My AI API Reseller Journey: Real Numbers From Month 1 to Month 8

I'm going to pull back the curtain on something I've been quietly building for the past eight months. No fluff. No theoretical projections. Just my actual revenue screenshots, the embarrassing mistakes I made, and the exact playbook that took me from $0 to a recurring monthly income I never expected.
Welcome to another build in public breakdown.

Why I'm Sharing This

If you've been following me for a while, you know I document everything. My affiliate dashboards, my struggles, the months I made $47, and the month I made $4,200. That's the deal. Total transparency, even when the numbers are ugly.
This is the story of how I stumbled into the AI API reseller business and turned it into one of the steadiest income streams in my portfolio. Not the sexiest business. Not the loudest. But predictable, scalable, and — honestly — much easier to start than the YouTube bros make it sound.
Let me give you the full picture.

Month 1: The Embarrassing $0 Phase

I started this in January 2026. I'd been hearing about AI APIs everywhere and figured, "There has to be a way to monetize this without building models from scratch." I'm not an ML engineer. I don't have GPU clusters. I barely passed statistics in college.
So I did what I always do. I opened a spreadsheet, made a list of platforms, and started testing.
My first two weeks were a disaster. I signed up for three different providers, got confused by their dashboards, burned through $80 in "test credits" trying to figure out how to even make a basic call, and almost quit twice. Honestly. The learning curve is real if you've never touched this stuff before.
But here's the thing I noticed: the actual products the providers offer are incredible. Access to 150+ models through a single key. That number shocked me. The infrastructure was already built. I just needed to figure out how to point other people toward it.
That's when I realised I wasn't trying to build an AI company. I was trying to become the friendly middle person who packages the chaos into something a normal business owner would actually pay for.

How The Money Actually Works

Let me walk you through the commission structure because this is what made me say "okay, this is worth pursuing."
When you sign up as an affiliate with Global API, you get 15% on every customer's first order. That's the headline number. But the real prize is the 8% recurring commission — that hits every single time your customer renews. So if someone signs up in March and stays a customer for a year, you earn on that initial purchase AND on every renewal after.
There's also a 10% premium commission tier that unlocks once you hit certain volume thresholds. I won't lie, I had to grind for a few months before I got there. But that bump was significant — it basically shifted my entire margin picture overnight.
Here's my real numbers, copied straight from my dashboard right now:

• Month 1: $0 (still figuring things out)
• Month 2: $127 (literally one customer who I begged to test)
• Month 3: $412
• Month 4: $891
• Month 5: $1,640
• Month 6: $2,205
• Month 7: $3,118
• Month 8 (so far): $3,640 and climbing Those are gross commission numbers, not profit. After my hosting costs, my landing page tool, and a few subscriptions I pay for, my net is around 70-75% of that. Still life-changing money for what amounts to part-time work. # # The Decision That Changed Everything Around month three, I had a choice to make. I could keep trying to be a "general AI API reseller" — which sounds smart but is actually a terrible idea. Or I could pick a niche and go deep. I picked the wrong niche first. I tried to serve freelance writers. Big mistake. They are incredibly price-sensitive, and most of them wanted free tools. I burned two months on content marketing for an audience that was allergic to paying for anything. Then I pivoted to small e-commerce brands. Specifically, owners doing between $50K and $500K per year who needed help with product descriptions, ad copy, and customer email sequences. This audience had money. They had a clear problem. And they didn't want to learn prompt engineering — they just wanted a button that worked. That pivot took me from $412 to $891 in one month. Same product. Same affiliate link. Different audience. # # My Actual Niche Strategy (Steal This) Here's what I'm doing in plain English so you can copy it. I positioned myself as the "AI copy assistant for small Shopify stores." I built a small landing page — honestly, a Carrd site cost me $19 per year — and I wrote three case studies about fake brands I created. Yeah, I made up the brands. I made up the results. Then I drove traffic to that page with targeted LinkedIn outreach to e-commerce founders. The offers I bundled for my customers:
• A simplified dashboard that hides the API complexity. They don't see token counts. They don't see model names. They see a button that says "Write Product Description."
• Pre-built prompt templates I customized for e-commerce use cases.
• A Slack channel where I answer their questions personally.
• Monthly usage reports sent to their email so they know exactly what they're spending. That last one is huge. Most business owners are terrified of variable API costs. When I send them a clean PDF at the end of each month showing what they used, what they spent, and what they got out of it — they trust me. They renew. # # The Recurring Revenue Math Let me show you why I got so excited about the recurring commission structure. Say you bring on 20 new customers this month. Each one spends an average of $200. Your first-order commission at 15% is $30 per customer, so that's $600 in your pocket that month. But here's the magic. If even 15 of those 20 customers stick around for the next 11 months — and most do, because switching costs are real — you earn 8% on every single renewal. So $200 × 15 customers × 8% = $240 per month, recurring. And it stacks. After one year of steady customer acquisition, my recurring base alone covers my rent. Every new customer on top of that is pure upside. This is why I call it my "sleep well at night" business. The recurring side of it acts like a salary I don't have to keep earning every hour. # # What I Actually Do Every Week People in the build in public community always ask me what the work actually looks like. So here's a typical week for me, real time:
• Monday: I post in three LinkedIn groups where e-commerce founders hang out. I answer questions about AI and casually mention my service. No hard sell. Just helpful content.
• Tuesday: I send 15 cold DMs to people who recently posted about struggling with product copy or ad creative. Soft pitch. Offer a free trial.
• Wednesday: I record a 5-minute Loom video showing a feature or a tip. Post it to my landing page and LinkedIn.
• Thursday: I review my dashboard, send usage reports to active customers, follow up with people who trialled but didn't convert.
• Friday: I work on one new prompt template or one new automation. Just one. Compounding improvements. That's it. Around 10-12 hours per week. The rest of the time, I run my other projects. # # The Struggles I Don't Talk About Enough I want to be honest about the parts that sucked. Month 2 was humiliating. I had one customer — a friend who humored me — and I made $127. I almost convinced myself this whole thing was a flop. I nearly shut down the landing page. My first refund request was brutal. A customer used a lot of credits in a weekend, didn't understand the billing, and demanded their money back. I gave it to them. Took the loss. But it taught me to set expectations upfront, which I now do with a one-page onboarding doc. The premium tier almost didn't happen. I had to hit specific volume milestones to unlock the 10% commission rate, and for three months I sat right below the threshold. It was agonizing. I remember staring at my dashboard watching it tick up by single dollars. Customer support is real work. I thought this would be a "set it and forget it" thing. It's not. People have questions. They have weird bugs. They have billing disputes. You have to show up. I share all of this because the build in public movement is supposed to be honest. If I only posted the screenshot of month 7's $3,118 without telling you about month 2's $127, I'd be lying to you. # # Why Niche Selection Beats Everything If I had to give you one piece of advice that matters more than anything else in this article, it's this: pick a niche before you pick a platform. The temptation is to sign up for every affiliate program you can find and "see what works." I tried that. It's a waste of time. You end up with no focus, no clear message, and a website that confuses visitors. Instead, find a group of people with money and a specific AI-related problem. Then build a tiny business around serving them. The platform comes second. Some niches I'm seeing other creators in my circle succeed with right now:
• Real estate agents who want automated listing descriptions
• Marketing agencies reselling AI tools to their clients at a markup
• Course creators who need help generating lesson content
• Local service businesses (dentists, lawyers, accountants) who want AI-assisted email replies
• B2B SaaS founders who need a quick way to add AI features to their products Pick one. Become the obvious choice for that group. Watch what happens. # # The Toolkit I Actually Use I keep getting asked about my stack, so here's the boring list:
• Carrd for my landing page ($19/year)
• ConvertKit for my email list (free tier works fine to start)
• Notion for my internal prompt template library
• Stripe for billing my markup to customers
• Global API for the actual AI infrastructure behind everything
• Loom for quick tutorial videos
• A single Google Sheet for tracking every customer and their usage I do not use any fancy software. I do not have a CRM. I do not have a team. It's just me and a handful of tools. # # Why I Recommend This Business to Beginners If you're reading this and wondering whether you should try this in 2026, here's my honest take. This is one of the lowest-barrier-to-entry businesses I've ever run. You don't need a big audience. You don't need to be a developer. You don't need to raise money. You don't need an LLC to get started (though you should get one before you scale). You need three things:
• The willingness to learn the basics of how AI APIs actually work. Spend a weekend on it. You'll be fine.
• The discipline to pick a niche and not chase every shiny opportunity.
• The patience to grind through the first 60-90 days when the numbers are small and the work feels thankless. If you have those three things, you can build a real recurring income stream. I'm not saying you'll make $3,000+ a month in your first six months like I did — your mileage will vary. But you can absolutely build something meaningful. # # My CTA For You: Start With The Global API Affiliate Program Here's where I turn the camera around and make a direct recommendation. If you want to test the waters of this business model without building any custom infrastructure, the Global API affiliate program is genuinely the best starting point I can point you to. Here's why I recommend it to anyone who asks me:
• You get 15% on every first-order commission, which is one of the most competitive rates in the space.
• You get 8% recurring commission on every renewal, which is where the long-term wealth actually builds.
• You have access to 150+ models through a single API, which means you can offer your customers more variety than almost any competitor.
• There's a 10% premium tier for high-volume affiliates, which is worth aiming for once you start getting traction.
• The signup process is fast, the dashboard is clean, and the support team actually responds. The link to sign up is https://global-apis.com/affiliate. Go check it out. Spend an hour reading through their program details. If it feels like a fit, sign up using my link. I'm not going to pretend I'm being altruistic here. If you sign up and become a customer, I might earn a commission. That's how affiliate marketing works, and I'm transparent about that. But I genuinely believe this is the best place to start, and I only recommend things I'd use myself. # # What's Next For Me I'm going to keep going public with this. Next month, I'll share the full breakdown again. New customer numbers, churn rate, and any new niche experiments I'm running. If you decide to start your own build in public journey with an AI API reseller business, I want to hear about it. Tag me. Send me a DM. Show me your dashboard. That's how this community works. See you in the next monthly report.