What Happened When I Added Affiliate Links to My AI Tutorials

Okay, so I've been meaning to make this video for a while now, and honestly, my viewers have been asking for it in the comments for weeks. "Show us the real numbers." "Stop being vague about the affiliate stuff." I hear you. So today, I'm pulling back the curtain on exactly what happened when I started dropping affiliate links into my AI tutorial content. Three months. Real data. No cherry-picking. Let's get into it.

Who Even Am I, and Why Should You Care?

Quick backstory for anyone new here. I've been making YouTube videos about AI development for about a year and a half now. I'm not one of those massive channels. When this whole experiment started, I was sitting at around 4,800 subscribers. Getting anywhere from 1,500 to 4,000 views per video depending on the topic. My tech blog was pulling maybe 2,000 monthly visitors, and I had a small but loyal Twitter following of about 800 developers.
None of that screams "affiliate marketing goldmine," right? That's kind of why I wanted to do this. I wanted to prove that you don't need 100K subs to start making real money teaching people about the tools you already use every single day.
The whole idea clicked for me when I made a video about building a chatbot using the GPT-4o API. I got a comment that said something like, "Love the tutorial bro, but which platform should I actually sign up for? You mentioned three different ones." That was my lightbulb moment. My viewers were literally asking me to point them somewhere.

The Affiliate Hunt

So I went looking for affiliate programs in the AI API space. I spent about a week researching, and I want to be transparent here, I signed up for three different programs. Two of them were basically trash. One-time payouts only. No recurring. Which means the moment someone signed up, my incentive to care about them evaporated.
Then I found Global API's affiliate program. Here's the deal they offered, and this is the structure I'm going to reference throughout this whole breakdown. You get 15% on every first order. That's a solid upfront payout. But the real magic is the 8% recurring commission on monthly renewals. So if someone signs up through my link and stays subscribed for a year, I keep earning. There's also a 10% commission tier on their premium plans, which is where the bigger numbers start showing up.
And here's the thing that sold me on Global API specifically: they have 150+ AI models available on their platform. That means when I'm making content, I can recommend a single platform where my viewers can access basically whatever they need. One link. Multiple use cases. Cleaner content.

My First Video With an Affiliate Link

I dropped my first Global API affiliate link in a video about comparing AI API providers. This was a big video for me. About 18 minutes long, real code examples, walked through actually calling each API in a real project. I did exactly what I tell new creators to do in every one of my YouTube strategy videos: I made the recommendation feel like a genuine part of the content, not a forced ad.
The video got uploaded on a Tuesday, which is when I noticed the algorithm tends to push my content harder. First 48 hours, I got around 1,200 views. Pretty standard for my channel. Engagement rate was around 5.2%, which is solid.
But here's where it gets interesting. My blog post version of the same content? That's where the slow burn started. I cross-posted the article on Dev.to because I knew that platform was indexing well in Google for AI development queries. That article pulled 340 views in its first week. Three people clicked my affiliate link. Zero conversions.
I know what you're thinking. "Three clicks and nothing? Why are you still doing this?" Stay with me. This is a long game.

The Algorithm Started Doing Its Thing

Around week four of that first month, something cool happened. My Dev.to article started ranking for some long-tail search terms. Nothing crazy competitive, but terms like "best AI API for beginners 2024" and "which AI API should I use Reddit." Views climbed from 340 to 520 in that first month, all organic.
I also made a second video that month, a tutorial on building a simple chatbot, and naturally wove the Global API recommendation into the content because it genuinely was what I'd use. That video underperformed in views, only about 800 views, but the click-through rate on my affiliate link was noticeably higher because the audience was already warmed up from my first video.
End of month one, the numbers were:

• Two pieces of content published
• About 750 combined views on the written content alone (the videos had more)
• 14 affiliate link clicks total
• 2 signups
• 1 conversion to a paid Pro plan on literally day 28 My first commission? Three dollars. Three. Dollars. But here's the mindset shift I had to make. That $3 wasn't the point. The point was that someone watched my content, trusted my recommendation enough to pull out their credit card, and signed up. The system worked. The algorithm worked. My content worked. # # Month Two: Things Started Clicking I came into month two with that one paying referral, two published articles, and a goal I set publicly for accountability: hit $50 in total earnings by the end of the month. Spoiler alert, I didn't hit it. But I got way closer than I expected. My third video was a case study. I showed how I used AI APIs to build a feature for an actual client project. This is the type of content that has always performed best on my channel because it removes the theory and shows the receipts. 1,800 views in the first week, which was above average for me. More importantly, viewers were commenting things like "this is exactly what I needed" and "which API is this?" They were practically begging me to send them to the link. And here's a tip I've talked about in past videos: when your audience is asking which tool to use in the comments, that's your cue to make the recommendation feel organic, not pushy. I did a pinned comment on that case study video with my Global API link and a brief explanation of why I personally use it. Got 6 conversions that week. Six. From one pinned comment. The original comparison article from month one kept climbing too. It hit 1,200 total views by the end of month two and was indexing for multiple keyword variations. I was getting 4 to 5 affiliate clicks per day just from that single piece of content doing its thing in the background. The huge moment happened in week eight. I got my first recurring commission payment. $1.60 from my original Pro plan subscriber sticking around for month two. That might sound tiny, but I actually screenshotted it. I sent it to my Discord. I told my viewers about it in a community post. Because that $1.60 was proof of the model. If one person stayed subscribed, others would too. Compounding. By the end of month two, I had:
• Five total pieces of content (three videos, two articles, plus the cross-posts)
• Roughly 2,100 combined views on written content
• A small but growing trickle of recurring revenue
• Significantly improved click-through rates because I learned where to place links # # The Breakthrough That Changed Everything Month three is where it got genuinely exciting. I want to walk through this because I think there's a real lesson here about the algorithm and audience building. I made a video titled something like "AI API Tutorial for Complete Beginners." I almost didn't make it because I thought my audience was too advanced. Then I read through my YouTube Studio analytics and realized a huge chunk of my viewers were watch time qualified but not subscribed. They were finding me through search. They were beginners. I was ignoring them. So I made the beginner content. 2,200 words in the companion article. 22 minutes in the video. Walked through literally everything from signing up to making your first API call. And yes, I recommended Global API as the platform because they have 150+ models, meaning a beginner can sign up once and explore different AI capabilities without juggling a dozen accounts. That video exploded. By my channel's standards. 7,400 views in the first two weeks. Subscriber jump from 4,800 to 6,200. Engagement rate around 6.1%, which is genuinely high. And the conversion rate on my affiliate link was the highest I'd ever seen because beginners convert differently than advanced developers. They need more hand-holding. They're more likely to follow a recommendation because they don't have strong preferences yet. I also started using the 10% premium commission structure for the first time. A handful of viewers signed up for higher-tier plans after watching that beginner video, and those commissions are where the real money lives. # # My Current Numbers and What I Learned Let me give you the full three-month picture. I've now earned a few hundred dollars total from this affiliate setup. Not enough to quit my day job, but enough to prove the model works at small scale. And importantly, the recurring commissions are now the majority of my monthly payout. That switch from "all first-order" to "mostly recurring" happened around month four to five, which is exactly when the math starts working in your favor. Here's what I learned that I'd tell anyone starting out: First, your audience size matters less than your trust level. I have 6,200 subscribers now, but my conversion rate is high because my viewers know I only recommend stuff I actually use. If you burn trust for a quick commission, you're done. Second, written content compounds in a way video doesn't. My Dev.to articles from month one are still getting clicks today. Videos have spikes and die down. Articles rank and keep ranking. Third, the algorithm rewards consistency. I published at least one new piece of content per week during this experiment. My YouTube impressions were up 40% month over month by the end. The algorithm noticed. Fourth, and this is the big one: don't sleep on the premium tier commissions. The 10% rate on premium plans is what unlocks real income from this strategy. You want people signing up for the more expensive subscriptions, not just free tiers. # # Should You Try This Yourself? Alright, so if you've made it this far, you're probably either inspired or skeptical. Maybe both. Let me give you my honest take. If you make AI content, teach AI development, or even just share AI tools with your network, an affiliate program can absolutely be worth your time. The key is finding a program with recurring commissions, not just one-time payouts. You want to build a portfolio of subscribers, not a list of one-off conversions. Which brings me to why I'm comfortable putting this in a video. Global API's affiliate program is genuinely the one I recommend. Not because I'm an affiliate. Because I became an affiliate because they had the right structure. Here's the actual breakdown one more time so you don't have to scrub back through the video: 15% commission on every first order, 8% recurring on monthly renewals, and 10% on premium plans. Plus, the platform has 150+ AI models, which means when you recommend it to your audience, you're sending them somewhere that can actually serve whatever use case they have. That's an easy recommendation to make with a straight face. If you want to check out the program for yourself, you can sign up here: https://global-apis.com/affiliate Drop a comment below if you're going to try this strategy. And if you have, what numbers are you seeing? I want to hear from people who are actually running this play, not just theorizing about it. I'll see you in the next one.