Marketero · Entrevistas
What’s Working (And What’s Not) Now on Meta and Google Ads With Nick Miller | PT EP 585
16 ideas de marketing sacadas de «What’s Working (And What’s Not) Now on Meta and Google Ads With Nick Miller | PT EP 585», de Perpetual Traffic. Sobre…
- 16 ideas despiezadas
- Fuente Perpetual Traffic
- Dura 39 min
- Publicada en 2024
Ver la entrevista entera en YouTube
De qué habla
Las 16 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 16 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 adquisiciónA diferencia de las campañas de conversión (donde el creativo puede filtrar a una audiencia amplia optimizada para ventas), en campañas de reconocimiento o alcance superior no se debe confiar ciegamente en el algoritmo con audiencias demasiado abiertas. De lo contrario, Meta optimizará para obtener las impresiones más baratas posibles, alcanzando perfiles de baja calidad que difícilmente convertirán a futuro.
well just why don't we just combine them all together if it's a website conversion campaign and let the creative decide you know what message works for that specific audience but it's more that a bit if you go too broad you're just wasting money because mea will spend your money you won't get anything funnel uh don't lean into the algorithm for those awareness and reach campaigns because yeah as we said meta will very very happily uh show that video to anywhere they can find an impression and it'll likely not be the people you're trying to reach yeah and probably the lowest common denominator…
Escúchalo en el minuto 21:54
Canales de adquisiciónA diferencia de las campañas de conversión (donde el creativo puede filtrar a una audiencia amplia optimizada para ventas), en campañas de reconocimiento o alcance superior no se debe confiar ciegamente en el algoritmo con audiencias demasiado abiertas. De lo contrario, Meta optimizará para obtener las impresiones más baratas posibles, alcanzando perfiles de baja calidad que difícilmente convertirán a futuro.
well just why don't we just combine them all together if it's a website conversion campaign and let the creative decide you know what message works for that specific audience but it's more that a bit if you go too broad you're just wasting money because mea will spend your money you won't get anything funnel uh don't lean into the algorithm for those awareness and reach campaigns because yeah as we said meta will very very happily uh show that video to anywhere they can find an impression and it'll likely not be the people you're trying to reach yeah and probably the lowest common denominator…
Escúchalo en el minuto 21:54
Evitar la segmentación algorítmica amplia en campañas de alcance y brand awareness
Canales de adquisiciónEn campañas de parte alta del embudo (alcance, reproducciones de video o reconocimiento de marca en plataformas como Meta o YouTube), no se debe delegar la búsqueda de audiencia en la IA. Al optimizar por volumen barato, el algoritmo mostrará los anuncios a usuarios de baja relevancia; por ello, en estos objetivos es indispensable aplicar una segmentación deliberada y precisa basada en el perfil del cliente.
changes is when you go up the funnel because here if you're running awareness campaigns and um we'll stick running reach video View awareness brand consideration campaigns for example you actually want to be really smart with your targeting this is where at this level you don't want to lean into the AI to find the right people because meta will happily show your ad to whoever in in the world if they can get those Impressions and you can usually get the Impressions at a really low cost um however that's not what you want to do that's you actually at the top of the objectives you want to Target…
Escúchalo en el minuto 18:12
Medición y experimentosLas pruebas A/B clásicas entre dos anuncios o campañas separadas ya no ofrecen resultados estadísticos fiables en Meta, ya que el algoritmo asigna dinámicamente audiencias distintas a cada pieza creativa según su contenido. Para testear de forma efectiva, se deben introducir múltiples creatividades y ángulos temáticos dentro de un mismo conjunto de anuncios, permitiendo al algoritmo optimizar la entrega mientras el…
no longer audience Centric if you try to split test creative you're not sending the same creative to the same audience not running a true split test because there's no longer continuity with the audience there's no control group is split testing creative you're not going to get confidence level data from running strict tests like that that's works with creative testing you're looking for recurring patterns and may be different but the overall theme in the overall angle and that comes down know don't turn it off have more creative give give the machine more inputs to work with and you know Let…
Escúchalo en el minuto 15:11
Medición y experimentosLas pruebas A/B clásicas entre dos anuncios o campañas separadas ya no ofrecen resultados estadísticos fiables en Meta, ya que el algoritmo asigna dinámicamente audiencias distintas a cada pieza creativa según su contenido. Para testear de forma efectiva, se deben introducir múltiples creatividades y ángulos temáticos dentro de un mismo conjunto de anuncios, permitiendo al algoritmo optimizar la entrega mientras el…
no longer audience Centric if you try to split test creative you're not sending the same creative to the same audience not running a true split test because there's no longer continuity with the audience there's no control group is split testing creative you're not going to get confidence level data from running strict tests like that that's works with creative testing you're looking for recurring patterns and may be different but the overall theme in the overall angle and that comes down know don't turn it off have more creative give give the machine more inputs to work with and you know Let…
Escúchalo en el minuto 15:11
Evaluar la publicidad de marca mediante nCAC combinado y no por ROAS de plataforma
Medición y experimentosJuzgar las campañas de 'top of funnel' o reconocimiento según el ROAS o CPA reportado en la plataforma publicitaria hace que parezcan ineficientes, debido a ventanas de atribución cortas (como los 7 días de Meta) frente a ciclos de compra reales de varios meses. La efectividad de estas campañas debe medirse observando si el costo combinado de adquisición de nuevos clientes (nCAC o blended CAC) a nivel de negocio…
because if you're looking in the ad to judge the Performance Based on Ras or CPA in the ad platform these campaigns are going to look terrible yeah because for stars meta has a 7day attribution window so and if you're going top a funnel where it may take someone one month two months three months from the first time to have a touch point on an ad um and you run an awareness campaign it's just going to look like you're that money's going out the window and this is where roas as where we're seeing you know roas has so moving to enac what it's costing you know to the amount it's costing to…
Escúchalo en el minuto 22:47
Aplicar exclusiones de audiencia en Meta Advantage+ para asegurar adquisición real de clientes nuevos
Canales de adquisiciónLas campañas Advantage+ de Meta tienden a inflar sus métricas de rendimiento dirigiéndose a tráfico tibio y visitantes recurrentes en lugar de clientes nuevos; para utilizarlas eficazmente en prospección fría y proteger el costo de adquisición de clientes nuevos (nCAC), es necesario configurar exclusiones estrictas de públicos existentes.
kind of in this phase right now where is it actually going out and finding new customers or is it just regurgitating sort of the customers and the people that are familiar with your brand and not really helping your enac which is new customer acquisition cost which powerful but the potential gotcha is that for example if you run it to cold traffic you may think it's actually performing better than it is because it's going after a lot of warm traffic repeat visitors etc etc so that a potential gotcha however what we've seen is that if you use your exclusions in the right way and you get…
Escúchalo en el minuto 8:15
Excluir visitantes y compradores previos en campañas Advantage+ de tráfico frío
Canales de adquisiciónEn campañas de adquisición en frío con Meta Advantage+, es imprescindible excluir activamente clientes, compradores y visitantes de la web; de lo contrario, el algoritmo destina presupuesto al retargeting, inflando artificialmente el rendimiento y falseando la métrica de CPA.
wow Nick you mentioned as long as you get your exclusions in place is that just Usual Suspects existing customers press converters exactly and even even website visitors because uh what we campaigns and we had an Advantage Plus campaign that was performing well above campaigns but when we looked at the amount of new visits in that campaign it was very low so it was actually what we thought was a cold traffic campaign was actually doing a lot of retargeting which brought up you know lowered the CPA gave us a misleading impression of the results so again it's comes back to that really…
Escúchalo en el minuto 10:35
Canales de adquisiciónEn lugar de fragmentar presupuestos testeando múltiples intereses y audiencias similares (lookalikes), conviene consolidar en audiencias amplias para acumular más datos y salir antes de la fase de aprendizaje, dejando que sea el propio anuncio creativo el que filtre y atraiga al público objetivo.
seeing is that with conversion campaigns and especially if the account has a good amount of data that broad audience targeting like let let the let the algorithm find find the right people and use the creative as as the targeting you know know your audience have the right previously we might have three different audience sets we might have lookalike audiences interest audiences and then you know testing against a broad audience and then we finding consolidate use a broad audience that gets more data gets out of the learning phase quicker because we're just not seeing that audience testing is…
Escúchalo en el minuto 13:49
Canales de adquisiciónEn lugar de fragmentar presupuestos testeando múltiples intereses y audiencias similares (lookalikes), conviene consolidar en audiencias amplias para acumular más datos y salir antes de la fase de aprendizaje, dejando que sea el propio anuncio creativo el que filtre y atraiga al público objetivo.
seeing is that with conversion campaigns and especially if the account has a good amount of data that broad audience targeting like let let the let the algorithm find find the right people and use the creative as as the targeting you know know your audience have the right previously we might have three different audience sets we might have lookalike audiences interest audiences and then you know testing against a broad audience and then we finding consolidate use a broad audience that gets more data gets out of the learning phase quicker because we're just not seeing that audience testing is…
Escúchalo en el minuto 13:49
Priorizar el MER sobre el ROAS de plataforma en estrategias multicanal
Medición y experimentosEl ROAS reportado dentro de cada plataforma publicitaria tiende a sobre-atribuirse ventas y puede reflejar un rendimiento positivo falso mientras el negocio global decae. En entornos donde conviven múltiples fuentes de tráfico de pago y orgánico, se debe medir el MER (Media Efficiency Ratio: ingresos totales divididos entre el gasto total de marketing) para obtener una visión agregada y realista de la rentabilidad.
mix right you need to understand what job each one is doing and then you're going to be allocate appropriate amounts to spend to make your Global your Blended business metrics improve not days so when you're having multiple sources of traffic and then you're adding in organic traffic as well uh you know that that Rass metric suddenly doesn't become so reliable and actually account or even more of a problem if you have a great roas but the business metrics are declining you may not know interesting thing is if you take me which is you know total revenue just divided by your total ad spend it…
Escúchalo en el minuto 30:05
Las campañas automatizadas priorizan tráfico templado para inflar el ROAS objetivo
Canales de adquisiciónAl configurar campañas automatizadas como Google Performance Max sujetas a objetivos de ROAS o CPA, el algoritmo tiende a capturar tráfico templado y búsquedas de marca porque es la vía más rápida y fácil de cumplir la meta asignada, inflando el rendimiento reportado a costa de no generar incrementalidad en audiencias frías.
when you think about I mean going back to Advantage Plus and pmax that they're they they pmax has to have a Ras Target or a CPA Target you can't run it with manual bidding and so the roas target has to so say you've hit you know you've set a pmax campaign at a 3X Google calls it a 300% Rass Target and and how what's the best way for it to hit that Target it's going to go after warm traffic and you right let's see hard guys to sell over here easy guys to sell over here I think I'm going for the easy guys yeah yeah and oh by the way Advertiser don't I look brilliant look at my 300% row ass or…
Escúchalo en el minuto 37:03
Medición y experimentosPara obtener una métrica indiscutible del costo de adquisición sin requerir herramientas avanzadas de atribución, divide el gasto publicitario total de todas las plataformas entre el número de clientes nuevos registrados en la fuente de verdad del negocio (como Shopify) en un periodo determinado.
store um there's going to be some numbers which are indisputable I like to call them you know being on firm ground you're going to be able to see your sales you're going to be able to see your the number of new customers in a certain time frame and that's that's inarguable that's the source of Truth and then you just take your total ad spend divide that by the number of new customers in that time frame total ad spend across all platforms and then you have your your enac that's what you're paying you pay for each customer
Escúchalo en el minuto 26:28
Medición y experimentosGuiarse exclusivamente por el ROAS que reporta cada plataforma publicitaria genera malas decisiones presupuestarias, ya que canales de última milla (como Google) suelen atribuirse retornos muy altos (ej. 13x) simplemente cerrando conversiones generadas previamente por canales de prospección como Meta.
prove their value because the whole idea is you put this much money into meta ads and you get this return and you put this much money into Google ads and you get this supposed return the problem is is that it usually doesn't match up to what's actually happening in the source of Truth which is say the Shopify store or or the business So Meta both meta and Google have no interest in promoting we need to put more money into meta pmax happens all the time look at the RAS on pmax uh look this happens a lot look at the row on Google ads we have like a 13x row us and we only have a 1x on meta…
Escúchalo en el minuto 27:48
Medición y experimentosEl ROAS es una métrica condicionada por los intereses de las plataformas publicitarias que asume de forma engañosa que toda la facturación atribuida se debe exclusivamente al gasto en anuncios, distorsionando la comprensión holística del crecimiento del negocio.
roas implies that the revenue is a return all due to your ad spend it's an incredibly self-serving platform based way of looking at your business growth from your digital marketing isn't there
Escúchalo en el minuto 0:00
Canales de adquisiciónLa combinación de campañas Meta Advantage+ con optimización de landing pages enfocada en móvil permitió reducir el costo por nuevo donante de más de 900 dólares a menos de 300 dólares y aumentar las conversiones respecto al gasto publicitario del 40% al 76%.
using Advantage Plus compared to the standard campaign types uh we're able to increase conversions versus ad spend from 40% to 76% and their cost per new donor their donor has we've brought that down from over $900 to less than $300 and there was some mobile first landing page optimization there as well but the combination of those two things
Escúchalo en el minuto 10:02
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.