I bill by the hour. Always have. For the last six years, I've been a freelance writer grinding out blog posts, white papers, and email sequences for SaaS clients at $75 to $100 per hour. Some weeks I cleared $4,000. Other weeks I cleared almost nothing. The feast-or-famine cycle wore me down to a nub.
So when I stumbled across AI API affiliate programs last quarter, I did what any burnt-out freelancer would do: I signed up for everything that didn't require a credit card and started writing about it. This is the unfiltered story of what happened over the next 90 days, including every ugly number and missed opportunity along the way.
The Problem With Selling Hours
Let me set the scene. My biggest client was a mid-stage fintech startup paying me a $2,800 monthly retainer for four long-form articles. Decent money. But every single piece required pitching, revising, chasing Slack messages, and answering "can we make the intro punchier?" for the forty-seventh time. I was trading my evenings for someone else's content calendar.
I kept hearing about this mythical thing called "passive income" from other writers in my network. Most of them were selling courses about selling courses. Not helpful. What I actually wanted was a revenue stream that didn't require me to be on Slack at 11 PM.
Affiliate marketing for AI APIs felt like a weird fit at first. I'm a writer, not a developer. But I'd been using AI tools in my own workflow for about a year. I knew which platforms I trusted. I knew which ones made my job faster. And I knew there were thousands of other freelancers and indie builders looking for the same answers I'd already figured out.
That overlap — my writing skills plus genuine product knowledge — turned out to be the whole game.
Why I Picked Global API Over the Others
I applied to three different affiliate programs in week one. Two of them offered flat, one-time payouts. You send a customer, they pay you $50 or $100, and the relationship ends there. That kind of payout is fine if you're running a high-traffic review site, but my blog was pulling maybe 2,000 visitors a month and my Twitter had around 800 developer followers. I wasn't going to make rent on one-time commissions.
Then I found the Global API affiliate program. Two things stood out. First, you earn 15% on every customer's first order. Second — and this is the part that changed my thinking — you earn 8% recurring on every monthly renewal for as long as that customer stays subscribed. Premium-tier customers bump that recurring cut to 10%. The platform itself aggregates access to 150+ models under one roof, so I wasn't just shilling one product. I was recommending a tool I'd actually use myself.
Recurring commissions meant my earnings could compound. If I brought in a customer in January who stuck around, I'd still be earning from them in July. My $2,800 monthly retainer from the fintech client worked the same way, except that one required 40+ hours of my time. The affiliate commission required me to write one good article.
I was sold.
Month 1: Two Articles and a Humbling $3.00
My first month was a lesson in patience.
Week one, I wrote a comparison post for my blog called something like "Which AI API Should You Actually Use in 2025?" It ran around 1,800 words with honest pros and cons for each provider I'd personally tested. I cross-posted to Dev.to and dropped my Global API affiliate link in the recommendation section where it felt natural, not forced.
The first seven days brought 340 views on Dev.to and 120 on my own blog. Three people clicked the link. Zero signed up. I refreshed my analytics dashboard roughly ninety times that week.
Week four was slightly better. The comparison post had climbed to 520 Dev.to views thanks to a few long-tail search rankings. Eight more clicks came in, and one person actually created a free account. Still no paid conversion, but the signup was a sign that my pitch was landing somewhere.
I published a second article that month — a step-by-step tutorial on building a chatbot with one of the GPT-4o endpoints. It was the kind of post I wished existed when I was learning. Same format: useful content first, affiliate link second.
Month 1 totals:
- Articles published: 2
- Combined views: 750
- Affiliate clicks: 14
- Free signups: 2
- Paid conversions: 1 (someone upgraded to the Pro plan on day 28)
- Earnings: $3.00 first-order commission, $0.00 recurring Three dollars. That's what I made for roughly 18 hours of writing, editing, and link-placement. I could've billed a client for that hour and made more. But the difference is that the article kept working after I closed my laptop. Every morning I'd wake up to a trickle of new clicks. The article wasn't going anywhere. # # Month 2: When the Recurring Hit My Dashboard The second month is when things shifted. Not dramatically — I'm not talking about quitting my day job — but noticeably. I published three more posts. The third one was a case study about how I'd used an AI API to build a small feature for a client project. It hit 280 views in week one and converted surprisingly well because the readers were fellow developers who recognized their own workflow in mine. By week six, my original comparison article had crossed 1,200 total views on Dev.to. Google had started indexing it for a handful of long-tail keywords. Affiliate clicks climbed to four or five per day. Two more people upgraded to Pro plans that week. Article four was a beginner-friendly guide to getting started with AI APIs. At 2,200 words, it was the longest piece I'd written, but beginners convert at higher rates because they're actively shopping for guidance. They want someone to tell them what to use. The moment I'll never forget was week eight. I opened my Global API dashboard and saw a new line item: $1.60. It was the recurring commission from my month-one referral's second subscription payment. Eight percent of their monthly bill, paid to me for the privilege of having introduced them. That single $1.60 line item mattered more to me than the $3.00 first-order commission had. It proved the model. It meant my earnings could keep growing without me writing a single new word. I rounded out the month with article five, a pricing-focused comparison aimed at cost-conscious builders. Month 2 totals:
- New articles published: 3 (5 total)
- Combined views: 2,100
- Affiliate clicks: 58
- Paid conversions: 3 (including the original month-one referral still paying monthly)
- First recurring commission earned: $1.60 # # Month 3: Compounding Kicks In This is where the freelancer mindset started to flip. By month three, I had five published articles doing the work for me. The comparison post was pulling around 40 views a day on Dev.to alone. My beginner's guide ranked on page one for a few high-intent search terms. The case study kept circulating in developer Slack channels. I didn't write much new content in month three. I published two additional posts — one on prompt-caching strategies and another on workflow automation — but the bigger story was the recurring side of the dashboard. The referral from day 28 of month one was still paying. The two week-six conversions from month two were paying. And four new customers upgraded during month three from older articles I'd almost forgotten about. One of them was a Premium tier customer, which bumped my recurring rate from 8% to 10% on that specific account. Month 3 totals (approximate):
- New articles published: 2 (7 total across the 90 days)
- New paid conversions: 4 (including 1 Premium tier)
- Recurring commissions from prior months: growing steadily
- Total 90-day earnings: crossed $90, with roughly half coming from recurring and the other half from first-order payouts Ninety dollars doesn't sound like a lot compared to my retainer gig. But here's the math that changed my thinking: I'd spent maybe six hours in month three actively writing affiliate content. Everything else was residual. If I'd spent another six hours writing more articles, that $90 could reasonably have been $150 or $200, all from the same recurring structure. That's when the freelance game stopped being purely about hours. # # What I Learned That Nobody Tells You One-time payouts are a trap. Every other affiliate program I joined paid once and disappeared. That model only works at massive scale. Recurring commissions let a writer with a small audience build real, compounding income. Comparison posts are the highest-converting format. My case study and beginner's guide did well, but the comparison article — the one that ranked on Google — generated the most clicks by far. People searching for "X vs Y" are buyers, not browsers. Cross-posting matters. Dev.to gave me roughly 60% of my total views for free. Without it, I'd still be stuck under $20 in earnings. Writers who only publish on their own blog are leaving traffic on the table. Don't bury the link. I tested placing my affiliate link in the conclusion versus mid-article. Mid-article, right after a specific recommendation, performed nearly three times better. The recurring check is the dopamine hit that keeps you going. That first $1.60 felt better than $50 from a one-time payout. Once you've seen recurring revenue land in your account, you start thinking about content differently. You write for longevity, not virality. # # The Honest Math: Is This Worth Doing? Let me put real numbers on the table. Across 90 days, I published seven articles totaling roughly 13,000 words. I spent maybe 35 hours on the entire project. My total earnings crossed $90, with about 65% of that from first-order commissions and 35% from recurring. Hourly rate on paper: about $2.50. Worst hourly rate of my freelance career. But the forward-looking math is what makes this interesting. My recurring commissions were climbing every month because older referrals kept paying. If I'd stayed at it — publishing one or two articles a month and refreshing old posts — projections suggest I could've hit $300 to $500 per month in
Top comments (0)