Getting to a hundred customers is rarely an idea problem. It is an order-of-operations problem: teams build for months, launch into a void, and only then start looking for the people who were supposed to care. This playbook runs the sequence the other way round.
Most builders I know skip validation, launch into a void, and then panic-post on Reddit asking why nobody cares. They had no framework. I had no framework either, until I sat down and reverse-engineered the playbook from people who did get to a hundred — Pieter Levels, Marc Lou, Sahil Lavingia, Tyler Tringas. The patterns are surprisingly consistent. The first hundred customers are not a marketing problem. They are a sequencing problem.
This piece is the playbook I wish I had two years ago. It is a 12-week sprint, broken into six two-week blocks, written for a vibecoder shipping their first product and for the bootstrapped SMB founder who finally decided that "we will figure marketing out later" was a mistake.
The short version — spend the first month proving the problem is real, the second month building distribution before you need it, and the third month converting attention into paying users. Most people invert that order and wonder why nothing compounds.
Week 0 — validation prerequisite (10 problem interviews)
Before week one starts, you owe yourself a hard gate — ten conversations with people who fit your target customer. Not surveys, not Twitter polls, not "I asked my cofounder." Ten thirty-minute calls with real humans who have the problem you think you are solving.
The vibecoder mistake is to skip this because the build feels productive and the calls feel awkward. The Marta mistake is to assume the existing customer base is enough signal — usually it is not, because the people who already pay you have selected into a specific use case.
Use the Mom Test framework — ask about past behavior, not future intent. "Walk me through the last time you ran into this problem" beats "would you pay for a tool that solves this?" The first reveals real workflows. The second invites flattery.
What you are looking for after ten interviews:
- Three people who described the same workaround using their own words
- At least one person who already pays for something adjacent (signal of budget)
- A clear "moment of pain" you can put a number on — hours wasted, dollars lost, deal closed late
If you do not have those three signals after ten calls, you do not have product-market fit potential — you have a hobby idea. Do another ten or pivot.
Weeks 1-2 — landing page and waitlist
Build the landing page before you build the product. I know that sounds backwards. Do it anyway.
Your landing has one job — capture emails from people who self-identified as having the problem you described in your interviews. Not "subscribe to our newsletter." Specifically — "I want early access because I have this exact problem." The headline should mirror the language your interviewees used, verbatim.
What goes on the page:
- Headline in the customer's own words
- One paragraph of context — what the product does, in plain language, no jargon
- A 30-second Loom video of the product working — even if "the product" is a Figma mockup
- A waitlist form with two fields — email and one qualifying question
- Social proof if you have it — interviewee logos who agreed to be named, advisor quotes
The qualifying question matters. "How are you solving this today?" filters tire-kickers and gives you copy ammunition for every email you will send for the next six months.
By end of week 2 you should have 50 emails on the list. If you do not, the landing page is wrong. Iterate the headline, not the form.
Weeks 3-4 — niche communities and build-in-public start
Now you go where your customers already congregate. Not Twitter generally. Specific subreddits, specific Slack groups, specific Discord servers, specific industry forums.
The mistake — posting "check out my product" links and expecting upvotes. Communities punish that fast.
The pattern that works — spend three weeks giving before you ask. Answer questions in your area of expertise. Share what you learned from your ten interviews (anonymized). Write a "I noticed everyone in this space has problem X — here is the data" post with zero product mention. By the time you mention your product in week 4, you are a known contributor, not a drive-by promoter.
Simultaneously, start building in public. Pick one channel — Twitter or LinkedIn — and post three times a week about what you are building, what you learned, and what is broken. Do not write threads pretending to be a thought leader. Write short, specific posts.
Two metrics matter — waitlist growth (target 50 to 200 by end of week 4) and reply rate on your build-in-public posts (target >2%).
Weeks 5-6 — content engine (3 posts per week)
This is where most vibecoders die. They get bored of posting because nothing seems to compound. The first six weeks of content always feel like screaming into a closet. It is normal. The compounding starts at week 8-10.
Set up a content engine that produces three posts per week minimum:
- Monday — a tactical post. Something concrete a reader can implement today.
- Wednesday — a behind-the-scenes post. What you are building, what failed, what you are questioning.
- Friday — a synthesis post. A pattern noticed across customers, a piece of data, a hot take backed by numbers.
The format does not matter as much as the cadence. Pick one platform where your customers live. For B2B SaaS — usually LinkedIn or Twitter. For consumer — Reddit, Instagram, TikTok. Do not try to be everywhere. You will be mediocre everywhere.
By end of week 6 you should have shipped 18 posts. Some will flop. Two or three will hit. Those are signal — write more like them. Marc Lou shipped daily on Twitter for two years before anything compounded. The engine has to start somewhere.
Weeks 7-8 — pre-launch traffic surge
You are eight weeks from week 0. You have a landing page, a waitlist of 300-500 people, a built-in-public presence, and a content engine. Now bigger levers.
Three plays in this window:
- Guest content swaps — find five creators in your niche, roughly your size. Trade newsletter posts and interviews. Each of you has 500-2000 subscribers. A swap puts you in front of 2,500-10,000 new people at zero cost.
- One Product Hunt-adjacent launch — pick BetaList, Pioneer, Tiny Launch, or a niche aggregator. Not Product Hunt yet. Test landing page conversion on cold audience and tune copy.
- Targeted outreach — identify 100 ideal customers by name. Send 20 personalized emails per week. "I am working on X, you wrote/said Y about this problem, I would love your feedback." Expect 15-25% reply rate if the personalization is real.
End of week 8 target — 500-800 waitlist, 20-30 active beta users giving feedback, one piece of content that did 5x your average reach.
Weeks 9-10 — launch and early customers
Now you launch publicly. Product Hunt, Hacker News, a tweet thread, a LinkedIn post, an email to the entire waitlist. All in the same week, coordinated.
The mistake — treating launch day as a one-shot. It is not. Launch is a week of activity, not a single Tuesday morning post. Stagger your channels — Product Hunt Tuesday, Hacker News Wednesday afternoon (lower competition), Twitter thread Thursday, LinkedIn long-form Friday, email blast to waitlist on Monday before any of it.
What you measure during launch week:
- Signups — not just visits, signups gated behind email
- Signup-to-paid conversion — tells you if your pricing/positioning is off
- Top-of-funnel cost — if you spent any money, what did each signup cost
- Qualitative feedback from the first 50 users — read every word, patterns emerge fast
By end of week 10 you want 20-40 paying customers. If you have 5, your pricing or positioning is wrong. If you have 100, you got lucky — re-read the qualitative feedback because the next 100 will be harder.
Weeks 11-12 — first retention and referral activation
Most playbooks stop at launch. This is where they go wrong. Acquisition without retention is a leaky bucket — you can keep pouring water in, but the bucket never fills.
Retention check. Of the customers who signed up in week 9-10, how many are still active in week 11? Define "active" by your product's core action — a login is not active. If retention is below 50% at two weeks, stop acquiring and fix the activation flow.
Referral activation. Your first 40 paying customers, if happy, will refer the next 40 if you make it easy. Most products miss this because they bolt on referrals six months too late, after early enthusiasm has cooled. Activate referrals in week 11, not month 6.
The mechanics — a one-click share link per customer, a reward worth offering on both sides (15-20% commission or one month free is a tested floor), and an email triggered after the customer hits their first "aha moment" inviting them to share. A simple two-sided reward, one-click share, triggered at the right moment, will convert 10-15% of happy customers into referrers.
Common pitfalls
I have watched dozens of vibecoders run versions of this playbook and screw up the same four things.
- 01No product-market fit, but launching anyway — if your week 0 interviews did not produce three "I have this exact problem" signals, do not proceed.
- 02Premature launch — vibecoders launch in week 3 because they got impatient. Result: 200 visits, two signups, zero paid, demoralization. The launch is the reward for eight weeks of work.
- 03No referral mechanic built in — your first 100 customers are the cheapest source of your next 100. If you do not activate them, you are restarting from zero every cycle.
- 04Confusing activity with progress — three posts a week that produced two replies, one DM, and zero waitlist signups is not progress. Audit metrics weekly.
The first 100 customers are not a marketing problem. They are a sequencing problem.
Validate in week 0, build distribution before you need it in weeks 3-6, launch in weeks 9-10 with everything you have built ready to fire, and activate referrals in week 11 so the next 100 are cheaper than the first.
Most playbooks fail because they invert this order — they build the product first, then panic about distribution. By the time the panic hits, the founder is six months in, out of runway, and convinced "marketing does not work for us." It works. The order matters.
Ship one thing this week. Talk to one customer this week. Post one thing this week. Compound it.
Sources
- Pieter Levels — Nomad List growth essays
- Marc Lou — ShipFast launch breakdowns
- Sahil Lavingia — How Gumroad got its first customers
- Tyler Tringas — Storemapper bootstrap journey
- Rob Fitzpatrick — The Mom Test (validation framework)
- Lenny's Newsletter — first 100 customer breakdowns
- IndieHackers — community case studies