Why a Media Buying Team Closed Its Nutra Department After 21 Months: The Write-Up, Fact-Checked

"In our industry it's customary to show only wins... But this post isn't about wins. This post is about a defeat." On July 30, 2026, a Ukrainian media buying holding, Legends, published how it closed its nutra department after 21 months: a record spring, a dead setup, banned accounts, a partner who walked away and 9 months without profit.

This is our review of the write-up for team owners and media buyers in health offers: the story in translation, our math on the few numbers it gives, and what it leaves out. It's an anti-case, so the score rates how honestly and in detail the failure is analyzed. Editorial fact-check: 2/5.

Short version: the timeline is dated, the mistakes are named, and the post doesn't dress up the ending. But there's no GEO, no offer, no network, no spend and no total loss, and not a single screenshot. It's a well-told story with the numbers left at arm's length.

Who wrote it

Legends describes itself as an international media buying holding; its job ads claim hundreds of specialists across iGaming, nutra, finance, crypto and sweepstakes. The write-up is unsigned, written in the first person by whoever ran the department, and opens with a boast before the confession: "today Legends is team #1 in gambling by results." Its public Telegram presence is a private invite-only channel; the post itself has no links or contacts. Every quote below is our translation from Ukrainian.

Our take: a holding that is hiring publishes a loss with "no excuses, no pretty fairy tales." That's good for the industry and good for the brand. "#1 in gambling" is a self-claim with no basis given, and it frames the loss as a side bet of a winner.

The start: 1 partner, 1 setup, 9 juniors

"The nutra direction at Legends started on November 1, 2024. I partnered with a very strong solo buyer, who had many years of experience in nutra, consistently generated good profit and led a team of 3 juniors." His setup "had been alive for about 2 years. It gave good ROI, worked steadily without bans and looked maximally reliable. In short, it was certain approaches in catalogs."

The risk was discussed and dismissed: "'What happens if the setup dies?' Back then it looked unlikely." The plan: "There's a scheme, you just need to scale it by hand." About 30 interviews, 9 juniors hired and trained in the office.

Our take: "certain approaches in catalogs" is all the mechanics the post gives, most likely catalog-based Facebook ads. Everything else, the GEOs, offers, networks and payout model, stays private. That's normal for a working setup. For a dead one, it means a reader can't check the lessons against the method.

The peak: "$500K revenue at 30-40% ROI"

"In the first month we closed 15K of net profit." Then January, "the famous nutra January, a month of high demand and an almost empty auction," and "spring and summer 2025 were the peak. We were reaching about $500K revenue with ROI of 30-40%."

Our math: the post doesn't say whether $500K is a month or the whole season. If ROI is profit over spend, 30-40% on $500K means $357K to $385K of spend and $115K to $143K of profit. If that's monthly, it's a big operation; if it's the season, it's about $20K to $30K a month for a team of roughly 12 buyers. The 2 readings differ by about 6 times, and the write-up never resolves it.

The break: autumn 2025 and January 17

"In autumn Facebook made its corrections and the eternal setup did die." Tests followed: "A month passed. Then another. No result." The team set January as the test: "if January goes, great. If not, radical changes will be needed."

"The first 15 days of January were very strong... But around January 17 everything died. The old accounts got banned, and nothing came out of the new ones. By the end of the month we'd lost practically everything we'd earned in the first 15 days."

Our take: that's the most informative paragraph in the post, and the ban story every nutra team will recognize. A setup that lived 2 years on old, trusted accounts didn't move to new ones: once the aged accounts went, the method went with them. The post never says what the accounts were, where they came from or what replaced them, which is the part a nutra buyer would need.

The end: Google, a partner exit and 9 months without profit

A new hire proposed a nutra Google department, launched in April 2026. "In parallel, the FB team looked more and more exhausted. It felt like the partner had stopped believing in the project." "In April we reached a very unpleasant point: the project had fully eaten all the profit it had earned before. Funding the direction out of pocket began." "In May the partner decided to leave the project. The team was left without a head."

"June closed at about minus 25K net. That was already the 9th month in a row without profit." The last target was July, at least 15K gross. "It didn't work out. Google: +7K gross. FB: -$2.6K gross."

Our math: July's 2 departments together made about $4.4K gross, 29% of the target. 9 months without profit back from June starts in October 2025, which matches the autumn collapse. The total loss isn't given: "fully eaten all the profit" means the 21 months netted out to about 0 by April, and then June added about -$25K. Units are mixed: some figures carry a dollar sign, some are just "K."

Not shown: spend, payroll, office costs, what happened to the roughly 12 juniors and the new offices, and any number for the Google department beyond July.

The 4 lessons

The post ends with its mistakes, and they're the reason to read it:

1. "Sitting on 1 setup is death. Any approach will die sooner or later. If your team is holding on 1 setup today, already look for the next one."
2. "1 head is good. 3 are better. All the expertise was concentrated in 1 partner. When his system stopped working, the team stopped working too."
3. "In nutra it's much harder to hire strong people. Everything is much more closed, many teams live on the principle of 'godfather, brother, matchmaker.'"
4. "Aggressive growth isn't always the right decision. At the peak we opened new offices and put juniors in team lead positions. They hadn't yet become strong specialists themselves but already had to teach others."

And the question every reader asks: "Is there money in nutra? Yes. I personally saw good ROI at decent volumes. Will I return to this vertical? Yes."

Our take: all 4 lessons are true and expensive, and 2 of them are really 1: the team scaled hands, not knowledge. 9 juniors trained to copy 1 setup is a team with 1 point of failure and 10 people standing on it. When Facebook killed the setup in autumn 2025, nobody had a second method, and the person who might have found one left in May.

Fact-check: what the author showed and what they hid

We rate every case on 10 points: what an outsider needs to judge the result and repeat it. For an anti-case the score rates how honestly and in detail the failure is taken apart. Like most case studies, this one is selectively true.

Score: 2/5. It's an honest, dated story with the right lessons, and publishing a closed department is rarer than publishing a profitable one. But without a GEO, an offer, a spend figure, a total loss or a single screenshot, it's a narrative a reader can't check, and its 1 technical clue, the January bans, stops at "nothing came out of the new ones." Honest, readable and unverifiable: a 2.