Marketero · Entrevistas
How Meta Ads Manager Is Lying to You & What You Can Do About It | Perpetual Traffic EP 675
23 ideas de marketing sacadas de «How Meta Ads Manager Is Lying to You & What You Can Do About It | Perpetual Traffic EP 675», de Perpetual Traffic. Sobre…
- 23 ideas despiezadas
- Fuente Perpetual Traffic
- Dura 1 h 01 min
- Publicada en 2025
Ver la entrevista entera en YouTube
De qué habla
Las 23 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 23 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.
Riesgo de clasificación errónea de audiencias nuevas en Meta Ads por ventanas de atribución y listas desactualizadas
Medición y experimentosMeta Ads no define a las 'audiencias nuevas' de forma histórica o de todos los tiempos, sino según la ventana de retrospección activa (por defecto 7 días o hasta 28 días) y las listas de clientes cargadas. Si las listas de clientes no se sincronizan continuamente o si un cliente interactuó fuera de la ventana de atribución, compradores reales preexistentes serán catalogados erróneamente como público nuevo en los…
new Now new as defined these are people who have not interacted with your products or services so it's dependent here another big factor we are not looking at all time these are not people who have not interacted with your products or service all time these are people just that are in your audiences but also really in a 28 day look back have not interacted with your products or services in the last seven days or would you default to say in the last 28 days what's your guess here um it's G to be seven days because in order to get audiences correctly correct because if you only uploaded that…
Escúchalo en el minuto 30:43
Desglosar las campañas de Meta Ads por segmentos de audiencia para auditar el CPA real de adquisición
Medición y experimentosEn Meta Ads Manager, se debe utilizar el desglose por segmentos de audiencia para evaluar por separado el costo de adquisición de clientes completamente nuevos frente a audiencias tibias o existentes; basarse únicamente en el CPA combinado (blended CPA) enmascara adquisiciones frías sobrecosteadas con compras de usuarios que ya interactuaban con la marca.
good cheat if you're auditing an ad account to know if the media buyers you go into breakdown you go to demographics you click on audience segments then what that will do is add now you're empowered to understand the distribution of your media across the various audience types to get a better look at that your new is 149 Blended is 128 so you're already given a different handicap if you're not breaking this down because you can see existing don't do this you're not looking at the 150 number you're looking at the 128 and
Escúchalo en el minuto 35:23
Recopilar consultas de atención al cliente para descubrir nuevos casos de uso e insights del mercado
Investigación de mercado y clienteEl valor del soporte y la moderación no radica únicamente en responder preguntas frecuentes, sino en recopilar y analizar los patrones de esas consultas para alimentar a marketing con insights no detectados previamente (como nuevos casos de uso o contextos de consumo sugeridos por la propia audiencia).
train it on your FAQs you're paying for in-house or offshore Town it doesn't matter to do low priority tasks like low productivity tasks versus like okay what if you collated all of those responses and use it for Consumer insights hey we're getting a lot of comments about this product working at the beach we've never marketed it at the beach resource here's an idea that we can take something with that data and run with it that's what a customer service person if they're just replying needs to then be providing consumer Insight but they're
Escúchalo en el minuto 11:20
Diferenciar el CAC de clientes nuevos (nCAC) del CAC combinado (aCAC)
Medición y experimentosEs necesario distinguir entre el costo de adquisición de clientes estrictamente nuevos (nCAC) y el CAC global (aCAC). El CAC combinado casi siempre resulta más bajo porque mezcla audiencias templadas, recompras y ventas ascendentes, ocultando la fricción y el costo real de captar usuarios que nunca han comprado.
actually see AAC which is different AAC returning customers and that aack or the all customer acquisition cost because even lower at $82 so always be lower you but your enaga new customer always has extra friction unless you have some spent into that audience your Alla because it Blends your evangelists and together so 90 9.9% of the time it'll be exception but enac is new they are not in your existing customer database they but you really do need to differentiate the two between you know everybody CU
Escúchalo en el minuto 45:45
Retención y relación con el clienteEl equipo de atención al cliente y gestión de redes sociales no debe limitarse a responder dudas o validar comentarios de forma reactiva considerándose un centro de costos; apoyado en IA y automatizaciones, debe funcionar como una extensión directa de ventas enfocada en calificar leads, convertir conversaciones en transacciones y mitigar el arrepentimiento de compra.
a year are still ignoring comments on Facebook they're still having terminal conversations their customer service team is like liking and validating comments but they're not turning those conversations into conversions even at minimal for like lead generation and Lead qualifying so it was like a an for me my language is if your customer service team is a liability they're an expense you've hired the wrong person or you're they're doing customer service wrong because customer service by itself with AI with all the automations and like scalable systems you can do your customer service team…
Escúchalo en el minuto 7:25
Medición y experimentosEs necesario efectuar cruces periódicos (semanales) entre lo que reportan las plataformas de anuncios (Meta, Google) y la fuente de verdad directa (el CRM, la tienda de comercio electrónico o la cuenta bancaria) para verificar con precisión el volumen real de clientes nuevos vs. recurrentes y el costo efectivo de adquisición.
way that we used to do this is we used to compare what we were doing inside of ads manager as well as Google with the CRM and then pull manual reports and it wanted to just make sure are we bringing in customers that actually are make sense for the business at a price that comparing it with you know the CRM and it's incredibly manual but that's the source of Truth the CRM or the bank account where you're like looking at the right right what we would do what we would do is on a weekly basis we would do a true up with the CRM or you know Shopify store and we would just double check all right…
Escúchalo en el minuto 51:57
Las interacciones en comentarios deben estructurarse para convertir y hacer sentir escuchado al usuario
Retención y relación con el clienteDar un simple «me gusta» a los comentarios en redes solo hace que el usuario se sienta visto, pero no escuchado; para generar disposición de compra y mover a los usuarios a través del embudo (de comentarios a suscriptores, luego a leads y prospectos), las respuestas deben profundizar y atender activamente sus inquietudes.
what I'm like not at all comments should become subscribers subscrib should become leads leads should become prospects and conversions and like we and like yeah we're just we're liking we're making people seen like feel seen I'm like cool they feel seen because you liked it but they don't feel heard and if I don't feel heard I don't feel inclined to buy so that that was just
Escúchalo en el minuto 10:24
Oferta y precioAl implementar un chatbot de IA que indexa todo el dominio, cualquier artículo de blog o página antigua con precios o condiciones obsoletas queda disponible para el modelo; si el bot presenta una oferta desactualizada a un usuario, la empresa se ve obligada a respetar ese precio por haber estado publicado.
like if it's on website it's fair game for the AI chatbot to pull and crawl from so the only danger is do you have outdated pricing and if you do that's fair game for the bot so someone can find access to it but at the same time is a page on my website that shows like really outdated pricing and like oh my God I better remove that before this is launched because a few times someone has found that and I have I'm honoring it because it's what was as
Escúchalo en el minuto 13:57
Medición y experimentosEvaluar las creatividades y anuncios individuales midiendo específicamente el porcentaje de adquisición de nuevos clientes (tráfico frío / eNCAC) permite identificar qué piezas creativas traccionan adquisición genuina y optimizar las campañas con datos precisos.
enca by campaign by adet by individual break that down even more so and you can see which creatives are the ones that are getting the majority of cold traffic that are actually you know getting the best uh enac and you know in this uh one particular ad about 42% of this is is new which is pretty good that's this creative here is probably one that we've iterated on multiple times and results because we're making those decisions based upon nearly 100%
Escúchalo en el minuto 55:33
Medición y experimentosLas métricas nativas dentro de plataformas como Meta o Google (in-app ROAS y CPA) no son completamente fiables debido a restricciones de cookies, regulaciones legales y a la tendencia de cada plataforma a reclamar crédito por las conversiones; deben tomarse como indicadores de tendencia y contrastarse con una fuente de verdad independiente.
bottom line is this is that the platforms are not accurate and you have to take that with a grain of salt and if you are crosschecking it with your source of truth then you can Benchmark where say well you know Ina Raz sucks because it does because it's a faulty metric inapp CPA is not accurate because they can which is why there's attribution there's an attribution war competing to claim credit right for the but there's legal restrictions um cookie restrictions and setup restrictions that will never provide accuracy so you again we say this all the time you cannot be solely dependent on…
Escúchalo en el minuto 56:56
Calibrar métricas en plataforma mediante benchmarks contra el costo real de adquisición (eNCAC)
Medición y experimentosDado que las plataformas de anuncios (como Meta) suelen reportar métricas inexactas de costo por compra, los gestores de medios deben definir factores de referencia o benchmarks (por ejemplo, saber que un CPA en plataforma de $128 corresponde en realidad a un eNCAC de $92 verificado externamente) para tomar decisiones operativas sin pausar prematuramente campañas que son rentables.
nothing away from this know that inapp numbers are inaccurate and providing the media buyer about this is like we use something that's called benchmarking and we know he probably knows that if I'm getting a cost per purchase in this adet that's about $128 which is in essence 30 40% above what I'm actually getting for enac then that's okay so even though inside ads manager 128 per purchase or or a CAC of 128 is unacceptable for the client we know that the inapp metrics are faulty and so that Benchmark he probably my to talk about this is hey if I'm getting a cost per purchase between 12 and…
Escúchalo en el minuto 50:40
Desglose de audiencias en Meta Ads y dependencia de audiencias personalizadas
Medición y experimentosPara que Meta pueda segmentar el rendimiento entre audiencia nueva (new audience), audiencia que interactuó (engaged audience) y clientes existentes (existing customers), el anunciante debe definir explícitamente en el administrador de anuncios sus listas de compradores y públicos de interacción; de lo contrario, Meta no puede categorizar adecuadamente a los usuarios nuevos ni aislar el tráfico desconocido (unknown).
this ad set is new audience so again The people who have not interacted with your products or services so again that this is engage audience is an audience that you defined key you the advertiser you the brand defined in the ad account settings as people who are aware of your business or interacted with your products or services but have not made a purchase so have existing customers so existing customers as defined by meta is an audience that you defined in an ad account settings as people who have purchased your products or services or here engage audiences and and your existing customers…
Escúchalo en el minuto 25:28
Meta mezcla tráfico frío y retargeting en la entrega a pesar de las exclusiones
Medición y experimentosEl administrador de anuncios de Meta tiende a inflar la proporción reportada de clientes nuevos (por ejemplo, reportando un 88% de tráfico frío cuando el seguimiento independiente muestra solo un 42% nuevo y un 57,4% de usuarios previos), ya que el algoritmo impacta a usuarios templados y existentes aun cuando se hayan configurado exclusiones de audiencia.
first off to say like the ad account itself is not giving you a true measure of what's actually happening it's using a 7-Day window it's blending audiences stating that 88% of your traffic is new that is totally not the case so and and roas is a faulty metric because of this know that that even with all my exclusions in this ad set and I know all the exclusions are made the audiences are set up correctly if I if I exclude all of those audiences for people to try I still can't do it because meta is going to blend those audiences together determine about 42% of these uh individuals who are…
Escúchalo en el minuto 41:21
Medición y experimentosAntes de escalar presupuesto en paid media, se debe determinar con precisión el eCAC (costo efectivo de adquisición de nuevos clientes); las métricas nativas dentro de plataformas publicitarias como Meta suelen mezclar audiencias nuevas, enganchadas y clientes existentes, mostrando datos alejados de la rentabilidad real del negocio.
know what enek is it is the cost to acquire a new customer just got off a if we don't know what our number is and if we don't know the accuracy of that number which is two things know the number know how accurate it is then we could run a lot of traffic and spent a lot of money on paid media but put you out of business that's not the goal the goal is for you to have profitable I said and that enac number is like new visits new customers getting that actual what they're seeing inside the apps is the correct number when in fact it isn't there is a way to show new audiences engaged audiences and…
Escúchalo en el minuto 17:14
ConversiónIntegrar un chatbot o widget conversacional en las páginas de destino de tráfico pagado puede generar un incremento de conversión de aproximadamente el 25%, ya que permite resolver dudas puntuales y fricciones individuales en tiempo real que el copy estático no logra abordar.
accessible than ever we've seen pretty much across the board 25% conversion lift when you install a chap out on your you don't install like a 5060 widget a month to help those that you assumed wrong right someone has a question maybe sales page and they just want to ask direct questions and by opening that conversation that two-way dialogue you can address their immediate points of friction for them without assuming because pages are made to talk to one to many but customers buy as individuals so
Escúchalo en el minuto 15:24
Uso del desglose por segmentos de audiencia en Meta Ads para auditar la inversión real en prospección vs. retargeting
Medición y experimentosPara auditar si el presupuesto de un conjunto de anuncios realmente se destina a prospección fría o si está impactando a clientes y usuarios ya comprometidos, se debe aplicar el menú de 'Desglose' (Breakdown) por segmentos de audiencia dentro de Meta Ads Manager para ver la distribución vertical (eje Y) del gasto y las conversiones.
seven here so where you actually find how to enter these in and actually show in your adset you can also see this inside campaigns we're just using adset it is I believe under the breakdown breakdown goes on your why so that's why we can see under each adet underneath the adset is new audience engaged existing unknown and then you might also have uncategorized so the breakdown is going down in your y- axis so it's breakdown so what you want to then do is if you scroll down you could either search and type in audience segments but if you scroll down on breakdown you go to demographics and then…
Escúchalo en el minuto 33:17
Medición y experimentosPara tomar decisiones reales de optimización y apagado de anuncios de adquisición, se debe medir el coste de adquisición de clientes nuevos (eNCAC) mediante herramientas de atribución externa, en lugar de confiar en el ROAS o CPA reportado dentro de la plataforma publicitaria, el cual combina clientes nuevos y recurrentes.
is okay if you double check it with a thirdparty attribution tool and so the single Friday we're like you know inapp roas is a faulty metric because of this which is fine as long as you can then break out the most important metric of all here which is enac or cost to acquire a new customer unless we know what we are paying to acquire a new but we separate out the new versus The Returning by using this enac metric and
Escúchalo en el minuto 41:12
Las plataformas publicitarias impactan tráfico templado aunque reporten prospección pura
Medición y experimentosDebido a restricciones de privacidad y atribución, plataformas como Meta y Google impactan de forma inevitable a audiencias que ya conocen la marca mientras reportan que son usuarios completamente nuevos, por lo que no se debe confiar ciegamente en las métricas nativas de la plataforma publicitaria.
you are always going to be as long as you're using meta platform and the Google platform they're always going to Target some of your warm traffic no matter what but they are going to claim as we see over an ads manager that most of it is brand new here and that's the learning from today is like you can't just look at these platforms and assume legislation with privacy policies and cookies and all that type of stuff it's restricting what meta can do so it's to take it with a grain of salt and that the assumptions and dependency on in apppp numbers only is Never Gonna paint
Escúchalo en el minuto 47:24
Definición técnica de audiencia «nueva» en la segmentación de Meta Ads
Medición y experimentosEn los desgloses de audiencia de Meta (nueva, comprometida y existente), un usuario se clasifica como «nuevo» si no ha visitado el sitio web ni ha interactuado con la tienda o catálogo de la plataforma; esto significa que es nuevo para el producto o servicio específico, no necesariamente que desconozca la marca por completo.
so the three categories are basically new engaged or existing new it's new to your product or service not necessarily new to your brand so the new piece is that they haven't visited your site they haven't left meta to see a product or service of yours so it's like new visitors to your website but then you have if you're in e-commerce and you have catalog or shop it's excluding visitors to that shop or catalog interaction because it's new visitors to the product or service you have most
Escúchalo en el minuto 20:30
Medición y experimentosEn una campaña de anuncios con un KPI objetivo de CPA de entre $80 y $90, el CPA combinado reportaba $128, pero el desglose reveló que el costo real por cliente nuevo era de $149 (un 50-60% por encima de la meta) frente a $66 de la audiencia interactuada o templada.
$149 which is not within their kpi their kpi is is 80 to 90 ideal is sort of a scale range but definitely under 100 so customers at $149 that would be W very bad that is 50 60% higher than it needs to be so but look at that your new is 149 Blended is 128 so you're already given a different your acquisition cost is
Escúchalo en el minuto 38:51
Medición y experimentosIncluso utilizando la API de Conversiones (CAPI), una parte significativa de los datos reportados dentro de plataformas publicitarias como Meta o Google sigue estando modelada por algoritmos y no refleja con total precisión lo que ocurre en realidad, a menos que se vincule externamente el identificador del usuario y el clic con la compra.
to be accurate data and we all know that even with the addition of Cappy or conversions API which we talked about here many many times the data is still faulty and it's not a true depiction of what's actually going on when you're platform especially in light of the iOS 14 updates from 2021 so we're still sort degree there is a certain amount of visibility that is just being blocked by it's meta the reality is is that a lot of the data that comes back into the platforms is modeled it is modeled based upon what meta actually thinks is happening as opposed to what's actually happening until you…
Escúchalo en el minuto 22:01
Canales de adquisiciónEn cuentas de Meta Ads de alto volumen, una única campaña puede ejecutar un presupuesto de 250.000 dólares al mes, concentrando aproximadamente 141.000 dólares en un solo conjunto de anuncios en un periodo de 30 días.
well as amount spent for the last 30 days so this campaign has spent um you $250,000 uh just in the past uh month and the adset that we're looking at here has actually spent about $141,000 in the last 30 days so take through these
Escúchalo en el minuto 24:40
Canales de adquisiciónUn conjunto de anuncios en Meta Ads operando a escala registró más de 11 millones de impresiones con una inversión aproximada de 141.000 USD en un periodo de 30 días.
particular ad set we've got in the last 30 days okay we've got over 11 million about $141,000 in the last 30 days on this
Escúchalo en el minuto 29:49
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.