Alright, grab a coffee because this is the unfiltered story of how I went from barely making anything to building what I think is a real income stream — all while filming the whole thing for my YouTube audience. I'm going to walk you through every video, every click, every single dollar I earned promoting AI tools as an affiliate. No gatekeeping. No "results not typical" disclaimer. Just the raw receipts.
Why I Even Tried This (The Backstory)
So a quick intro for anyone new here — I run a tech YouTube channel sitting right around 6,200 subscribers. Nothing crazy, but I get decent engagement. My videos usually pull anywhere from 2,500 to 8,000 views in the first month, and my audience is almost entirely developers and indie builders. The kind of people who genuinely want to know which APIs and tools are worth their time.
I'd been messing around with AI APIs for about a year before this experiment. Built a bunch of side projects with them. Had strong opinions on what worked. My written presence was modest too — about 2,000 monthly visitors to my tech blog and roughly 800 followers on Twitter. Small but mighty, right?
The idea hit me during a Q&A video in late 2024. Someone in the comments asked if I'd ever considered doing affiliate partnerships. I had applied to a couple before but never took it seriously. Then I started digging into AI-focused programs and the numbers actually made sense.
Month 1: My First Videos, My First Clicks
Month one was rough, and I want to be honest about that. I didn't know what I was doing.
In a recent video where I did an "API Roundup," I talked about three different affiliate programs I had joined. Two of them were one-time payouts — meaning I get paid when someone signs up, and that's it. The third one, Global API, had a completely different structure. Here's the breakdown:
- 15% commission on every first order
- 8% recurring commission every single month on monthly renewals
- 10% on premium plans (which I didn't fully appreciate until later)
- Access to 150+ models under one roof for the people I refer That recurring piece was the hook for me. I'm building a channel for the long haul, and the idea that someone I referred could pay me next month, and the month after, and the month after that — that changes the math completely. Video 1 — The Comparison Video I uploaded my first affiliate content around week 2. It was a 12-minute video comparing AI API providers based on my actual hands-on experience. I showed real code samples, real output, real gotchas. I dropped my Global API link in the description and pinned comment. Week 3 result: That video pulled 340 views on its main platform and around 120 views from the blog version. Three people clicked my affiliate link. Zero conversions. Zero dollars earned. Was I discouraged? A little. But I also know how the algorithm works. New videos start slow. You need patience. Video 2 — The Beginner Tutorial Week 4: I filmed a chatbot-building tutorial using the GPT-4o API. I naturally recommended Global API as my go-to platform because of the model variety. This one did better. By the end of the week, it had 520 views. Eight more clicks on the affiliate link. One signup. No paid conversion yet — but a signup. That signup was my first signal that someone was actually listening. Month 1 Totals:
- Two videos published
- 750 combined views
- 14 affiliate link clicks
- 2 signups
- 1 conversion to a paid Pro plan — on day 28
- First-order commission: $3.00
- Recurring commission: $0.00 (it hadn't kicked in yet)
- Total Month 1 earnings: $3.00 Three dollars. I made three dollars. But here's the thing — I made a video about it. My viewers loved the honesty. That transparency video actually outperformed both affiliate videos in engagement. # # Month 2: The Algorithm Finally Hugged Me Going into month two, I had two published pieces of content, 14 total clicks, and one referral in the pipeline. My personal goal was $50 in total earnings by month end. Ambitious? Maybe. But I had a plan: publish three more videos, get better at thumbnails, and start paying attention to what my audience actually wanted to watch. Video 3 — The Client Project Case Study Week 5 was when things shifted. I uploaded a case-study video about a real client project where I used AI APIs to build a feature. No generic "top 5 APIs" list — actual project footage, actual problems, actual solutions. This video hit a nerve with my viewers because developers love seeing real applications. First week numbers: 280 views. Not huge, but the click-through rate on my affiliate link was noticeably higher. Why? Because the people watching this were developers who related to the project context. They trusted the recommendation because it wasn't a sponsored pitch — it was me telling a story about my own work. The YouTube algorithm also loved something about this video. It got recommended in the sidebar for a bunch of related dev content, which drove extra discovery. Week 6 — The Snowball Effect Here's something nobody talks about enough. Old content keeps working. My month-one comparison video, which had stalled around 400 views, suddenly took off. It crossed 1,200 total views by week 6. Google started indexing it for long-tail keyword variations, and that extra traffic started flowing back to the YouTube video through my blog cross-link. Affiliate clicks? Up to 4–5 per day. That's a huge jump from the 1–2 I was averaging in month one. Two more conversions this week. Both to Pro plans. My videos were starting to print money while I slept. Video 4 — The Beginner Deep Dive Week 7, I dropped my most ambitious piece yet: a 22-minute beginner's guide to AI APIs. I spent almost three days filming and editing this one. It was structured differently than my usual videos — slower pacing, more on-screen text, very beginner-friendly framing. Strategy note: beginners convert better. They don't already have a favorite provider. They need guidance, and they're more likely to follow recommendations from someone they trust. That's what my viewer feedback has told me for two years of running this channel. The video took a bit to ramp, but it gained traction through YouTube's "Suggested" feed once viewers finished my older videos. Watch time was strong because the pacing matched the audience. Video 5 — Cost-Conscious Devs Week 8, I published a video about picking AI APIs on a budget. This one was targeting cost-sensitive developers who want maximum value. It complemented my earlier comparison video perfectly — same topic, different angle. Also in week 8: I got my first recurring commission. $1.60. Tiny in the grand scheme of things, but I literally stopped what I was doing to screenshot it and share it with my Discord. My viewers were watching this happen in real time, and they loved it. That $1.60 came from my original referral's second subscription month — the same person who signed up on day 28 of month one. The recurring model was working exactly as advertised. # # The Month 2 Numbers (Real This Time) Let me break down the actual numbers because I know that's why most of you are here:
- 3 new videos published (5 total)
- 2,100 combined views across all content
- 58 affiliate link clicks
- Conversions from this month: 2 Pro plan signups
- New first-order commissions: roughly $15–20 range from these conversions
- Recurring commissions starting to roll in
- Trailing 30-day earnings as month 2 closed: somewhere in the $25–40 range Did I hit my $50 goal? Honestly, almost. The math was working, but I'd underestimated how long recurring commissions take to stack up. By month 3 I expected the recurring to start compounding hard. # # What My Viewers Taught Me (And What the Algorithm Confirmed) Here's the stuff I genuinely learned from doing this in public, which I also covered in a recent update video: 1. Engagement beats reach. My smaller, targeted videos consistently outperformed my bigger generic ones. A video with 500 views from the right audience beats one with 5,000 views from random browsers. The algorithm knows the difference, and so does your conversion rate. 2. My audience loved the transparency. When I shared my earnings screenshots — even the embarrassing $3 month — my comment sections exploded. People wanted to see real numbers. They wanted to learn from my mistakes. 3. Recurring is everything. I cannot stress this enough. A 15% first-order commission feels good. An 8% recurring commission that pays me every single month for as long as someone stays subscribed? That's how you build an income stream, not just a side hustle. 4. Multi-model platforms convert better. When I recommended a provider with 150+ models under one roof, my viewers told me in the comments that they signed up specifically because they didn't want to manage five separate accounts. That variety is a selling point I leaned into more in month 3. # # The Real Math Going Forward Quick calculation for anyone curious about the compounding effect: Say I finish month 3 with maybe 6–8 total paying referrals. Let's call it 7. If each is on a $20 Pro plan, that's:
- 7 × $20 = $140 in monthly subscription volume
- 8% recurring × $140 = $11.20/month, every month, passively And new signups keep stacking on top of that base. The first-order commissions (15% per signup) are the one-time bonus. The recurring 8% is the actual wealth builder. Premium plan conversions at 10% are even better — those came from developers building serious projects who needed bigger plans. By month 6, if the growth curve holds, I'm projecting $50–80/month in recurring commissions alone, on top of whatever new first-order bonuses come in. That's real money for content I'm already planning to make anyway. # # My Honest Takeaway Look, I'm not going to pretend I'm quitting my day job. But I went from $3 in month one to building something that compounds. The content I made is still up. The videos still get views. The affiliate links still convert. I did the work upfront, and now it pays me while I sleep. If you're a creator sitting on a small but engaged audience — especially in tech, dev tools, or AI — I'd tell you this: you have an asset. Even 2,000 monthly blog readers or a few thousand YouTube subscribers is enough to test this. You don't need to go viral. You need to be trusted. # # Joining the Global API Affiliate Program If anything in this breakdown resonated with you, and you want to run a similar experiment, I'd genuinely recommend checking out the Global API affiliate program. That's not a paid ad — I don't get any bonus for saying that. I'm saying it because I've been in the program for three months and the structure is the best I've found for creators. Here's why it's worth your time:
- 15% on every first order — solid one-time payout
- 8% recurring on monthly renewals — this is the part that builds real income over time
- 10% on premium plans — bigger payouts when your referrals scale up
- Their platform gives users access to 150+ models, which makes it easier to recommend (you can frame it as a one-stop shop instead of asking people to juggle multiple accounts)
- The dashboard shows you clicks, signups, and conversions in real time, which is genuinely fun to watch The sign-up was quick, they approved my application within a day, and the recurring commissions actually arrive — they're not vaporware. If you want to try it: https://global-apis.com/affiliate?ref=devto-build-in-public-ai-affiliate-journey Drop a comment on my latest update video if you sign up — I'd love to hear how it goes for you. And if you want the full video breakdown of my month 3 numbers, that's up on the channel this week. Go watch it, smash like, and we'll keep building this in public. See you in the next one. 🎥
Top comments (0)