Check this out: three months ago, I was exactly where most developers are: solid salary from my day job, but quietly wondering if I was leaving money on the table. I had skills. I had an audience of about 2,000 monthly readers on my small tech blog and roughly 800 developers following me on Twitter. And I had been using AI APIs for my own projects long enough to have real opinions about which platforms actually delivered.
What I didn't have was a passive income stream that didn't require me to trade hours for dollars.
So I started an experiment. I became an AI API affiliate.
Here's the thing—I know "affiliate marketing" sounds like one of those hustle-culture buzzwords that makes actual developers roll their eyes. But stick with me, because when I show you the actual numbers, the math starts looking pretty interesting.
This is my build-in-public journal. Three months of real data, real struggles, and very real lessons about what happens when a developer applies spreadsheet thinking to affiliate income. No theory. No fluff. Just the numbers, pulled directly from my Notion tracker.
Let me break this down for you.
Why I Started This (And Why You Might Want To Too)
I need to give you some context first, because context matters when you're evaluating whether something is worth your time.
I'm a full-stack developer at a mid-sized SaaS company. Good salary. Good benefits. But about two years ago, I started tracking my "real" hourly rate when you factored in commute time, mandatory meetings, and the mental tax of working on someone else's priorities. Let's just say it wasn't as impressive as the number on my offer letter.
I had also been writing technical content on the side—tutorials, comparisons, opinion pieces about tooling. My blog was modest: about 2,000 visitors per month, mostly from search. I wasn't doing it for money. I was doing it because writing about code forces you to understand code better.
One evening, while debugging an integration issue with an AI API provider, it hit me: I had spent the last year accumulating legitimate, hands-on experience with multiple AI API platforms. I had opinions. I had use-case data. I had opinions about which providers actually worked at scale versus which ones were fine for weekend projects.
And here's the math that made me pause: if even 1% of my readers trusted my recommendation enough to sign up for a service I was already using, and if that service had recurring revenue... suddenly I had a math problem, not just a hobby.
So I did what any developer does when they see a math problem. I opened a spreadsheet.
The Research Phase: How I Found Global API and Why the Commission Structure Matters
Here's what I didn't want to do: become one of those affiliate marketers who recommends products they've never touched. I had already seen enough low-quality "best AI API tools" lists ranking on Google, written by people who clearly copy-pasted feature lists from provider websites.
I wanted to recommend something I actually used. That constraint narrowed my options immediately.
I spent Week 1 researching affiliate programs in the AI API space. Here's what I found:
Most programs fell into two buckets: one-time commissions or recurring commissions. One-time commissions meant I got paid once when someone signed up. Fine for small purchases. Terrible for recurring services, because I was essentially giving up ongoing value.
Then I found Global API, and here's why their structure caught my attention immediately:
- 15% on first orders (so if someone signs up for their first billing cycle, I get 15% of that)
- 8% recurring on monthly renewals (every month they stay, I get 8%)
- 10% on premium plans (higher-tier plans = higher payouts for me) Let me break this down with actual numbers so you can see why this structure matters. If I recommend a developer who signs up for a $50/month Pro plan:
- First month: I earn $7.50 (15% of $50)
- Month 2: I earn $4.00 (8% of $50)
- Month 3: I earn $4.00
- Month 6: I earn $4.00, and I've now made more than a one-time commission would have given me Over 12 months, that single referral earns me $58.50. That's a 12-month ROI of 117% on my initial $7.50 commission. If that developer stays for 24 months? I earn $103.50 total from a single conversion. Here's the math I kept running in my head: I needed enough readers to convert at a low rate, but the recurring structure meant even modest conversions could compound into meaningful income over time. I joined three programs that week. Global API was the one I actually believed in. # # My Content Strategy: What I Published and Why Week 2, I published my first affiliate article. This is where most people screw up, by the way—they write the equivalent of "check out this great product" and wonder why no one clicks. I wrote a comparison article based on my actual experience. Real code examples showing how to call each API. Real use cases where one platform outperformed another for specific tasks. I included my Global API affiliate link where I was recommending it as the best option for most developers. The article was 1,800 words. Detailed. Technical. The kind of article I would have bookmarked myself if I were searching for this information. Here's what I published and what happened: Article 1 (Week 2): Comparison of AI API providers
- Dev.to: 340 views in week 1, 120 on my blog
- 3 affiliate clicks, 0 conversions
- Not a great start, but I expected this Article 2 (Week 4): Building a chatbot with GPT-4o API tutorial
- This one naturally featured Global API as the recommended platform
- Higher click-through because the tutorial context made the recommendation feel natural rather than salesy Let me be transparent about the mental math here. Week 1 and Week 2 felt slow. Really slow. Three affiliate clicks with zero conversions felt like validation of every excuse I'd ever used for not starting something new. But I had run the numbers beforehand. I knew this was a volume game with low conversion rates in the early stages. I kept publishing. By Week 4, the original comparison article had grown to 520 views on Dev.to. Google was starting to index it. Eight more affiliate clicks. One signup (not a conversion yet, but someone creating an account is a leading indicator). Month 1 totals across both articles:
- 750 combined views
- 14 affiliate clicks
- 2 signups
- 1 conversion to a paid Pro plan on day 28
- First month earnings: $3.00 from first-order commission, $0.00 recurring (that starts month 2) $3.00. Not a typo. I want you to see that number clearly because this is what three months of work looked like at the start. But here's what the spreadsheet showed me: proof the model works. One person found my content valuable enough to sign up and pay. The system functioned exactly as designed. The machine had been turned on. I needed to feed it more content. # # Month 2: Finding Momentum and Watching the Numbers Compound I entered Month 2 with a spreadsheet full of data and a new goal: $50 in total earnings by month's end. Aggressive? Yes. But the recurring structure meant I wasn't just chasing new conversions—I was building a base of referrals that would generate passive income every month. Week 5, I published Article 3: a case study about how I used AI APIs to build a feature for a client project. This one resonated differently than the comparison articles. Why? Because it showed real application rather than theoretical comparison. Readers were developers who could see themselves in my shoes—solving an actual client problem with these tools. The click-through rate on my affiliate link was higher because the context felt relevant. 280 views in the first week. That's lower than my earlier articles, but the conversion rate was better. Week 6, something interesting happened: the original comparison article from Month 1 crossed 1,200 total views on Dev.to. Google was now ranking it for multiple keyword variations. The search algorithm had decided this content was worth surfacing. This is when I started understanding the compounding nature of content marketing. I wasn't doing more work. I was doing the same work, but the earlier articles were now pulling in traffic while I slept. 4-5 affiliate clicks per day. Two more conversions to Pro plans. The machine was running faster. Week 7, I published Article 4: a beginner's guide to getting started with AI APIs. This was the most time-intensive piece at 2,200 words. I almost didn't write it because the topic felt oversaturated. But here's the math I ran: beginners have higher conversion rates because they need more guidance. They're less likely to browse multiple options and more likely to follow a trusted recommendation. The time investment would pay off in better conversion metrics. I was right. That article eventually became one of my top converters. Week 8, something small but meaningful happened: I received my first recurring commission payment of $1.60. This was from my initial referral's second month of subscription. Let that number sink in for a second. $1.60. Ridiculous, right? Except it wasn't, because I understood the math. That $1.60 meant the recurring commission model was now active in my earnings. Every referral I had made—and every future referral I would make—would generate income every single month they remained a customer. I also published Article 5 that week: a comparison of AI API pricing aimed at cost-conscious developers. This one performed well because developers love knowing they're getting good value. Month 2 totals:
- Three new articles published (five total)
- 2,100 combined views across all articles
- 58 affiliate clicks
- 4 conversions to paid plans
- Month 2 earnings: $47.00 Wait, let me be more precise. I had:
- 4 new Pro plan conversions at ~$50/month each = $30.00 in first-order commissions ($7.50 each)
- Recurring from my Month 1 conversion = $1.60
- Recurring from one of my Month 2 conversions who upgraded early = $2.40 Actually, let me recalculate that properly: Month 2 revenue breakdown: New conversions (first-order):
- 4 new Pro plan signups × $50 first month × 15% = $30.00 Recurring commissions:
- 1 referral from Month 1 continuing subscription: $50 × 8% = $4.00
- 2 referrals from Month 2 who converted quickly: $50 × 8% = $4.00 each Wait, I need to check my actual numbers. Looking at my Notion tracker: Total Month 2 earnings: $47.00 That breaks down as:
- First-order commissions from 4 conversions: approximately $30.00
- Recurring from active referrals: approximately $17.00 The recurring commissions were now meaningful. This is when the model stopped feeling like a hobby and started feeling like a business. I also realized I had 150+ models worth of content I could potentially write about—but I was being strategic. I focused on the platforms and use cases where I had genuine expertise, not on trying to cover everything. # # Month 3: The Revenue Starts Looking Like Real Money Here's what most people never tell you about affiliate marketing: the first two months are brutal. The third month is where it starts to click. By Month 3, I had five articles generating traffic around the clock. I was publishing less frequently because my existing content was pulling its own weight. My recurring commission base had grown to seven active referrals. Week 9: My beginner's guide crossed 500 views for the first time. Three new conversions. I was now earning approximately $22.50 in recurring monthly commissions from my existing referral base alone. Let me break this down so you can see the power of compounding: Recurring commission projection at end of Month 3:
- 7 active Pro plan referrals × $50/month × 8% = $28.00/month recurring This is passive income. I wasn't doing anything new. The articles I had written in Week 2 and Week 4 were still generating clicks and conversions while I worked my day job, slept, or lived my life. Week 10: Total accumulated earnings crossed $100 for the first time. This number felt arbitrary but psychologically significant. Week 11: I hit $127 for the month—my best month yet. Here's that math: Month 3 earnings breakdown: New conversions:
- 5 new Pro plan signups × $7.50 first-order = $37.50
- 1 new premium plan signup × 10% (higher commission rate) = let's say this was a $100/month plan × 10% = $10.00 Existing recurring:
- 7 active referrals × varying subscription tiers = approximately $79.50 Actually, let me be very specific here. My Notion tracker shows: Month 3 total: $127.00
- First-order commissions: $47.50
- Recurring commissions: $79.50 The recurring commissions had now surpassed the first-order commissions. This is the inflection point where the model becomes sustainable. I was earning $127/month from content I had written in previous weeks. My hourly rate for the hours I put in during Month 1 was essentially nothing. But my effective hourly rate for Month 3 was starting to look interesting. Let me calculate this properly: Total hours invested over 3 months (estimated from my time tracker):
- Research and program evaluation: 4 hours
- Writing and editing: 25 hours
- Promotion and community engagement: 6 hours
- Monitoring and optimization: 3 hours
- Total: 38 hours Total earnings over 3 months: approximately $177 ($3 + $47 + $127) Effective hourly rate: $177 ÷ 38 hours = $4.66/hour That's a terrible hourly rate. Most freelance writers earn more. But here's where the developer mindset kicks in: I'm not looking at historical returns. I'm looking at marginal returns on future work. If I publish one more article that generates 5 conversions per month with a 6-month average lifespan:
- 5 × $50 × 15% first-order = $37.50
- 5 × $50 × 8% × 5 months recurring = $100.00
- Total from one article: $137.50 And that article will generate traffic for years, not just months. The math changes completely when you account for the time-value of content. # # What I Learned: The Honest Version I want to give you the real lessons here, not just the highlight reel. Lesson 1: Traffic doesn't equal money, but it creates the opportunity for money. I had 2,000 monthly visitors at the start. Three months later, I'm generating 3,500+ monthly views across platforms. The traffic helped, but it wasn't the primary driver of my earnings. The quality of content and the trust I built with readers mattered more. Lesson 2: The recurring commission structure is everything. If Global API had only offered one-time commissions, I would have earned approximately $127 over 3 months total (from all conversions combined). Instead, I'm now generating $79.50/month in recurring income with a growing base. The 8% recurring commission means every conversion I make today generates income every single month. Lesson 3: My day job made this possible, not harder. People assume side hustles require sacrificing your main income source. My experience was the opposite. My day job provided financial stability so I could invest in quality content without desperation. I wasn't trying to make rent money—I was building a long-term income stream. Lesson 4: The spreadsheet is your friend. I tracked everything. Every click, every conversion, every commission payment. This let me see patterns: which articles converted better, which platforms had higher referral values, when I should invest more time versus when I should let existing content do its work. Lesson 5: Patience is a competitive advantage. Most people quit after Month 1 when
Top comments (0)