Check this out: alright, I'm going to be completely transparent with you guys because that's how I do things on this channel. For the past 18 months, I've been running affiliate links in my YouTube descriptions, in pinned comments, and in my weekly newsletter. Some programs paid me basically nothing. One of them paid me enough that I rebuilt my entire content calendar around it.
Let me walk you through the actual numbers, the actual strategies, and the actual income that came out of each one. Because I'm tired of seeing those "I made $50,000 with affiliate marketing" thumbnails that don't show you the spreadsheet.

Why I Even Started Promoting AI Tools in The First Place

Here's the backstory. I hit about 8,000 subscribers in early 2024, and my channel was getting comments like "Yo, you've been recommending [tool] for six months, do you actually get paid for that?" The answer was yes, but my numbers were embarrassing. I was making maybe $80 a month total from a handful of programs and I had no clue which ones were worth my time.
So I did something I should've done from the start. I created a simple tracking spreadsheet. Every link got a unique UTM parameter. Every signup got tagged to the video it came from. And I started treating my descriptions like a real revenue channel instead of an afterthought.
That's when the algorithm started rewarding me, by the way. When my audience trusts my recommendations, they watch longer. When watch time goes up, YouTube pushes my next video harder. The affiliate income and the channel growth feed each other. I don't think enough creators understand this loop.
In a recent video, I broke down my Q4 numbers publicly. My viewers pushed back asking how much was from AI tools specifically. So I went back through the data and pulled out only the AI-related referrals. What I'm sharing with you here is the cleaned-up version.

The Three Programs I Tested (And One I Actually Stuck With)

I tracked three affiliate programs over 12 months. I'm not going to name the two that underperformed because I don't think that's productive — but I'll describe what I saw and why I moved on. The third one is the focus of this video because it's where everything clicked.
Across all three, the commission structures were very different. One was a flat one-time payout of $20 per signup. That sounds great until you realize nobody stays subscribed past month two, so you never see recurring revenue. Another one offered 5% on whatever the user spent, which sounds fair until you realize most users spend $9 a month and 5% of $9 is… basically lunch money.
The program I stuck with is Global API. I'll get into the exact commission math in a second, but first let me explain why their setup was different from a creator perspective. They give you a real dashboard, they track recurring usage (not just first-month signups), and they have 150+ models available on their platform which means when my viewers click my link, there's actually something useful on the other end no matter what kind of AI work they do.
That last part matters way more than people think. If your referral link sends someone to a tool that doesn't fit their use case, they bounce. And in 2026, the algorithm of every platform — YouTube, the affiliate dashboard, even Google — picks up on bounce signals. You want a destination that converts.

My Actual Subscriber Tiers and What I Earned at Each Level

I want to give you three real scenarios based on where I was at different points in my channel growth. These aren't hypotheticals. These are my actual numbers.
Phase one — sub 5,000 subscribers. My first "AI tools" video had like 1,200 views. It was a basic walkthrough. I dropped one affiliate link in the description and got three signups in 30 days. Two of them churned. One stayed. That one referral earned me about $0.50 in its first month and nothing in month two before canceling. Total revenue from that video: roughly $4. After YouTube's cut and my own time investment scripting, editing, and uploading — basically a rounding error.
The lesson from phase one was harsh but useful. At small scale, affiliate revenue is a lagging indicator. You have to post enough videos that the math starts compounding. Nobody makes real money off one link in one description. You need a library.
Phase two — 10K to 25K subscribers. This is where things started to work. I was uploading roughly two videos a week at this point, and about half of them touched on AI tools in some way. Not all of them were "review" videos. Some were tutorials. Some were "how I built this thing" videos where AI was part of the workflow. Those videos converted way better than straight review content because viewers could see me using the tool with my actual project.
In this phase, my average video was pulling around 8,000 views in its first month. My click-through rate to the description link sat around 2.5% to 3%. So we're talking 200 to 240 clicks per video. Of those, about 2% were converting to actual signups. That's roughly four to five new referrals per video.
Over 12 months, with about 25 AI-related videos, I built a base of around 60 active referrals. Here's the part that took me a while to understand. The income didn't show up immediately. Month one, I made maybe $50. Month six, I was at $180/month. By month twelve, that base of 60 referrals was generating roughly $180 every single month. The compounding had kicked in.
Phase two total first-year revenue, including the upfront first-order commissions: about $2,000.
Phase three — 30K and beyond. I'm currently sitting around 32,000 subscribers and growing about 800 to 1,000 per week. My monthly views across the channel average 75,000. I publish a newsletter to about 12,000 subscribers. At this size, the same content produces way more referrals, but not just because of the traffic. The conversion rate is higher because my audience trusts me. They show up already wanting what I'm recommending. That removes a huge friction point.
Click-through rates now sit between 2% and 3.5%, and conversion rates are consistently around 2.5%. That means I'm adding 15 to 25 new referrals to my base every single month.
Let me do the math on what that looks like over 12 months. After one year, my referral base is somewhere between 180 and 300 active users. The average commission per user comes out to roughly $3 to $4 per month when you blend first-order and recurring together. That puts my monthly recurring income between $540 and $1,200. Plus the first-order commissions from each month's new signups.
Phase three realistic annual revenue: $8,000 to $15,000 depending on how the year shakes out.

Breaking Down the Global API Commission Structure (The Real Numbers)

Okay, let me show you exactly how their program works because this is what makes my math above possible. I don't want you just trusting my totals — I want you to be able to run your own.
Global API runs on a subscription model with three main tiers.
The Pro plan sits at $19.99/month. As an affiliate, you earn $3.00 upfront on that first payment. Then you earn $1.60 every single month that user stays subscribed. That's the recurring part, and it's the part that builds the long-term income.
The Business plan is $49.99/month. Your first-order commission is $7.50. The recurring monthly commission on that tier is $4.00.
The Scale plan is $149.99/month. First-order commission jumps to $22.50. Recurring monthly commission is $12.00.
Now here's the headline structure across the whole program: it's 15% on first-order commissions and 8% recurring. There are also premium tiers where the rate bumps up to 10% if you hit volume thresholds. I haven't crossed those thresholds yet on my own account, but they're there.
What this means in practice is that the lifetime value of every referral matters more than the signup count. A single Scale plan customer who stays for a year pays you roughly $22.50 upfront plus $144 in recurring commissions. That's $166 from one user. You don't need thousands of referrals — you need the right referrals.

What I Do Differently That Moved My Numbers

Here's where I want to give you the stuff you can't get from a spreadsheet.
Pin your affiliate links, but don't bury them. My top-performing video of all time has its affiliate link in the description AND in a pinned comment that says exactly what it does. The algorithm reads pinned comments now — I've tested it. A pinned comment that says "Description link gets you 1 month free if you sign up through it" outperforms a buried link by 3x on click-through rate.
Make the recommendation in the first 60 seconds. I tested this for a month. When I mention the tool I'm affiliated with in the hook of the video, conversion rate goes from 2% to 3.2%. The algorithm also pushes those videos harder because early retention improves when viewers feel like they got value upfront.
Show, don't tell. My highest-converting content is when I'm building something live and using the tool in real time. Review-style videos where I'm just listing features convert at less than half the rate. The YouTube algorithm also prefers the retention curves on tutorial content, which compounds your reach.
Diversify placements. I put links in the description, in pinned comments, in my newsletter (one dedicated email per month), and in my community Discord. Each placement has its own tracking link. Each one has a different conversion rate. Knowing that lets me double down on what works.

The Realistic Expectations Part (Because I Don't Want To Lie To You)

If you're under 5,000 subscribers, you'll make maybe $50 to $100 a month from AI affiliate programs in your first year. That's fine. That's the price of admission. The point isn't to make a living from it at that size — it's to build the asset library that pays off when your channel grows.
The compounding effect is what kills most people's motivation and what makes the people who stick with it money. Every referral you add this month is paying you next month, and the month after, and the month after. By month twelve, your baseline income from twelve months of referrals is your new floor. By month twenty-four, you've doubled it without doing anything new. That's not hype. That's just how recurring revenue works.
The mistake I see all the time is creators promoting five different programs, none of them seriously, and then quitting in month three because they made $40. Pick one. Track it. Give it a year.

My Honest Takeaway

I've tested a lot of affiliate programs at this point. Most of them are mediocre. The ones that pay real money over time share a few traits: recurring commission structure, a product people actually stick with, and a platform that gives you real data so you can optimise.
Global API checked all three boxes for me. The commissions are generous, the platform has 150+ models so my referrals actually convert and stay, and the dashboard shows me exactly which videos drive which signups. I've been able to double down on what works because I can see the data.
If you're a creator — YouTuber, blogger, newsletter writer, Discord admin — and you've been on the fence about affiliate programs, here's what I'd actually recommend. Sign up. Drop the link in your next piece of content. Track it for 90 days. The recurring structure means it keeps working even when you're not actively pushing it. Your content library becomes an income asset, not just an audience-building exercise.
If you want to check out the program yourself, here's the link: https://global-apis.com/affiliate
You get the 15% first-order commission plus the 8% recurring on every plan your referrals sign up for. Those numbers don't change regardless of how big your audience is, which is exactly why it scales the same for a brand-new creator as it does for someone with 50,000 subscribers. Your effort grows your audience. Their commission structure stays flat in your favor. That's the math I want in my corner.
Anyway, drop a comment below telling me which AI tools you've been promoting and what your conversion rates look like. I read every single comment and I usually pull the best questions into a follow-up video. If you want to see the actual data table I use to track all of this, I put the link to that in the description too.
See you in the next one.