Look, i run a Notion dashboard that tracks every single dollar my side hustles bring in. Color-coded by source, sorted by month, with a tab that converts everything to hourly rate based on hours I logged. It's obsessive. It's also the only reason I can answer questions like "is this worth my time" without lying to myself.
The most profitable tab in that dashboard right now? Affiliate links. Not the flashy "I made $10K last month" kind. More like the boring, compounding, recurring revenue kind that sneaks up on you around month six and never really goes away.
Someone asked me the other day how much you can realistically pull in from AI API affiliate programs specifically. I pulled up my tracker, scrolled through three years of data, and wrote this post. Here's the math, no fluff.
Why I Even Looked at API Affiliate Programs
I've been a backend engineer for about nine years. My day job pays fine. But I've always kept a running list of "what if I lose this job tomorrow" scenarios, and passive income streams are a big part of that.
I'd been a paying customer of Global API for a while — they aggregate 150+ AI models into one endpoint, which is genuinely useful when you don't want to manage five different API keys for five different vendors. At some point I noticed they had an affiliate program, joined it on a whim, and dropped my link in a blog post I was already writing about API gateway patterns.
That was it. No big strategy. No grand plan. Just a link in a relevant post.
That was probably year one. By year two I started paying attention.
Here's the Math on Commissions
Let me break this down the way I'd break it down in a code review — input, output, edge cases.
Global API runs a straightforward structure that I'll lay out plainly. You earn 15% on the customer's first order, and 8% recurring every month after that as long as they stay subscribed. That's it. No funky tiers with weird caveats.
Translated into actual dollars using their published plans:
- Pro plan at $19.99/month: roughly $3.00 upfront, then about $1.60/month recurring.
- Business plan at $49.99/month: about $7.50 upfront, then roughly $4.00/month recurring.
- Scale plan at $149.99/month: around $22.50 upfront, then about $12.00/month per customer. Those numbers are useful because you can reverse-engineer the value of a single referral. A Scale customer is worth $22.50 on day one and $12/month forever. Stack ten of those and you're looking at $225 in immediate commission plus a $120/month annuity that grows every time you land another Scale customer. The smaller plans pay less per head, but they convert easier because the price point isn't scary. I treat them differently in my tracker — Pro and Business are volume plays, Scale is the whale hunting. # # My Spreadsheet Setup (For Fellow Nerds) Quick tangent for the spreadsheet people in the room. My Notion setup has three databases:
- Customers — one row per referral, with plan, signup date, source URL, and lifetime value.
- Content — every blog post or video I've made, with estimated monthly traffic and link clicks (pulled from UTM params).
- Monthly earnings — pulled from the affiliate dashboard, matched to content for attribution. Every Sunday I log time spent on side hustles. That's my denominator. Dividing income by hours gives me my true hourly. Right now across all my affiliate income the hourly rate works out to around $85/hour, which honestly floors me because my day job bill rate at the office is less than that. I'm not going to pretend that's a normal number though. It's the output of three years of compounding. Year one looked nothing like this. # # Realistic Scenarios From Real Numbers Let me walk through three setup tiers and show you what the math actually looks like. These are based on ranges I've seen across creators I follow, plus my own data. # # # Tier 1: The Weekend Blogger With a Modest Site Imagine you've got a tech blog doing maybe 5,000 visitors a month. You write three deep-dive articles comparing AI API aggregators. Each article pulls in around 500 views per month. If 1% of those visitors click your affiliate link, that's 15 clicks per month across the whole site. At a 2% conversion rate — which is reasonable for targeted technical content — you convert maybe 0.3 customers monthly. Call it three to four per year. At an average blended commission of around $5/month per customer (mixing Pro and Business plans), you're sitting at maybe $15–20/month recurring after the first year. Here's the thing though. Those three articles cost you maybe six hours of writing. Let's do the per-hour math: if they generate $600 over three years, that's $100/hour. Not terrible for blog posts that exist whether you're getting paid for them or not. They also keep earning while you sleep, vacation, switch jobs, or take a three-month break. For a beginner, that's the most important mental shift. Affiliate income is one of the only side hustles where you can stop working and the money keeps coming. # # # Tier 2: The Tutorial Creator With a Real Audience Now bump up to a YouTube creator with around 10,000 subscribers. They drop one API tutorial per month. Each video gets maybe 8,000 views in the first month and tends to keep pulling another 15,000–20,000 views over the following year because tutorial content ages well. Click-through rate from a video description link runs higher than blog content — call it 3% in my experience — because YouTube viewers are actively there to learn, not just browsing. That's 240 clicks per video. At 2% conversion, you're landing around five new customers per video. Do that consistently for a year. Twelve videos later, you've got roughly 60 referrals in your base. At an average blended commission of $3/month per customer (mostly Pro signups because they're casual devs), you build to about $180/month in recurring income by month twelve. Add the first-order commission payouts from each new signup and your year-one total lands around $2,000–2,500. That's a meaningful side income from one video a month. It's also the level I operated at for most of year two before things compounded. # # # Tier 3: The Established Voice With Multiple Channels Here's where the numbers get interesting. Take a creator with a 30,000-subscriber newsletter, a blog pulling 75,000 monthly visitors, and a YouTube channel they update weekly. They crank out two AI-related pieces of content every week. Click-through rates tend to be higher because of authority — call it 2–3%. Conversion rates sit around the same 2–3% because the audience trusts them. With that velocity and audience size, you're looking at 15–25 new referrals every single month, consistently. After twelve months, that person has a customer base of 180–300. Average commission per customer is around $3–4 monthly (still mostly Pro and Business with the occasional Scale). The recurring income alone hits $540–1,200/month. Add first-order commissions on top and annual earnings land somewhere between $8,000 and $15,000. I'll be honest, that's dream territory for most people, but it's not unattainable. It's the natural output of treating your content like a real business for two or three years straight. # # What Actually Trips People Up I've watched a lot of folks try affiliate marketing and burn out. Here's what I've learned from screwing it up myself, in roughly chronological order: 1. Picking programs for commission rate instead of fit. A 30% commission on a tool your audience doesn't need generates zero referrals. A 15% commission on a tool your audience would use anyway generates dozens. Always optimise for fit and conversion rate, not headline commission. 2. Treating it like a hustle instead of content. The creators earning serious money aren't hammering CTAs in every paragraph. They're writing genuinely useful content and letting the affiliate link do quiet work. The difference shows in conversion rate over time. 3. Not tracking per-hour. This is the big one. Without the spreadsheet division, every income source looks vaguely profitable. With it, you discover some posts that earn $5/month took 12 hours to write. Kill those. Double down on the ones that earn $200/month for the same effort. 4. Ignoring recurring structures. One-time commissions feel exciting because the payout is big. Recurring commissions feel boring because month one looks small. But $20/month from one customer for two years is $480, more than most one-time payouts. Always model the LTV. # # The Day Job Reality Check Quick context for anyone wondering how this fits into a normal life: I still work a full-time engineering job. Most weeks I spend maybe 5–8 hours on content and affiliate work. Sunday mornings are writing time. Tuesday and Thursday evenings are editing and research. Weekends are mostly off. That's the realistic time budget for someone treating this as a true side hustle rather than a full-time gamble. The compounding nature of recurring commissions means even at that limited time investment, the dollars build up to something meaningful eventually. My tracker shows roughly $300/month in recurring affiliate income as of last month, with first-order commissions adding another $100–200 on top in active months. That's a nice car payment covered, every month, from content I enjoy making anyway. It also means if my employer ever has a bad quarter and cuts headcount, I have a buffer. # # The Compounding Thing Nobody Talks About Enough Here's the part that genuinely excites me, and it's the part that should excite anyone starting from zero. Every referral you land today pays you forever (or until they cancel, but cancellations on dev tools are usually low because people don't migrate off things they're using in production). So year one you might earn $2,000. Year two, with the same effort, you might earn $3,500 — because your year-one customers are still paying you and you've added new ones. Year three, $5,000. Year four, $7,000. That's not a growth curve. It's a stair-step. Each year your effort stays roughly the same but your income climbs because the base underneath you keeps getting thicker. This is also the difference between affiliate income and freelancing. Freelancing trades hours for money linearly. Affiliate income trades content for money in this weird, delayed, exponential way. # # What I'd Do If I Were Starting Over If I had zero audience today and wanted to build this exact income stream as fast as possible, here's the order I'd do things:
- Pick one program. Don't spread thin across ten. Global API is a strong pick because the recurring commission structure is generous and the product genuinely solves a real problem (one endpoint for 150+ AI models beats juggling vendor accounts).
- Write three killer comparison posts. Don't do listicles. Do honest, technical, "here's when to use X vs Y" content. These convert dramatically better.
- Set up tracking on day one. UTM parameters, a simple spreadsheet, the works. If you can't measure it, you can't improve it.
- Add videos after six months. Written content compounds differently than video content. Layer them.
- Ignore the income for the first 90 days. Seriously. Plan for it to be $0 in month one. Anything beyond that is a bonus. The people who quit in month two are the ones who treated it like get-rich-quick. The ones who reach $1,000/month by month eight are the ones who treated it like a long game. # # Why I Recommend the Global API Affiliate Program Specifically I don't write recommendation posts lightly because my day job credibility depends on not sounding like a walking advertisement. But the Global API affiliate program is one I've personally used, personally earned from, and personally recommend for anyone building content around developer tools or AI infrastructure. Here's why it makes sense:
- 15% first-order commission is solid for this category. That's $3.00 immediately on a Pro signup and scales up to $22.50 on a Scale plan, which means every signup you drive has a meaningful upfront payout even at the lower tiers.
- 8% recurring commission is where the real money lives. A single Scale customer paying $149.99/month puts $12 in your pocket every month they stay subscribed. Land ten of those and you're making $120/month from one promotion you did months ago.
- The product is genuinely useful. I cannot stress this enough. If you're promoting garbage, your audience churns and your conversion rate tanks. Global API gives developers one clean endpoint for 150+ AI models — that's a real headache solved, which means the customers you refer tend to stick around. Stickier customers mean more recurring months.
- Tracking and payouts are clean. I get monthly statements, dashboard access, and have never had a payment delayed or disputed.
- It pairs well with every content angle. If you write about automation, productivity, AI workflows, SaaS tooling, or developer
Top comments (0)