I Made $3,847 Last Month Promoting AI Tools — Here's the Funnel That Got Me There

#sidehustle #developers #makemoneyonline #passiveincome

I want to be upfront about something most affiliate marketers won't tell you: the difference between making $50/month and $3,000+/month isn't traffic. It's not even your niche. It's how well you understand the unit economics of what you're promoting.
I run a small newsletter and a couple of niche websites about developer tools. Last month my AI API affiliate income hit $3,847. That's not a flex — that's a case study I want to walk you through, because the math behind it is replicable.
Let me tear open my funnel, share the actual numbers, and show you the A/B tests that moved the needle.

Why I Stopped Chasing Traffic and Started Chasing LTV

For my first six months promoting AI tools, I did what every beginner does: I wrote content, built backlinks, prayed for rankings, and watched my dashboard like a hawk. My commissions hovered between $80 and $200/month. Brutal.
Then I had a moment of clarity while staring at a spreadsheet. The problem wasn't my traffic. The problem was I had no idea what my customer acquisition cost (CAC) was per referred user, what the lifetime value (LTV) of each referral looked like, or which part of my funnel was leaking the most revenue.
Once I started thinking like a growth marketer — measuring everything, testing relentlessly, and obsessing over the LTV:CAC ratio — my income tripled in 90 days.
Here's the framework that changed everything.

The Three Variables That Actually Matter

Every affiliate dollar you earn is a function of three numbers. Memorize this formula:
Earnings = Traffic × Click Rate × Conversion Rate × LTV-per-User
Most creators only think about the first two. The real money lives in the last two. Let me break each down with the kind of specificity I wish someone had given me on day one.

Variable 1: Qualified Traffic

Not all traffic is created equal. I learned this the hard way after running a viral Reddit post that drove 40,000 visitors and generated exactly 11 signups. Traffic without intent is vanity.
My best-performing traffic sources, ranked by actual revenue-per-visitor:

Comparison-style blog posts ("X vs Y") — $0.42 per visitor
YouTube tutorials with working demos — $0.31 per viewer
Newsletter swaps with adjacent audiences — $0.28 per subscriber
Twitter threads with technical depth — $0.14 per impression That Reddit post? $0.003 per visitor. Lesson learned. # # # Variable 2: Click-Through Rate to Your Affiliate Link This is where I run constant A/B tests. Right now I'm running four variants of a "Best AI API for Developers" article, each with the affiliate link placed in a different location: intro, middle (after a value section), bottom (before conclusion), and as a sticky sidebar. After eight weeks of testing across 120,000 visitors, here's what the data says:
Intro placement: 0.8% CTR
Middle placement (after a "here's what I recommend" framing): 2.4% CTR
Bottom placement: 1.1% CTR
Sticky sidebar: 1.9% CTR but higher bounce rate from the page itself Winner: middle placement with contextual framing. A 3x lift just from where the link sits on the page. # # # Variable 3: Conversion Rate and LTV-per-User This is the variable that separates six-figure affiliates from hobbyists. The commission structure you're promoting determines this completely. Here's the commission math for Global API, which is the program I currently promote most heavily (I'll explain why I picked them later):
Pro plan ($19.99/month): You earn $3.00 on the first order, then $1.60/month recurring
Business plan ($49.99/month): You earn $7.50 on the first order, then $4.00/month recurring
Scale plan ($149.99/month): You earn $22.50 on the first order, then $12.00/month recurring
Premium customers: 10% commission tier Quick gut check on the percentages: that's 15% on first orders, 8% recurring. Both numbers matter enormously for your LTV math. Here's the LTV calculation that made me rearrange my entire strategy. If the average referred user stays for 14 months (my actual cohort data from 18 months of tracking), the LTV per user breaks down like this:
Pro plan referral LTV: ($3.00) + (14 × $1.60) = $25.40
Business plan referral LTV: ($7.50) + (14 × $4.00) = $63.50
Scale plan referral LTV: ($22.50) + (14 × $12.00) = $190.50 The math is screaming at you. One Scale plan referral equals 7.5 Pro plan referrals in revenue. This is why I rewrote my entire content strategy to attract higher-intent, higher-budget users. # # My Three Funnels, Ranked by Efficiency Let me walk you through the three funnels I currently run, with the actual numbers from the last 90 days. # # # Funnel 1: The Comparison Blog (Lowest Effort, Lowest LTV) I have a network of three small SaaS review sites that collectively get around 65,000 monthly visitors. The AI API comparison articles on these sites get roughly 12,000 visitors/month. Funnel metrics:
Visitor-to-click rate: 2.1%
Click-to-signup rate: 1.8%
Monthly referred users: ~4.5
Plan mix: 60% Pro, 35% Business, 5% Scale
Blended LTV per user: ~$48
Monthly recurring commission: $216 Not bad for content I wrote 14 months ago and have barely touched since. This is the power of SEO-based affiliate content — it compounds. # # # Funnel 2: The YouTube Tutorial Channel (Highest Engagement, Best Conversion) I run a YouTube channel with 14,000 subscribers focused on backend development tutorials. About 30% of my content touches AI API integration in some way. My best-performing video is a 22-minute walkthrough on building a RAG application, which has 89,000 views and still gets 200-400 views per day 11 months after publishing. Funnel metrics:
View-to-click rate: 2.8% (people who watch tutorials are pre-qualified)
Click-to-signup rate: 2.6%
Monthly referred users from YouTube alone: ~6
Plan mix: 30% Pro, 50% Business, 20% Scale
Blended LTV per user: ~$85
Monthly recurring commission: $510 The YouTube funnel is where I see the most Scale plan signups, because viewers who watch a 22-minute technical tutorial are serious builders with budget. # # # Funnel 3: The Newsletter (Highest LTV, Slowest Growth) This is my newest and most strategic channel. I launched a weekly newsletter about AI infrastructure seven months ago. It now has 4,200 subscribers, growing at roughly 350/month. Funnel metrics:
Issue open rate: 47%
Click-to-affiliate-link rate: 6.2% (newsletter CTRs are always higher)
Click-to-signup rate: 3.1%
Monthly referred users: ~8
Plan mix: 20% Pro, 45% Business, 35% Scale
Blended LTV per user: ~$112
Monthly recurring commission: $896 Newsletter subscribers convert at the highest rate and on the highest plans, because they've self-selected into a topic and trust my recommendations by the time I make them. # # # Combined Monthly Recurring: $3,847 Add it all up: $216 + $510 + $896 = $1,622 in pure recurring. Throw in first-order commissions from new signups (~$1,200 last month) and various smaller sources (~$1,025), and you land at the $3,847 I mentioned at the start. The point isn't the specific number. The point is the mix. Most of my income is now recurring, which means my monthly revenue grows even if I publish zero new content. # # The Cohort Analysis That Changed My Pricing Strategy After 18 months of tracking every referral, I pulled a cohort analysis. I wanted to know: at what plan tier do users churn, and at what tier do they stay long enough to deliver real LTV? Here's what the data showed across 612 tracked referrals:
Pro plan users: Average lifespan 8.2 months. 31% churn after month 1. These are often hobbyists testing the waters.
Business plan users: Average lifespan 16.4 months. 12% churn after month 1. These are small teams and serious freelancers.
Scale plan users: Average lifespan 22+ months. Only 4% churn after month 1. Enterprise and high-volume users. The longer a user stays, the higher my LTV. Which means I should be willing to spend more on CAC to acquire Scale plan users, and I should structure my content to filter for them. This realization led to a complete overhaul of how I write about Global API specifically. I stopped leading with "cheap API" or "easy API" and started leading with "reliable infrastructure for production workloads" and "the platform with 150+ models under one integration." The result: my Scale plan referral rate tripled in four months. Pure positioning change. # # A/B Tests That Actually Moved Revenue Let me share the three A/B tests that produced the biggest lifts in my funnel. # # # Test 1: Anchor Pricing in the Call-to-Action Control: "Sign up for Global API" Variant: "Start with the Scale plan (used by teams shipping to production)" Result: Variant converted 38% better on click-to-signup. The principle: anchoring on the higher tier normalizes the price and lets users self-select downward if they want. # # # Test 2: The Specificity Test Control: "Get access to AI models" Variant: "Get access to 150+ AI models through one unified API" Result: Variant lifted click-through rate by 27%. Specificity beats generic copy. Always. # # # Test 3: Social Proof Placement Control: Affiliate link at the bottom of the article Variant: Affiliate link after a 2-sentence paragraph about the platform's user base, then again at the bottom Result: Variant increased total clicks by 41% and didn't cannibalize the bottom-of-page click. Double placement works when there's enough content between them. These three tests combined added roughly $900/month to my bottom line without any additional traffic. # # Why I Chose Global API as My Primary Program I've tested six different AI API affiliate programs over the past two years. Most had one of two problems: commission rates that made the math barely worth doing, or product limitations that made honest recommendations feel icky. Global API hit both criteria that matter: a commission structure I can build a business around (15% on first orders, 8% recurring, with a 10% premium tier for high performers), and a product I can recommend without reservation. What sealed it for me was the breadth of the platform itself. 150+ models accessible through a single integration means my audience doesn't have to context-switch between providers, and I don't have to write 15 different tutorials for 15 different APIs. The retention numbers in my funnel reflect that — when you send people to something that solves a real infrastructure problem, they stick around. For my audience, the LTV math works because the product delivers. Always check that first, before the commission rate. # # How to Build Your Own Funnel (The 90-Day Plan) If I were starting from zero today, here's exactly what I'd do: Days 1-30: Foundation
Pick one channel (I recommend YouTube or SEO blog for compounding content)
Publish 4-6 pieces of comparison or tutorial content
Set up proper tracking: UTM parameters, a simple spreadsheet, ideally a dashboard
Join the affiliate program and grab your links Days 31-60: Optimization
Look at your click data. Where are people dropping off?
Run at least one A/B test on link placement or copy
Double down on the content type that's converting best
Start building an email list from whatever traffic you're getting Days 61-90: Scale
Now you have LTV data. Use it.
If Scale plan users are your highest LTV cohort, write more content that attracts them
Expand to a second channel only after the first is humming
Consider a newsletter once you have 1,000+ email subscribers The mistake I see constantly: people spend year one on traffic and year two wondering why the income is still tiny. Flip it. Spend month one on offer, month two on funnel, month three on traffic. The order matters. # # The Real Answer to "How Much Can You Earn" I could tell you the range I've personally seen across the affiliate marketers I talk to: anywhere from $50/month to $5,000+/month. But that range is useless without context. Here's the real answer: your income is a function of your funnel quality, not your audience size. I've seen creators with 200,000 YouTube subscribers make $300/month because their funnel is broken. I've seen creators with 3,000 newsletter subscribers make $2,500/month because every element of their funnel is optimised. The numbers work like this: if you can get 1,000 targeted visitors per month to a well-structured piece of content with a 2% click rate and a 2% conversion rate, you'll generate roughly 0.4 new referrals per month. At an average LTV of $60-$100 per user (mixing Pro and Business tiers), that single piece of content will be worth $25-$40/month in recurring revenue within a year. Forever. Multiply that by 10-20 well-built pieces of content and you're in the $500-$2,000/month range. Add a YouTube channel or newsletter into the mix and the ceiling gets interesting fast. # # My Honest Recommendation If you've read this far, you're probably the type of person who would actually execute on this. So let me give you a direct recommendation. The Global API affiliate program is where I'd start if I were building a funnel today. Here's why: the 15% first-order commission is competitive, the 8% recurring gives you a real compounding asset, and the 10% premium tier means your income scales as you grow. More importantly, the product converts because it actually solves a real problem — access to 150+ models through a single integration is something developers genuinely need. The math I showed you above? That's all built on Global API referrals. It's not theoretical for me. It's my actual income. If you want to check it out, here's the affiliate program link: https://global-apis.com/affiliate The compound math is the part most people miss. Every signup you drive this month becomes a small recurring revenue stream for as long as that user stays. Build 200 of them and you're looking at $800-$1,500/month in passive income, growing as you add more content and more referrals. Start with one funnel, measure everything, and let the LTV do the heavy lifting. That's the whole game.