Marketero · Entrevistas
4PI Analysis, You're looking at FACEBOOK ADS DATA Completely WRONG! Stop Looking at ROAS, CPC & CTR
14 ideas de marketing sacadas de «4PI Analysis, You're looking at FACEBOOK ADS DATA Completely WRONG! Stop Looking at ROAS, CPC & CTR», de Professor…
- 14 ideas despiezadas
- Fuente Professor Charley T
- Dura 26 min
- Publicada en 2023
Ver la entrevista entera en YouTube
De qué habla
Las 14 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 14 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.
Dirección creativa de anuncios dinámicos basada en el rol del anuncio en el embudo
Medición y experimentosPara optimizar una cuenta publicitaria sin segmentaciones artificiales, analiza qué rol cumple cada anuncio y define la siguiente iteración creativa (DCT) entre tres caminos: buscar prospección con nuevo tipo de atención, crear un anuncio de fondo de embudo de mayor eficiencia o encontrar un anuncio de retargeting con menor CPM.
and increase spend? Well, looking at just these four metrics, we have the answer to literally all of those things. And we can give very specific, laser focused creative direction for what we're going to do in our DCT to make this campaign better. Very specifically, we're going to go in one of three directions, depending on our resources, our needs, our creative team, a backlog of testing and other data. We can pick whatever one of those we want to focus on. But it's really one of three things. Number one, we want to find another prospecting ad that might also bring in a slightly different…
Escúchalo en el minuto 20:52
Medición y experimentosPara diagnosticar si el algoritmo de Meta utiliza un anuncio como prospección (top of funnel) o retargeting (bottom of funnel), se debe desglosar el rendimiento por día durante los últimos 7 días en lugar de mirar métricas semanales acumuladas. Evaluar periodos largos distorsiona la frecuencia real, ya que variaciones en el presupuesto aumentan las impresiones y la frecuencia aparente sin reflejar el comportamiento…
information pretty much the only thing that matters. Now, when we go in to do this analysis of the frequency to figure out where that ad is sitting, the only thing that matters is when we go to breakdown, time, and day. We don't care about week over week because the spend is going to dramatically change how often that ad is shown to people. If it spends five times as much, the frequency might be higher even though it's a prospecting ad because it's literally shown five times more. So when we're analyzing how broad is using the ad, we only need the last seven days.
Escúchalo en el minuto 14:30
Medición y experimentosEn campañas con segmentación amplia (broad targeting), la función de cada anuncio en el embudo se diagnostica cruzando el CPM relativo y la frecuencia diaria: los anuncios de prospección (top of funnel) presentan CPM más bajo que el promedio y frecuencias diarias cercanas a 1.0, mientras que los anuncios de retargeting o mid/bottom funnel concentran CPMs elevados y frecuencias diarias promedio superiores a 1.15 (más…
$18 CPM. Now our average is 24 across these three. Remember, there's no such thing as a good or bad CPM. It's all relative to each other. So this ad out of these three, according to the CPM says that people like it. And when we look at the frequency, 1.06, 1.1, 1.04, 1.09, a 1.2 on one day, this is top of funnel. So people like it. The CPM is low. So that's also telling us it's top of funnel. When we look at the frequency by day, it's a top of funnel ad. Now let's go to the next ad. 1.18. This is averaging at least a 1.15 or higher. If your ad averages over a 1.15, it is retargeting. That…
Escúchalo en el minuto 5:13
Medición y experimentosEn cuentas de Meta Ads con segmentación abierta (broad), la relación entre el CPM y la frecuencia indica la etapa del embudo en la que el algoritmo coloca el anuncio: los anuncios de prospección (TOFU) presentan menor CPM y menor frecuencia al impactar audiencias amplias, mientras que los anuncios que funcionan como retargeting (BOFU) muestran mayor CPM y mayor frecuencia sobre un público más reducido.
we're going to talk about is CPM and frequency. Now, what is CPM? This is cost per milli or honestly, important, not because there's a good or a bad CPM, but because it's going to help us identify how Facebook is using your ad. When we index that against frequency, we're going to know even at broad, is this a prospecting ad? Is this a retargeting ad? Is it mid funnel? Is it bottom funnel? Is it doing we have to remember is that generally speaking, what is a prospecting ad? A prospecting ad is something that people see at the top of the funnel and it's generally cheaper to show to people…
Escúchalo en el minuto 1:45
Medición y experimentosMétricas intermedias como CPC, CTR y hook rate son redundantes porque su efecto real ya se refleja en el CPM, frecuencia y CPA; además, el ROAS atribuido por la plataforma no representa dinero real para tomar decisiones de negocio.
that you've been told is important actually doesn't mean anything. If you notice, we don't have anything about CPC in here. Why? Because it's Facebook. CPC does not matter. We don't have CTR in here. Why? Because it's Facebook. CTR does not matter. We don't have hook rate in here. Why? Because there's nothing that hook rate's going to do with telling us whether this ad is good or not, because hook rate will only inform the CPM, the frequency, the spend in the CPM. By that point, it's completely useless information. Anyway, we also don't have ROAS in here. Why? Because that's not real money…
Escúchalo en el minuto 24:17
Medición y experimentosAl analizar el rendimiento de campañas publicitarias a lo largo del tiempo, se debe ignorar la ventana de datos más reciente debido a la atribución diferida y el retraso en el reporte de conversiones, los cuales presentan información incompleta.
the first thing we're going to do is just ignore this top line because that's what have you done for me lately and delayed attribution that happens just because of delaying of passing information back and forth. We know that this is incomplete data. So we can more or less ignore that data. Now,
Escúchalo en el minuto 15:45
Medición y experimentosUn incremento en el costo por adquisición (CPA) al multiplicar el presupuesto publicitario de un anuncio no representa un problema si el nuevo costo se mantiene por debajo del CPA promedio objetivo de la cuenta.
it does better. So now one of the things we're going to say is, wow, it went from a $60 to an $80. Okay. It also went from $800 to $5,600. Okay. It went from $860 in spend to nearly $5,700. And our average cost is $84. And on that week, it went from a $61 to an $80. Do we care that we just spent more money at below our average cost? Of course not. That's a good thing. So let's look at the
Escúchalo en el minuto 17:51
Medición y experimentosSi un anuncio rinde mal tanto en días de baja frecuencia (mostrando CPM y CPA altos como anuncio de prospección) como en días de alta frecuencia (fallando como retargeting) y no logra captar gasto de forma consistente, debe apagarse inmediatamente sin importar la opinión creativa o resultados previos en pruebas.
So if this is actually going to be used as a prospecting ad, it's really bad at it, right? Because the other one, this other ad is a prospecting ad that is half as much to show to people based on CPM. And the cost per purchase is better on average than 132. It's almost half that at 73. So objectively speaking, this ad isn't doing prospecting very well at all. And then we take a look about that day where it spent $367? Our cost per purchase, 91. When this thing's a retargeting ad, it does a really bad job. So with this, we know that this ad isn't good for prospecting, isn't good for bottom of…
Escúchalo en el minuto 11:50
Medición y experimentosSi un anuncio rinde mal tanto en días de baja frecuencia (mostrando CPM y CPA altos como anuncio de prospección) como en días de alta frecuencia (fallando como retargeting) y no logra captar gasto de forma consistente, debe apagarse inmediatamente sin importar la opinión creativa o resultados previos en pruebas.
So if this is actually going to be used as a prospecting ad, it's really bad at it, right? Because the other one, this other ad is a prospecting ad that is half as much to show to people based on CPM. And the cost per purchase is better on average than 132. It's almost half that at 73. So objectively speaking, this ad isn't doing prospecting very well at all. And then we take a look about that day where it spent $367? Our cost per purchase, 91. When this thing's a retargeting ad, it does a really bad job. So with this, we know that this ad isn't good for prospecting, isn't good for bottom of…
Escúchalo en el minuto 11:50
Dependencia del gasto de retargeting respecto a la prospección
Canales de adquisiciónLos anuncios de retargeting o fondo de embudo no pueden escalar su gasto de manera independiente; la cantidad de presupuesto que pueden absorber está estrictamente limitada por el volumen de audiencia nueva y tráfico que ingresan los anuncios de prospección.
why doesn't this ad get more spent? Well, because it's a retargeting ad, right? It's not designed to drive more spend. The more money that the prospecting ad spends, the more money that this ad can spend.
Escúchalo en el minuto 10:56
Medición y experimentosPausar un anuncio de prospección que concentra gran parte del gasto publicitario solo porque muestra un CPA directo elevado (por ejemplo, $168 frente a una media global de $73.83) puede colapsar el rendimiento de la cuenta, dado que los anuncios interactúan como un ecosistema donde el anuncio de captación alimenta la demanda que los anuncios de retargeting luego convierten.
this is number one, spend is going to tell us what is the contribution of this to our bottom line. And CPA is also going to tell us, does this thing do well at its job? Now we're also going to find in a lot of ad accounts, the ad that does the best never gets any spend. We're going to break down why that happens almost all the time and why you shouldn't be concerned about that at all. And we're going to talk about why sometimes that bad ad that gets all the spend isn't actually that bad. And you turning it off might be the worst thing that you could do. Back to camera B, let's dive back into…
Escúchalo en el minuto 8:20
Canales de adquisiciónOperar campañas publicitarias en Meta con segmentación totalmente abierta (broad) elimina la necesidad de crear públicos de intereses, lookalikes o retargeting manual, reduciendo la carga de trabajo y permitiendo que el algoritmo optimice y distribuya los anuncios de forma más eficiente.
absolutely have to look at. Now, before we get any further, we are assuming that you have an account set up in broad only, that you're not wasting your money in lookalikes and retargeting and interest groups. And if you have those, that's fine. We'll dissect that stuff too, but understand that you don't need any of that stuff. What we're going to talk about today will actually help you solve the problems that you think those things are solving in a way that makes Facebook smarter, makes you more money, and ultimately massively reduces your workload while also making scaling your business
Escúchalo en el minuto 1:05
Canales de adquisiciónUna cuenta de Facebook Ads escaló de un gasto inicial de 50.000 dólares al mes a una facturación de 50 millones de dólares al año en menos de dos años, apoyándose en la simplificación de estructura y dirección creativa en lugar de hacks o audiencias segmentadas.
spend more money, which ultimately means we're going to scale our business. And we don't need lookalikes. We don't need retargeting. We don't need cost caps. We don't need advantage shopping. We don't need any of those hacky hacks. Remember, Facebook hacks are for Facebook hacks. And the ad account that I showed today, it started less than two years ago at $50,000 a month. Now we're doing $50 million a year, and it'll probably double by the end of the year. It is quite literally
Escúchalo en el minuto 23:31
Canales de adquisiciónUna cuenta de Facebook Ads escaló de un gasto inicial de 50.000 dólares al mes a una facturación de 50 millones de dólares al año en menos de dos años, apoyándose en la simplificación de estructura y dirección creativa en lugar de hacks o audiencias segmentadas.
spend more money, which ultimately means we're going to scale our business. And we don't need lookalikes. We don't need retargeting. We don't need cost caps. We don't need advantage shopping. We don't need any of those hacky hacks. Remember, Facebook hacks are for Facebook hacks. And the ad account that I showed today, it started less than two years ago at $50,000 a month. Now we're doing $50 million a year, and it'll probably double by the end of the year. It is quite literally
Escúchalo en el minuto 23:31
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.