The Strange Problem of the Finished Product
There is a peculiar moment in the life of a startup that doesn't receive nearly as much attention as it deserves. The product works. The website is live. Stripe is connected. The founder has tested the signup process three times, asked a friend to test it twice, and perhaps even posted an announcement on LinkedIn. Then the founder opens the analytics dashboard and waits. Three people have visited the website. One of them was probably the founder.
Nothing is technically wrong. The servers are running. The buttons work. The product may even be genuinely useful. The problem is something much more fundamental: nobody knows it exists.
This is where the mythology of startups collides with reality. We have developed an elaborate vocabulary for building products. We talk about artificial intelligence, product-market fit, agile development, user experience, rapid iteration, and minimum viable products. Entire industries exist to help entrepreneurs build software faster. But once the software has been built, the founder encounters a question that sounds almost embarrassingly simple: where, exactly, are the customers supposed to come from?
That question is distribution. And it may be the most underestimated problem in entrepreneurship.
Why the First 100 Are Different
Imagine two companies selling exactly the same product. The first has ten customers. The second has 100,000. Technically, they are in the same business. Practically, they live in different universes.
The company with 100,000 customers can calculate conversion rates to the second decimal place. It can run advertising experiments, segment email campaigns, measure retention cohorts, build referral systems, and optimize landing pages against thousands of visitors. The company with ten customers has none of those luxuries. If its conversion rate doubles tomorrow, the difference might be one person.
This leads to one of the great misunderstandings of early-stage growth. Founders often study how large companies acquire customers and then attempt to imitate them. They see a successful company investing in SEO, paid acquisition, partnerships, social media, affiliates, conferences, and content, so they construct a miniature version of the same machine. They create six social accounts, start a newsletter, write three blog posts, experiment with ads, and launch an affiliate program. Three months later they have become extremely busy without becoming particularly good at acquiring customers.
The first 100 customers require a different strategy because the objective is different. At this stage, the founder is not optimizing a machine. The founder is trying to discover one.
"Your first 100 customers aren't merely revenue. They are information."
Those customers tell you who actually wants the product, which problem they believe they're buying it to solve, what objections almost stopped them, what words they use to describe the problem, and where other people like them can be found. Customer acquisition at this stage is therefore less like marketing and more like detective work.
Start With the Complaint
Suppose you have built software that uses artificial intelligence to organize customer interviews. There are at least two ways you might begin looking for customers. The first is to announce to the world that you have created an "AI-powered customer intelligence platform." You post about it on LinkedIn, launch on Product Hunt, send emails, and hope that somewhere in the enormous ocean of the internet there are people who find those words interesting.
There is another approach. You search Reddit and discover a product manager complaining that she has forty customer interviews sitting in Google Drive and no reasonable way to organize them. You find another founder asking how people summarize customer calls. Then you discover a discussion in a product-management community where half a dozen people are comparing tools for extracting insights from interviews.
Something important has changed. You are no longer searching for people who might be interested in your solution. You have found people who are already interested in the problem.
This distinction sounds small, but it sits at the heart of distribution. Entrepreneurs naturally begin with their product because that is what they have spent months thinking about. Customers begin somewhere else entirely. They begin with their lives. They have a report due Friday, a spreadsheet that has become unmanageable, a customer who is angry, a boss demanding results, or a tedious process they have repeated for the hundredth time. Your product enters the story somewhere in the middle.
The earliest distribution strategy, then, can begin with a surprisingly simple question: where are people already complaining about the problem I solve?
Find the Watering Holes
Before cities existed, finding animals on a vast landscape presented an obvious problem. You could wander randomly through thousands of acres hoping to encounter them, or you could find the places they had to visit. A watering hole transformed a search problem into a location problem.
Customers have watering holes too.
Software developers congregate on GitHub, Hacker News, Reddit, Discord servers, technical forums, and Stack Overflow. Entrepreneurs gather in founder communities, accelerators, LinkedIn discussions, Slack groups, and niche subreddits. Musicians, accountants, real-estate agents, designers, physicians, teachers, restaurant owners, and nearly every other professional community have developed their own gathering places. Sometimes these places are obvious. Sometimes they are surprisingly obscure.
A founder's job is not to be everywhere. It is to identify the handful of places where the probability of encountering the right person rises dramatically. If you sell software for independent coffee shops, ten thousand random Twitter impressions may be less valuable than a conversation inside a private community containing two hundred coffee-shop owners.
This is why "How do I get more traffic?" is often the wrong question for a young company. Traffic is an abstraction. People are not. A thousand visitors who don't have your problem are mostly noise. Twenty people who are actively trying to solve it can change the trajectory of a company.
The Advantage of Doing Things Badly
There is an irony at the beginning of a startup. The founder desperately wants a scalable customer-acquisition system at precisely the moment when scalability may be least useful.
Automation is seductive because it feels like leverage. Why send ten emails when software can send ten thousand? Why personally onboard a customer when you can build an onboarding flow? Why answer the same question repeatedly when you can create documentation?
Because those inefficiencies contain information.
When you personally email twenty prospects, you notice which subject lines get responses. When you onboard customers yourself, you hear where they become confused. When you answer questions manually, you discover that everyone keeps asking the same question. What appears to be inefficient work is actually research disguised as labor.
Imagine a founder who personally recruits ten customers. During those conversations, six customers independently mention the same problem. Four describe the product using a phrase the founder has never used. Three say they almost didn't purchase because the website made them think the product was designed for larger companies. Those observations might lead to a new homepage, a new positioning statement, and an entirely different target market.
An automated funnel might have recorded seven abandoned signups.
The founder would know what happened, but not why.
The Three-Experiment Rule
The internet presents entrepreneurs with a peculiar form of abundance. There are too many ways to reach people. You can publish articles, make YouTube videos, send cold emails, build a newsletter, post on Reddit, participate in communities, optimize for Google, recruit affiliates, create partnerships, host webinars, appear on podcasts, launch on marketplaces, or build referral loops into the product itself.
This abundance creates the illusion that successful marketing means doing many things simultaneously. Usually, the opposite is more useful at the beginning. Pick three plausible channels and treat them as experiments.
Perhaps the first experiment is direct outreach. Find fifty people who appear to experience the problem and contact them individually. The second might be community participation. Find two places where those people gather and become genuinely useful there. The third might be search-driven content. Identify questions customers are already typing into Google and write the best answers you can.
Then wait for evidence.
The important word is evidence, because early founders are particularly vulnerable to vanity metrics. A post receiving 20,000 impressions feels successful. A hundred likes feel encouraging. A sudden spike in website traffic looks impressive on a chart. But if none of those people become customers, you have learned something important.
Distribution should eventually produce behavior, not merely attention.
Customers Write Better Copy Than Founders
There is another benefit to talking directly with early customers that is easy to overlook. They teach you what your product is called.
Not literally, of course. They teach you the language surrounding it.
A founder who has spent a year building software might describe it as an "AI-powered workflow orchestration platform for distributed organizations." This sounds sophisticated because the founder understands every word. A customer may describe exactly the same problem by saying, "I need a way to stop tasks from falling through the cracks."
Which sentence belongs on the homepage?
Probably the second one.
Companies tend to describe products from the inside out. They describe architecture, capabilities, features, and technology. Customers describe their lives. They talk about frustrations, deadlines, money, wasted time, embarrassment, uncertainty, and things they wish would simply disappear.
This makes early customer conversations a remarkable source of marketing copy. Save the phrases people use in emails. Read reviews of competing products. Study Reddit discussions. Write down how customers explain the problem before you explain your solution. Eventually those phrases become headlines, landing pages, articles, sales scripts, and search keywords.
The paradox is that one of the best ways to learn how to speak to your next hundred customers is to listen carefully to your first ten.
Look for the Channel That Pulls Back
Most distribution experiments fail. This isn't a defect in the process. It is the process.
You might send fifty carefully written emails and receive three polite replies. You might spend two weeks posting on LinkedIn and discover that almost nobody clicks. Then you answer one unusually specific question in a niche community and six people visit your website, three create accounts, and one becomes a customer.
That is interesting.
Do it again.
If it happens again, it becomes very interesting.
This is the moment founders should become disproportionately curious. Instead of asking, "What else should we try?" ask, "Why did this work?"
Perhaps the community contains an unusually high concentration of your ideal customers. Perhaps the question revealed a problem more urgent than the one you thought you were solving. Perhaps the language you used resonated. Perhaps customers trusted the recommendation because it appeared inside a community rather than an advertisement.
Whatever the explanation, you have found something worth investigating.
The purpose of testing multiple channels is not to maintain multiple channels forever. It is to discover where the market is pulling back.
One Customer, Ten Customers, One Hundred Customers
The journey to 100 customers becomes easier to understand when you stop thinking of 100 as a single milestone.
The first customer proves something extraordinary: a stranger values the solution enough to act. Customers two through ten begin revealing whether that person was unusual or part of a pattern. Customers eleven through thirty start showing you which objections repeat, which features matter, and which types of customers stick around. Somewhere between thirty and one hundred, you may begin to see the outlines of a repeatable system.
The questions should evolve with the company. In the beginning, ask who desperately needs this. Then ask why those people are buying. Only after you understand those answers should you become obsessed with finding more people like them.
There is a tendency in technology to think exponentially from the beginning. How do we reach a million users? How do we build something viral? How do we scale this?
Those are wonderful questions when you know what you are scaling.
Before that, they can become distractions.
Distribution Is Not What Happens After the Product
Perhaps the deepest mistake is the assumption that distribution begins when the product is finished. In reality, distribution and product are intertwined.
Consider two applications that perform roughly the same function. One costs $19 per month and is designed for freelance designers. The other costs $50,000 per year and is designed for Fortune 500 companies. The software might share enormous amounts of code, but the businesses surrounding that code are fundamentally different.
The $19 product cannot afford a six-month enterprise sales process. The $50,000 product probably cannot depend entirely on impulse purchases from Google searches. Price influences distribution. Customer influences distribution. Positioning influences distribution. The product itself can influence distribution if collaboration, sharing, referrals, or user-generated content cause new customers to discover it.
This is why distribution cannot simply be bolted onto a company after the engineering is finished. The way customers discover a product should influence the product you choose to build.
The Thirty-Day Test
If you have built something and customers are not arriving, the natural instinct is often to return to the product. Add another feature. Redesign the dashboard. Improve the onboarding flow. Perhaps the product simply isn't good enough yet.
Sometimes that is true.
But there is another possibility: not enough people have encountered it for you to know.
For the next thirty days, try treating distribution with the same seriousness you give product development. Spend the first week talking to potential customers and finding the places where they already discuss the problem. Spend the second sharpening your positioning and offer until a stranger can understand what you sell and why it matters. Spend the third running three focused distribution experiments. Spend the fourth studying what happened and concentrating your effort on the channel that produced the strongest evidence of genuine demand.
You may discover that the product needs to change. You may discover that your target customer is wrong. You may discover that the market describes the problem completely differently than you do. Or you may discover one small, obscure channel that consistently produces customers.
Any of those outcomes is useful.
There is a simple rule I return to throughout Mastering Distribution: don't spend another month building something until you've spent at least a week trying to distribute it. The internet is already filled with beautifully engineered products that work exactly as intended. Their buttons function. Their servers stay online. Their founders continue shipping features.
What they are missing is not another feature.
They are missing a path between the product and the person who needs it.
Go Deeper Into Distribution
Mastering Distribution: How to Get Customers in the AI Era explores the entire journey from finding a market that hurts to positioning, offers, messaging, customer acquisition, content, partnerships, conversion, retention, and building a distribution moat.
If you're building something and wondering where the customers are supposed to come from, that's exactly the problem the book was written to solve.