What I actually sold

With that clear, the revenue lines almost designed themselves, and I built three.

The lowest-hanging fruit was spaces. Venues and businesses want to be seen, so I wired up an advertising system based on rankings and clicks. They pay to rank higher and to be discovered, which is the most natural thing in the world on a platform whose whole purpose is discovery.

Second, the usual verification charge. A trusted, verified badge that says this business is real and accountable.

Third, I looked at my users and reasoned that the ones who could afford it would want a premium signal and extra visibility. And before you tell me people will not pay for something so intangible, let me remind you that, for reasons I will go to my grave not understanding, people pay actual money to promote their Snapchat. My research also showed that platforms like Tinder happily charge a premium purely for more visibility. So I built that in too.

Utility is not always practical. People pay for status and visibility as readily as they pay for goods. Charge for the intangible thing your platform makes people feel, not just the tangible thing it does.

Choose your payment rails carefully

Here is where the actual teaching begins, because how I take the money turned out to matter far more than what I charge for it.

I started crude. Embarrassingly crude. Businesses would fill out a form and send it to me along with proof of payment, and I would sort them out by hand. And I did this deliberately at first, because I was trying very hard not to step on Apple and Google, who both collect rent, a cut of as much as thirty per cent, on sales made through an app downloaded from their stores. My honest reaction to that was: I should hand Apple thirty per cent of my money? I am just a kid, please.

But on second thought, the manual form-and-receipt system was obviously never going to work. It does not scale, it is full of human error, and it makes you look like exactly the amateur you are trying not to be. So I went looking for what would actually work.

Does the Apple and Google 30% apply to you?

Before I tell you what I found, one accurate note that will save you grief, because the Apple and Google fear is half-true and the half people get wrong is expensive.

That thirty per cent applies to digital goods and services consumed inside the app. Real-world goods and services — a ticket to an actual event, promotion for a physical venue, a voucher you redeem in person — are generally not subject to that in-app commission at all, and you are allowed to use an outside processor for them. That is precisely why my model is clean. But if you sell a purely digital, in-app premium upgrade to an iPhone user, that one can still fall under Apple's rules, so check your specific case.

Which leads me to my single most important piece of advice in this whole section, and I will repeat it at the end: research the laws of your sector properly, so you do not end up in jail or broke.

Why Paystack was the answer

What I found was Paystack, and it was perfect.

It offered the easiest integration known to man. They now even have an AI helper you can ask to hand you the exact code to paste into your AI chat, which means the integration becomes the same conversation as everything else in this journey. You build and test with a test key so no real money moves while you are checking everything works, and then, when you are happy, you simply swap in the live key and you are taking real payments. That test-key-then-live-key pattern is your safety net; use it every single time.

One discipline to add on top: keys never belong in your source code. Treat sk_test_xxx and its live sibling as environment variables your app reads at runtime, so the secret never lands in a repo, a screenshot, or a chat with an AI assistant.

But the integration being easy was the least interesting reason I chose it. The real reasons were structural.

How Paystack split payments work for a marketplace

I did not only want to charge businesses for ads. I wanted businesses to be able to sell tickets and vouchers to my users directly. And the moment you allow that, you create a nightmare for yourself, because now money is flowing in for dozens of different vendors, and somebody has to pay each of them their share.

I did not want that somebody to be me, manually sending out a hundred transfers, when I could be on a beach, or in Monaco, or, ideally, on a beach in Monaco.

Paystack's subaccount feature solves this completely, and here is the beautiful part. You can use a webhook so that a vendor simply fills out a form, and that submission automatically sets them up as a subaccount on your Paystack, pending your approval. Once you approve them, every payment made to that vendor goes straight to them, while your agreed share is split off to you automatically. You never touch their money. You can have the Paystack API handle this whole webhook flow for you.

In plain English: a vendor fills a form, the system enrols them, you approve, and from then on the money divides itself correctly, forever, without you lifting a finger.

The right infrastructure does not just process your payments. It removes entire categories of work and risk from your life. Always ask not "can this take a payment" but "what work does this delete."

The quiet wins of a good payments partner

Choosing Paystack handed me a stack of advantages I did not fully appreciate until later, and I want you to see them because they are the kind of thing beginners miss.

You avoid becoming a money-handler. Because you never hold anyone else's money, you sidestep a whole regulatory swamp. Holding other people's funds turns you into something the law takes very seriously. Letting the money split directly to vendors means Paystack carries that responsibility, not you. This is also why it travels so well across borders. My wife lives in South Africa and I am there often, so it is a market I have genuinely considered, and these subaccounts mean I can expand without wading into each country's money-handling regulations myself. (The exact rules still differ by country, so confirm each one, but the principle holds.)

Fast settlement is a sales pitch. Paystack settles in about twenty-four hours. That means the businesses on HVNG get their cash fast, and fast cash is its own selling point. It is another reason for a vendor to choose us.

You borrow trust you have not earned yet. Do not underestimate this one. Paystack is a trusted brand, and HVNG gets to borrow that trust. When your users see Paystack at checkout, they relax, because they know Paystack has done its own KYC and is not going to disappear with their money. A young platform cannot manufacture that kind of trust quickly. Borrowing it from a credible partner is one of the smartest free advantages you can give yourself.

And once more, with feeling: please research the laws in your sector before you build any of this. The fastest way to kill a promising platform is a legal problem you could have read about in an afternoon.

What to charge, and how to know your numbers

Let me end the building part with the maths, because a dream without numbers is just a daydream.

Here is my monetisation dream for HVNG. One thousand businesses, each spending twenty thousand naira with us a month. That is twenty million naira a month, and it is genuinely all I need to scale the thing properly.

But here is the more important number, the one that lets me sleep. I do not need a thousand businesses to survive. At just ten businesses paying twenty thousand naira each, HVNG is already sustainable. Why? Because I know my costs cold. Roughly twenty dollars for Claude Code, twenty for Vercel, twenty-five for Supabase, twenty for Resend. That is eighty-five, and let us call it a hundred dollars a month with fifteen for extras. A hundred dollars is around a hundred and forty thousand naira. (Exchange rates move, so re-run this with today's rate, but the shape is what matters.)

So ten businesses cover the lights with room to spare. And stretch it out: a hundred dollars a month means twelve hundred dollars buys me a whole year of operation, even if growth completely stalls. Even in the worst case, the lights stay on for twelve months while I figure it out.

Yes, marketing is not in those numbers yet, and we will get to it. But knowing your minimum run rate, the smallest amount that keeps you alive, is one of the wisest things you can do early. It turns panic into a plan.

Hands-on lab

The chapter above is the why. These guides are the how. The one rule never changes: if a step stumps you, screenshot it and paste it to your AI with "assume I have no technical experience; tell me exactly what to click next." Nothing here cannot be undone.

G7.1 · Integrate Paystack (test key, then live key) — core to monetise

What you'll have: your app taking real payments.

Before you start: Beginner–Intermediate · ~1 hr · free to integrate, Paystack takes a transaction fee · you'll need a Paystack account and business details.

1. Create a Paystack account and complete business verification (KYC).
2. Get your TEST API keys from Settings → API Keys. Store them as environment variables (for example PAYSTACK_SECRET_KEY=sk_test_xxx), never in your source code.
3. Use Paystack's docs or AI helper to get integration code, and hand it to Claude to wire into your app.
4. Test payments with the test key (no real money moves).
5. When it works, swap in your LIVE key — again through the environment, not the codebase.
6. Note: real-world goods and services (event tickets, venue promotion, in-person vouchers) are generally not subject to Apple/Google's in-app 30%; a purely digital in-app upgrade on iOS may be, so check your case.

Check you did it right: a test payment completes and your live key is ready.

If something looks off: verification or webhook issues → see G7.3; always verify webhook signatures.

Unlocks: → G7.2

G7.2 · Set up Paystack subaccounts and split payments — advanced

What you'll have: money from vendor sales routes to vendors automatically, with your cut split off.

Before you start: Intermediate · ~2 hrs · free, fees apply · you'll need Paystack live and vendors.

1. In Paystack, learn Subaccounts and Transaction Split.
2. Create a subaccount per vendor with their bank details and your split percentage.
3. On a payment to that vendor, Paystack settles their share to them and your share to you; you never hold their money.
4. This eases the regulatory burden and helps cross-border expansion (for example, South Africa); confirm each country's rules.

Check you did it right: a test split sends the right amounts to the right accounts.

If something looks off: wrong split → fix the subaccount percentage.

Unlocks: → G7.3

G7.3 · Auto-enrol vendors with a form and a webhook — advanced

What you'll have: vendors onboard themselves; no manual transfers, ever.

Before you start: Intermediate · ~3 hrs · free, fees apply · you'll need G7.2.

1. Build a vendor form for business and bank details.
2. Ask Claude to use the Paystack API so a submission creates a subaccount pending your approval.
3. Use a webhook so Paystack validates and confirms the subaccount.
4. Approve it; thereafter payments auto-split to that vendor.
5. Settlement is about 24 hours, so vendors get cash fast, which is a selling point.

Check you did it right: a test vendor self-enrols and receives a split payment.

If something looks off: webhook not firing → check the webhook URL and secret, and verify signatures.

Unlocks: → hands-off operations

Now make them glad they paid

So the charity is officially over. HVNG has rails, it has revenue lines, and it has numbers that say survival is cheap and scale is reachable.

But there is a catch. If businesses are the ones I want to pay me, then I cannot just point a till at them and wait. I have to sweeten the deal. I have to make them feel genuinely well served and well valued on HVNG, so that paying me feels less like a cost and more like a smart move on their part.

And notice that the single biggest decision here — the payment rails — was a decision about which partner to trust, not about being a lone genius. You cannot research every law, vet every processor, and dodge every regulatory trap by yourself. This is exactly why we move as an ecosystem and not as a thousand solo founders each learning the hard way that Apple wants thirty per cent.

Come build with us at AIStoryLab. Twenty thousand naira a year, still too cheap, and inside you get the how-tos, the integration walkthroughs, the prompt techniques, and the sector-by-sector notes other builders have shared on staying legal and getting paid. Bring what you have learned, take what you need.

What's next

You can take money now. Next, we make sure the people paying you are smiling while they do it: how I optimised HVNG for business accounts with contextual search, offline-proof event check-in, lawful attendee data and social gifting.