All articles
Startup BuildingCustomer AcquisitionFounder Marketing

What Building My Own Products Taught Me About Acquiring Customers

Customer acquisition lessons from building my own products: why distribution beats a great product, and how founders earn customers before launch.

On this page

Building the product was maybe half the work. Getting customers meant building a way to reach them before launch. I now budget at least as much effort for making a product findable as I do for building it.

I came to this from the marketing side, having built several small products of my own. The pattern that repeats every single time is that building is the easy half. Getting anyone to care is the craft nobody warns you about.

What the post-mortems keep saying

If distribution were optional, the failure data would look very different. Distribution is simply how people find out your product exists and get to it, and it’s the part founders understand least.

CB Insights’ failure analysis, updated in March 2026 from hundreds of startup post-mortems, still puts product at the scene of the crime: 43% of shut-down companies cited poor product-market fit, and 29% blamed bad timing. Meanwhile 70% listed running out of capital as the final cause, which is usually the symptom, not the disease.

Peter Thiel put it more bluntly in his Stanford course notes back in May 2012: “Most businesses actually get zero distribution channels to work. Poor distribution (not product) is the number one cause of failure.”

I’ve come to read those two findings as one finding. Products rarely fail in the lab. They fail in the quiet, after nobody finds them.

It’s also worth noticing how these numbers get misread. “No market need” doesn’t usually mean the product was useless. It means nobody was out there looking for it, because the builder never joined the conversation the market was already having. That distinction is the whole difference between a product failure and a distribution failure.

1. Nobody buys what they can’t find

Before quality comes discovery. If buyers can’t find you in the places they already look, the quality argument never even gets to start.

Gartner’s buyer research found that B2B buyers spend only about 17% of their buying time meeting with potential suppliers, all of them combined (September 2020). The rest of the journey happens alone: searching, reading reviews, comparing notes internally, and increasingly asking chatbots. G2’s Buyer Behavior Report (July 2026) found 82% of B2B software buyers had sourced software recommendations from an AI chatbot in the previous two years.

For a solo builder, that reframes what marketing is. You’re not interrupting people into attention. You’re standing, in advance, exactly where they already look when the problem shows up.

2. Your product’s own search terms are your best marketing brief

The cheapest market research you will ever get is the exact words people type when they have the problem. Those words should drive your homepage, your docs, your feature names, and your blog.

When I skip this, and I have, more than once, I end up describing the product the way I built it instead of the way people look for it. The pages that actually brought customers to my products were never the clever ones. They were small, specific answers to questions real people were already asking.

This carries straight into AI search. Chatbots recommend what they can quote, so plain claims and concrete detail travel further than slogans. Don’t take that as “churn out more posts”, I’ve written before about how much low-value AI content I had to unlearn producing. And reviews still matter: in G2’s July 2026 data, review sites (38%) actually edged out AI chatbots (37%) as the biggest influence on which vendors made the shortlist. Be quotable, be listed, be specific.

3. The audience compounds, but only if you start early

Start distribution before there’s anything to sell, because almost every marketing channel pays out on a delay. Search rankings take months. Relationships take months. An audience takes. You get the idea.

The mistake I’ve made more than once is building in private for half a year and then “launching,” which in practice means asking strangers to care about something that arrives with zero context. Distribution first isn’t a slogan; it’s a schedule. The work that compounds (content, connections, reputation) has to be old by the time you need it.

Paul Graham named the underlying truth in “Do Things that Don’t Scale” (July 2013): “startups take off because the founders make them take off.” His image is a car engine you have to crank by hand, once it’s running it keeps running, but nothing happens until someone does the cranking. The launch isn’t the start of your marketing. It’s a checkpoint, and by then the cranking should already have calluses.

There’s a pleasant asymmetry here, too: audience work almost never dies with the product. The writing, the reputation, and the relationships carry over to whatever you build next. The code doesn’t. That’s the strongest argument I know for starting early, the asset outlives the thing you built it for.

4. Talking to users is a distribution strategy, not just research

Most founders treat user conversations as research that comes before marketing. Early on, that’s backwards, the conversations are the marketing.

Graham’s essay has the canonical examples. Stripe’s founders did the “Collison installation”: instead of sending a link, they’d say “Right then, give me your laptop” and set the thing up on the spot. Airbnb’s founders went door to door in New York recruiting hosts one by one. Neither move scales. Both worked.

Every one of those conversations does double duty. You learn the exact words buyers use, which feeds your pages and your roadmap, and some fraction of the people you talk to buy, or send someone who does. Thiel’s line fits solo founders perfectly: if you don’t see any salespeople around, you are the salesperson. For a first-time builder with no budget, this is the whole founder marketing playbook. Be useful in the communities where the problem gets discussed, then show up again tomorrow.

The etiquette matters more than people expect. Answer the question completely, even if that means recommending something other than your product, and mention what you’re building once, at the end, like a person rather than a pitch. Communities forgive builders. They don’t forgive marketers wearing builder costumes.

5. Free tools and boring channels beat ads on a small budget

When money is short, don’t rent attention, build assets that keep working. Engineering as marketing means building a small free tool that solves one slice of the problem, and it has beaten every paid experiment I’ve thrown at the same job. A tool demonstrates competence instead of claiming it.

The usual suspects work: calculators, graders, generators, checklists that actually do something. They earn links and trust, and they keep showing up in search long after a paid campaign would have blinked out. Even product-led growth (where a free tier or trial does the selling) leans on this, because the product can’t demo itself to someone who never hears about it.

The rest of my list is boring: pages that answer real search terms, an email list, the directories and communities your buyers actually read. TrustRadius’s 2023 survey of 1,600-plus B2B buyers found demos, reviews, and free trials were still the most-used sources of information. None of this is clever, and that’s the point. Gartner’s June 2025 survey found 61% of B2B buyers prefer a rep-free experience anyway, and 73% actively avoid suppliers that send irrelevant outreach. Nobody is waiting for your cold email.

6. Shipping publicly is marketing

Building in public means sharing the process of building a product as it happens (decisions, progress, setbacks included) and it’s the closest thing a solo builder has to a free sales team. It works because trust forms before the pitch. When people watch you solve the problem week after week, the eventual product page isn’t a cold pitch; it’s a continuation.

It’s slow, it’s uneven, and it only compounds if you keep at it for months. Share specifics, not victory laps, the details are what make following you worth anyone’s time. Done well, the audience you build this way is the only distribution asset that’s truly yours, and nobody can raise your CAC on it.

What I believed then vs. what I believe now

My beliefs about customer acquisition didn’t change through insight. They changed through correction, usually by watching something I was sure about quietly not work. These are the beliefs that cost me the most time:

DimensionWhat I believed when I startedWhat I believe now
What makes a product winA better product eventually wins on its ownThe product that’s easiest to find and easiest to trust wins
When marketing startsAfter launch, once there’s something to sellBefore there’s anything to sell
What a launch isThe moment customers arriveOne checkpoint in a long distribution effort
Where first customers come fromStrangers discovering the siteConversations, communities, and people already listening
What marketing work looks likeClever campaignsUnsexy assets: pages, tools, replies, relationships

Where I’d start today

If I launched a new product tomorrow, here’s the order I’d work in, learned mostly by doing it in the wrong order first.

  1. Collect the exact words people use for the problem (search terms, community threads, real conversations) before writing any code.
  2. Publish the ten most useful answers to those words as pages, even while the product is half-built.
  3. Show up in the two or three communities where the problem gets discussed. Be useful first; be promotional rarely.
  4. Build one small free tool that solves a slice of the problem and quietly points to the product.
  5. Ship publicly on a schedule, sharing decisions and progress, so trust accumulates before you ask for anything.
  6. Watch which single channel actually produces customers and pour everything into it. Thiel’s warning is the backdrop: most businesses get zero channels working, and one is plenty.

FAQ

Why is distribution harder than building?

Because building gives you honest feedback (the code works or it doesn’t) while distribution gives you slow, noisy feedback and punishes impatience. Builders also assume quality is self-evident, and that engineering bias is exactly why distribution gets skipped. It’s also the more uncomfortable half, because it means asking instead of polishing.

How do solo founders get their first customers?

One at a time, by hand. Show up where the problem is discussed, talk to the people who have it, and do the unscalable version, personally onboarding someone beats sending them a link. Pair that with one findable channel built on the words they already search, plus a small free tool, and you have a first-ten-customers playbook that costs time instead of money.

What do founders get wrong about marketing?

They treat it as a later task (something to hire for once the product is “ready”) and they try five channels at once instead of one until it works. They also describe the product in builder’s language instead of the buyer’s words. The deepest mistake is assuming quality is self-evident. It never is.

How early should you start marketing?

Before the product exists. Collect search terms and start publishing answers, join the conversations, and share the build from day one. Channels compound on a delay, so if your first real marketing push happens on launch day, you’re already months behind your own best assets.

Does building an audience replace marketing?

No, an audience is one asset inside a strategy, not the whole strategy. It converts when you have something relevant to sell and a way for outsiders to still find you; a big following with no findability is a stage with no doors. Build it, but don’t wait for it. You can sell to ten people long before you have ten thousand followers.

Is SEO still worth it when buyers ask AI chatbots?

Yes, but the job has widened rather than disappeared. AI answers are built on findable, quotable content, and G2’s 2026 data shows buyers mixing sources, chatbots heavily, but review sites still slightly ahead as shortlist influencers. Write clear, specific, sourceable pages and you serve the crawler, the chatbot, and the skeptical human at once.

Sources

Questions & answers

What is "What Building My Own Products Taught Me About Acquiring Customers" about?

Customer acquisition lessons from building my own products: why distribution beats a great product, and how founders earn customers before launch.

Who wrote this article?

Aditya Mallah is a growth marketer for SaaS, AI tools, and fintech. Full bio: https://adityamallah.com/about

About this article

The content on this blog is for educational and informational purposes. These are my own views and do not constitute professional, financial, legal, or business advice. I can be wrong; do your own research before making decisions. Third-party links are provided for convenience and do not imply endorsement.

Was this article helpful?

Get my growth notes

Notes on growth, AI, and finding customers. Delivered to your inbox.

All articles

b2b community 12 min read

Community Platforms for B2B Founders: Circle vs Slack vs Discord and Alternatives

Circle vs Slack vs Discord compared for B2B founders: where each platform wins, what free actually costs you.

ai coding 13 min read

AI Coding Assistants for Nontechnical Founders: What Can You Build Without a Developer?

What can a nontechnical founder actually build with AI coding tools? Compare eight assistants, with October 2026 pricing and practical advice on logins and payments.

online courses 12 min read

Online Course Platforms for Consultants: Pricing, Payments, and Audience Ownership

Compare course platforms for consultants on pricing, payments, and audience ownership, Teachable, Thinkific, Kajabi, and more.