent every new user a handwritten thank-you note.
This feels wrong to most founders because it does not scale, and that is exactly the point. In the beginning you are not building a growth machine, you are recruiting humans one at a time. Here is what that looks like in practice.
Recruit your first hundred users personally. Message them, email them, and go into the exact communities where they already hang out. Post where your buyer already is, the specific subreddit or Discord or forum, and answer questions without pitching until you have earned trust. Launch to a warm list that you built before you launched, through building in public or a waitlist, so launch day converts instead of echoing into an empty room. Do unscalable onboarding: jump on calls, record personal walkthrough videos, give people direct access to you. And build a free tool that ranks and feeds your main product, what Weinberg calls engineering as marketing.
It is slow and it is manual and it is the only way it has ever worked.
What it looks like when it works
The clearest proof is the founders who won on distribution, not just product. A quick honesty note first: almost all the revenue numbers these people share are self-reported, posted to their own audiences. They are directionally believable because these founders built their reputations on being transparent, but treat them as roughly right, not audited.
Pieter Levels built Nomad List and Remote OK as one person and reports a few million dollars a year. He did it by building in public to a large following and by programmatic SEO, generating thousands of genuinely data-rich pages, one per city, that pull in long-tail search traffic. Marc Lou won Product Hunt’s maker of the year in 2024 by combining a fast string of launches, an audience on X that he grew from tens of thousands to roughly double that, and SEO. Justin Welsh built a one-person business reportedly past four million dollars a year almost entirely on a LinkedIn audience and a newsletter. Lenny Rachitsky turned one newsletter into the biggest paid business newsletter on Substack, with reportedly over two million dollars a year, by owning his audience directly. And products like Dub have used a sharp Product Hunt launch to get an initial burst of attention and credibility.
Different people, different channels. What they share is that each one picked a channel, went deep, and treated distribution as the main job, not an afterthought.
The mistakes
- Spreading across every channel at once. You get weak signal everywhere and master nothing. Test a few, then commit to the one that works.
- “If we build it, they will come.” They will not. This is the exact belief that kills good products. Distribution is something you do, not something that happens to you.
- Treating a launch as a strategy. Product Hunt or a Show HN post is a spike, not an engine. After the bump fades, you still need a channel that brings people in repeatably.
- Chasing vanity metrics. Followers, upvotes, signups, and impressions feel like progress. They do not pay. Track the things that mean someone got value and stuck around: activation, retention, and revenue. Distribution only counts when it turns into paying users you keep.
Build half. Distribute half. Pick one channel and go deeper than feels reasonable. That is how products that deserve to win actually win.