## 1. Executive Summary
Iago Peixoto built a multimillionaire trajectory with direct marketing before age 20, focusing on offers, math, and testing over hype [OPERATOR CLAIM]. Core: direct-response fundamentals, tight unit economics, rapid testing, and reinvesting. All figures below are operator-stated claims, not independently verified.

He was 20 at the time of the episode, five years into the market, having started at 15 including a stint trading the stock market at 15 and he claims eight figures in digital sales before turning 20. The episode's spine is one repeatable move: model an offer in 24 hours instead of copying it, validate it with a small paid budget, and let the numbers, not the excitement, decide what gets scaled. He also argues hard against the volume-affiliate model and for owning a single, deeply improved vehicle.

## 2. Alpha / Sauce - what actually moves the needle, and exactly how to run it
- **Direct marketing first principles.** List, offer, hook, response, tracking math > magic. Master response mechanics early.
- **Test velocity > perfection.** Small, fast tests to find winners before competitors. Kill fast, scale hard.
- **Unit economics discipline.** Know allowable CPA, breakeven, net per sale, LTV before scaling. Decisions by numbers.
- **Offer engineering.** Irresistible offer beats traffic quality at same spend. Strengthen offer until it pulls.
- **Reinvest compounding.** Early profits recycled into winners, not lifestyle creep. Capital velocity builds moat.
- **Model, do not copy in 24 hours.** His headline process: research the niche, pull the competitor's sales video off their page, transcribe it, translate it if needed, and feed it to a custom GPT trained on that niche to identify and rebuild a unique mechanism. The distinction he insists on: modeling means understanding why the mechanism works and building your own; copying means lifting the words.
- **Read the mechanism's shape before you write yours.** The tooling question he asks is concrete was the mechanism explained in 15 words or in 2,000? A short mechanism wants a short, punchy claim; a long one wants a path. In the dating niche he describes mechanisms as a route or map ("a technique that moves the person's brain"), so the offer sells a path plus an accelerator, for example a pack of 100 distinct phrases meant to speed up the result.
- **Follow with a content GPT, not a blank page.** Once the mechanism is rebuilt, a second GPT produces the delivery he asks for a 50-page ebook that explains the method and how to apply it so the product exists in the same 24-hour window as the offer.
- **Validate cheap, then decide.** His framing of the downside: the most you lose is the R$2,000 you put in over three days of work. If it validates, the upside dwarfs that; quitting over R$300 of test spend is how people lose millions by fear of losing crumbs.
- **Hit rate is a math problem, not a talent problem.** He launched roughly 20 products across different niches and mechanisms back pain, weight loss, seduction, smoking cessation and about two worked, roughly 10%. He calls that normal, which is exactly why test volume and kill speed matter more than any single launch.
- **One vehicle beats a hundred.** He rejects the affiliate-flood model: 100,000 students, 1% doing well, each closing a sale a day, produces a headline of a million reais a day while wrecking the market for everyone. His arithmetic for ownership: 100% of nothing is zero, while 20% of a million is 200,000 so take the smaller share of the real pile.
- **Know your 80/20 and refuse to leave it.** His daily 80/20 list: study and apply, ship one cost-reducing tech update in the operation, read forums and refresh market knowledge, protect his own self-knowledge (podcasts and reading), and block all of it in the agenda. Everything else, including other people's 80/20s, he deliberately ignores.
- **The bio link is a conversion asset.** Give the product's Instagram about a dozen posts success cases, authority, how the product was made, the mechanism plus highlights, and point the bio link at a short written summary instead of a closed sales video. He reports around 15% more revenue from that change alone, sometimes enough to cover the week's ad spend.
- **Work the commenters.** People who comment on ads or message through the inbox often never open the message box; a sales team working those threads added roughly 15%, and in some months 20%, to revenue.
- **Cap CPA with a rule, not a feeling.** Set an acceptable cost per purchase in the platform and create a rule that shuts the campaign down if it does not reach it, so the system hunts within your number instead of spending blindly.
- **Cut geography instead of guessing.** One campaign, 26 ad sets for the 26 Brazilian states; cut the weak, keep the strong. In his scaled periods Sao Paulo, Rio and the South/Southeast carried the volume, with Sao Paulo alone around half of sales, and he notes Sao Paulo's 10 million people behave like a country.
- **Face on camera is a legitimate multiplier.** An A/B test he ran doubled conversion when a face was added, and a single change to the lead path can generate around 12 testable variations from one edit.
- **Ethics is a filter, not decoration.** He walked away from a R$10,000 job he considered unethical after doing similar work earlier and disliking it, and he puts it as a rule: when an offer arrives that asks you to override your values, review your character. He also owns the failures: almost 100% of the problems in his operations, he says, were created by him and believing that is more useful than blaming the market.

## 3. Methods & Setup - Step-by-Step
1. **Master offer + hook set.** Setup: 10 hooks, 3 offers minimum per angle. Test to small spend.
2. **Track everything.** Setup: campaign-level tracking (angle, creative, list). No untrackable tests.
3. **Find breakeven fast.** Setup: compute allowable CPA + net, test until statistically directional.
4. **Scale winners aggressively.** Setup: horizontal/vertical scaling only on proven net-positive.
5. **Build list + backend.** Setup: capture, email follow-up, upsell to lift LTV.
6. **Rapid iteration loop.** Setup: test daily, decision grid (kill/iterate/scale) every 24 48h.
7. **Model the offer in 24 hours.** Research the niche, pull the competitor's sales video, transcribe and translate it, then run it through a niche-trained GPT that identifies and rebuilds a unique mechanism; setup: transcript plus niche GPT ready before writing.
8. **Ship the delivery the same day.** Hand the rebuilt mechanism to a second GPT and ask for a 50-page ebook explaining the method and the application steps; setup: outline with the promise, mechanism and how-to in it.
9. **Put the cheap test on the table.** Fund three days and about R$2,000 of traffic, define the acceptable CPA as a rule that kills the campaign, and let the result decide; setup: budget cap + CPA rule in the ad account.
10. **Map the states.** Build 26 state-level ad sets, cut the underperformers weekly, keep Sao Paulo, Rio and the Southeast running; setup: state report updated every seven days.
11. **Build the bio funnel.** Twelve posts of proof and mechanism on the product's Instagram, highlights, and a bio link pointing to a short letter rather than the sales video; setup: profile rebuilt in one afternoon.
12. **Assign the comment desk.** Have someone from the sales team reply to every ad comment and inbox message, because those buyers never open the message box on their own; setup: daily list of unanswered commenters.

## 4. Tools & Stack - what each one is for here
| Tool | Role | In this episode |
|---|---|---|
| Direct-response tracking | Decision engine | Angle/creative/list-level truth. |
| Offer matrix | Testing engine | Systematic hook x offer testing. |
| Email list | Asset | Follow-up + LTV lift. |
| Unit math sheet | Gatekeeper | CPA/net/breakeven before scale. |
| Rapid test pipeline | Velocity | Kill fast, scale winners. |
| Sales-video transcriber + niche GPT | Offer modeling | Competitor video rebuilt unique mechanism in 24 hours. |
| Content GPT | Delivery builder | Mechanism 50-page ebook the same day. |
| CPA rule in the ad account | Spend guard | Kills the campaign before the cost exceeds the number. |
| State-level ad set map | Budget allocation | 26 states, weekly cut of the weak. |
| Product Instagram + bio letter | Pre-sell surface | About 15% revenue lift from the bio link alone. |
| Focus-blocker app | Attention | Roughly R$9 a year; it showed him 40 Instagram opens in one day. |

## 5. Actionable Checklist
- [ ] Build offer matrix (3 offers 10 hooks).
- [ ] Track angle/creative/list on every test.
- [ ] Define breakeven + allowable CPA before spending.
- [ ] Run 5 10 micro-tests/week, kill < breakeven in 24 48h.
- [ ] Scale only proven net-positive with backend.
- [ ] Reinvest 80%+ of early profits into winners.
- [ ] Model one competitor offer end-to-end in 24 hours: video, transcript, rebuilt mechanism, delivery.
- [ ] Cap three days of testing at a fixed budget and set a CPA kill rule before launch.
- [ ] Split one campaign into 26 state ad sets and cut the losers weekly.
- [ ] Rebuild the product profile and point the bio link at a short letter instead of the sales video.
- [ ] Assign someone to answer every ad comment and inbox message within the day.