Business Plan — Afiliatex OS

Afiliatex OS is an information and distribution HUB for affiliate operators:

• Ingestion — connectors collect and normalize offers from affiliate platforms (Hotmart, ClickBank, MaxBounty, OfferVault and other networks) into a single central catalog.
• Matchmaking — the OS crosses offers with each user's profile and delivers the information needed for campaigns on each platform.
• Distribution — via the user's affiliate ID injected into every link, the OS distributes campaigns on 200+ platforms (social, messaging, ads, email) through a 4-layer Integration Engine.
• Tracking — every click comes back via subID, feeding the hub with real performance.

What we are asking for: US$ 5–10M in two investment tracks (Patron and Partner, detailed in §9).
Goal: 1,000,000 active users, revenue of ~US$ 5B/year at target and an IPO in 2037, with personal control preserved (30% + super-voting shares).

Why now

• The product foundation is written and operating end to end: operator console (web) + API with ~40 modules, a billing model already implemented in code (7-day trial + US$ 400 + progressive revenue share) and fiat payments (Stripe/Transak) + USDT.
• The hub was architected to connect hundreds of affiliate networks — the technical foundation (normalized catalog, connector and distribution contracts) is already defined in code.
• The affiliate market keeps growing with no neutral aggregation + cross-platform distribution layer — the OS fills exactly that void.

The OS is not an AI content generator. It is information and distribution infrastructure:

Moat: the integration layer (normalized catalog + 200 destinations + tracking) is an asset competitors cannot copy over a weekend. AI is an auxiliary layer (suggested angles, offer research, reports) with negligible marginal cost.

3.1 What is already implemented (code written today)

3.2 What the round expands (coverage and scale)

3.3 Execution roadmap — year zero (2027)

Revenue per user: the professional operator on the full fee pays ~US$ 1,000/month to the OS (and receives ~US$ 10–15k/month in commissions). The plan's effective blended ARPU is ~US$ 420/month (US$ 5k/year) — lower than the full fee because it accounts for the month 1–2 ramp at 0%, the trial, churn and smaller profiles. At the target of 1M active users: ~US$ 5B/year (≈ US$ 420M/month).

5.1 Principles

5.2 Target architecture

• Edge: Cloudflare (WAF, CDN, DDoS, click redirects on Workers)
• App: stateless autoscaling multi-region NestJS API + Next.js frontend
• Data: Postgres sharded by tenant + replicas · Redis cluster · ClickHouse (tracking) · object storage (S3/R2)
• Async: Kafka (events) + worker fleets (scanner, distribution, tracking, financial reconciliation)
• Security/ops: OAuth token vault (tens of millions), 24/7 observability, blue-green deploys, DR with RTO < 15min / RPO < 1min

5.3 Infrastructure ramp (cost per year)

The OS works 100% from day one — there is no "20-platform version" that later becomes another one. The same platform that serves 100 operators serves 1 million: asynchronous workers outside the API, shared catalog (1 scan serves all users), tracking at the edge and sharding by tenant (§5) — the structure is sized for the target scale from the foundation. If 1 million users arrive in 7 months instead of 7 years, the infrastructure scales elastically; nothing needs to be rebuilt, replanned or granted "time to breathe".

The 200+ platforms are breadth of coverage, not product stages — each integrated platform adds reach to the same operating product and becomes immediately available to all users:

Coverage curve: ~60 platforms by month 6 → ~120 by month 14 → 200+ by month 24. No layer is a prerequisite for the system to work — they are parallel fronts that add reach on top of a product that already operates.
Program cost: integrations squad (4–6 devs) US$ 1.5–3M + aggregator licenses + RPA fleet (US$ 5–20k/month at scale).
Continuity: third-party connector marketplace — coverage grows beyond the 200 with no internal cost.

Average loaded cost: ~US$ 130k/head/year (Brazil + global mix).

Assumptions: 2027 = year zero (expenses only, revenue not counted); users = paying active; blended ARPU = US$ 420/month (US$ 5k/year); revenue = average users in the year × ARPU; CAC US$ 250–350/user acquired; payment processing fees ~3% of revenue; investor distributions per §9.

Cash flow: the round's capital has a single destination: turning the OS that already works into an operation at scale — opening billing to the public, connecting hundreds of networks to the hub, reaching the 200+ distribution platforms and assembling the infrastructure and team to sustain it. None of the capital funds a loss: every dollar is mapped to these fronts, with headroom for the operation's financial health. The investor comes in knowing that: (1) year zero is the only pure-investment year, covered by the round itself; (2) there is no survival round in this plan; (3) from the first commercial year on, the operation pays for itself. The "Result" column is the annual operating result, before investor distributions and taxes; projected net profit for 2037, after distributions and taxes: ~US$ 2.5B/year.

Sensitivity: every ±US$ 100/month of ARPU = ±US$ 1.2B/year at 1M users. If the blended reaches the full US$ 1k/month fee, revenue doubles.

9.1 Track A — Patron (tickets up to US$ 1M) — revenue participation, no equity

• Track cap: 25 units (US$ 2.5M) — the sum of all participations never exceeds 15% of revenue.
• Decay: the participation falls by 1/6 of its initial value per year of growth ≥15%; when revenue grows less, it freezes. Single floor: 1% per US$ 1M invested (the same as the partners') — equal to 1/6 of the initial value, reached in year 6. No promise of absolute value — the return is 100% the revenue the OS bills, auditable.
• No equity, no vote. Term: 7 years (2028–2034) — the contract ends 3 years before the IPO. Buyback is a pre-IPO process: the company may buy back (extinguish early) the contract at any time before the IPO; and, in any case, the patron contract is mandatorily extinguished before the IPO — no patron reaches the IPO with revenue participation and no % of the company, a conflict that an investment fund's compliance will not accept.
• Platform Net Revenue (single contractual definition): everything the OS receives from users (access fee + revenue share received). Never GMV, never the user's commission.

How much the patron receives per year (money in the account, US$)

Year 1 = 2028 (first commercial year; 2027 is year zero). Highlight: in year 1 itself, the 50k patron already receives US$ 60k — 120% of the contribution.

Honest reading: values use the US$ 420/month blended ARPU; if the blended reaches the full US$ 1k/month fee, multiply by ~2.4. The single floor (1% per US$ 1M invested) is reached in year 6 — that is why 2033 drops versus 2032: the participation reaches the floor before revenue grows further; in 2034 the floor multiplies by the higher revenue. The program cap (pool ≤ 15% of revenue) limits the sum of all patrons in any given year — the table shows one individual patron. The contract ends in 2034, before the IPO.

9.2 Track B — Partner (US$ 1M to 5M) — equity + profit bonus

Anchor: pre-money US$ 20M — reflects the current stage: pre-revenue, product under construction, with the technical foundation already written and the year-zero execution roadmap (§3.3).

• Same decay engine as the patron (falls by 1/6 of the initial value per year of growth ≥15%), with the same single floor: 1% per US$ 1M invested — it applies to both the profit bonus and the equity: US$ 1M → 1% · US$ 3M → 3% · US$ 5M → 5%.
• Dilution with a floor: the ownership stake is not an eternal percentage — every issuance event (subsequent rounds, the IPO's primary issuance) dilutes everyone proportionally, including the founder (standard anti-dilution against down rounds). The scale carries the stake from entry down to the equity floor: 1% per US$ 1M invested — and never below it.
• Dividends only after the IPO, and only them: before the IPO there are no dividends — profit is reinvested in the operation. At the IPO, the profit bonus ends and the partner receives exclusively dividends: equity floor × total annual revenue, annually, growing with revenue. There is no partial payout policy — net profit is already after costs; the base is revenue. A single post-IPO income stream — no parallel bonus, which would complicate an investment fund's compliance and reading.
• Cap: equity sold in this round ≤ 30% — the founder never loses control.
• Above 10 units (US$ 1M), the investor leaves the patron track and becomes a partner.
• Partner income = profit bonus exclusively pre-IPO (decays to the floor and ends at the IPO) + dividends exclusively post-IPO (on the IPO stake).

How much the partner receives per year (money in the account, US$)

Projected net profit of the OS (after costs, distributions and taxes): 2028: US$ 8M · 2029: US$ 50M · 2030: US$ 140M · 2031: US$ 320M · 2032: US$ 700M · 2033: US$ 1.2B · 2034: US$ 1.7B · 2035: US$ 2.1B · 2036: US$ 2.3B · 2037: US$ 2.5B.

2028–2036: profit bonus only — dividends exist only after the IPO (profit is reinvested). At the IPO the bonus ends: from 2037 on, income is exclusively dividends = equity floor (1% / 3% / 5%) × total annual revenue — a single post-IPO income stream, growing with revenue. No partial payout: net profit is already after costs; the dividend base is revenue.

Plus the equity itself: the partner's stake rides with the company to the IPO and is governed by the same program dilution scale that applies to everyone — including the founder (§10.1). The entry (~5% / ~12% / ~18%) is the starting point; subsequent issuances dilute it proportionally down to the equity floor (1% / 3% / 5%), which is the stake that sustains the post-IPO dividends — and which participates fully in the market valuation formed in the 2037 IPO process.

9.3 Milestone tranches (both tracks)

10.1 Cap table evolution

• IPO plan: 30 / 15 / 55 — 30% founder, 15% investor partners, 55% floated for investment fund(s) to acquire in the process.
• Personal control: 30% economic + super-voting share structure (Google/Meta pattern) — voting control maintained.
• Equal dilution for all: issuances between round 1 and the IPO dilute founder and partners on the same scale — it is that scale which carries the stakes from entry to the floors (1%/3%/5%) and to the final 30/15/55 design.

10.2 Results in the IPO year (2037)

1. Blended ARPU of ~US$ 420/month (US$ 5k/year) — the professional operator on the full fee pays ~US$ 1k/month; the blended discounts the ramp, the trial, months 1–2 at 0% and smaller profiles. Sensitivity: ±US$ 100/month = ±US$ 1.2B/year at 1M users.
2. Revenue at the 1M-active-user target: ~US$ 5B/year; net profit ~40% of that revenue.
3. CAC US$ 250–350 in paid channels; mix with organic/referral growth.
4. Active user = paying account operating (not an inert signup).
5. Infra cost per active user: ~US$ 15–20/year at scale; email/messaging driven by usage.
6. Payment processing fees ~3% of revenue.
7. Post-IPO dividends = equity floor × total annual revenue — no partial payout policy: net profit is already after costs, there is nothing that "consumes" part of it; the dividend base is revenue.
8. "100k users in 2030" and "1 million by 2034" are targets for paying actives, with 1–2 years of schedule headroom in the IPO timeline (2037).