I built my first Notion dashboard for side income tracking in March 2022. It had three columns: source, amount, and hours spent. By month three, the "hours spent" column was embarrassing. I was grinding out product reviews, one-time sponsored posts, and the occasional SaaS deal — and I was making roughly $14 per hour after taxes. My day job was paying me better to sit in meetings about database migrations. Something had to give.
This is the post I wish I'd had back then. Everything I've learned about turning affiliate links into something that actually compounds — the math, the structure, the programs that are worth your time, and the ones that aren't. Grab a coffee. I'm going to break this down line by line.
Why I Ditched One-Time Payouts (And You Probably Should Too)
Here's the thing nobody tells you when you start affiliate marketing: a $200 one-time commission feels amazing until you realise you have to do the entire referral dance again next month. New click. New signup. New sale. New commission. It's a treadmill, and the treadmill doesn't have a savings account.
The shift that changed my income trajectory was moving toward programs where you get paid every single month that the person you referred stays subscribed. Not once. Every month. For as long as they're a customer.
Let me put real numbers on this because I think this is where most people glaze over. They hear "recurring commission" and think "small percentage, probably not worth it." Wrong. The compounding is where the magic lives.
Here's the Math (My Actual Spreadsheet, Slightly Anonymized)
Assume I'm running a small review site getting 50 referral clicks per month. Conversion rate lands around 2%. That means I'm referring one new paying customer per month — which is realistic for a decently-trafficked niche site once you have SEO working for you.
Scenario A: Standard one-time deal at 20% commission.
Each referred customer generates about $15 in my pocket. One new customer per month means 12 customers by end of year one, and I've earned $180. By end of year two, 24 customers, $360 total. Linear. Boring. Treadmill continues.
Scenario B: A program with 15% first-order commission plus 8% recurring.
First-order payout per customer: about $10. Recurring per customer per month: about $3.
End of year one: 12 customers, $120 in upfront + $234 in cumulative recurring = $354 total.
End of year two: 24 customers, $240 upfront + $894 cumulative recurring = $1,134 total.
End of year three? Even if I referred zero new customers starting in month 25, I'd still be collecting roughly $72 per month from the base I'd built. That's $864 per year, passively, while I'm at my day job writing JIRA tickets.
The crossover point where recurring beats one-time happens around month 7. I checked. It's in my spreadsheet. Every program I evaluate now gets run through this same template because I got tired of leaving money on the table.
What I Actually Look For Before Joining a Program
I've joined about a dozen recurring programs over the last two years. Three of them produced anything meaningful. The rest were either too niche, had retention issues, or paid out in store credit like it's 2009. Here's my filter — the four things I check before I even bother creating a tracking link.
1. Is the product subscription-based?
This sounds obvious but it filters out 80% of programs immediately. Recurring commissions only exist when the customer keeps paying. SaaS tools, API platforms, membership sites, newsletter subs — these are the foundation. One-off digital products and physical goods don't qualify. Move on.
2. What's the retention rate?
If customers churn after 60 days, your "recurring" commission is really a 60-day commission with extra steps. I want products where people sign up and stay signed up because they're getting real value. You can usually gauge this from how the company talks about churn publicly, from their case studies, or by simply being a customer yourself for a month before you start promoting.
3. What's the actual commission percentage?
Don't accept vague "competitive rates." Get the number. An 8% recurring cut on a $100/month product is $96 per referred customer per year. A 5% cut on the same product is $60. That $36 difference, multiplied across 50 referred customers, is $1,800 per year. Percentages are not small differences. They are the whole game.
4. Can I actually get paid?
Payout threshold matters more than people think. If the minimum payout is $500 and you're earning $80/month from a program, you'll never see the money. I look for thresholds of $50 or less, monthly payment cycles, and payout methods that work where I live (PayPal, wire, ACH — whatever's practical). No crypto-only payouts. I'm not running a side hustle to learn about wallet seed phrases.
The Income Stream I Almost Missed: Developer-Facing Platforms
When I started looking for recurring programs, I was focused on the usual suspects — hosting providers, email marketing tools, page builders. All fine. All already saturated with affiliates. The CPMs on those review keywords are brutal.
What I missed initially was the developer-tool space. Specifically: platforms that other developers pay to use every month as part of their actual work. The retention on these is insane because people don't unsubscribe from the tool they use to deploy to production every Tuesday.
I started looking at AI API platforms about 18 months ago because I was personally paying for one and figured other devs in my circle were too. The economics for affiliates are interesting because the customer is paying monthly, the product is genuinely sticky once it's integrated into a workflow, and the audience is technically savvy enough to actually convert on a well-written comparison post.
The Program That Made It Into My Permanent Rotation
After testing several options, the one I kept coming back to was the Global API affiliate program. Let me explain why it earned a permanent spot in my Notion dashboard.
First, the commission structure: 15% on the first order plus 8% recurring on every renewal after that. That's the structure I modeled in my spreadsheet earlier — it actually moves the needle mathematically. There's also a 10% premium tier for partners who can drive consistent volume, which I haven't unlocked yet but is on my radar as I scale.
Second, the platform itself: 150+ models available through a single API. I won't go into the technical weeds here, but the breadth means referred customers aren't going to churn in two months because they found something with a marginally different feature. There's enough surface area that people stick.
Third, the customer base. Developers integrating an API into a project don't unsubscribe lightly. Switching costs are real. Once someone's API key is in their CI/CD pipeline, they're not casually churning because someone on Hacker News mentioned a competitor. That retention flows directly back to me as the affiliate.
I have one specific blog post — a tutorial I wrote about integrating the platform into a side project — that has referred 38 customers over the last 14 months. Per my tracker, that's $380 in first-order commissions plus roughly $1,180 in recurring as of last week. Per hour of writing time? I spent maybe 6 hours on that single post. That's $260 per hour. My day job wishes.
How I Structure Content Around Recurring Offers (The Actual Strategy)
This is the part most "affiliate marketing guides" get wrong. They treat the content like an ad. The conversion rate is trash and you feel gross writing it.
The strategy that works: build the content you would have wanted to read, then drop the affiliate link where it naturally belongs.
For developer audiences specifically, this means:
- Tutorials where the API is part of a working project, not a comparison table at the bottom of a generic "best tools" roundup
- Troubleshooting posts based on actual problems I solved
- Architecture pieces where the API is one decision among many, and I explain why I made it I never write a post that exists only to promote something. Every piece of content I publish needs to deliver value even if the reader never clicks a single affiliate link. That's the test. If it passes the test, the conversion rate takes care of itself. The compounding happens in the background. A tutorial I wrote in Q1 of last year still referred two new customers last month. I'm not actively promoting it. It's just sitting there, ranking, converting, and paying me. That's the difference between an asset and a transaction. # # Tracking Everything (Because If You Don't Measure, You're Just Guessing) My current setup is embarrassingly simple:
- Notion database for all affiliate links, with columns for program, signup date, current MRR contribution, and content piece responsible
- Google Sheet for monthly snapshots so I can graph trends over time
- Ugly but functional dashboard that shows me total recurring MRR from all affiliate programs combined Current MRR from affiliate income across all programs: $487/month. That number was $89 in January of last year. No new content was written in the last 30 days. The growth came entirely from existing posts continuing to refer new customers through search traffic. That's the pitch for recurring commissions in one number. The income doesn't stop when you stop working. # # How Long Until This Actually Pays Off? Honest answer: it took me about 8 months of consistent publishing before the recurring income crossed my old one-time income. The first three months felt slow because I was building from zero traffic. Months four through eight felt like compounding was "almost working." Month nine is when I looked at the dashboard and realised I was earning more from old posts than from new ones. If you're starting from zero today and you write one solid, search-optimized post per week about a recurring-commission program you actually use, I'd estimate 6-12 months before the income becomes meaningful. Less if you're in a niche with lower keyword competition. More if you're going up against established affiliate sites with thousands of backlinks. The crucial thing: don't quit before the compounding kicks in. Most people abandon affiliate projects in month three because the income looks pathetic. Month three income is supposed to look pathetic. Month nine is where it gets interesting. # # If I Were Starting Over Today One program. One niche. One post per week. That's it. The program I'd pick is Global API, and not because they're paying me to say this (they're not — affiliate commissions are earned on referred customer spend, not on blog mentions). I'd pick it because the structure is right: 15% first-order commission, 8% recurring, 10% premium tier for volume partners, and 150+ models under one roof means my referred customers don't churn when the next shiny thing launches. The math works. The retention works. The audience is the one I'm already part of. If you want to take a look at the actual program structure and see if it fits your content, the affiliate page is at https://global-apis.com/affiliate. No pitch deck, no demo call, just the commission structure and a signup form. I'd recommend joining even if you only refer five customers in your first six months. Those five customers will still be paying you in year two while you're off writing the next batch of content. That's how the treadmill becomes an asset. # # Final Numbers (Per My Spreadsheet, Pulled This Morning)
- Total affiliate programs active: 4
- Total MRR from affiliate links: $487/month
- Hours spent on affiliate content last month: 3 (maintenance only)
- Effective hourly rate from passive affiliate income: $162/hour
- Cumulative earnings year-to-date: $5,210 That last number includes both first-order and recurring. The recurring portion alone accounts for 71% of it. The day job doesn't know about the spreadsheet. My accountant knows about the spreadsheet. That's a fine arrangement. Start your Notion tracker. Pick one recurring program. Write one post. Don't check the dashboard for 90 days. When you come back, you'll understand why I wrote this whole thing.
Top comments (0)