The Most Valuable Meta Ads Training You'll Ever Watch
Jeremy Haynes breaks down what actually controls Meta ad targeting, why your pixel isn't as smart as you think it is, and the exact 'Thunderdome' testing system he uses to find the 1-3% of ads that truly scale.
Posted
2 days ago
Duration
Format
Tutorial
educational
Views
13.4K
338 likes
57 · 43
Big Idea
The argument in one line.
Messaging, not the pixel, decides who your ads reach first, and scaling profitably means testing creatives at high volume, cutting the roughly 97% that lose fast, and pouring spend only into the rare winners that keep converting instead of running dry.
Who This Is For
Read if. Skip if.
READ IF YOU ARE…
You're actively spending on Meta ads, anywhere from a few hundred dollars a day in testing up to an account already trying to scale past its current ceiling.
You've noticed cost per result creep up every time you try to raise the daily budget on a winning ad or ad set.
You run a lead-gen or call-booking funnel and lead quality has swung wildly for stretches of a few weeks at a time.
You want a structured, high-volume creative testing process instead of guessing which of your ads deserves more budget.
SKIP IF…
You've never run a paid Meta campaign and don't yet have a pixel, a funnel, or creative to test.
You're looking for organic social growth tactics; this is entirely about paid Meta advertising mechanics.
TL;DR
The full version, fast.
Meta ad performance is set first by messaging (the copy in the ad and the funnel it points to) and only second by pixel conditioning, which holds about 180 days of data and weights the most recent two weeks most heavily, so a bad two-week stretch can poison targeting even after months of good data. The fix is to withhold conversion events from reporting until the right audience returns. Because Meta's prediction engine reliably turns up only 1-3% of tested creatives as true scaled winners, the practical path to scaling is the 'Thunderdome': give every creative its own ad set with a 3-second view exclusion, split budget evenly across all of them, cut the roughly 90%+ that lose within a few days, and pour that freed spend into the winners without ever moving them out of the campaign where they're already working.
Free for members
Chat with this breakdown — free.
Sign in and you get 23 free chat messages on us — ask for the hook, quote a framework, find the exact transcript moment, generate a markdown action plan. Bring your own key when you want unlimited.
Cold open promising the video covers how million-dollar-a-month Meta advertisers scale without wrecking ROAS, immediately followed by an on-screen earnings disclaimer graphic citing a 0.1% probability of ever hitting $10M/year.
01:02 – 04:14
02 · What Controls Targeting: Messaging Pockets & The Scale Ceiling
Targeting is controlled by two things in order: messaging first, pixel conditioning second. Each messaging angle has a finite, replenishing 'pocket' of people who'll convert on it per day; scale past that pocket's supply and costs spike at what he calls the scale ceiling. Illustrated with a Carvana 'don't go to the dealership' ad example.
04:14 – 11:02
03 · The Pixel Conditioning Myth: Recency Bias & The Turd Pocket
The Meta pixel only holds 180 days of data and weights the most recent ~2 weeks heaviest (recency bias). A short run of bad leads (a 'turd pocket') retrains the pixel toward more bad leads even after months of good data. The fix: withhold conversion events from reporting back, switch to manually-fired Conversion API events, or duplicate the campaign, until the right audience returns.
11:02 – 14:49
04 · Learning Mode: 50 Reported Events Per Ad Set Per Week
Each individual ad set, not the campaign as a whole, needs roughly 50 reported conversion events per week to exit the learning phase and stabilize on one audience pocket.
14:49 – 19:45
05 · Why Your Messaging Is More Sensitive Than You Think
Every word in an ad, funnel page, or VSL script actively steers targeting. A student's ad using the words 'author' and 'J.K. Rowling' to sell a copywriting course pulled his entire funnel into an audience of aspiring novelists instead of buyers. Also covers standard events vs. custom conversions and Meta's claimed 52,000 data points per user.
19:45 – 22:32
06 · Meta Is a Prediction Machine: The Thunderdome Setup & Tribe V2
Introduces the Thunderdome testing strategy by first reframing Meta itself as a prediction machine, then explaining Tribe V2, Meta's MRI-trained model that predicts which part of the brain a piece of content will activate more than 90% of the time.
22:32 – 32:22
07 · Why Only 1-3% of Ads Become True Scaled Winners
Explains why broad/Advantage+ targeting is Meta's default (trusting the prediction engine beats manual targeting restrictions), why true ROAS doesn't show up at small test budgets, and why only a single-digit percentage of ads, ultimately 1-3%, ever become scaled winners. Introduces the oil fields vs. aquifers analogy: some winning messages run dry, true scaled winners keep replenishing.
The mechanics of the Thunderdome: one unique ad per ad set, a 3-second video-view exclusion so no one sees the same creative twice, budget split evenly across every ad set (e.g. $3,000/day across 30 ad sets = $100/day each). Cut the roughly 90% of losers within about three days and immediately redirect their budget into the winners. Never move a winning ad into a separate scaling campaign.
39:40 – 44:08
09 · Thunderdome vs Trusting Meta, Outro & Offers
Frames the Thunderdome as a deliberate trade-off (intentionally wasteful in testing to expose true cost-per-result) versus simply trusting Meta's broad targeting. Closes by admitting the video withheld 80-90% of what he knows, then pitches Jeremy AI, Jeremy's Inner Circle, the Master Internet Marketing Program, and his older pixel conditioning video.
Atomic Insights
Lines worth screenshotting.
Messaging determines which audience pocket an ad reaches before any pixel data exists; pixel conditioning only reinforces that audience once enough data accumulates.
A Meta pixel only retains about 180 days of data, so treating years of ad history as protection against a bad targeting stretch is a myth.
The pixel weights roughly the most recent two weeks of reported conversions most heavily, so a short run of bad leads can retrain targeting toward more bad leads.
Restabilizing a contaminated pixel means withholding conversion events from reporting back, or switching to manually-fired Conversion API events, not sending it more data.
Each individual ad set needs roughly 50 reported conversion events per week on its own to exit the learning phase, not the campaign as a whole.
A single ad mentioning the words 'author' and 'J.K. Rowling' pulled an entire copywriting course's ad account into an audience of aspiring novelists instead of buyers.
Standard events are generally a better targeting signal than custom conversions because more advertisers use them for what they actually represent.
True ROAS doesn't exist at small test budgets; cost, click-through rate, and conversion rate all shift once real scale hits an account.
Meta's Tribe V2 model was built from MRI brain-scan data on tens of thousands of users and predicts which part of the brain a piece of content will activate over 90% of the time.
Typically only a single-digit percentage of tested ad creatives, and only 1% to 3% overall, ever become creatives capable of holding cost while absorbing serious scale.
Meta defines a 'winner' as the ad that can absorb the most spend while holding its cost, not necessarily the ad with the single best cost-per-result.
A winning message can behave like a non-renewable oil field and run dry, while a true scaled winner behaves like a self-replenishing aquifer.
The Thunderdome puts one unique ad in its own ad set with a 3-second video-view exclusion, so nobody sees the same creative twice as budget rotates.
In a 30-creative Thunderdome test, expect roughly 27 of them to lose and get cut within about three days, with their budget redirected into the winners.
Moving a winning ad into a separate 'scaling' campaign is one of the most common ways advertisers kill their own best performer; a winner keeps winning where it already is.
Takeaway
What Actually Controls Your Meta Ad Targeting
META TARGETING
Messaging decides what Meta targets first, pixel conditioning only reinforces it, and scaling means testing high volume, cutting losers fast, and funding the rare winners that behave like aquifers, not oil fields.
02What Controls Targeting: Messaging Pockets & The Scale Ceiling
Messaging determines which audience pocket an ad reaches before any pixel data exists, and that pocket has a finite daily supply of people ready to convert.
Pushing more daily spend into one messaging angle than its pocket can supply is what creates a 'scale ceiling' where costs suddenly become inefficient.
Once an ad set hits its scale ceiling, the fix is a new messaging angle in a new test, not more budget on the same angle.
03The Pixel Conditioning Myth: Recency Bias & The Turd Pocket
A Meta pixel only retains about 180 days of data, so treating years of ad history as protection against a bad targeting stretch is a myth.
The pixel weights roughly the most recent two weeks of reported conversions most heavily, so a short run of bad leads can retrain targeting toward more bad leads.
Restabilizing a contaminated pixel means withholding conversion events from reporting back, or switching to manually-fired Conversion API events, until the right audience returns.
Consistency in who converts matters more than any single stat, because every downstream metric is meaningless if the type of person coming through keeps changing.
04Learning Mode: 50 Reported Events Per Ad Set Per Week
Each individual ad set needs roughly 50 reported conversion events per week on its own to exit the learning phase, not the campaign as a whole.
An ad set that never hits that per-ad-set threshold keeps bouncing between audience pockets instead of settling into a stable one.
05Why Your Messaging Is More Sensitive Than You Think
Individual words in ad copy, funnel pages, and VSL scripts actively steer targeting, not just the overall offer or headline.
A single ad mentioning 'author' and 'J.K. Rowling' pulled an entire copywriting course's ad account into an audience of aspiring novelists instead of buyers.
Standard events are generally a better targeting signal than custom conversions because more advertisers use them for what they actually represent.
Losing a long-running pixel and starting a fresh one exposes whether messaging alone was ever doing the targeting work, or the account was just riding pixel conditioning.
06Meta Is a Prediction Machine: The Thunderdome Setup & Tribe V2
Meta's targeting system is fundamentally a prediction engine trained on which pieces of content are most probable to move a specific person to act.
Meta's Tribe V2 model was built from MRI brain-scan data on tens of thousands of users and predicts which part of the brain a piece of content will activate over 90% of the time.
Distrusting that prediction engine and forcing manual targeting constraints is generally a losing strategy against a system built on that scale of data.
07Why Only 1-3% of Ads Become True Scaled Winners
Reported cost, click-through rate, and conversion rate at small test budgets are not the real numbers an ad will produce at scale; scaling always erodes performance slightly.
Typically only a single-digit percentage of tested ad creatives, and only 1-3% overall, ever become creatives capable of holding cost while absorbing serious scale.
A winning message that converts today can behave like a non-renewable oil field and run dry, while a true scaled winner behaves like a self-replenishing aquifer.
Meta defines a 'winner' as the ad that can absorb the most spend while holding its cost, not necessarily the ad with the single best cost-per-result.
The Thunderdome puts one unique ad in its own ad set with a 3-second video-view exclusion, so nobody sees the same creative twice as budget rotates across every ad.
Daily test budget gets divided evenly across every ad set, so a $3,000/day test across 30 creatives runs at $100/day per ad set.
Expect roughly 90% or more of tested ad sets to lose; cut them within a few days, then redirect their budget into the ad sets that are winning.
Never move a winning ad into a separate 'scaling' campaign; an ad wins because of the audience and data context it's already in, and pulling it out usually kills the result.
09Thunderdome vs Trusting Meta, Outro & Offers
Running the Thunderdome is a deliberate trade-off: it's intentionally inefficient in testing in exchange for exposing the true cost-per-result of every individual creative.
The alternative, broad/Advantage+ targeting with many ads in one ad set, lets Meta's own prediction model concentrate spend on the ads it already believes will win.
Glossary
Terms worth knowing.
Pixel conditioning
The process by which Meta's tracking pixel learns who to target based on which visitors trigger the advertiser's chosen conversion event.
Recency bias (pixel)
Meta's tendency to weight the most recent one to two weeks of conversion data far more heavily than older data when deciding who to target next.
Scale ceiling
The daily spend level for a given ad or messaging angle above which cost per result starts climbing because the profitable audience pocket is exhausted.
Standard event
A pre-built Meta conversion action, like Lead or Purchase, that most advertisers use for its intended purpose, giving Meta more reliable comparison data than a custom conversion.
Custom conversion
A conversion rule an advertiser builds from URL or event rules of their own, rather than using one of Meta's standard actions.
ABO (Ad Set Budget Optimization)
A campaign setup where the daily budget is set and controlled at the individual ad set level instead of the campaign level.
CBO (Campaign Budget Optimization)
A campaign setup where Meta distributes one shared daily budget across all ad sets in a campaign automatically.
Advantage+ / broad targeting
Meta's default targeting approach that removes manual audience restrictions and lets its prediction system choose who sees an ad.
Conversion API
A server-side method of sending conversion events to Meta directly from a business's own systems, instead of relying only on the browser pixel.
Thunderdome
Jeremy Haynes's high-volume creative testing structure that puts one ad in its own ad set with a short view-time exclusion, forcing spend across every creative being tested.
TRIBE V2
A Meta-built machine learning model trained on MRI brain-scan data that predicts which part of the human brain a piece of content will activate.
Learning phase
The early period of a Meta ad set during which the algorithm is still gathering enough conversion data, roughly 50 events per week, to stabilize delivery.
Resources
Things they pointed at.
11:40videoJeremy Haynes's Pixel Conditioning video
“Different messaging pockets act as oil fields... a true and literal scaled winner will be more like an aquifer.”
original analogy that reframes the whole video→ IG reel cold open↗ Tweet quote
38:25
“Where a winner is winning is where it's probable to continue winning. We are not going to take the winners out of this campaign and drop them somewhere else.”
See every word as it's spoken — crank it to 2× and still catch all of it. The same dual-channel trick behind Amazon's Kindle + Audible.
17px
metaphoranalogystory
The most valuable meta ads training that you're going to watch. In today's video, we're going to be talking about how million dollar a month earners or a couple million dollar a month earners are scaling the hell out of their ad accounts and maintaining quality and solid ROAS. as they scale whether you're just getting started at a few hundred dollars or a few thousand dollars a day in test budgets or whether you are actively trying to scale the hell out of your current campaigns this video is going to have you covered we're going to be talking about everything as it relates to the latest and greatest best practices on meta to be clear quick income disclaimer there's not a probability in hell that you go out there and hit a million dollars a month as a result of watching any video or anything that you'd ever buy from me let alone a couple million dollars a month according to research there is a 0 .1 % probability that you'd ever crack $10 million a year.
There's obviously an even smaller probability than that that you'd crack $12 million a year, a million dollars a month, or any amount greater than that. This is just lessons from people that are actively in the trenches on meta, including myself, actively trying to scale the hell out of their businesses and doing so successfully.
So without further ado, let's get started. The first thing and one of the most important things that you really have to consider nowadays that the algo actively optimizers around is what controls targeting.
What actually gets you in front of and keeps you in front of the right people on a consistent basis. It's a combination of two things and they move in order. The first of which is messaging.
Surprisingly, this is one of the most under talked about things in the advertising space. Before there's any pixel conditioning data, messaging is what controls who you are probable to go after. Pixel conditioning plays into who you are probable to stay in front of.
But messaging also controls how long you are probable to stay in front of the right type of person, and how long you are probable to maintain particular costs. One of the things that I never see literally anybody ever talk about is the fact that each messaging pocket has a finite amount of people that are actively in it, and each messaging pocket has what's called a replenishing rate.
There's a certain quantity of people that actively enter and exit a market in a given day that are probable to convert with this specific messaging combination and funnel type that you have actively. put in front of them. As an example, if I'm actively selling cars and somebody doesn't want to go to the dealership and I'm advertising on behalf of Carvana, I'm only probable to reach that specific market who is probable to go to my website, enter their information, look at cars, and choose to buy that particular way.
As soon as I go to scale the hell out of my ads, if the specific messaging angle that I've chosen Don't go to the dealership. Buy out of my car vending machine.
You could do it all online, 100 % A to Z. We'll even bring the car to you and pick up your trade all at your house without you ever having to leave. You could do everything online.
That specific angle has a finite pocket, a messaging pocket of people that are probable to convert on it per day profitably. Whereas if I try to extract too many people at once from that specific messaging pocket per day, I'm going to run into inefficiencies where my costs start to bloat and where I start to have issues.
You'll notice that if you go to scale a specific campaign or scale a specific ad set that you run into a clear ceiling. This ceiling is where your costs start to become inefficient. That ceiling is what I'm referring to.
It's referred to as the scale ceiling. This is how you know that every dollar that is spent above this amount is inefficiently spent. The scale ceiling is where you stay for that specific message and funnel combination.
Your goal from that point forward is to find additional messaging angles that you can incorporate into new testing ad sets either within that same campaign or within a new campaign depending on how you choose to structure it. We'll talk more about that here soon enough. But I want you to understand what most advertisers don't get is right off the bat, whatever you choose to launch with, messaging is going to determine who you're probable to get in front of.
Once you start to scale that particular set of messaging, it better be broad enough to actually have some serious spend absorb into that set of messaging. If not, you're going to hit a scale ceiling right away. The other thing that obviously controls targeting, and this is our number two, is pixel conditioning.
There's a lot of myths about pixel conditioning that I need you to understand. Those myths include how long the data is actually held on the pixel. A pixel, confirmed by Meta by the way, this isn't a theory, this isn't some made up sh** like most people make up.
Pixels hold data for 180 days currently. That's six months worth of data that are actively held on that pixel and that's it. There is no more data.
that gets held on that pixel. So if you've had a pixel for like a decade now and you've been advertising that entire duration of time, stop operating under the illusion that you have a decade worth of pixel conditioning. You don't.
The pixel operates off what's known as recency bias. Not only does it only contain 180 days worth of data, usually it factors in, in terms of who it targets and goes after next, the people that have most recently hit that pixel. So over the course of this 180 day timeline, Let's use the example that you're optimizing like any high ticket product or service based business for scheduled calls or qualified leads.
Maybe you've selected submit application or schedule as an example of the standard event. Maybe you're running webinar ads as an example. And on the most recent two webinars, you've had a bunch of donkeys come through and hit that specific lead or complete registration standard event you're optimizing around or your custom conversion, whatever the hell it is.
This is important to understand. If this is day 180, meaning six full months ago.
And this is the current day in terms of the timeline that we're operating in. What we found time and time again is usually the most recent, like two weeks worth of data is factored in the most in terms of who it goes after next. So I'm going to give you a very direct example of this because this comes from experience.
When I say that it's only the most two weeks -ish worth of data that gets factored in, let's say that we had a business that's been running a webinar. for a full six months. And let's use the example that the messaging that they've picked, they haven't hit a scale ceiling with it yet, meaning they have more people that they could actively target.
But let's use the example that naturally is what most of us experience on MetaOccurs. We reach a period of time where we just dip into the wrong pocket of people. There is a period of time where our messaging all of a sudden gets hijacked and we're in a turd pocket.
And let's use the example that we run a weekly live webinar in this example or maybe an automated daily scheduled webinar. If I have a full six months where everything's been absolutely awesome and then in that most recent two weeks that's when I've been in the turd pocket, there's actually a very high probability that if I continued to advertise from today forward, The pixel's most probable to keep me in front of more of that turd pocket because I'm optimizing around the people that are hitting the results column.
So if I'm optimizing around lead or complete registration as an example, I am actively chumming the waters. I am loading up that specific standard event with a bunch of donkeys if I'm getting the wrong people through. Now, here's where most people make a very common mistake, and this is very important to consider.
Targeting is where I'm starting in this conversation because targeting is arguably the most important thing. If I can successfully and consistently get in front of the right people over a broad period of time, I'm going to have stability in my ROAS. I'm going to have consistency in my scale.
I'm going to have predictability in all the KPIs and the statistics. It is chaotic in most advertisers' accounts if they can't keep a consistent type of person actively coming through. Every single stat will fluctuate.
You'll make decisions when you are trying to improve things, whether it be CRO actions, conversion rate optimization actions, or whether it be new messaging tests. You're just going to flail around and try a bunch of random shit when everything is chaotic, unpredictable, and inconsistent. If I can start the entire process of restabilizing your business and giving you an opportunity to truly scale the hell out of your meta ads, Every time I'm going to talk about targeting to start, I want consistency first.
That way I know if the stats that I'm later going to optimize around are actually improving or are simply changing because the type of person coming through is changing. Understand what I just said. If I have completely different people come in all of a sudden, I'm going to have completely different stats on every single step of my funnel all of a sudden.
And if I go reactively try to take improvement -oriented actions around those stats that are... completely subjugated by the type of person who's coming through and I'm getting inconsistencies or randomness in who's coming through, none of what I do on the improvement side of things is actually probable to improve it and stick because as soon as the pocket changes again, all my stats are subject to change again.
So if I have an issue with who is conditioning my pixel and I'm seeing that hold for about a two -week period of time, I'm going to withhold data that hits the pixel. I might even change the pixel and have a new one completely.
I'm going to go through a reconditioning process. And this is important you understand this, okay, because I really want to make sure you get this. If messaging controls targeting first, and then the pixel data rolls in, the pixel conditioning, the pixel data that rolls in has a secondary effect on who I'm targeting.
So if I know I have the right messaging, let's use the example, let me give you some circumstances. You've had good messaging. The messaging has gotten you in front of the right people.
All of a sudden you get in front of a pocket of donkeys. Those donkeys and those turds have conditioned your pixel now to go after more donkeys and turds. The very first thing I should do is withhold data going to the pixel.
That way if I'm still running that same messaging and if I know it's actually the right messaging, all I simply can do is just let the messaging control the targeting again. How do you let the messaging control the targeting again? withhold the data that's reporting back to the pixel.
So as an example, sometimes we get people that run call funnels, they'll run schedule standard events or submit application standard events or potentially a custom conversion, but they're optimizing around a scheduled qualified call reporting into the business. Let's use that same example. I'm getting the right people coming through historically, and all of a sudden I'm in a donkey pocket.
I'm in the wrong pocket of people that are actively reporting in. That's bad. I don't want that.
The very first thing I'm going to do is I'm going to stop allowing whatever source of data that reports back to that pixel from reporting to it. If I'm using, as an example, conditional logic inside of my application where people answer the right questions the right way, they are allowed to book. And then those bookings are what gets sent back.
But again, in this example, those people have been lying and they're donkeys or they're just unqualified people. I'm going to change what data goes back to the pixel because the pixel operates off recency bias, okay? I'm potentially just going to change the pixel completely.
When I say I'm going to withhold data from the pixel, you have to understand how all this works. You've heard of this before when it comes to the learning mode. The learning phase of every campaign is asking you one thing.
It's asking you to get 50 reported events per week. per ad set. So that means, let's use the example, I have a campaign.
And in that campaign, let's say I have three ad sets. Okay, and in those three ad sets, again, whatever you got on the ad level, you got on the ad level. In those three different ad sets, each one of these, not one of them, not just on the campaign level, each one of these needs 50 reported events per week, per week.
ad set. Otherwise, it's not going to work. It's going to start flailing around and it's going to go into different pockets of people.
So we'll talk more about this here shortly. For now, just put a pin in this. What I want you to understand as it relates to what I was just talking about is if I'm getting the amount of data per week, per ad set that I need, I'm going to stay in the same pocket audience.
So if I'm all of a sudden in the wrong pocket audience, the first thing I'm going to do is I'm going to intentionally This up. I'm going to withhold the amount of people that report back to the pixel.
I'm going to try to fall below that specific number. That way the algorithm starts to try to find different people again. I could also relaunch the campaign.
There's plenty of different ways to do things. That's one thing you'll learn about meta time and time again. I have advice that I can give one person and I will give literally the exact opposite of that advice to somebody else.
Because the advice that you take is circumstantial. One of the most important things you have to learn is when to do what. Because two different things can be that are polar opposite truths can be true at the exact same time.
So again, let me give you multiple ways you can do this. I'm either going to intentionally withhold data from reporting back on my pixel. And from there, that's going to cause that specific ad set to start trying to find different pockets of people so it can get the right people reporting back.
If I do it that way, once I start to see the right people reporting back, that's when I allow the pixel floodgates to open up again and I'm going to happily shove as much data as I can into this. When I am in the wrong pocket, I'm withholding data. When I'm in the right pocket, I'm giving in as much data to that results column as I possibly can because that keeps me in the right pocket.
You understand? The other way I can do this is I can duplicate the campaign or I can duplicate the ad sets depending on if I'm running CBO or ABO or depending on how f***ed my campaign has been. Here's the moral of the story.
If I choose to duplicate the campaign, same logic, I'm not going to allow whatever was reporting back from the pixel to report back again. I'm going to try to withhold Stop allowing data to report back either autonomously and use the conversion API to manually fire it off.
So in that example I was telling you about, it kind of was in the middle of that story. I'm running a call funnel. I'm getting the wrong people to all of a sudden report back.
I'm using conditional logic that allows the right people to go and schedule a call and then I'm automatically sending that data back to the pixel in that example. I'm going to stop doing that. I'm going to switch to the conversion API and I'm going to have my sales people manually mark when they have the right person they're talking to, that'll fire that person off to the pixel instead.
There's all different kinds of ways that you can solve this. By the way, I have one sole video dedicated to pixel conditioning. It's at this point probably about a year old, and it still follows all the same best practices, but in a more expanded version.
I want to move on from what we're talking about. But to be clear, the most important thing to understand is how to get in front of the right people in the first place, and how to stay in front of the right people. There's one more thing I want to talk about as it relates to messaging, which is important to consider here.
And it's this, the messaging pocket that you pick, and again, you pick it. That's something that most people really don't consider. The messaging pocket that you pick, you have to remember this word, sensitive.
Okay, so we had a million dollar a month award winner out of my Jeremy's Inner Circle program that got up in front of us here in our facility, told us about how he did it. And he was historically an organic first business. He had only gotten to about maybe five to 600K on his best months via organic.
And so he had to incorporate paid to get to the million a month. When he started incorporating paid and he was giving us these lessons to my group about what got him there, he used the word, man, the algorithm is so sensitive. He was teaching a copywriting course.
It was a biz op to become a copywriter. This guy in his ads is telling somebody AI makes it easier than it's ever been before to write world -renowned copy.
famous authors like J .K. Rowling. Those two words, author and J .K.
Rowling, put him in front of people that were wannabe authors. All of a sudden, for his copywriting course, he's getting people coming through his call funnel. He's getting people that show up to his webinars that are wannabe authors.
How? Why? Well, he tracked it all back to that ad that had the most spend out of any ad that he had run.
And that one specific ad used the word author and JK Rowling. Those two specific words got him in front of the wrong pocket of people.
You have to understand, there's a pro tip here to be acknowledged. That what you say in your videos controls your targeting nowadays. What you write in the copy of your ads, including the body copy and the headlines, control your targeting nowadays.
What you write on the funnel itself And what you say inside of VSL is on the funnel itself. Yeah, they track that too.
So you have to remember everything on the ad level plus everything on the funnel itself. That's what controls the targeting. In addition to that, in terms of messaging, these are the specific messaging variables that we have.
We also have to consider what specifically we have selected as a standard event because that also plays into it. whether it's a standard event, whether it's a custom conversion. I personally prefer standard events because more advertisers use them and most advertisers use them for the thing that they actually represent.
So therefore, Facebook Meta has more data already on what people are probable to do. Meta, they released this statistic a few years ago at this point, claims they have 52 ,000 revolving data points per user. that update in real time, give or take the specific trade of course.
When you look at the profile data points that every user on their platform has, they have every single action that somebody's took in the past. Every single action that an advertiser has optimized for historically that they've gotten a person who's been targeted by ads to successfully do is logged and is available for you to have at your disposal to have a higher probability to get what you're after.
So when you select your standard event or when you select your custom conversion, that's also going to add a little bit of bias towards who it goes after and who it doesn't go after. But I can't stress this enough. In terms of weight, what you say on the ad level and what you say on the funnel level does a tremendous amount of the targeting.
And then as time goes on, you start to get into pixel conditioning and you start to have to naturally acknowledge that that's playing a factor as well. Occasionally, we'll get somebody who, as an example, comes into my Jeremy's Inner Circle program or comes into my Master Internet Marketing program or becomes a Jeremy AI user and they have this specific issue.
They've had a pixel for a long time. They've consistently been in front of the right people for a long time. Maybe their ad account goes down or maybe their business manager goes down and they lose access to that pixel.
They go relaunch their campaigns inside of a new account and then they'll say, man, I'm having such a tough time getting in front of the right people again. That means your messaging sucks. What you were reliant on in that example was the pixel.
So it demonstrates exactly what I'm saying to be true, which is at the beginning when you launch something, the messaging does all the targeting in addition to what standard event or custom conversion you select. As time goes on, the pixel conditioning heavily biases who you're probable to go after next. So even with messaging, you can lean into the pixel as time goes on.
But I want you to remember this. It's very dangerous to exclusively operate on the pixel by itself. Very dangerous.
You cannot do that. What you have to do. is you have to have great messaging because if the pixel has issues, that's what's gonna be what you lean into to get in front of the right people versus the wrong people.
So remember the word sensitive. I always want you to consider that. It is such a regular occurrence.
That's why I'm harping on this so much to start this video off for you for how many people have the wrong types of people coming through or inconsistency of the right types of people coming through. It's the simplest words that'll throw it off. Let's get into testing a bunch of messaging.
So I have a specific strategy. It's referred to as the Thunderdome. The Thunderdome is an incredible high -volume creative testing strategy.
And I'm going to help break this down for you so you can really understand this. This is an approach that forces spend onto every single creative. And you have to remember something.
What meta technically is, is it is a prediction machine. That's all it is. It is a gigantic prediction machine of super intelligence that takes all these different data points that they've accumulated in addition to what you specifically are optimizing around, in addition to what you're saying on your creative level and your funnel level, and it says who's probable to take this action the most.
One thing that Meta has that surprisingly a lot of people aren't aware of is a model that they created called Tribe V2. So Tribe V2 is a pretty cool thing from them. This is actually a free model from them specifically.
What they did is they put a bunch of people underneath an MRI and they just had people scroll through their phones like normal when they did this MRI to these tens of thousands of people. And while they're sitting there scrolling on their devices, all they wanted to see was what parts of the brain were firing off when people saw what.
With those tens of thousands of people that they had this done with, they then took all that data, threw it into a machine learning model and created an AI with it that successfully 90 plus percent of the time predicts exactly what part of the human brain is probable to fire off when somebody sees what. Now, I really need you to understand this because it shows the power of Meta's prediction machine.
They took people, put them under an MRI. mapped their brain activity. After they mapped the person's brain activity, they put an AI model together that 90 plus percent of the time predicts to the T what part of the human brain was going to fire off.
That's emotions, probability to act as an example, benefits us advertisers, and all kinds of other shit. Then... After they had that model created and they officially had a human brain that was lighting up with different things as people took action or not in the news feed and just sat there scrolling, they put people back under an MRI machine.
And they had them specifically look at the same pieces of content that their model said would fire off a specific part of the brain and validated it. That's insane. So one thing that most advertisers do is they lack trust.
And when you lack trust, You are not going to have the prediction machine play into your favor as much as it otherwise would. I want to be very direct before I put you on game to this Thunderdome high volume creative testing strategy.
One of the only reasons you're going to do it is because you believe that there are specific creatives that are not getting spend that should be. If you load up and your structure for your campaign matters a lot nowadays, let's use the example you do something very simple. Let's say that you launch a campaign and in that specific campaign, Let's just say you have one ad set.
And let's use the example that it's a broad targeting ad set. And by the way, broad just means nowadays that you open up the constraints of targeting to allow it to go after the people that are the most probable to convert. Facebook got so fed up with dog sh** advertisers who didn't trust them and who have no idea what they're doing, trying to force targeting.
Facebook... did a simple study that said, if we didn't allow these advertisers to target people, and we just went after the people that we know our trillion dollar prediction machine could get them in front of that are actually probable to convert at the highest probability level, would that benefit all the advertisers? And they said, yeah.
So then they naturally rolled out that test. That's why advantage audiences exist. That's why broad targeting is the default now, by the way, is because if you actually trust the prediction machine.
you would have a very high probability to get in front of the people who are the most probable to convert. If you launched with messaging, you do not get that probability to play into your favor. So back to my point.
Assuming you launch a simple campaign, assuming you do an ad set that's broad, and again, in this example, it could be CBO. Most of the time nowadays, by the way, we're doing ABO just to be direct. But anyway, back to my point.
Let's say that from here, you have a total of 25 ads. or let's say, just to keep it super simple, five ads that are on the ad level. What's probable to happen inside of this structure is maybe one to three of these ads in either scenario, no matter how many you have, one to three are probable to get all the spend and all the reach.
Realistically, one of them is probable to get the most spend and reach. Now, I want you to remember something very, very important. They run a trillion dollar company.
with super intelligence on the back side of it that a large by far and large majority of the time successfully predicts who's the most probable to convert at any given time what they do is they say well yeah majority of these ads this guy gave us suck ass like we don't want to waste his money we don't want to put spend towards the ads that aren't actually probable to work so they just don't spend money on them occasionally for some of the best advertisers that exist and have super high levels of clarity on what ads and what messaging are actually probable to work for them, you'll see a larger percentage of spend go towards a wider range of the total ads you have on the ad level.
So if your ads are actually all probable to get conversions, you'll see a majority of your ads get spend. That is a rare blue moon event I just described. A majority of the time, you're gonna see one to three ads get all the reach and all the distribution.
Why? Because there is an unknown number for most advertisers for what their win rate is. So win rate on the ad level is super important to actually quantify.
Usually it's a single digit percentage of total ads. So what that means is, is that nine ads or less out of 100 are probable to win.
So let's use the example you launched with 10 ads. One ad is probable to be the winner out of 10 in that example. One.
If I launched with 100 ads, that means that somewhere between one and nine ads out of 100 are probable to actually be winners. What does a winner actually mean? Because it's not up to you to define what a winner means.
It's up to them to define what a winner means when you allow them to distribute the creatives in that campaign structure I articulated to you. What they define a winner to be is something that gets not typically the best cost per result. It's something that can have the most spend put into it that has the most scale.
So there's a bias here at play. Meta's bias and your bias, okay? So here's what happens.
Advertisers, when they spend a small amount of money on a campaign, which let's be honest, every single one of us spends, in most instances, a small amount of money on a campaign when we first get started with it. And because of that, we're not probable to see the true stats because we spent too little. The true ROAS does not exist at test budgets.
So if I'm launching, let's say, at $100 a day or $500 a day, something low as a test budget, every single statistic is not real. Okay, my CPMs might not be what they'd actually be at scale. My link click -through rate might not actually be what it's going to be at scale.
My conversion rate on my page might not be what it's actually going to be at at scale. Therefore, my cost per result is not probable to be what it is in testing versus what it's going to be at scale. And so what you really have to understand about what I just said there is very simple.
You will be deceived when you go to scale. So when you go to scale, you'll typically see things inflate a little bit in cost. Not a lot, but a little bit.
And you'll see what's called the trough of scaling, which is something very important to consider. We'll talk about that here shortly. For now, I just want you to understand this.
If a single digit percentage out of 100 creatives is probable to actually win, meaning be something that can actually have scale and actually be something that returns positive ad dollars back to you, how many total creatives are probable to be true scaled winners becomes the next question. So let me help you understand this.
It's typically 1 % to 3 % out of 100 ads that are probable to be scaled winners. What does that actually mean? So back to the point of this original conversation I've had with you.
Messaging controls targeting. Not every ad is created equally. I gave you the concept.
I refer to it as oil fields and aquifers. Different messaging pockets act as oil fields. Oil is a non -renewable resource, supposedly.
What that concludes to is that if you strike oil, meaning you found a specific message that works really well, you go to scale the hell out of it. Eventually, if it's an oil pocket, you'll run out of people to convert. And what that means in actual dollars and ad statistics is your cost per result's gonna fly.
You'll chase that high forever of saying, damn, that messaging at one point worked and it's not working for me anymore. Meta's not getting me in front of the right people anymore. Man, they suck.
But in reality, you just ran out of people that were probable to convert with that specific messaging. A true and literal scaled winner. will be more of like an aquifer.
An aquifer replenishes itself if it has a spring connected to it. So if I go down and I dig a well and bam, I lock into an aquifer and that aquifer is replenished by a spring. It may be a really aggressive amount of water that's consistently replenishing that aquifer.
It may be a small little trickle. that's actively replenishing that aquifer. The rate that I extract water out of the ground, I also have to factor in the rate that the spring is replenishing the aquifer.
So if I extract too quickly and the spring is barely adding new people into the aquifer, then I will also reach a period of time with my scaled winters where they will also become inefficient. so what you have to understand Facebook biases towards they are attempting to find out every creative that you give them what specifically has the chance to be the scaled winner.
The thing that if you went to spend a lot more money on it, it's the most probable to hold in its cost consistently. It's the most probable to continue to deliver results. And guess what?
Out of every single creative you give it, let's say you give it five, let's say you give it 25, it's probably not enough. You probably don't have enough spend to justify a lot more than that, especially if you're testing. But to be clear on top of that, it has a very low statistical probability to be a true scaled winner.
So when you find a winner, one very important thing to consider is where's the scale ceiling at for that winner? Where does the original cost per result peg? Where is the original profitability peg?
Because as I go to scale something, I need to know, naturally as I go to scale, I'm gonna erode my profitability a tad. That's natural. Not by a lot, but it's natural to erode it by a little bit.
The true scaled winners that you find, they'll barely budge on lowering the profitability when you go to scale it. The loser winner ads that you find, meaning let's say out of those nine out of a hundred or less that you maybe have found as a winner, if they have tiny little pockets, tiny little oil fields, tiny little aquifers that they've tapped into.
Well, as soon as you go to scale those, you might find a ceiling at like a hundred to a thousand dollars a day if they're small enough. What that infers is, is that a tiny, tiny minority out of every creative that we ever launch and run are probable to actually have serious amounts of spend put behind them. A majority of our ads that we launch are probable to have a tiny amount of spend behind them.
And our goal is to do one key thing, accumulate as many winners over time as possible. find their scale ceiling, chalk them into a pile that I refer to as foundational campaigns, and let them continue chugging along at the scale ceiling we ran into. The scale ceiling means that if I spend a dollar above a certain amount per day, it's inefficient dollars being spent.
If I go above the scale ceiling, It doesn't make sense. If I keep the ad at the scale ceiling, I can continue to run it at that current spend per day and it's not going to negatively impact the business.
Everything will be fine. if I can maintain it at the level where it's most profitable. If I go to scale it above that certain level of profitability, it's inefficient dollars being spent.
Those dollars should be risked on finding new tests that are probable to be true scaled winners. So the goal of testing is to rapidly move through creatives in a cost -effective and profitable way to reveal the true scaled winners. That's what Meta's trying to do, and that's what they're trying to save you from when you set up a campaign structure in a more efficient way.
way that biases towards trusting them. Trusting them meaning I'm willing to put as many ads as I want into an ad set. I'm willing to let them spend on what they deem may be my true scalable winners.
And then from there, the other ones that don't get spend, I'm going to conclude that they likely saved me money on the losers that they determined with their MRI prediction level analysis that just simply weren't probable to work. And so if I don't trust that system, the alternative is to run what I refer to as the Thunderdome.
So the Thunderdome is an incredibly valuable campaign structure if you don't trust Facebook and if you wanna cycle through a bunch of creatives rapidly. I'm gonna help you understand this in a way that gives you perspective on it without making the entire video just about this Thunderdome strategy. So this is my Thunderdome SOP.
I give this to people in my Jeremy's Inner Circle program, my Master Internet Marketing program. Jeremy AI is also trained on this. This entire thing is like the exact how -to of how to run this specific strategy.
But here's the premise of it. I want to put one ad inside of each ad set. I'm running ad set budget optimization in this example.
And I'm going to add a three -second exclusion to whatever ad is within that ad set. So let me help visualize this for you so you can understand it a little better. Because it's real easy to understand in practice.
I've got my campaign. And in this specific campaign, let's say I have 30 different ads, right? So I'm gonna have 30 different ad sets.
I'm gonna have ad set one, ad set two, ad set three, et cetera, through ad set 30. And inside of each one of these ad sets, I'm gonna have one ad. And of course, it's gonna be a unique ad.
So I'm gonna have ad one, ad two, ad three, ad four, et cetera. Now for this strategy to work, I just wanna put it in perspective that If you have images you're going to run, ideally you turn them into videos.
That way, if somebody sits there and looks at the image for more than three seconds, the exclusion would apply. What I'm doing on the ad set level specifically is I am excluding three second viewers. These are video viewers to be clear of that ad within that.
ad set. Now what that does, and this is really important that you get this, is once somebody sees that creative, they're never going to see it again. They're going to get rotated to see all the other creatives that you have.
In terms of targeting, we're typically running broad on that ad set. If you have a better targeting preference, like interest stacks, or maybe lookalike audiences, or even warm audiences, you can choose to do that if you prefer, but I like to do it broad. Now, what this does...
This is gonna cause excessive bloat in the cost per result for a short period of time, and then it makes it one of the most cost effective results that you'll see. It's important to consider this real quick so you can understand it. If I'm testing 30 different creatives, there's a high probability that 27 of them, assuming 10 % or less, actually deserve spend.
There's gonna be a high probability that 90 % of those ads, so in this case, if I'm launching 30 ad creatives to test it once, and there's a 10 % success rate or less, that means 27 of them are inefficient and shouldn't have spend going towards them at all, but are gonna get spent. So I'm gonna give you an example here. Let's say that I have a total of $3 ,000 a day that I'm looking to test with, okay?
So in this case, it'd be a total of 90 grand a month in test budgets. I'm simply going to divide that by the total ad sets. So if I have 3K divided by 30, that'd be my specific ad set budget.
If I have $1 ,000 a day and I have 30 different creatives I want to test, again, I'm taking the $1 ,000 a day divided by the total amount of unique ad sets that I have. That's my budget per day per ad set with ABO set up specifically in this specific structure. Now, I want you to understand this.
When you are spending on all of the losers, you will have a slightly inflated cost per result when you look at the summary analytics at the bottom. If you look at the campaign level or if you look on the ad set level, the summary analytics of what your cost per result is, that will look inefficient. If you click into the campaign and you look at the ad sets individually, you'll see that there will be less than 10%.
that are gonna be winners. They're gonna be a cost per result that's typically below your regular cost per result you'd get if you did it in the Facebook trusted way. Now, the longer that you have money going towards the losers, the more weight you're giving for your cost per result to be up and bad.
The less time that you're spending on the losers, You want to cut them rapidly. This is a very cutthroat campaign structure.
I'm trying to cut the turds fast. And when I cut the turds fast, I'm taking that spend and I'm putting them into the winners because I want to see what each winner can absorb before it starts to get inefficient. I want to reveal the scale ceiling rapidly of my winners.
So again, 10 % or less of the total creatives, in this case, three out of 30 are probable to win. All 27 of those ad sets with those individual ads that are being tested are going to get cut quick within like three days. In this example, spending a good amount per day per ad set.
If I'm spending less, whatever your cost per result is, try to give that amount per spend per ad set before you choose to cut something off. But once you choose specifically to cut something off, here's what I want you to know. You need to take that spend and put it towards the winners now.
Because like I said, the second half of the Thunderdome is I want my winners to absorb spend rapidly. A lot of people will choose to inefficiently test and scale. They'll create a separate testing campaign and a separate scaling campaign.
They'll take a winner out of where it was winning. They'll try to drop it into a different campaign that they call a scaling campaign. And they'll start to see it lose and they'll be like, I wonder why.
Where a winner is winning is where it's probable to continue winning. We are not going to take the winners out of this campaign and drop them somewhere else. We're going to leave them literally right where they are, where they're already winning, because that's where they have the highest probability to win.
Don't do that dumb shit. So anyway, here we go. Once we start cutting the losers and once we start funneling the spend towards the winners, the winners are going to start to average down the cost per result on that summary set of analytics, like all the ad sets combined, your campaign level of the campaign when you look at it.
The more time you have exclusively spending on winners that have lower costs, the lower the overall cost per result will be of that specific campaign. You are choosing to inefficiently test if you run the Thunderdome because you are biasing towards, I want to put spend towards every individual ad that with a very high probability, a 90 plus percent probability doesn't deserve spend.
But there's a lot of advertisers that we work with. They do prefer that specific Thunderdome structure that we just articulated to you here. because they don't trust Facebook enough.
They don't believe in the prediction machine. They wanna expose the true cost per result per ad. They think that every ad they create is a special snowflake and deserves the spend and is probable to win.
They don't know the actual stats. They don't know the actual true and honest, what are the probabilities for winning versus losing on an ad creative testings. So they want to force spend towards every creative.
Now, I wanna be very direct when I say this. The campaign as time goes on, you are going to delete the losers out completely.
That way you open yourself up to more opportunity to be able to add those new testing ad sets as time goes on. So this campaign is very efficient. You are rapidly testing.
You're cutting losers and forcing spend to the winners. You're revealing the scale ceiling of each winner sooner than later. Some will be lower.
A very, very few handful will be your largest winners you've ever had. You're going to continue just putting more spend on each individual ad set in this example that's winning to see how big they can individually win. After that, and you've officially had a good bit of time where your winners are scaling, you can then start testing again.
And you can just work new creatives into that exact same campaign structure. cut your losers quick, repeat the process, put spend into the new winners. And your goal, like I said, is to accumulate as many winners as you can.
As time goes on, you'll notice that a majority of your winners actually had a lot smaller scale ceilings than what you previously thought. And again, you'll have a tiny portion of the total creatives you've ever tested that have a large scale cap. Be very mindful of that.
So anyway... When you look at all of this and you factor all of what I just articulated together, this gives you a lot of incredible perspective on what you can do with Meta specifically to get the most out of it right now for what is literally working best and most today. There's naturally a ton of other tips and best practices that I could give you.
There's a lot more information that I could sit here and discuss with you on each one of these individual topics that we just went through right now. But here's what you're going to do. You're gonna go out there and you're gonna try to apply what you just learned and you're gonna get better results even with what we just discussed here.
And when and if you become ready to take the next step and learn all of what I can sit here and discuss because I know this may seem shocking to you because you've paid for a lot of programs in your life. You've paid for a lot of coaches. You've paid for a lot of mentors.
You've had agency owners that have known less than what you just learned here in a YouTube video today. My paid stuff. has about 80 to 90 % more than what I just discussed with you here.
I withhold 80 to 90 % of what I could sit here and talk to you about. I'm not one of those guys that puts it all on display. This is 10 to 20 % of what you need to know, but this is very high impact stuff for what you need to know.
Check out the links down in the description. Jeremy AI is a clone of me that you can talk to, connect to your ad account, and get the exact plan and system and structure you need right now to start scaling or finally escape test mode. Jeremy's inner circle, you'll find a link for it down in the description, is for rich people trying to get a hell of a lot richer.
That's where I do one -on -one calls. That's where I do quarterly in -person masterminds. That's where we give out our million dollar a month, three million dollar a month, or five million dollar a month trophies.
The highest earner in that group does 12 million dollars a month, by the way. You ain't gonna be the big dog when you join that, I promise you that. We also have my master internet marketing program, which we're currently in the middle of right now.
I do it on an annual basis. It is three -ish to sometimes upwards of four hours per class, one topic per week. large homework libraries in between, a tremendous amount of SOPs to make you walk out exactly as described, a master of internet marketing, not just somebody who's good, average, or mediocre, a person who's truly great, knows exactly when to do what.
Either one of those three offers will help get you the certainty and the direction that you're looking for so you don't run around like a chicken with your head cut off anymore inside of the ad accounts and know exactly what to do and when to do it. I look forward to seeing you inside of any one of those paid programs that, as I just mentioned, there's links for down in the description.
the very least, do yourself a favor. Go check out my pixel conditioning video I posted a little over a year ago so you can learn all the latest and greatest about that. And in addition to that, check out some of the content ad strategies, offer best practices, or really any one of the other videos on my channel because everything that we talk about around here is dedicated to one topic and one topic alone.
how to crack million dollar months or how to add that next million a month. And like I said at the beginning, you have no chance in hell of making that possible. It's less than a 0 .1 % probability according to research.
But all that we talk about on this channel are the lessons from people that have been there, done that. And the people who had been there, done that, they're abusing the out of what I just talked about here today. Go get richer.
Talk soon.
The Hook
The bait, then the rug-pull.
Before he ever gets to pixels or ad structures, Jeremy Haynes opens with a legally-mandated gut check: less than 0.1% of people will ever hit a million-dollar month. Then he spends the next 43 minutes laying out exactly how the few who do keep their Meta ad accounts from falling apart as they scale.
Frameworks
Named ideas worth stealing.
01:59concept
Messaging Pockets & The Scale Ceiling
Each messaging angle has a finite, replenishing daily supply of people who will convert on it. Scaling spend past that supply creates a hard ceiling where cost per result becomes inefficient.
Steal fordiagnosing why a winning ad set suddenly stops scaling profitably
28:19concept
Oil Fields vs Aquifers
A messaging pocket that converts today can behave like a non-renewable oil field and run dry. A true scaled winner behaves like an aquifer fed by a spring, replenished by a steady stream of new buyers, and can absorb scale without eroding cost much.
Steal fordeciding which winning ads are worth aggressive scale vs. milking briefly
32:22list
The Thunderdome
One unique ad per ad set
3-second video-view exclusion on each ad set
Broad/ABO targeting on every ad set
Divide total daily test budget evenly across all ad sets
Cut losing ad sets within roughly 3 days
Redirect freed budget into the winning ad sets
Never move a winner into a separate 'scaling' campaign
A high-volume creative testing structure that forces Meta to spend on every single ad instead of concentrating spend on the 1-3 ads it would otherwise pick on its own.
Steal forany account with more untested creative than the advertiser trusts Meta to fairly evaluate
11:02concept
50 Events Per Ad Set Per Week (Learning Phase)
Meta's learning phase asks for roughly 50 reported conversion events per week, per individual ad set, not pooled across the campaign, before delivery stabilizes.
Steal fordiagnosing why a specific ad set (not the whole campaign) keeps flailing
CTA Breakdown
How they asked for the click.
VERBAL ASK
41:10product
“Check out the links down in the description. Jeremy AI is a clone of me... Jeremy's Inner Circle... my master internet marketing program...”
Stacks three paid offers (Jeremy AI, Jeremy's Inner Circle, Master Internet Marketing) back-to-back at the very end, framed by openly admitting the free video withheld 80-90% of what he knows.
Add Modern Creator as a preferred source and Google shows you more of our breakdowns in Search, Top Stories, and AI Overviews. It only changes what you see, and you can undo it in your Google settings anytime.
Add to Preferred SourcesOpens your Google source preferences with us pre-loaded. Tick the box and you're done.
A creator who says he nets seven figures a month explains why he built his content engine around revenue instead of views, and why one channel still does 90% of the work.
A media buyer pulls up real Ads Manager screenshots to prove one account really went from $30,000 to $1.1 million a month in Meta spend, then breaks down the three things behind it.
A real consulting call, unedited, where a Meta ads consultant walks a 15-person email marketing agency through the creative strategy and production system it's missing.
Two direct-response operators trade notes on the offer changes that took a low-ticket start to nine figures, and why most health offers stall at the same monthly number while a handful pull in millions more.
Matthew Larsen spends three unscripted hours scoring an Australian ad-training business, page by page, against the same 1-to-5 rubric he applies to every audit.
00:40
01:10
01:55
03:14
03:31
04:08
05:47
06:14
06:53
07:11
08:02
08:33
09:14
09:26
09:59
10:34
11:44
12:24
12:57
13:30
14:17
14:26
15:01
15:37
16:44
17:10
17:55
18:17
19:01
19:43
20:28
20:59
21:47
22:29
22:49
23:25
23:59
24:16
25:20
25:50
25:59
26:45
27:18
28:09
28:58
29:30
29:55
30:37
31:01
31:38
32:38
33:16
33:46
34:34
35:14
35:55
36:36
37:14
37:34
38:20
38:53
39:26
39:59
40:30
40:51
41:25
42:29
43:13
43:51
Video of the Day45:11
20:48
55:40
12:57
22:21
2:57:49