How I got my first customers, what actually pulled the credit card out of their pocket, and every unscalable hustle in between — with the real story behind each decision.
No paid ads. No growth team. One founder, one wedge.
I've read a lot of “how I made $X” posts, and most of them are confetti. They throw the number in your face, sprinkle a few screenshots, and quietly skip the part where the actual work happened, which is the only part you can learn anything from. I always close those tabs feeling like I watched the highlight reel of a game I needed the full footage of.
So I wrote the full footage.
Kortex is a Chrome extension that adds the features Google forgot to put in NotebookLM: exports, imports, search, folders, bulk actions, and a podcast feed for Audio Overviews. I built it alone. No paid ads, no growth team, no co-founder to split the panic with. Just one person, a clear wedge, and a distribution engine I laid down one boring brick at a time, usually at hours I'm not proud of.
The question I get more than any other, in DMs, in Reddit replies, over coffee, is some version of: “The product's fine, but how do I actually get customers?” That question is the whole reason this exists. So the book leans hard into it: how I found my first users, how I earned their trust as a no-name solo dev, and what actually happens in someone's head in the three seconds before they decide to pay. Every chapter gives you the tactic and then the real story behind it: how I stumbled into it, what I almost did instead, and what happened when I shipped it. The tactic is the cheap part. The reasoning is the part worth stealing.
The one-line version
Find a tool people already love, build the thing it's missing, and talk to the humans who need it one by one until you understand them better than they understand themselves. Then build every price, page, and pitch around how people actually decide to spend money.
Receipts from inside
The channels behind chapters 07 & 09 — proof, not theory
Give-first on Reddit
Creators on YouTube
Hundreds of thousands of views, earned with answers and honest demos — never a paid ad. The exact playbooks are in chapters 07 and 09.
What's inside
Three parts. Fifteen moves.
Part I is the foundation, the boring-but-load-bearing decisions most founders rush past. Part II is the part you probably skipped ahead to look for: how to actually get customers and traction — the longest section on purpose. Part III is everything that turns a visitor who likes you into a customer who pays you.
Part I— The Foundation
01
The wedge: pick a host, not a market
02
How your customer actually thinks
03
Build a product that does the marketing for you
04
The ‘aha’ moment: your real conversion event
Part II— Getting Customers
05
Your first 100 users: do things that don't scale
06
Go where your people already gather
07
The give-first engine: Reddit, content & SEO
08
The launch: treat it as a campaign, not a day
09
The creator engine (where the real money came from)
10
Three quieter channels that punched above their weight
Part III— Turning Visitors Into Revenue
11
What actually brings the credit card out
12
Pricing psychology: design the decision, not the number
13
The trust stack: why people believe a one-person company
14
Keep them: retention is the real growth engine
15
The whole playbook, condensed
Part One
I
The Foundation
No growth tactic in Part II works if you get this part wrong. Before any of it, you need the right product standing in the right place, and you need to understand the one human you're building for better than they understand themselves.
01
The wedge
Pick a host, not a market.
For about a week I had two tabs open that were quietly arguing with each other.
One was a half-built “second brain” app I'd been tinkering with at night, the kind of thing every developer eventually tries to build. The other was NotebookLM, which I was actually using almost every day for my own research. And every single day NotebookLM made me want to throw my laptop. I'd spend an hour building this beautiful notebook full of sources and insights, and then I'd have no way to get any of it out. The chat was a roach motel: ideas checked in, they never checked out.
One night I closed the second-brain tab for good. Not because the idea was bad. Because I finally saw what the two tabs were telling me. The app I was building meant paying Google's ad prices, forever, to convince strangers they wanted a thing they'd never heard of. The app I was frustrated by already had millions of people inside it, and a pile of them were as annoyed as I was about the exact same missing feature. One of those is a war. The other is a doorway someone left open.
So I built a one-click export and shipped it as a Chrome extension that lived inside NotebookLM. Within a few days the reviews started, and one of them has stuck with me: “no NotebookLM user should be without it.” That's not the review you get when you've invented demand. That's the review you get when the pain was already there, screaming, and you just walked over and turned it off. I didn't create the market. I stood in front of one Google was already paying to fill.
★★★★★
“Love! love! love! — and I'm only using the free version. Huge time saver — no NotebookLM user should be without it! I'm definitely considering the Pro version once I get my business beyond the early start-up stage.”
Jenni Duncombe · Feb 19, 2026 · Chrome Web Store review
That decision — where to stand — did more for the $200K than any clever tactic later in this book. I think of it as the remora-on-a-shark move: attach to something bigger that's already swimming, and let it pull you.
Manufacture demandA war.
Create awareness — you pay for every bit of it
Convert
Borrow demandA doorway.
Host already paid
Convert & build the product
Where your energy goes — borrow demand and almost all of it lands on the part that makes money
Why it works
Borrowed demand. Every time Google ships a NotebookLM feature or a creator posts a video, a fresh wave of “how do I do more with NotebookLM” searches rolls in. You didn't pay for a cent of that traffic. You're just standing in the doorway it pours through.
Pre-qualified pain. Everyone inside NotebookLM has already decided AI research tools are worth their time. You're not dragging a skeptic from zero. You're upgrading someone who already loves the category, which is a far shorter and warmer conversation.
A built-in surface. Your product literally appears inside the thing they already opened, so distribution and product stop being two separate jobs. That's the rarest thing a solo founder can get for free.
A moat made of annoyance. Big platforms are slow to ship “nice-to-have” features. The gap between a magical demo and something usable every day is your territory, and it stays open far longer than you'd expect.
The risk, said out loud
I won't pretend this is free of danger. The host could build your feature next quarter or change the rules overnight, and you'd be standing in a doorway that just got bricked up. I lost some sleep over it early on. My answer is the whole of Chapter 12: own an email list and a brand, so that even if NotebookLM shifts under me, the relationship with the customer is still mine. You rent the soil; you own the seeds.
The test for your own wedge
Run a candidate idea through these four questions
Question to ask
If yes…
If no…
Is there a popular product whose fans complain in public?
You have a host. Build the missing layer.
You'll have to manufacture demand. Expensive and slow.
Can you ship the #1 complaint as a v1 in weeks, not quarters?
Ship it. Speed is the only moat early on.
Narrow the scope until you can.
Will the host's own growth pull you along for free?
You found a tailwind. Ride it.
You're paddling upstream forever.
Can you reach the host's users without the host's permission?
You have a distribution path. Go.
Reconsider. You're a hostage, not a partner.
Unlock the four questions in the full playbook
The takeaway
The biggest lever in this whole story wasn't a growth tactic. It was choosing to attach to a product people already loved instead of launching another AI note app into a bloodbath. Borrowed demand, pre-qualified users, and a surface inside the host beat building an island, every time you can pull it off.
02
Know the one human
How your customer actually thinks.
For a few weeks my landing page was aimed at “anyone who uses AI,” and it converted like wet cardboard. I'd stare at the analytics trying to figure out what was broken. The copy was clean. The product worked. People just bounced.
The thing that fixed it wasn't a better headline. It was giving up on “anyone” entirely and writing to one specific person instead. I called her Riya — a grad student who'd fallen hard for NotebookLM, had something like forty notebooks going, and had just smacked into the wall where the magic stops being usable…
The day I rewrote the whole page as if I were talking only to Riya, in her words, about her exact forty-junk-notebooks problem, conversion jumped. That was the lesson that reorganized everything: a message built for everyone lands on no one. So the real first job is answering one question with uncomfortable specificity—
Chapter 2 continues in the full playbook
+9
The rest of the playbook
Nine more chapters, behind the cover.
You just read the opening chapter in full — and chapter two is already pulling you in. Here's everything waiting in the complete edition: the exact moves that turn loved-but-free users into paying customers.
Locked
Chapter 11
How to bring the credit card out
Loved-but-free becomes paid
Thousands of happy free users and almost no revenue taught me the six specific triggers that actually move someone from "great, free" to a paid plan, starting with the one most founders get backwards…
Locked
Chapter 12
Design the pricing decision, not the number
Anchor · decoy · charm · loss-aversion
I stopped picking a price and started engineering the decision around it — the anchor that resets expectations, the decoy that makes the middle feel safe, the charm digit, and the loss-aversion line that…
Locked
Chapter 05
Get your first 100 users by hand
Conversations are R&D and marketing at once
The unscalable part is the point. I recruited the first hundred one message at a time, and every conversation quietly doubled as product research, copy testing, and…
Locked
Chapter 06
Go where your people already gather
Relevance crushes reach
Reach is a vanity number. What actually moved installs was showing up in the three small rooms where the exact people who needed this already spent their…
Locked
Chapter 07
The give-first engine
Give value 10× before you promote once
Before I ever linked the product I spent weeks just being useful in public, and the contribute-to-promote ratio that made it work without getting me flagged was…
Locked
Chapter 09
The creator engine
Trust transfer that can’t lose
One niche creator outperformed every ad I ever considered, and the affiliate structure I used meant the marketing spend literally could not lose money because…
Locked
Chapter 10
Three quieter channels that compound
Store ASO · cold email · build-in-public
Three boring channels stacked into most of my volume: treating the store listing like a landing page, a cold-email motion that felt personal at scale, and…
Locked
Chapter 14
Retention & the list you own
Ship weekly · own the channel
Getting the sale is the start line. Shipping something visible every single week and owning an email list is what turned one-time buyers into people who…
Unlock the full Growth Playbook
All 15 chapters — the complete system that took Kortex from 0 to $200K, with every story, table, and tactic.
