Marketero · Entrevistas
Meta Ads expert reacts to $1M/month Strategy
24 ideas de marketing sacadas de «Meta Ads expert reacts to $1M/month Strategy», de Professor Charley T. Sobre todo de canales de adquisición y medición y…
- 24 ideas despiezadas
- Fuente Professor Charley T
- Dura 46 min
- Publicada en 2026
Ver la entrevista entera en YouTube
De qué habla
Las 24 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 24 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.
Canales de adquisiciónEn la publicidad moderna en Meta, la segmentación ya no se controla mediante la configuración técnica a nivel de conjunto de anuncios (intereses o demografía), sino a través del propio mensaje y el contenido creativo, que guían la entrega algorítmica.
Again, targeting is controlled by messaging. Nowadays, targeting is rarely controlled on the ad set level. Like it's historically done. That's an example of an archaic outdated belief. Messaging controls targeting nowadays. So again, we. Yeah. Creative controls delivery. That's been true since 2018. And if messaging and targeting controls delivery, why are we using interest groups
Escúchalo en el minuto 4:34
Canales de adquisiciónAl probar lotes de 25 a 30 creativos en plataformas pagadas, los algoritmos suelen concentrar el gasto en solo 1 a 3 anuncios. Intentar forzar la distribución separando cada anuncio en su propio conjunto complica la cuenta, genera solapamiento de audiencias y perjudica el rendimiento del negocio.
single individual ad that exists. That would be exhausting. Long story short, we also typically find that about one to three of these ads are going to get all the reach and all the distribution. that view. Outdated belief is every creative deserves distribution. It doesn't. Okay. You're going to feel like it's wasted effort when you go launch 25 to 30 ads and one to three of them get all the reach. Okay. That means just to put it in perspective, let's say that we launched with 30. That means that we are going to have 27 to 29 with little to no reach. Like they'll have spent pennies. fucking…
Escúchalo en el minuto 16:46
Canales de adquisiciónSegmentar mediante públicos similares (lookalikes) o intereses fragmenta el presupuesto y reduce la densidad de datos del algoritmo, funcionando en la práctica como filtros de exclusión costosos basados en muestras reducidas (por ejemplo, usar 1.000 registros para excluir al 99% del mercado). Su aparente rendimiento superior suele ser una ilusión de atribución de último contacto que canibaliza el mérito del tráfico…
lookalike audience. Now in the United States, a 1% lookalike audience is around 2.1 million people. What you're doing is saying use a thousand data points to produce something 2000 times bigger by eliminating something that is an order of magnitude greater than that of 100 because it's a 1% lookalike. So you're saying, let me use a thousand data points to exclude 99% of people. Remember that lookalikes are an expensive exclusion audience. They're not an inclusion audience. especially knowing that the messaging does the targeting anyway. Lookalikes are essentially an sentence, as a fully built…
Escúchalo en el minuto 5:45
Canales de adquisiciónAl probar lotes de 25 a 30 creativos en plataformas pagadas, los algoritmos suelen concentrar el gasto en solo 1 a 3 anuncios. Intentar forzar la distribución separando cada anuncio en su propio conjunto complica la cuenta, genera solapamiento de audiencias y perjudica el rendimiento del negocio.
single individual ad that exists. That would be exhausting. Long story short, we also typically find that about one to three of these ads are going to get all the reach and all the distribution. that view. Outdated belief is every creative deserves distribution. It doesn't. Okay. You're going to feel like it's wasted effort when you go launch 25 to 30 ads and one to three of them get all the reach. Okay. That means just to put it in perspective, let's say that we launched with 30. That means that we are going to have 27 to 29 with little to no reach. Like they'll have spent pennies. fucking…
Escúchalo en el minuto 16:46
Canales de adquisiciónNo se deben copiar las tácticas publicitarias actuales de empresas multimillonarias si no se cuenta con su volumen de distribución orgánica, recurrencia o ventas minoristas; los negocios en crecimiento deben aplicar lo que esas marcas hicieron para escalar, no lo que hacen cuando ya pueden permitirse el desperdicio.
And to be fair, for a lot of people, that's all they need. Alex and Mosey's ad strategy isn't good. It's objectively bad, but his organic strategy is so strong, his ads don't need to be good. Same thing with Grooms or with AG1 or with Nike. People follow all these huge brands, we're like, we're just going to do what they do. You don't have the business model they do. You don't have millions coming in every month from repeat buyers. You don't have organic distribution or retail outlets in the way that they do. So you can't follow their path. Don't do what a hundred million dollar business or…
Escúchalo en el minuto 3:36
Medición y experimentosLanzar decenas de anuncios y combinaciones dinámicas a la vez dispersa el presupuesto e imposibilita alcanzar significancia estadística, dejando la distribución en manos del azar inicial en vez de un aprendizaje científico validado.
it looks like to move from ad to ad to ad. When you have 25 to 30 ads, there is no way for you to look more money to exclude a bunch of good people just so that you can have dumber machines with 25 to 30 ads. That's nearly a hundred ads running. Do you have the budget to have any statistical significance across 75 to 90 ads? No, you don't. I don't care what your budget is. That number is astronomical. The amount of variations that you would need to get through just 50. Let me give you a concept of what that looks like. Let's say you did one variation a second. You would need 78 billion years…
Escúchalo en el minuto 10:18
Medición y experimentosLanzar decenas de anuncios y combinaciones dinámicas a la vez dispersa el presupuesto e imposibilita alcanzar significancia estadística, dejando la distribución en manos del azar inicial en vez de un aprendizaje científico validado.
it looks like to move from ad to ad to ad. When you have 25 to 30 ads, there is no way for you to look more money to exclude a bunch of good people just so that you can have dumber machines with 25 to 30 ads. That's nearly a hundred ads running. Do you have the budget to have any statistical significance across 75 to 90 ads? No, you don't. I don't care what your budget is. That number is astronomical. The amount of variations that you would need to get through just 50. Let me give you a concept of what that looks like. Let's say you did one variation a second. You would need 78 billion years…
Escúchalo en el minuto 10:18
Optimizar la captación de leads por calidad y rentabilidad del equipo de ventas
Medición y experimentosAumentar ciegamente el volumen de leads sin evaluar su calidad sobrecarga y desmotiva al equipo comercial, reduciendo su tasa de cierre. La optimización publicitaria en generación de leads debe vincular el concepto del anuncio con las métricas posteriores al clic y la conversión final en llamadas, maximizando el valor económico del tiempo de los vendedores.
the value of the time spent by the call center. And I think if you're avoiding the scientific method, and not getting down to the unit economics of somebody's time, that's actually responsible for generating revenue for your business, you're not running a good business at all. And if the idea is, well, I'm just going to get more leads, I'm going to get more leads, I'm going to get more leads, you're going to burn out your call people. Like if they're just instead of taking three calls, now you want to take five. Do you think they're going to close even the same rate they did when they were at…
Escúchalo en el minuto 41:21
Superar el techo de escala mediante monetización posterior e incrementalidad
Medición y experimentosOptimizar campañas publicitarias solo por eficiencia en el front-end impone un techo de inversión donde el CPA se dispara; la verdadera capacidad de escalar depende de generar demanda incremental o de aumentar el valor del cliente (conversión comercial, email marketing y recompra) para permitirse pagar más por la adquisición.
passed, usually we hit a ceiling. Okay. And with that ceiling, there's just a hard limit where we don't get the opportunity to spend above it. Every single dollar that we spend above that ceiling, to add incrementality to the funnel. All of this is essentially hitting a limit because it can only take so much credit for the work of everything else that's going on outside of Facebook. So the way how do I get rid of all of this shit so I can understand what would scale my search, my email, and my conversion rate and the people that come back and buy again and again in my system or improve the…
Escúchalo en el minuto 27:37
Optimización de campañas de leads mediante eventos de conversión offline y métricas de calidad
Medición y experimentosEn lugar de optimizar campañas de captación por volumen bruto o basarse en el feedback subjetivo del equipo comercial, se deben enviar los eventos de conversión offline (aperturas de correo, llamadas agendadas, asistencia y calificación) a la plataforma publicitaria para alinear los anuncios con el valor y el retorno del tiempo de ventas.
that call showed up. Well, you can pass that back as offline conversion events, and then see that in the campaign, see that as additional columns, see that this ad's getting a low quality lead. This ad's getting a high quality lead. You shouldn't be running lead gen around anything other than what is the value of the salesperson's time optimized for the dollar of their hour to the call to the showed up to the quality, all of these are qualifications that can be mapped digitally. Why are you not measuring all of that stuff and optimizing down through the funnel to the stuff and not even…
Escúchalo en el minuto 35:31
La trampa de optimizar el embudo publicitario basándose solo en el último punto de contacto
Medición y experimentosOptimizar anuncios atribuyendo el éxito exclusivamente al último clic favorece los impactos de fondo de embudo y elimina los activos que generan demanda inicial, impidiendo escalar el negocio de forma incremental.
other than saying which one is able to take credit at the very last touch point. You're not able to see what scales the funnel. You're not able to see what drives incrementality. You're not able to see why am I getting new customers versus returning ones. You're not able to actually measure the impact on any of the business. You're just focused on what gets credit for the very last touch point. And then you optimize out everything that sucks, which is why you can't scale the business because you're basically just optimizing for the lowest funnel, least incremental efforts. That's why you're…
Escúchalo en el minuto 10:27
Validar los anuncios ganadores por la calidad del prospecto y no solo por el coste por lead
Medición y experimentosUn anuncio con métricas técnicas favorables y bajo coste por resultado no debe considerarse ganador si los clientes potenciales que atrae no son cualificados; la optimización debe guiarse por la retroalimentación de ventas y eventos de calidad de llamada.
winners within that ad set sucked. Okay. Meaning they're technically working like they're bringing in the result you want. They're technically bringing in a cost per result that might be favorable, but you go listen to the calls or you see who shows up to your webinars, or you just look at the leads in general. And you hear the feedback from your sales team and they're saying like, nah, these people suck that are coming in. If they suck. Okay. And you don't like who's coming process. Why aren't you optimizing to quality call? If this is truly for lead gen, that right now. Why not have…
Escúchalo en el minuto 34:07
Canales de adquisiciónLanzar entre 20 y 30 anuncios únicos dentro de un mismo conjunto de anuncios dispersa el presupuesto e impide que el algoritmo de Meta aprenda la secuencia óptima de impactos entre los distintos puntos de contacto por falta de volumen de datos.
level. So on the ad set level, I usually launch with about 25 to 30 unique ads. If you have not fucking cooked. Okay. Yeah. You're getting absolutely fucking cooked, especially if you're running 25 to 30 ads. Do you really think the machine has any ability to understand the right sequence of one touch point to another touch point to another touch point when you've got dozens of them? Let me tell you, you don't have enough money for all of that. If you've got 20 ads, I don't think you understand
Escúchalo en el minuto 9:04
Canales de adquisiciónLanzar entre 20 y 30 anuncios únicos dentro de un mismo conjunto de anuncios dispersa el presupuesto e impide que el algoritmo de Meta aprenda la secuencia óptima de impactos entre los distintos puntos de contacto por falta de volumen de datos.
level. So on the ad set level, I usually launch with about 25 to 30 unique ads. If you have not fucking cooked. Okay. Yeah. You're getting absolutely fucking cooked, especially if you're running 25 to 30 ads. Do you really think the machine has any ability to understand the right sequence of one touch point to another touch point to another touch point when you've got dozens of them? Let me tell you, you don't have enough money for all of that. If you've got 20 ads, I don't think you understand
Escúchalo en el minuto 9:04
La optimización a nivel de conjunto de anuncios (ABO) provoca canibalización y solapamiento
Canales de adquisiciónConfigurar campañas frías utilizando optimización de presupuesto por conjunto de anuncios (ABO) combinando audiencias amplias, intereses y públicos similares (lookalikes) compitiendo entre sí fragmenta la información y genera canibalización interna, en lugar de permitir que el algoritmo de Meta distribuya el presupuesto de forma eficiente a nivel de campaña (CBO).
to get delivery. Now this is starting to make sense. This is a fucking complicated system of a shit ton of overlapping cannibalization mapped to an ABO. So really think of this as there is no campaign. It's just three campaigns with one ad set apiece with a shit ton of ads, all competing against each other, not sharing any data at all, all competing to be the last touch point according to now you have ad fatigue and then you have a whole bunch of cannibalization and audience overlap. And like he mentioned that audience overlap was a bad thing and then structures a system to have So this…
Escúchalo en el minuto 20:31
La iteración creativa exige hipótesis estratégicas en lugar de producción aleatoria
Canales de adquisiciónGenerar anuncios de reserva de forma indiscriminada sin una dirección creativa basada en datos equivale a jugar a la lotería. Las nuevas creatividades deben diseñarse bajo el método científico para abordar objeciones concretas, dirigirse a cohortes con mayor probabilidad de cierre o captar etapas superiores del embudo, en vez de duplicar conjuntos de anuncios y probar variantes al azar.
off the ad that sucks? If an ad is getting you results that you don't like, why are you still running that ad? Why duplicate the ad set? How many fucking ad sets are live in these ad accounts? you slapping on shit instead of just removing the thing that sucks? Why are you not eliminating waste and measuring for quality? Why is the scientific method and the value of somebody's time and dollar I've been waiting for it, but he never got to. What should those ads be? Are they just fucking random shots in the dark with somebody who built a system out of It's still a $5 fucking scratch-off. Like…
Escúchalo en el minuto 42:37
El gasto algorítmico temprano premia señales de interacción, no la incrementalidad del negocio
Medición y experimentosLas plataformas publicitarias concentran el presupuesto inicial basándose en señales inmediatas de experiencia de usuario y probabilidad de conversión rápida, lo que premia creativos que capturan crédito de demanda existente en lugar de aquellos que generan ingresos netos incrementales.
Full disclosure. I like where Jeremy's going, but the reason the ad got spent early is because there was some signal in the first few impressions that that's what end users wanted to see the most. And because you've grouped all of these ads with a million permutations, it's just going to go off the early positive signals. It's not saying that that's the good ad. It's not saying that's the bad ad. It's the ad that gave the most highest quality with highest confidence and user experience in a way that that trillion dollar company knew how to deliver to the end user. Now, if that's good or bad…
Escúchalo en el minuto 18:07
Canales de adquisiciónDuplicar conjuntos de anuncios para forzar que los creativos que el algoritmo no distribuyó reciban presupuesto genera canibalización, solapamiento de audiencias e inestabilidad en la cuenta publicitaria, en lugar de un crecimiento incremental.
that got choked out by them be able to spend in some other ad set, essentially creating a new campaign to launch all the other ideas that is again, cannibalistic and not complimentary or compounding in value, why not just turn off the ad that sucks? That's soaking up the budget so you don't create more cannibalization, more overlap, more problems, more issues, But again, you basically just have a completely new environment that cannibalizes the other environments that has to compete with them without adding something to it. So you're just
Escúchalo en el minuto 31:25
Canales de adquisiciónDuplicar conjuntos de anuncios para forzar que los creativos que el algoritmo no distribuyó reciban presupuesto genera canibalización, solapamiento de audiencias e inestabilidad en la cuenta publicitaria, en lugar de un crecimiento incremental.
that got choked out by them be able to spend in some other ad set, essentially creating a new campaign to launch all the other ideas that is again, cannibalistic and not complimentary or compounding in value, why not just turn off the ad that sucks? That's soaking up the budget so you don't create more cannibalization, more overlap, more problems, more issues, But again, you basically just have a completely new environment that cannibalizes the other environments that has to compete with them without adding something to it. So you're just
Escúchalo en el minuto 31:25
Desactivar las optimizaciones automáticas de Advantage+ Creative excepto el destaque de comentarios positivos
Canales de adquisiciónAl configurar creativos en Meta Ads, es preferible apagar las funciones automáticas de Advantage+ Creative (como enlaces de sitio o modificaciones de formato) para evitar distorsiones del anuncio, preservando únicamente la opción que resalta comentarios positivos.
Let's keep going. We'll go down on the ad level. One more little pro tip, and we'll make sure none of that stupid shit when it comes to the AI advantage creative stuff is turned on. No site links, no site links, none of the little fucking talk. The only one that I like right now, to be fair, with the AI creative fucking advantage stuff that they're pushing is the one that highlights a positive comment. I do like that one. All the other ones can get fucked. Got to make sure all
Escúchalo en el minuto 23:08
Canales de adquisiciónLa segmentación por intereses en Meta es imprecisa porque no refleja intención ni sentimiento de compra: agrupa a personas que interactuaron negativamente con un tema o que fueron asignadas por error, lo que introduce ineficiencia en campañas de adquisición directa.
Really in the year of our Lord, 2026, we're still using interest groups. The engineering team at Facebook that runs interest groups was disbanded in 2018. And fun fact on that at best, those interest groups had at least one third of the audience there by mistake. Also interest is not sentiment or intent. So if you were to say something like, I hate cats, you're in the cat interest group. So really one third of that audience is there by mistake. One half of that audience might be there because they feel negatively about something. And the people who do feel positively about something, maybe…
Escúchalo en el minuto 2:24
Canales de adquisiciónDuplicar conjuntos de anuncios colocando un solo anuncio dentro de cada uno para forzar la distribución genera costes inflados artificialmente y no ofrece aprendizajes científicos replicables sobre el rendimiento creativo.
Because what you think you're doing is you think you're doing creative testing. When you duplicate out that single ad set a bunch of times over and you put one ad inside of each ad set in order for that ad to get all the distribution. First of all, you're getting artificially inflated costs. That's one thing to really consider. That's what we find time and time again. When we test, we just get an artificially inflated cost compared to what this structure would produce. And on top of that, this is the most Okay. Your creative testing is scratch off tickets. There's nothing scientific about…
Escúchalo en el minuto 25:28
Canales de adquisiciónDuplicar conjuntos de anuncios colocando un solo anuncio dentro de cada uno para forzar la distribución genera costes inflados artificialmente y no ofrece aprendizajes científicos replicables sobre el rendimiento creativo.
Because what you think you're doing is you think you're doing creative testing. When you duplicate out that single ad set a bunch of times over and you put one ad inside of each ad set in order for that ad to get all the distribution. First of all, you're getting artificially inflated costs. That's one thing to really consider. That's what we find time and time again. When we test, we just get an artificially inflated cost compared to what this structure would produce. And on top of that, this is the most Okay. Your creative testing is scratch off tickets. There's nothing scientific about…
Escúchalo en el minuto 25:28
La fatiga de anuncios como error operativo de canibalización
Canales de adquisiciónLa fatiga de anuncios suele ser producto del error del operador al crear estructuras fragmentadas y lanzar constantemente nuevos anuncios que compiten por el mismo segmento en la parte baja del embudo, restando presupuesto y resultados a los anuncios ganadores en vez de expandir la audiencia real.
occurs, we can relaunch that ad set. The reason ad fatigue occurs, which is operator error, is because you have this entire cannibalistic environment built around the efficiency of taking credit for the rest of the work of the machine and you're not optimizing the quality. If you didn't constantly create more and more ad sets with more and more audiences with 25 permutation per ad within it and dozens of ads with those 25 permutations, you wouldn't have ad fatigue. You wouldn't have new winners coming at the expense of creating a loser. Like the reason ad fatigue occurs, by the way, 99% of ad…
Escúchalo en el minuto 39:38
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.