Marketero · Entrevistas
$100K Meta Ads Hack: Simple Change, Big Payoff
27 ideas de marketing sacadas de «$100K Meta Ads Hack: Simple Change, Big Payoff», de Professor Charley T. Sobre todo de canales de adquisición, medición…
- 27 ideas despiezadas
- Fuente Professor Charley T
- Dura 1 h 00 min
- Publicada en 2025
Ver la entrevista entera en YouTube
De qué habla
Las 27 ideas de esta entrevista se reparten así entre las etapas del marketing. El reparto es lo que distingue una entrevista de otra: dice en qué se detuvo de verdad quien hablaba.
Las 27 ideas, con su cita y su minuto
Cada una está sacada de un punto concreto de la entrevista. El enlace del titular lleva a
la página de esa idea, donde se ve quién más la dice por separado; el del pie salta al
segundo exacto del vídeo para que la compruebes tú.
Estructura de testeo 3:2:2 para anuncios dinámicos
Medición y experimentosEl formato óptimo de experimentación publicitaria es el 3:2:2 (3 creativos, 2 textos principales y 2 titulares): ofrece una decisión binaria simple en el copy y 3 oportunidades para el formato visual (que representa el 85% del impacto para frenar el scroll), permitiendo validar o descartar hipótesis de forma rápida y eficiente.
Like, no. No. We teach 322 ads. 12 3 * 2 * 2 this is a total number This is basically the sweet spot. And it's done because a 322 ad is two headlines, two primary text. So, it's a binary choice. Which one is better? Or do neither of them matter? And then the three, the creative, you give the machine a few chances because the creative is 85% what stops the scroll. And if the ad sucks, you can remove one creative if you want to and give it a second chance. And if it still sucks, you know that it's bad. Move on. You've eliminated 479 million permutations. That is [ __ ] efficient.
Escúchalo en el minuto 28:17
Consolidación de cuenta en Meta Ads mediante creatividades ganadoras y segmentación abierta
Canales de adquisiciónPara simplificar y escalar una cuenta publicitaria de forma rentable, selecciona de 5 a 20 de tus mejores anuncios históricos y colócalos en una campaña optimizada para compras o nuevos clientes con segmentación amplia (broad) sin exclusiones. Usa esta campaña como un motor de consolidación que absorba gradualmente el presupuesto de cualquier conjunto de anuncios con peor rendimiento o CPA.
We start by pulling your top 10, 15, even 20 ads that spent the most and perform the best. From that pool, start with your five strongest performers. Remember the math that we ran earlier? Start with five. campaign optimized for purchases or if you're advanced, new customers, right? this stuff. But you keep targeting broad. No exclusions, no overthinking. offer, right? So, use this campaign as a consolidation tool. Let it absorb budget from anything performing worse on CPA or GPT. This shifts the spend to the most profitable lane while cleaning up your ad account. Basically, if it's the…
Escúchalo en el minuto 46:06
Reutilizar los 10-15 mejores anuncios históricos para escalar campañas en Meta
Canales de adquisiciónEn lugar de crear creativos nuevos constantemente, conviene duplicar y reactivar los 10 a 15 mejores anuncios históricos de la cuenta publicitaria, ya que el algoritmo de Meta ya tiene datos acumulados sobre qué audiencias responden mejor a sus textos y formatos, lo que permite escalar más rápido y a menor costo.
We used the best performing ads in the account. The ones that had already spent big and proven themselves. These ads are like allstar salespeople. Meta's algorithm has seen them win hundreds, sometimes thousands of times. It knows exactly who responds to that headline, that text, that creative. And when you duplicate them, you're not just starting from scratch. You're leveraging all of the expensive hardearned data that you've already paid Go into your account, pull in your top like 10 to 15 ads. These aren't just good ads. They're trained assets. When you redeploy them, they scale faster…
Escúchalo en el minuto 42:46
Canales de adquisiciónCuando un conjunto de anuncios 3:2:2 funcione, escala aumentando el presupuesto de la campaña en lugar de intentar escalar el anuncio ganador por separado. Solo aísla la publicación en un conjunto exclusivo para ganadores si absorbe tanto presupuesto que impide seguir probando nuevas variaciones, asumiendo que al aislarlo su rendimiento individual puede disminuir respecto al conjunto dinámico.
If you don't have a winning, you you don't scale the winners. Okay, you you scale the campaign. If your 322 ads are working, add budget to the campaign. If they aren't, turn the ad off. Eventually, you're going to find some 322 that's so good, that's so strong that you just can't test anything else and like you're in this weird spot where like you just got to isolate it. Grab that post, throw in your winner's headset. Now, it won't perform as well isolated because it's one ad instead of 12, but you can test more things and you can throw some restrictions on the other stuff to inhibit some of…
Escúchalo en el minuto 36:51
Regla de los tres clics para evitar rechazos de anuncios y penalizaciones de CPM
Canales de adquisiciónUn anuncio publicitario puede ser rechazado y provocar aumentos de CPM o suspensión de la cuenta aunque la creatividad cumpla las políticas, si el usuario puede acceder a contenido no conforme a tres clics de distancia dentro del mismo dominio. Para resolver penalizaciones persistentes, se recomienda migrar la pauta a un nuevo dominio completamente limpio.
terms and services. Now remember your ad could be completely compliant but if somebody clicks on that ad and within three clicks on your domain they can reach anything that is not compliant your ads will get rejected your CPMs will go up and you could be banned from running ads on platform. So, in a very the way to prevent your ads from getting rejected is to either if you just can't seem to overcome the issue, make a new domain, run your ads to that domain, make sure it's totally clean and law-abiding and and legit, and uh you should see all of your issues
Escúchalo en el minuto 5:24
Consolidación de conjuntos de anuncios en campañas CBO según el presupuesto diario
Canales de adquisiciónEn campañas con presupuesto optimizado a nivel de campaña (CBO) de escala moderada (como 300 USD al día), dispersar el presupuesto en muchos adsets (por ejemplo 9) fragmenta el aprendizaje; si el costo por adquisición oscila entre 30 y 50 USD, conviene consolidar los anuncios en un máximo de 1 o 2 conjuntos.
turn them off. Yes, Sebastian. At $300 a day, unless your cost per purchase is like five bucks, you can't afford to probably have more than two adsets. Um, absolutely, honestly, at $300 a day, if your cost per purchase is less than is is more than like 30, 40, 50 bucks, I'd probably take your 322 ads um, and drop them all in one ad set. I would probably
Escúchalo en el minuto 9:32
Estrategia multicanal con VSL largo, recortes para anuncios y retargeting por tiempo de visualización
Canales de adquisiciónGraba un video de ventas (VSL) extenso de 30 a 60 minutos alojado en YouTube dentro de tu landing page, y recorta piezas de 3 a 4 minutos para usarlas como anuncios en Meta. Esto permite hacer retargeting en Google y Meta a quienes interactúan con el video o la página, e incluso retrasar la activación del píxel (por ejemplo, 2 minutos) para capturar únicamente a los usuarios con alto compromiso.
is get one long ass VSSL video, like 30, 40, 50, 60 minutes, like a proper [ __ ] VSSL, and then cut it down and run all the cut downs that are three or four minutes long as ads. I would have the VSSL on that landing page be a YouTube video. And that way, if I'm running Google ads in addition to Facebook ads, I can retarget the individuals that watch that video. I can retarget the people that have been to the site. And I could even delay pixel fires on that website to say the pixel doesn't even fire for 2 minutes and assume that that's essentially somebody that was watching the video.
Escúchalo en el minuto 35:22
El algoritmo publicitario se entrena con modelos de atribución, no segmentando por intereses
Canales de adquisiciónEl píxel de seguimiento no almacena inteligencia por sí mismo ni se entrena usando grupos de intereses; la optimización algorítmica se guía mediante la ventana y modelo de atribución configurados, que le enseñan a la máquina qué tipo de conversión debe maximizar.
here we go. 4PI. Is it true that for new pixels you have to use interest to train it? How true is that since this video debunked a lot of concepts? That is 100% false. Um, first off, the pixel doesn't know anything. Um, the pixel is like if you're voting, the pixel is the paper that the vote is on. It doesn't carry any information. It's just code. Uh, but outside of that, the algorithm that you're teaching using interest groups does nothing to train it. training it, which I'm a big fan of doing, is all about showing it what successful looks like. The way you train the algorithm is with…
Escúchalo en el minuto 15:36
Medición y experimentosAumentar indiscriminadamente la cantidad de anuncios activos (por ejemplo, 10, 20 o 50 creativos a la vez) multiplica exponencialmente las permutaciones del recorrido del usuario, haciendo matemáticamente y financieramente imposible obtener significancia estadística real sobre qué funciona.
So if we have 10 ads and we want to get one subset of permutations. Five is 120. 10 is three million. What If you have 20 ads, four, five, six commas. It's 2.4 What if you have 50 ads? six times more money in the entire than the entire global economy If you think running more ads is a way to have good data to understand how to scale a business, you've lost the plot.
Escúchalo en el minuto 25:26
Cómo desbloquear la entrega de nuevos anuncios en campañas CBO
Canales de adquisiciónSi Meta no asigna presupuesto a anuncios recién incorporados en una campaña CBO, la solución no es seguir agregando más creativos: si el rendimiento general lo permite, incrementa el presupuesto; si no es rentable para subir presupuesto, apaga los anuncios con bajo rendimiento para forzar a la máquina a distribuir gasto hacia las nuevas variantes.
spend. Uh you have way too many ads. simple. Uh you don't need new ads to get spend. You need the machine to be good enough to increase spend. If you can't increase spend because the performance isn't good enough, turn off the ads that Um, if your new ads aren't getting spend, but you can increase the budget, just increase the budget. If your new ads aren't getting spend, but your performance isn't good enough to increase the budget, turn off the ads that aren't good enough. Eventually, those new ads will have a chance or you will just keep scaling the budget.
Escúchalo en el minuto 30:15
Canales de adquisiciónUtilizar anuncios pagados para descubrir qué producto tiene demanda es una de las vías más costosas e ineficientes de validación. La estrategia correcta consiste en determinar primero qué producto ya se vende y luego emplear la inversión publicitaria para escalarlo.
sets? Edith, you don't run ads to test which product sells best. You run ads to the product that is selling the best. Running ads to test products is these one of especially on Facebook is one of the most expensive ways to understand what works. I would highly highly recommend that you don't invest in the most expensive way where 90% of people fail. My suggestion is that you figure out what sells and then you begin to scale that with ads. Otherwise, you're just investing a whole bunch of money into something that might just be a huge pit.
Escúchalo en el minuto 57:32
Medición y experimentosEl conjunto de anuncios ganadores debe funcionar como el grupo de control estable en la cuenta, escalando el presupuesto a nivel de campaña mientras el testing creativo busca responder únicamente si los nuevos activos permiten incrementar la inversión global de forma segura.
set? I believe that you put your best ads in a winner's ad set, but you scale the campaign. So, the purpose of the the winner's ads set is to be the most stable. It is the baseline. It is the control in your scientific method. The purpose of all testing is to beat that control or complement that control so that you can raise the budget so that Creative testing isn't do I get a winning ad. Creative testing is does this portfolio of assets allow me to increase my spend. If you're running are answering questions. The only question you need to answer is does this allow me to increase my budget?…
Escúchalo en el minuto 23:24
Pausar anuncios evaluando el porcentaje de presupuesto que concentran
Canales de adquisiciónPara apagar anuncios sin desestabilizar la entrega de una campaña, ordénalos por inversión acumulada: desactivar un anuncio que concentra la mayor parte del gasto (por ejemplo, el 80%) provocará una disrupción proporcional en el rendimiento, mientras que pausar creativos con gasto marginal no generará impacto estadísticamente significativo.
how to safely turn off ads. Which ads can I pause without breaking the funnel? Sort by amount spent. If you've spent 500 bucks, the ad that spent 400 is responsible for 80%. Turning that ad off is an 80% disruption. Conversely, if there's an ad that spent 10 bucks, that is statistically insignificant. And if you turn it off, it will make a statistically insignificant change in your results.
Escúchalo en el minuto 51:06
Canales de adquisiciónPara optimizar anuncios de catálogo (DPA), desglosa los reportes por ID de producto para evaluar el gasto real frente a las conversiones en la tienda, y excluye mediante etiquetas personalizadas aquellos artículos que absorben presupuesto sin generar ventas.
you help me? Um, so if you go to uh delivery and then in the breakdown section of your Facebook dashboard, uh post ID or product ID. And at that point, you'll be able to see the level of investment by product ID. Sometimes got clicks? You don't really care what products in the catalog got clicks. What you care about is what got spend. And then you go to your store and say, "What is the traffic per page?" And if the traffic per page doesn't match the spend, i.e. something got a ton of spend, but it got no clicks, uh, odds are it's not driving sales. If you do that and then go to the conversion…
Escúchalo en el minuto 20:41
Limitar conjuntos de anuncios y creativos cuando el CPA es alto respecto al presupuesto
Canales de adquisiciónCuando el costo por adquisición (CPA) objetivo es elevado en comparación con el presupuesto disponible (por ejemplo, $200 de CPA con $1,000 de presupuesto), no debes utilizar más de un conjunto de anuncios ni exceder 5 creativos activos, ya que no existe suficiente volumen de datos para optimizar múltiples variaciones.
Uh, $200 CPA and $1,000 budget. How many ad sets should you run? And if I add a new ad set, how many ads? How much increase budget? I mean, $200 CPA and $1,000 budget, you can't afford more than one adset, man. I would not try to have more than five ads and one adset. You You don't have the luxury of the volume of data to get out of that. Uh yeah, all
Escúchalo en el minuto 53:20
Dirigir tráfico de anuncios probados a la página principal en lugar de una ficha de producto
ConversiónEnviar tráfico pagado a la página principal (homepage) en vez de a una página de producto específica suele elevar la tasa de conversión y el ticket promedio (AOV), ya que transforma una decisión binaria de compra ('comprar esto o nada') en una opción de selección entre múltiples productos dentro de la tienda.
But if you asked which shirt would you like, everything changed. Instead of being a yes or no decision, it was a choice. And choice dramatically increased conversions. A product page forces a this or that decision. And basically people choose when it's this or nothing, they choose nothing. The homepage gives people the freedom to choose. And freedom almost always lifts conversion rate in AOV. proven assets, duplicate them, send that traffic to your homepage. Your store is already built to convert better, upsell better, and capture more value and sell Your homepage isn't just a page. It's a…
Escúchalo en el minuto 44:03
Medición y experimentosMétricas intermedias como el hook rate o el hold rate son secundarias frente al objetivo real de un anuncio publicitario: evaluar si el creativo permite aumentar de forma rentable la inversión y el volumen de gasto publicitario.
But here's the thing. Um, I don't care about the hook rate and the hold rate, your evaluation of an ad is does it help me spend more money. That's so if it the job's done, great. If the ad doesn't help you do that, then it's not great or it's not good. Uh I've
Escúchalo en el minuto 34:31
Medición y experimentosEl costo por lead objetivo no debe basarse en métricas superficiales de ROAS porcentual, sino en determinar el valor monetario real (beneficio bruto) de cada lead y balancear el margen unitario frente al volumen total de adquisición deseado.
a lead cost target is essentially we have to understand what is a lead worth. Um and then if we understand what a lead is worth and how much we can get. Let's say a lead is worth 10 bucks and it costs us five to get it. Well, if we want to make 50 bucks, we need 10 leads. margin not by percentages, right? ROAZ is [ __ ] but like what is the gross profit, the GPT of that lead times volume of leads, and then your ad spend, right? That's how you figure it out. Um, if you want to make more money with less money, you got to be more efficient. If you're willing to invest more to get a smaller…
Escúchalo en el minuto 17:29
Identificar creatividades ganadoras en anuncios dinámicos revisando la bandeja de comentarios
Canales de adquisiciónPara descubrir qué variaciones de post ID están ganando en anuncios flexibles o dinámicos de Meta, revisa la bandeja de entrada del Business Manager: los anuncios que concentran la mayor cantidad de comentarios son los que están recibiendo la mayor parte del presupuesto y tracción por parte del algoritmo.
ads works the same way. And if you want to know what your top ads are, go to the inbox in your business manager, look at the comments. The ads that get the most comments are the ones that are earning the most spend that are driving the most engagements. Find the outliers. Then it's really easy because hitting the comments and replying back to people, you're replying back to them on the winning post IDs. If you can't find
Escúchalo en el minuto 58:38
Medición y experimentosNo intentes clasificar anuncios en etapas rígidas del embudo (TOFU, MOFU, BOFU) apoyándote en el desglose de audiencias nuevas e interactivas de Meta; estos datos no son del todo precisos ni accionables para tomar mejores decisiones de optimización publicitaria.
how do we identify top, middle, and bottom funnel ads when the same ad is being shown to both new and engaged audiences? First off, um new and engaged audiences doesn't mean [ __ ] Don't worry about that. That breakdown in the ad account doesn't mean [ __ ] anything. It's not actionable data. it's not necessarily correct either. Um, and it has no value in your ability to make better decisions. So, with that in mind, never look at that information. Uh, data that isn't actionable is a
Escúchalo en el minuto 49:42
Riesgo de fragmentación de datos por exceso de conjuntos de anuncios en CBO
Canales de adquisiciónEjecutar demasiados conjuntos de anuncios dentro de una campaña con presupuesto a nivel de campaña (CBO) dispara exponencialmente el número de permutaciones posibles, fragmentando la entrega y haciendo que los resultados del test sean inconclusos.
And uh for those of you that are running nine adsets in a CBO, you can't know. Um it's not that you're not smart enough, it's not that you're not spending enough, it's that the volume of permutations is astronomical. Okay, so this is what happens if you have five ads to get a data set of one. There's 120 different ways. So if you
Escúchalo en el minuto 24:22
Canales de adquisiciónNo existe una etapa fija del embudo (superior, media o inferior) que sea universalmente óptima para concentrar los anuncios de pago. La estrategia debe basarse en potenciar lo que el negocio ya domina y cubrir sus puntos débiles; por ejemplo, una marca que destaca generando tráfico masivo con videos virales en la parte alta del embudo debe enfocar sus esfuerzos en desarrollar creatividades estáticas para convertir a…
the thing, Smiles. There is no best place in the funnel for you to focus your ads on. What really matters is that you are prioritizing um what you do well and what you need. here is this brand that joined crushes an upperfunnel video. All of their videos go viral. They're incredible at it. So the number one way to get their business better is to start making statics that appeal to those individuals. And if you watch
Escúchalo en el minuto 13:15
Canales de adquisiciónNo existe una etapa fija del embudo (superior, media o inferior) que sea universalmente óptima para concentrar los anuncios de pago. La estrategia debe basarse en potenciar lo que el negocio ya domina y cubrir sus puntos débiles; por ejemplo, una marca que destaca generando tráfico masivo con videos virales en la parte alta del embudo debe enfocar sus esfuerzos en desarrollar creatividades estáticas para convertir a…
the thing, Smiles. There is no best place in the funnel for you to focus your ads on. What really matters is that you are prioritizing um what you do well and what you need. here is this brand that joined crushes an upperfunnel video. All of their videos go viral. They're incredible at it. So the number one way to get their business better is to start making statics that appeal to those individuals. And if you watch
Escúchalo en el minuto 13:15
Canales de adquisiciónNo existe una etapa fija del embudo (superior, media o inferior) que sea universalmente óptima para concentrar los anuncios de pago. La estrategia debe basarse en potenciar lo que el negocio ya domina y cubrir sus puntos débiles; por ejemplo, una marca que destaca generando tráfico masivo con videos virales en la parte alta del embudo debe enfocar sus esfuerzos en desarrollar creatividades estáticas para convertir a…
the thing, Smiles. There is no best place in the funnel for you to focus your ads on. What really matters is that you are prioritizing um what you do well and what you need. here is this brand that joined crushes an upperfunnel video. All of their videos go viral. They're incredible at it. So the number one way to get their business better is to start making statics that appeal to those individuals. And if you watch
Escúchalo en el minuto 13:15
Escalar anuncios ganadores unificándolos bajo la misma oferta en CBO
Canales de adquisiciónEs válido seleccionar creativos individuales ganadores de distintos conjuntos de anuncios para escalarlos en una campaña CBO, siempre y cuando todos los anuncios correspondan a la misma oferta y no se apresure la extracción.
the idea of picking individual winning ads from all adsets and then scaling it. I've been teaching people to do that for 5 years. Um, so totally on board with that. But make sure they're all of the same offer and in a CBO campaign and don't be in a rush to grab those. I'll get into Jason.
Escúchalo en el minuto 34:01
La fatiga de anuncios como consecuencia de no estructurar un equipo de creativos complementarios
Canales de adquisiciónLa fatiga publicitaria a menudo se origina por un error de gestión al sobrecargar un único anuncio ganador y esperar que sostenga el rendimiento por sí solo; en lugar de buscar un ganador aislado, la cuenta publicitaria debe operar como un equipo donde distintos creativos apoyen diferentes alturas del embudo.
the same thing. The best ads are the ones that were running for a long time. The more new ads I launch, the faster my ads fatigue and the harder I work and the more I have to launch more new ads Here's the thing. Ad fatigue is operator efforts, where does that ad sit in the funnel? It might crush as a midfunnel ad according to 4PI. Support it with the rest of the team. Um, like if it's just Messi on the pitch, he's going to lose If you're winning, if you're focusing on finding a winning ad and then scaling it, what you're doing is saying, "Let me find the best player that I have
Escúchalo en el minuto 54:08
Medición y experimentosUna prueba de anuncios con presupuesto inicial de $100 al día escaló a $10.000 de gasto diario y generó hasta $100.000 en ingresos diarios en clientes de comercio electrónico.
last year I ran a $100 a day test for one of my clients. Nothing flashy, nothing complicated. But within days, it scaled to $10,000 a day, and on some days, it drove $100,000 in revenue. the
Escúchalo en el minuto 41:34
Seguir leyendo
Esta es una de las 1872 entrevistas despiezadas. Puedes
ver todas, leer las ideas por tema o
empezar por la ruta guiada si no sabes por dónde.