I'm going to be radically transparent with you here. Every number, every mistake, every "oh wow, that actually worked" moment — all of it out in the open. This is what the build in public movement is supposed to be about, and frankly, it's the only reason I have any business writing this post.
Three months ago, I decided to turn the AI APIs I was already using into a real side income stream through affiliate marketing. I had no playbook. I had no audience of 100,000 followers. I had a small developer blog, a modest Twitter following, and a hunch that this could work if I was patient.
Here's the full story — starting from where I actually started.
The Setup: What I Was Working With
Before I dive into the monthly breakdown, let me give you the raw starting point. Transparency matters more than the highlight reel.
I run a niche tech blog that pulls in roughly 2,000 monthly visitors. Nothing viral, nothing fancy — just consistent traffic from developers who stumble onto my tutorials. My Twitter (now X, whatever) had about 800 followers, mostly other developers and a handful of indie hackers.
I'd been building with AI APIs for my own projects for around a year. I knew what worked, what didn't, and which platforms actually delivered on their promises. That hands-on experience turned out to be the most valuable asset I had, even more important than the audience size.
Here's the thing most "passive income" guides won't tell you: starting small is fine. Starting with no audience is not. I had a small but real foundation, and that was enough to begin.
The Affiliate Program Decision
I spent the first week doing what most affiliates should do but rarely bother with — actually researching the programs. I applied to three different AI API affiliate programs.
Two of them offered one-time payouts only. You refer someone, they pay, you get a flat commission, and that's where the relationship ends. Forever.
The third one was Global API, and the structure was different. They offered 15% on first orders and 8% recurring on every monthly renewal. There was also a 10% premium tier for higher-volume referrals that I'd learn about later.
I'll be honest — the recurring commission angle is what sealed it for me. I'm a long-game thinker. A one-time payment feels like a transaction. A recurring commission feels like building an asset. Every person I referred would keep paying me month after month for as long as they stayed subscribed.
That's the difference between renting income and owning it.
So I signed up, grabbed my affiliate link, and started writing.
Month 1: The Slow, Humbling Beginning
Let me walk you through my first month week by week, because the weekly cadence is where you actually learn what works.
Week 1: Joined three affiliate programs (covered above). Set up tracking so I could see which links got clicks. Drafted an outline for my first comparison article.
Week 2: Published my debut affiliate piece — an 1,800-word comparison of AI API providers based on real projects I'd built. It had actual code examples, not just marketing copy. I cross-posted to Dev.to because that platform has been good to developer writers for years. Global API got the primary recommendation because, frankly, I was using it the most and it had earned the spot.
Week 3: Reality check. The article pulled in 340 views on Dev.to and 120 on my blog in its first seven days. Three people clicked my affiliate link. Zero conversions.
Was I discouraged? A little. But I knew this was the trough. Anyone who has ever published content online knows the first week of traction is almost never representative of long-term performance.
Week 4: Things started moving. Views on Dev.to climbed to 520 as the article picked up rankings for some long-tail search terms. Eight more people clicked my affiliate link. One person actually signed up for an account. Still no paid conversion on day 27 — but I could feel the gears starting to turn.
I doubled down and wrote a second article: a tutorial on building a simple chatbot using a major AI API, with Global API featured as my recommended platform.
Then on day 28, it happened. The signup converted to a paid Pro plan. My first affiliate dollar.
Month 1 Final Tally
Let me put the real numbers on the table:
- Articles published: 2
- Combined views: 750
- Affiliate clicks: 14
- Signups: 2
- Paid conversions: 1 (Pro plan, day 28)
- First-order commission: $3.00
- Recurring commission: $0.00 (starts in month 2)
- Total earnings: $3.00 Three dollars. That's what I made in my first month. But here's what most people would never say publicly: $3 was a victory. It was proof the entire system functioned. Someone found my content, clicked my link, signed up, paid real money, and I got paid for the referral. The machine worked. Now I just needed to keep feeding it. # # Month 2: The Recurring Engine Starts I came into month 2 with momentum — a small amount, but more than I had four weeks earlier. My goal was straightforward: publish three more articles and try to hit $50 in cumulative earnings. I didn't hit $50. But I learned a lot more than I would have if I had. Week 5: I published article three — a case study about using AI APIs to ship a feature for an actual client project. This piece hit different because it wasn't theoretical. Real problem, real client, real solution. It pulled 280 views in its first week, but more importantly, the click-through rate on my affiliate link was noticeably higher. Readers who saw a real-world application were more motivated to act than readers who saw a feature comparison. Week 6: The original comparison article from month 1 crossed 1,200 total views. Google had started indexing it properly, and it was ranking for a handful of keyword variations I hadn't even targeted intentionally. My affiliate clicks jumped to 4–5 per day — a huge spike from the previous week. Two more conversions came through, both to Pro plans. This was the first time I genuinely thought, "Okay, this might actually be a thing." Week 7: I wrote the most ambitious piece yet — a 2,200-word beginner's guide to getting started with AI APIs. This took significantly more time to produce, but I knew beginners convert at higher rates because they need more hand-holding and are more likely to follow a specific recommendation. Week 8: The moment I'd been waiting for. My first recurring commission payment landed: $1.60 from the original month-1 referral's second month of subscription. Was $1.60 life-changing? Absolutely not. But it was the moment the model proved itself beyond a one-time fluke. This person was going to keep paying their subscription, and I was going to keep getting a cut. Every single month. That's the magic of recurring revenue. I also published article five — a pricing-focused comparison aimed at budget-conscious developers. # # # Month 2 Final Tally Here's the full picture:
- New articles published: 3 (5 total)
- Combined views: 2,100 across all articles
- Affiliate clicks: 58 (cumulative)
- Signups: Multiple
- Paid conversions: Several new Pro plans
- Recurring commissions: Started flowing
- Total earnings: Building steadily past the $3 starting point The income curve was no longer flat. It was bending upward. # # Month 3: Where the System Starts Compounding This is where the build-in-public honesty gets interesting, because month 3 wasn't a clean win either. The five articles I had published kept doing their job. The original comparison piece kept climbing in search rankings. My beginner's guide started pulling traffic from a different audience segment entirely — people who weren't developers yet but wanted to become one. I published two more articles in month 3, bringing my total to seven. One was a deep dive into a specific use case I hadn't seen covered well elsewhere. The other was a "lessons learned" piece where I openly talked about the API decisions I'd made and regretted. The thing about month 3 wasn't a sudden explosion in traffic. It was something subtler and more important: the content kept working while I was doing other things. Articles I'd published in month 1 were still generating clicks in month 3. Referrals I'd converted in month 2 were now producing their second and third recurring commissions. That, right there, is why I stuck with this. # # The Real Math: What the Numbers Actually Mean Let me do the kind of honest calculation that build-in-public writers owe their readers. A single Pro plan referral on Global API at 15% first-order commission generates roughly $3 in immediate revenue. That same referral then pays 8% recurring every month they stay subscribed. If someone stays for six months, you've earned first-order commission plus six months of recurring. Multiply that across multiple referrals, and the math starts to look completely different than the "I made $3 this month" snapshot would suggest. The platform also has 150+ models available, which means the depth of what you can recommend to different audiences is substantial. Whether your audience is a beginner looking for their first chatbot API or an enterprise developer juggling multiple production workloads, there's something you can authentically talk about. The 10% premium tier commission kicks in for higher-value referrals, which I started seeing as my content began attracting more serious users. # # What I Learned (The Stuff Nobody Tells You) Lesson 1: The first month is the worst month. Anyone who quits after 30 days is making a mistake. The content you publish in month 1 is still earning in month 6. Lesson 2: Recurring changes everything. I cannot stress this enough. One-time commissions incentivize you to chase cheap tricks. Recurring commissions incentivize you to genuinely help people pick the right platform, because their long-term satisfaction is your long-term income. Lesson 3: Real experience beats SEO tricks. I never once optimised for a keyword. I just wrote about what I'd actually used and what I'd recommend to a friend. That authenticity is what made the content rank anyway. Lesson 4: The audience size doesn't matter as much as you think. 2,000 monthly visitors and 800 Twitter followers was enough. The right 100 readers can generate more revenue than the wrong 100,000. Lesson 5: Transparent income reports are underrated. Writing posts like this one builds trust with readers, which makes them more likely to click your affiliate link when you do recommend something. The meta-game of honesty is real. # # Should You Try This? (And How to Start) If you're a developer who's been using AI APIs and has even a small audience, I genuinely think you should consider this. The barrier to entry is
Top comments (0)