duma

.

[duma @nifdmd]

@johnvirality MEDVI

[John @johnvirality]

this telehealth company hit $1.8B with 2 people by selling d*ck pills with fake AI UGC doctors:

the all-mighty MEDVI

it ended with an FDA class-a lawsuit, a New York Times front page, and platform bans across the board.

but hey their methods were pretty cracked

so today i’ll be breaking down the ENTIRE clinical avatar arbitrage playbook behind their $400M run rate

including the legitimate version that doesn't end with federal regulators involved...

here's what's inside:

> the authority psychology framework that explains why AI-generated doctors converted better than real ones in health advertising (the neurological mechanism that makes this work in every high-trust category, not just health)

> the three-step production system MEDVI used to generate synthetic medical presenters, localize them across demographics and languages, and manufacture social proof through sheer volume

> the exact line between legal AI spokesperson advertising and what MEDVI did to cross into fraud (most brands don't know where it is)

> the legitimate version of clinical avatar arbitrage: three proven approaches that extract the same trust mechanism without fabricating credentials

> the cross-category breakdown of how authority-based AI UGC applies to SaaS, DTC, finance, and professional services, not just health

all reverse-engineered from a real $400M case study with real regulatory consequences.

like + comment "MEDVI" and i'll send EVERYTHING your way

(must be following + RT for priority access)

[Ethan Dobbins @Ethan__Dobbins]

$0-$1m dropshipping - day 28​

What's going on pookies, I just had an epiphany that's going to completely change the trajectory of this challenge.​

It's a hypothesis, but if right, it easily explains why every product I've tested up to this point has failed horrendously.​

Including this UTI product that's now spent $300 with only one purchase.​

Here's the theory…​

…I have been marketing to the wrong people.​

GASPPPPPPPP (FAINTS INTO VICTORIAN COUCH)​

Now you might be wondering, how does someone with eight years of ecom experience​

Make a boneheaded mistake this fundamental?​

And the answer is simple my dear Watson​

I have been dramatically overestimating the IQ of the people I'm actually marketing to.​

With this challenge being my first time getting into supplements​

I assumed that people I’m shoving my ads in front of are very knowledgeable people ​

who do hella research before putting something foreign in their body!​

But after spending yesterday doing market research on my new liver support product​

I realized I tend to base most of my marketing, my angles, my desire mapping​

off of Reddit threads and Amazon reviews.​

And that's a great place to dive deep into the weeds.​

You get the actual language your audience uses to describe their problem and the outcomes they want.​

But here's the part I never really questioned.​

The people who are actually leaving detailed reviews on Amazon and posting in deep threads on Reddit ​

are not the same people I'm trying to convert on Facebook and Instagram.​

Those Reddit and Amazon users are far more sophisticated buyers.​

They've already done the research.​

They've already gone down the rabbit hole.​

They are level four sophistication audiences in most cases ​

if you make the generalization that the audience you’re marketing to is just these people​

But the audience on Facebook or instagram doom scrolling at 3pm while eating a half frozen arbys pork shoulder sandwich​

In between checking their email and fantasizing about their manager, Debrah getting into a car crash after she went on a tear at the office ​

that impulse buy after seeing one ad on Instagram from a brand they’ve never heard of​

are not the same human beings as the redditors and amazon peeps​

That cold audience on fb/insta is almost always a level two or level three sophistication.​

They know they have a problem​

but they have not done tons of research on every mechanism under the sun.​

I've been writing 12th-grade-level copy for what is potentially more a third grade IQ ​

Let me give you a real example with the liver product I'm researching.​

When I had my Claude project map out the desire clusters based on the Reddit and Amazon data​

the top desire cluser was "get my ALT and AST numbers back into normal range."​

That's the most sophisticated, scientifically literate version of the goal.​

That's a desire that someone who has already googled and got their lab results ​

But when I look at the winning ads for every brand actually scaling this exact category, ​

not a single one of them mentions ALT or AST.​

They don't lead with "get your lab work back to normal."​

They take that underlying benefit of getting a healthier liver​

and they expand it outward to surface-level outcomes the average person actually cares about.​

Liver back to normal means you can lose weight easier.​

Liver back to normal means you can drink and live the same lifestyle you used to without paying the price.​

Liver back to normal means slower aging, more energy, better skin, better mood.​

Not one of them talks about lab numbers.​

That's the move every elite copywriter teaches.​

You take a feature or surface level benefit and ask why does that matter?​

You keep asking until you arrive at the outcome the person actually wants in their daily life.​

Camera with 1000 GB of storage → you don't have to delete photos → ​

you can keep every memory of your family forever.​

Liver lab numbers back to normal → you don't have to worry about your health anymore → ​

you can drink, eat, live, and look like the version of yourself you used to be.​

I was doing this to an extent ​

but I believe I was outsourcing too much of my avatar work to a Claude project whose entire input was sophisticated user content ​

from people way further along the journey than my actual target.​

The second desire cluster on the liver product was getting rid of the fear of their liver damage being permanent​

before fibrosis, cirrhosis, before liver failure.​

And in the past few products, I would have 100% targeted that angle.​

Bleeding neck problem, urgent, emotional, mass market.​

But again, when I look at the winning ads, none of them lead with that fear.​

They mention it briefly as a symptom​

but they always tie it back to bigger surface level desires.​

Looking your best, losing weight, being able to live and party the way you used to.​

So I went back to the UTI product and reviewed my own work through this new lens.​

And sure enough, I made the same choice there.​

My top desire cluster was "stop the cycle of getting UTIs over and over again because they feel like their body is broken."​

It's the kind of desire that comes out of someone who has been on the Reddit threads.​

But when I rewatched the winning competitor ads​

none of them went to that depth either.​

They were mostly level two and level three sophistication marketing.​

Simple, surface-level, emotionally direct.​

The Facebook UTI buyer is probably a woman who has gotten one too many UTIs recently ​

and is tired of the burning, the urgency, the embarrassment, and most importantly​

the inability to just have sex without dreading the next 48 hours.​

She has not necessarily tried 5 supplements.​

She has not done deep research on different mechanisms yet​

She has probably tried the basic natural solutions you'd google in 10 seconds.​

I came in assuming she was already at the deep end of the research rabbit hole​

and I wrote copy that explained my unique mechanism like she had already tried 4 other supplements.​

When there’s a chance she hasn’t done that yet (according to my hypothesis)​

She just wants to know if this product would let her have sex without the consequences. ​

GROSSSSS, FIND JESUS WOMEN​

But that's the unlock.​

So I'm not killing the UTI product yet.​

Gonna go back to the drawing board on the avatar and the desire clusters.​

Rewriting the entire angle stack to a level two or level three sophistication audience.​

I'm also making a bigger ad shift in general.​

Going forward, I'm leaning much harder into video ads, not image ads.​

I've been over-optimizing for testing efficiency by leading with image ads only on every product​

but when I actually study what works for supplements specifically​

the answer is consistently videos with a lot of education.​

Most successful supplement brands only use images for bottom and middle of funnel social proof/offer creatives.​

Pretty much their entire top of funnel is video, which I was avoiding like a pussss.​

So today I'm writing new scripts at the right sophistication level for the uti product​

Two unique video concepts with multiple hook variations, each targeting a new desire​

And for the liver product, I'm shifting to video ads from the start ​

along with a sprinkle of native images​

See you in the next one pookies

[duma @nifdmd]

@seamlesspro93 Nice

[Doug Kelsey @seamlesspro93]

If you want the best payment processing in the game, sign up here - https://t.co/3oqfpV9Fmx

[duma @nifdmd]

@SimScaler HK

[duma @nifdmd]

@giubuilt @PhantomStays @whop @stripe Yes I know but if someone fills out the credit card info and click the OB to add it he needs to introduce the card details again

[duma @nifdmd]

@giubuilt @PhantomStays @whop @stripe They can’t even do OBs

[duma @nifdmd]

@YDaftary your platform only accepts US people? I got rejected didn't even upload my documents

[duma @nifdmd]

@XaviScales HORMOZI

[XAVI @XaviScales]

claude opus 4.8 + a custom agent now takes any product, runs it through every alex hormozi book, and ships back a full set of ad angles ready to film...on autopilot

here's the pipeline:

- loaded with 100M offers, 100M leads, 100M money models
- the black books, playbooks, launch book + lost chapters
- you paste your offer, it scores it on the value equation
-rebuilds it with a bonus stack + risk-reversal guarantee
-turns that offer into 20 ad angles mapped to real buyer objections
writes the hooks, scripts and CTAs for each one

no generic chatgpt soup, it pulls the exact framework on demand

reply "HORMOZI" + RT and i'll send it straight to your DMs (must be following so i can DM you)

[duma @nifdmd]

@mrbrandowner PLAYBOOK

[Brandon @mrbrandowner]

We scaled our supplement brand to $3M/month with MRR

$35M+ run rate this year
30k+ subscribers
77.7% margins

I put our entire strategy into one doc + a 30 min video

I covered:

- the MRR model that pays you even when your ads are off
- outscaling your competitors
- the creative engine behind every winning ad
- how to actually scale past $1M/month
- the payment processors nobody's using yet
& more

Like + RT + Comment "PLAYBOOK" and I'll send it straight to you

[Evan @yllevan2]

What works for me is a 3 layer angle research method most affs dont do systematically.

Most ppl when they want new creatives just look at competitor ads and try to riff. The problem is you copy the WORDS but miss the underlying structure thats actually making the ad work in the first place

First, collect 20-30 winning ads in your niche. Ad Library, use spy folders shared in private groups. Whatever works. Get 20-30 examples that have been running 30+ days, the longer running the more proven the angle.

Second, transcribe all of them. Use whisper api or any other app

Third, feed them to claude with a prompt like: “Here are 20 winning ad scripts in [niche]. For each one identify the surface promise (what they literally offer), the emotional driver (what the person actually wants underneath), the objection addressed (what theyve been told before that didnt work), the hook formula (first 3 seconds structure), and the proof element. Then synthesize the most common patterns across all 20.”

This gives you the underlying framework not just the words. Like the surface promise might be “lose 20 lbs in 30 days” but the actual emotional driver is “stop feeling invisible” and the objection addressed is “tried diets that always fail.”

Fourth, now ask the AI to generate 30 fresh ad concepts for YOUR offer using these patterns but with completely different hooks, CTAs and so on

Fifth, cherry pick the best 10-15 concepts and send them into production

Whole workflow takes 2-3 hrs and gives you some good angles

I used to do this manually but eventually had a friend build me a bot that automates the whole thing, takes winning ads as input, outputs 100+ rendered video variations, auto launches to Meta, auto kills/scales ads.

[Joshua Gavin @joshdgavin]

this market makes so much money lol

crazy pain point

mass market

selling to woman

so much room for more info offers here and ecom

taking Cara babies is one of the biggest

I heard her numbers and was shocked

maybe this'll be my next publishing deal

[duma @nifdmd]

@drcopybymatt You run your own offers or network?

[duma @nifdmd]

@Bogzabs96 pet

[Bogdan | Ad Creatives for Meta & TikTok @Bogzabs96]

I took my pet brand client from $30k to $750k.

So I put all my learnings in a 5-page doc.

- How we’re planning to reach $1M/month
- The formats & why UGC doesn’t work anymore
- Why and how we position our clients in crowded markets

Here’s how to get it:

Like this post and comment “pet” below.

(must be following)

[MAX @maxxmalist]

a lot of questions about prompt so here it is

[duma @nifdmd]

@luagaroz @gagerunsads VIDEO