How I Made My First 300k Using AI With Just My Phone
April 28, 2026
I Found A Business Problem At 11pm.
By Morning, I Had A Product.
Now It Pays Me Every Week.
How I used AI to vibe-code solutions to real problems, charge $3/month, and wake up richer than when I went to sleep.
Let me tell you exactly how this happened. No “this is not the typical I discovered a secret” nonsense. It’s the real, embarrassingly simple thing I started doing that changed everything.
I was not a developer. I still cannot explain recursion without Googling it. I do not have a computer science degree. I have a phone, a decent internet connection, and I got genuinely tired of watching my bank alert stay the same number every Monday morning.
Something had to change. And when it did, it changed fast.
The Moment That Started Everything
It was a Tuesday evening, around 11pm. I was doing what most of us do at that hour: scrolling aimlessly, half-asleep, phone on my chest.
I stumbled onto a small business owner’s post on a Facebook group. She was complaining that she spends two hours every week manually compiling her WhatsApp orders into a spreadsheet. Two hours. Every week. Just copying and pasting customer names and order details into Excel.
She was frustrated. She said she had asked three developers for help. One quoted her ₦350,000. The other never responded. The third built something that crashed after one week.
I read that and something clicked. This was not a technology problem. This was a money-on-the-table problem. And she was not the only one with it. There were dozens of people in that same thread nodding and saying “same issue here.”
What Is Vibe Coding And Why It Changed My Life
Vibe coding is the practice of building software using AI tools like ChatGPT, Claude, or Cursor as your actual coding engine. You describe what you want in plain English. The AI writes the code. You test it, tweak it, describe the problem again, and the AI fixes it.
You are not a developer. You are a director. You know what the movie should look like. The AI handles the camera work.
That night at 11pm, I opened Claude on my phone and typed something like: “Help me build a simple web form where a business owner can paste raw WhatsApp messages and it organizes them into a clean order table automatically.”
By 1:30am, I had a working prototype. Rough, yes. Ugly, honestly. But it worked.
I want to be clear: I did not write a single line of code myself. Every line was generated, explained, and debugged by AI. My job was to describe the problem clearly, test the output honestly, and keep asking until it worked.
The Business Model That Now Pays Me Weekly
Here is the part most people skip over and then wonder why their “AI business” flopped. The tool alone is not the business. The subscription is the business.
I did not sell the tool once for ₦50,000. I priced it at $3 per month. That is roughly ₦4,500 at current rates. Cheap enough that people do not overthink the decision. Valuable enough that they keep paying it every month because it saves them two hours of work every week.
Do the math yourself. 127 subscribers times $3 is $381 per month. At today’s exchange rate, that is over ₦570,000 per month. Weekly, that clears ₦380,000 comfortably. And those subscribers keep growing because I keep finding new pockets of that same problem in different industries.
The part that still makes me blink sometimes is this: that money arrives whether I wake up at 6am or noon. Whether I am traveling. Whether I am sick. Whether I am just staring at the ceiling doing absolutely nothing. The subscribers pay. The platform collects. My account grows.
The Exact Process I Follow Every Time
I spend 30 to 60 minutes per week reading Facebook groups, Nairaland threads, Reddit forums, and Twitter complaints in specific niches. I am not looking for ideas. I am looking for pain.
I post a simple poll or question in the same community. “If there was a tool that did X automatically, would you pay ₦5,000 per month for it?” If at least 20 people say yes, I move forward.
Lemonsqueezy works smoothly for Nigerian creators. No complex setup. Literally 20 minutes to go live and start collecting dollar payments.
Their feedback shapes the next version. By the time I start charging, the tool already has testimonials, real data, and people who feel ownership over it.
Every month I improve one thing based on what subscribers say. Happy subscribers refer other subscribers. The list grows without paid ads.
Common Mistakes That Will Kill This Before It Starts
- Building a tool first before finding out if anyone actually has the problem
- Pricing too high — $3 to $10 beats ₦150,000 one-time every single time
- Waiting until the product is “perfect” before launching
- Solving a problem you personally have but that only five other people share
- Using complicated tech stacks because a YouTube tutorial said so
- Ignoring subscriber feedback after the first month
Hard Truths Nobody In The AI Space Will Say Out Loud
Most AI content teaches you to use AI as a productivity tool. Faster writing. Quicker research. Better summaries. That makes you a better employee. What I am talking about makes you an owner.
The $3 price feels embarrassingly small when you first set it. You will be tempted to charge more to feel important. Do not. One hundred people paying $3 is infinitely better than zero people refusing to pay $50.
The Thought I Want To Leave You With
The ₦380,000 is not the impressive part. The impressive part is that it arrived last week while I was visiting my aunt in Ibadan, eating jollof rice, and not thinking about work at all.
That is what I want for you. Not a hustle. An asset. Something that runs when you are not running it.
The only question is: what problem will you go looking for tonight?
Your First Step Is Free
Open Claude or ChatGPT right now. Type: “Help me find common frustrations that small business owners in Nigeria face that could be solved with a simple web tool.” Read what it says. Then go validate one of them in a real community tomorrow morning.
Start With Claude — Free
