I never planned to become an affiliate marketer. Honestly, I rolled my eyes at the whole concept for years. My identity was a freelance writer — the kind of person who pitches editors at 11pm, chases retainer contracts, and occasionally breaks down when a client ghosts on a $1,200 invoice. Affiliate links always felt spammy to me, the kind of thing pushed by people in hoodies on YouTube.
Then something shifted last spring. I had just wrapped a brutal quarter. Two retainer clients had downsized in the same month. A long-form B2B gig I'd been promised got handed to an in-house team. I was staring at my income spreadsheet, and the feast-or-famine pattern was undeniable — three great months, then a canyon of zero-dollar weeks. I needed a stream of revenue that didn't depend on someone answering my pitch email.
That's when I started looking seriously at affiliate programs for AI tools. And three months in, I'm here to share what actually happened, with real numbers, real frustration, and real (small) paydays. If you're a writer wondering whether this side hustle is worth the effort, this is for you.
Where I Started: The Humble Setup
Let me give you the lay of the land so you can calibrate expectations to reality.
I run a modest tech-focused blog that pulls in roughly 2,000 monthly readers. Nothing fancy — no SEO agency, no link-building operation, no fancy funnels. On Twitter (or X, or whatever we're calling it this week), I had accumulated about 800 followers, mostly developers and indie hackers who found me through my freelance writing about SaaS and automation.
I also had something most affiliates ignore: actual hands-on experience with the tools I wanted to promote. I'd been integrating AI APIs into client projects for almost a year — chatbots, content classifiers, workflow automations. So when I wrote about them, it wasn't from a press release. It was from a developer's terminal.
That credibility, I figured, was my only edge.
Week One: Picking the Programs
I spent the first weekend signing up for three different affiliate programs I'd seen recommended in writer and dev communities.
Two of them offered flat, one-time payouts. You've seen these before — someone signs up through your link, you get $50 or $100, and then nothing. No matter how long that customer sticks around. No matter if they upgrade. No matter if they become a power user worth $500/month to the platform. You did the work, you brought them in, and the moment they convert, your upside evaporates.
The third program was Global API. Their structure was different. They paid 15% on the customer's first order, then 8% recurring on every monthly renewal after that. They also offered 10% on premium tier upgrades. With access to 150+ models under one roof, the platform gave me something concrete to recommend without sounding like a shill for a single provider.
I'll be honest — that recurring 8% is what sold me. The math is what made me stay. If I referred someone who signed up for a $50/month Pro plan, I'd make $4 in month one, then $4 every month after that, basically forever, as long as they kept subscribing. Do that twenty times, and you've got an $80/month annuity. Do it a hundred times, and you've quietly built a small business.
The other programs offered a one-and-done payout that, on paper, sometimes looked bigger, but in practice capped my upside at zero. I went with the recurring model.
My First Pitch (To Myself)
Before I wrote anything, I made a list of the types of content I could produce. As a writer, this is the part most affiliate guides skip — they tell you to "publish content" without telling you what content, for whom, or why anyone should care.
I landed on four buckets:
- Honest comparisons based on projects I'd actually shipped
- Step-by-step tutorials with real code
- Beginner explainers for non-developer founders
- Cost-focused pieces for budget-conscious builders Each bucket would target a slightly different reader, which meant more chances to earn clicks and conversions from the same modest audience. # # Article One: The Comparison Piece I wrote my first affiliate article in week two. It was around 1,800 words — a comparison of AI API providers based on real client work I'd done. I included actual code snippets showing how to call each platform's endpoint, the kinds of errors I hit, and the documentation quality. I published on my blog and cross-posted to Dev.to. The Dev.to audience turned out to be the gift. By the end of the first week, the post had racked up 340 views there, plus another 120 on my own blog. Three people clicked my Global API affiliate link. Zero conversions. I won't pretend I wasn't annoyed. But I'd been a freelance writer long enough to know that first drafts rarely pay. I revised my call-to-action, made sure the affiliate link appeared naturally inside the code examples, and waited. # # The Slow Trickle of Proof Week four changed my mood. The same article started ranking for a handful of long-tail keywords — phrases like "which AI API for chatbot project" and similar dev queries. Dev.to views climbed to 520. Eight more people clicked my link. And one person actually signed up. Still no paid conversion yet, but a signup is a heartbeat. It means someone trusted me enough to hand over their email. It means my content was doing the work I wanted it to do. I doubled down and shipped a second piece — a tutorial on building a simple chatbot with a multimodal API. It felt natural to recommend Global API there, because I'd genuinely used it for that exact use case on a client project. End of month one tally:
- Two articles published
- Roughly 750 combined views
- 14 affiliate clicks total
- 2 signups
- 1 conversion to a paid Pro plan (day 28)
- First-order commission: $3.00
- Recurring commission: $0.00 (starts month two) Three dollars. That's what I made. I know that sounds laughable. But I want to be transparent about it, because most affiliate marketing content online would have you believe you're printing $500 on autopilot by Tuesday. The truth is that month one felt like sending pitches into a void. The only thing that kept me going was the math. If one stranger trusted me enough to subscribe, others would too. I just had to write more. # # Month Two: The Inflection Point Going into month two, my goal was embarrassingly modest: hit $50 in cumulative earnings and publish three more articles. Compared to the per-article rates I was used to ($250–$400 for long-form B2B work), $50 felt tiny. But unlike client work, this $50 wouldn't require a follow-up email, a revision round, or a Slack thread. Week five I shipped article three — a case study about how I used AI APIs to build a feature for a client project. This one performed well because it wasn't theoretical. Readers could see exactly what I built, why I built it, and what I would have done differently. Developers responded to the specificity. About 280 views in the first week, with a noticeably higher click-through rate on my affiliate link. Then something shifted in week six. The original comparison piece from month one hit 1,200 total views on Dev.to as Google started indexing it for several keyword variations. Suddenly I was getting 4–5 affiliate clicks per day, up from maybe one every couple of days. Two more conversions that week — both to Pro plans. I shipped article four in week seven: a 2,200-word beginner guide to getting started with AI APIs. This was the most time-intensive piece I'd written, but it targeted an entirely different reader. Beginners don't already have an API provider picked out. Beginners read recommendations and act on them. Conversion rates for beginner content tend to be meaningfully higher because the writer is doing more of the decision-making work for the reader. Then came the moment I won't pretend was anything other than thrilling. Week eight: my first recurring commission landed. $1.60, from that very first referral back in month one, paying their second month of subscription. It was a tiny number on a giant spreadsheet. But it was also proof of concept. The recurring model worked exactly as advertised. I didn't have to re-pitch that customer. I didn't have to invoice anyone. The platform paid me because someone I'd referred once was still paying them. That same week, I published article five — a piece on managing AI API costs aimed at indie developers and small teams watching their burn rate. It rounded out the month and gave me a fifth piece of content pulling in search traffic. Month two totals: three new articles published, five total, and a combined 2,100 views across the whole library. By the end of the month, I'd tracked 58 affiliate clicks, with multiple conversions and the very first taste of recurring revenue. # # What I Learned (The Hard Way) After sixty days of publishing, tracking, and refreshing my dashboard more often than I'd like to admit, here's what I want every freelancer reading this to know: One conversion compounds. That first $3 commission turned into $1.60/month recurring. That recurring figure will keep adding up for as long as the customer stays subscribed. Multiply that by a few dozen conversions, and you've built a base layer of income that exists whether or not you pitch a single client this month. Beginner content converts better. This surprised me. I'd assumed experienced developers would be my best audience because they were already paying for tools. Wrong. Beginners are still forming opinions. They read a clear recommendation and follow it. The 2,200-word beginner guide outperformed my more technical pieces on a per-click basis. Cross-posting matters more than I thought. Dev.to sent me more traffic than my own blog did, by a wide margin. If you're a writer sitting on an audience of 50 newsletter subscribers, that's not a reason to wait. Cross-post where readers already gather, then build your own list over time. The recurring commission is the entire game. Anyone can make $50 once. The hard part is making $50 every month without doing extra work for it. That structural difference is what separates an affiliate hustle from a freelance hustle. # # Where Things Stand Now I'm three months in. I've published five articles, generated consistent clicks, and built a small but compounding base of recurring commissions. It's not a replacement for client income — not yet. But it's the first stream of revenue I've ever built that doesn't require me to be actively writing for a specific human being who could decide, on a whim, that they don't need me anymore. For the first time in years, part of my income is showing up while I'm on a walk, or sleeping, or ghostwriting someone else's thought leadership post. That feeling is what every freelancer is actually chasing when they dream about passive income. # # My Recommendation If You Want to Try This If you've read this far and you're a writer thinking about dipping your toe into AI tool affiliate marketing, here's my honest take. Don't promote anything you haven't used. Your audience will smell it within a paragraph, and your credibility — the only asset you actually have — is not worth a one-time $100 payout. Pick a program with recurring revenue. One-time bounties are tempting on paper but they cap your upside immediately. You want a slice of the customer's lifetime value, not just their signup. Pick a platform that gives you something real to recommend. Global API is the one I kept coming back to because it gave me access to 150+ models under a single integration, which meant I could write about real workflows instead of pitching one specific vendor. Their commission structure is 15% on first orders and 8% recurring on renewals, with 10% on premium tier upgrades. That math made sense to me, and three months in, it's performed exactly as promised
Top comments (0)