## 1. Executive Summary

This is a roundtable with three of Brazil's biggest Meta Ads media buyers: Gabriel Saraiva (in the market since 2022, over eight figures invested, operation did R$22 million last month), Dairo Junior (since 2015, more than R$100 million invested), and Pedro Andrade (more than R$50 million invested, operation doing multiple seven figures every month). Combined they have invested more than R$1 billion in Meta alone. They go deep on account contingency: how to buy versus farm profiles, BMs, and pages; which BMs survive; how rollback works; how to warm up bought assets; how to test creatives across two accounts; and how to scale with lateralization instead of a single risky ad set. Investment totals, revenue, and conversion figures are operator claims that were not independently verified.

## 2. Alpha / Sauce

- **Account quality is one whole, not one piece.** Your score connects profile, BM, page, and account; plug a great BM into a bad profile and the profile drags the whole score down. Start from a real profile, not a farmed one.
- **Do not chase the most expensive, oldest BM.** A student bought a R$8,000 BM and lost it the next week. Age alone proves nothing the goal is isolating the account variable, and a R$2,300-2,500 BM already does that.
- **Buy BMs created 2020-2023 that only ran white.** Lead-gen, local business, and e-commerce traffic leave clean history; a BM full of cheap remarketing CPAs approves creatives better and releases limits faster.
- **Rollback is real and new.** Meta treats an admin as new for 90 days, and a BM can revert to its creator; never remove the creator, remove the other admins, and wait 7-15 days before pushing volume.
- **Warm up bought assets by approving more creatives than you get rejected.** Approving white creatives at cents per day dilutes the rejected-ad ratio, raises relevance, and makes the account last longer.
- **Account priming bites.** Uploading a random dog or statue image with no business relationship to the link can trigger restrictions; the creative must relate to the destination.
- **Test on two accounts, scale on more than one.** Two accounts isolate the account variable; a creative that validated 30% more on account B would have been discarded if tested on one account only.
- **Horizontal lateralization beats one big vertical push.** Duplicate many campaigns and scale the day's winners at 30% increments; it minimizes the risk of one ad set not selling and losing the whole day.
- **Pattern recognition is the number-one media buyer skill.** Document every rejected creative with a hypothesis; the best buyers never stop testing to beat their own control.

## 3. Methods & Setup - Step-by-Step

1. Build contingency from a real human profile used like a normal person, then create the BM from it; setup: a genuine profile plus a fresh BM.
2. Warm a fresh BM toward e-commerce-like trust for about seven days, running white traffic to a marketplace or known store; setup: a white-traffic warm-up stream.
3. Decide to farm or buy: farm profiles as if real, and buy BMs created between 2020 and 2023 that only ran white (lead gen or local business); setup: trusted suppliers and a 2020-2023 BM filter.
4. Avoid BMs older than 2020 (rollback risk) and brand-new ones (low trust); check time, spend per account, rejected-creative count, and payment rejections; setup: a BM scoring checklist.
5. After buying, remove the other admins immediately but keep the creator, and let the BM sit 7-15 days; optionally also buy the creator profile so rollback is safer; setup: an access-removal and hold schedule.
6. Warm bought assets with engagement campaigns or conversion-max-clicks campaigns, approving many white creatives at cents per day; setup: minimum-spend warm-up campaign running even after sales start.
7. Keep approved creatives higher than rejected ones, and never delete a rejected creative deactivate the ad and place a white ad in its place; setup: a rejected-versus-approved balance tracker.
8. Read the account tier (tier 1 low, tier 2 medium, tier 2 plus, tier 3, tier 5 with credit line) using a browser extension before committing scale; setup: a tier-checking extension.
9. Consider agency accounts only at advanced scale: they are rented, charged a percent of spend (3-5%) or a fixed fee, restore accounts same day, and have Meta partner status; never for beginners; setup: an agency account test running parallel to owned BMs.
10. Run creative tests with five creatives and three hooks each (fifteen ads) in ABO, one ad set per creative body; setup: an ABO test campaign launched Saturday at midnight.
11. Test on at least two accounts and read the verdict over the first two days, then scale from Tuesday toward the weekend; setup: two test accounts per offer.
12. Validate per vertical, marking a creative valid once it hits the sales threshold for that market; setup: vertical-specific validation thresholds.
13. Scale by first trying vertical (doubling budget every two hours, five times the tested budget, minimum ten CBOs), then horizontal by duplicating and raising 30% at a time; setup: vertical-then-horizontal escalation rules.
14. Lateralize many campaigns and pause only the day's losers, reactivating them the next day, to minimize risk and lift total scale; setup: a duplicate-and-reactivate routine.
15. Document rejections with hypotheses in a spreadsheet by format (Reels, podcast, POV, interview), feeding CTR, CPC, and CPA back to the copy team; setup: a rejection-and-format spreadsheet.
16. Keep testing constantly to beat your own control, and unify results in a global profit spreadsheet that links campaign to real profit; setup: a unified profit dashboard plus a constant test loop.

## 4. Tools & Stack

| Tool | Role |
| --- | --- |
| Meta Ads Manager | Campaign creation, scaling, and account management |
| Tier-checking Chrome extension | Reads the account's trust tier before scaling |
| Antidetect browser | Manage many profiles safely without being hacked |
| N8N | Automation and an AI agent that optimizes campaigns |
| Tracking platform | Server-side tracking, sale-time and funnel patterns |
| Supplier network | Trusted sources for BMs and agency accounts |
| Rejection spreadsheet | Documents rejected creatives and hypotheses |
| Unified profit spreadsheet | Links campaign results to real company profit |

## 5. Actionable Checklist

- [ ] Start from a real profile, then create the BM from it.
- [ ] Farm profiles as real and buy only white 2020-2023 BMs.
- [ ] Reject BMs older than 2020 or brand new, and score payment history.
- [ ] Remove other admins right away, keep the creator, and hold 7-15 days.
- [ ] Warm assets with cheap white creatives and keep approvals above rejections.
- [ ] Never delete a rejected creative; swap in a white ad instead.
- [ ] Check the account tier before committing to scale.
- [ ] Use agency accounts only at advanced scale, not as a beginner.
- [ ] Test five creatives with three hooks each in ABO on two accounts.
- [ ] Launch tests Saturday at midnight and read the verdict within two days.
- [ ] Scale vertical first, then lateralize at 30% increments.
- [ ] Document rejections and constantly test to beat your own control.

 EN Alpha