Marketero

Marketero · Entrevistas

How to Target ACTUAL Cold Traffic on Meta (Hint: It’s Not Exclusions) | PT EP 632

19 ideas de marketing sacadas de «How to Target ACTUAL Cold Traffic on Meta (Hint: It’s Not Exclusions) | PT EP 632», de Perpetual Traffic. Sobre todo de…

Ver la entrevista entera en YouTube

De qué habla

Las 19 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 19 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.

Estrategia en dos pasos para escalar tráfico frío evitando el sesgo de retargeting de la IA

Canales de adquisición

Los algoritmos de conversión impulsados por IA (como Meta Advantage+ o Google Performance Max) tienden a sobreoptimizar dirigiendo el presupuesto hacia retargeting encubierto para inflar el ROAS dentro de la plataforma, limitando el volumen de clientes nuevos. Para escalar la adquisición real, se recomienda aplicar un embudo en dos pasos: utilizar campañas con objetivo de tráfico o clics puros en la parte alta del…

But why they say that is because in-app measurement is not going to generally look favorably on top of the funnel, true cold traffic. So this whole thing comes back to if you're staring at in-app ROAS, then the AI is also saying, geez, I got to get this number up or I know people aren't going to spend. And then they say, okay, I got to retarget people more likely. So you got to force the algo. So it's a pretty big, in my opinion, development because you want at least half of your traffic coming from new clicks. And then all this awesome advantage stuff can go to work, but just not at the very…

Escúchalo en el minuto 10:34

Construir audiencias propias en Meta usando objetivos de tráfico o video con CPMs reducidos

Canales de adquisición

En lugar de competir exclusivamente por conversiones directas en Meta —donde los CPMs son elevados porque los algoritmos y competidores pujan por el mismo público con intención de compra inmediata—, conviene utilizar objetivos de tráfico, alcance o visualizaciones de video en la parte alta del embudo. Esta estrategia permite crear audiencias de retargeting propias a CPMs mucho más bajos (incluso por debajo de 15$ en…

So instead of fighting it, figure out a strategy around it. And it might be this traffic and or step up above, which is going to be much lower CPMs, by the way, as soon as you go into traffic campaigns, you go into video views or you go Sub $15 in the US. Like we've got like two, $3 audiences that we're going after in very competitive. So think about it this way. You're creating your own retargeting audiences with just a level up. It's like there's only so much traffic there, but you're also fighting everybody else for that audience that's likely to convert.

Escúchalo en el minuto 20:39

Alinear el CAC objetivo a la ventana de recuperación de LTV que el negocio puede tolerar

Medición y experimentos

Calcular el CAC permisible tomando como base el LTV total de ciclo de vida completo (a varios años) suele generar fricciones financieras; el CAC y el NCAC deben calibrarse frente a una ventana temporal específica de retorno (1, 3, 6 o 12 meses) según el flujo de caja y la tolerancia de la dirección.

A lot of talk on CAC to LTV ratio. But then when you dig in, is the CMO or CEO greenlining the budget really going to give you credit for all of the LTV? Like particularly if it's a subscription brand and they have like four years of subscriptions and they're going to say, oh, the customers are worth 10 grand. They're not happy because like the CAC's too high. The LTV is 10,000. So you can, what you need to do is figure out how much LTV are you going to be willing. Is the man or woman cutting the checks or business being willing to wait for it? And then that's the LTV target you got to have…

Escúchalo en el minuto 37:29

Medir el tráfico frío evaluando el ciclo completo de ventas y no el ROAS inmediato de la plataforma

Medición y experimentos

Muchos anunciantes apagan campañas de tráfico frío porque el ROAS inmediato dentro del administrador de anuncios parece deficiente frente al retargeting. Para obtener una ventaja competitiva, se debe medir el impacto mediante herramientas de atribución de terceros y esperar a que transcurra el ciclo de ventas completo del cliente antes de juzgar la rentabilidad de las campañas de captación superior.

What I like is that it gives you an edge because most people aren't going to take the time to create a click campaign at the top because they're going to see the in-app ROAS and say, oh, that's not as good as these retargeting ROASs. So therefore, if you have that strategy of, I'm just going to buy these clicks and rely on third-party measurement, not in-app, because other people see the same results as you. If they're bidding on the cold traffic, they see the crappy ROAS in-app, so they're going to not spend as much. If you're instead spending a ton on that and then waiting a customer sales…

Escúchalo en el minuto 23:00

Estructuración del análisis de tráfico publicitario separando métricas de clientes nuevos y recurrentes

Medición y experimentos

Para evaluar con precisión el impacto de las campañas publicitarias, conviene organizar las columnas y reportes siguiendo el flujo del embudo y desagregando las métricas (CAC, AOV, LTV e ingresos) entre clientes nuevos (NCAC y New Revenue) y clientes recurrentes, en lugar de analizar únicamente datos agregados globales.

Now, one thing I like to do is I like to think about it in the flow of the funnel. What do they cost for the new visits? Did they turn into anything in my first party data? And then segment it down. The new customers, what did they cost and what was their order value? And then what's their LTV? And then how much revenue did new bring? And then the repeats. What was the CAC, AOV, LTV and revenue there? So new first, all after that, ACAC. N, anything with an N, NCAC, for example, is new. I'm going to do how many new customers to new revenue was that? And then we'll do the same with repeat.

Escúchalo en el minuto 25:42

El ROAS de nuevos clientes en Meta parece significativamente menor al aislar la atribución multicanal

Medición y experimentos

Al auditar Meta Ads filtrando exclusivamente por adquisición de clientes nuevos en modelos de atribución multicanal, el ROAS suele verse notablemente más bajo porque canales de fondo de embudo (email, SMS, búsqueda de marca y Amazon) absorben la conversión final.

So the big one right here, optimize Facebook for short-term from new customers. Because this is going to really open your eyes to the new verse repeat situation and the new visitors. And then you'll be a little traumatized because you'll see, oh my God, the robust is a lot lower. One, because we're only looking at the news. And two, because we're looking multi-channel. Because email, SMS, and then branded search and Amazon for the bigger brands clean up a ton of bottom of the funnel. And so if you aren't already prioritizing or strategizing around top and middle funnel for meta, Facebook…

Escúchalo en el minuto 34:27

Rendimiento real de Meta Advantage+: 18% de visitas nuevas y 40% de compradores recurrentes

Medición y experimentos

En un análisis de 3.300 campañas Advantage de Meta contrastadas con datos reales de compras en ecommerce, el promedio de visitas nuevas fue de apenas 18% (1 de cada 5 clics) y el porcentaje histórico de clientes nuevos adquiridos fue del 60%, lo que significa que un 40% de las ventas generadas corresponden en realidad a retargeting de clientes preexistentes.

We had 3,300 advantage campaigns running. So what I found was that the average new visit percentage, for example, in July of these 3,300 campaigns was 18%. And then the percentage of new customers acquired versus repeat was 53%. The average is 60% all time. So 60% all time, new customers. No, it's good at retargeting and finding existing people almost 50% of the time. One out of every like five are actually brand new.

Escúchalo en el minuto 5:09

Los objetivos de conversión en Meta anulan automáticamente las exclusiones de audiencia

Canales de adquisición

En campañas optimizadas para Ventas, Clientes Potenciales (Leads) o Catálogo, Meta aplica automáticamente Advantage Audience e ignora las exclusiones manuales de públicos personalizados o tráfico web previo, ya que el algoritmo busca la vía más rápida para generar conversiones dirigiéndose a usuarios que ya conocen la marca.

So they have like this disclaimer type of language that tells you we're basically going to ignore any other audience segmentation you've established in the account minus anything you put in the account settings specifically. You're excluding website traffic. And so it turns out that certain campaign objectives, as told on their business help center, will override what you set up as an audience and automatically apply the advantage audience. Which then means what's happening is the AI, as we talked about at length in episode one, is going to say, hey, who's most likely to hit the objective?…

Escúchalo en el minuto 7:29

Medición de nCAC y coste por nueva visita con atribución externa

Medición y experimentos

Para escalar de forma rentable, no se debe depender únicamente de las métricas in-app de las plataformas publicitarias; es crítico medir con atribución externa el coste por visita nueva y el coste de adquisición de clientes nuevos (nCAC) para verificar que el presupuesto adquiere clientes netos y no solo recicla tráfico templado.

But as far as these individual metrics, you know, if you're a small to midsize to midsize to larger business, like the platform that you guys have built here is outstanding. Because you can go in and look at all these individual metrics and gauge exactly how well your campaigns are doing, not relying or not being over-reliant on in-app metrics. And one of the things that we talked about the last time we were here was estimated cost or effective cost per new site visit. And new site, as well as NCAC, which is new cost to acquire a customer, is highly important now because that is the goal of…

Escúchalo en el minuto 2:46

Proporción real de tráfico nuevo en campañas de Meta y Advantage+

Canales de adquisición

En campañas de adquisición y Advantage+ de Meta, el algoritmo tiende a apoyarse intensamente en retargeting encubierto para optimizar sus métricas; la proporción real de visitas o tráfico genuinamente nuevo suele situarse únicamente entre un 30% y un 43% en los mejores casos.

Like it's, don't expect it to be greater than 33% best case scenario of new traffic. We're just going to retarget the stuff that people, well, that makes the algorithm look as good as possible. It's two thirds dirty. I mean, I've seen new visits inside Wicked Reports and also. I think I actually in Wicked Reports, I've seen it as like 43 percent is like the highest for new visits I've ever seen, which is kind of insane. But if you're in and around the 30, 40 percent using these metrics, like these are MPIs.

Escúchalo en el minuto 17:42

Falsa precisión al excluir audiencias dependiendo únicamente del píxel

Canales de adquisición

Confiar exclusivamente en el píxel de Meta para excluir a clientes y visitantes previos es un error habitual: el uso generalizado de bloqueadores de cookies, navegadores que enmascaran identidad y pérdida de datos impiden que las exclusiones sean 100% efectivas, contaminando las campañas de tráfico frío.

I really liked how Scott brought up something where you can't rely on the pixel data, which I think again is what we see when we like audit accounts. People are excluding assuming that pixels works full percent, right? You assume that people aren't using cookie, cookie blockers. They're not using alternative search engines that they're masking who they are. So much of that data gets lost and people don't think about it and over rely on pixels. But it's not to the degree that we actually want. It's not customers. You can't exclude them.

Escúchalo en el minuto 18:38

Escalamiento de presupuesto con foco en nuevos visitantes reduce el costo por visita de $4.50 a $2

Canales de adquisición

Al escalar el presupuesto publicitario un 100% aplicando exclusiones para evitar que Meta retargetee exclusivamente a clientes existentes, el volumen de visitas nuevas aumentó un 400%, reduciendo el costo por nuevo visitante de $4.50 a $2.00 y multiplicando los ingresos en un 600%.

So you've doubled the costs. But what I really like is you've 400% increased new visits. So some exclusions get in there, some good hard work. So there's a lot more new visitors coming. You didn't just pay for more retargeting. Because you had more new visitors, your cost for a new visitor plummeted from four and a half bucks to $2. Well, as we covered quite a bit now, Meta loves to retarget people that are already customers. So your NCAC came down because you're bringing in way more potential new customers. And then revenue has gone up. When your costs go up 100% and your revenue goes up…

Escúchalo en el minuto 27:34

Porcentaje habitual de visitas netamente nuevas en Meta Ads

Medición y experimentos

En campañas de adquisición en Meta Ads, frecuentemente solo alrededor del 30-35% (un tercio) del tráfico corresponde a visitantes netamente nuevos, debido a la tendencia del algoritmo a impactar a audiencias templadas recurrentes incluso cuando existen exclusiones configuradas.

But one of the bigger problems is that when you start doing this on Meta specifically, Meta has a tendency to bring in a lot of those warm traffic audiences. And so when we looked at the account last session, we saw the new visits around 30 something percent. You know, and I see that a lot. It's around a third.

Escúchalo en el minuto 3:35

Usar promociones intermedias como pretemporada de prueba para Black Friday

Medición y experimentos

Las promociones de fin de verano (como Labor Day) deben utilizarse como banco de pruebas para evaluar la respuesta de las audiencias captadas en meses previos y calibrar los segmentos antes de las campañas principales de Black Friday y Cyber Monday.

As we talked about on one of the previous episodes, which Lauren discussed, Labor Day should be your testing ground, like your preseason to your Black Friday, Cyber Monday. You get people to opt into your brand in the summer and then you give them a great offer. And if they're not going to bite during Labor Day, you might have not acquired the right audiences in the summer. So then you can adjust audience segments for Black Friday. And then if they did bite, then you've now proven your runway.

Escúchalo en el minuto 39:35

Activación algorítmica de competidores directos tras la interacción del usuario

Investigación de mercado y cliente

Al generar interacciones en la parte alta del embudo en Meta, el algoritmo detecta la intención del usuario y de inmediato le muestra anuncios de cinco o seis competidores directos de la categoría, por lo que la verdadera competencia ocurre contra sustitutos cercanos y no contra el mercado publicitario general.

But when you opt into a brand or an offer of some type, now you're served their five competitors. So often when people are afraid of doing top of funnel content, it's because they think that they're going up to bat against products that are completely unrelated. When in fact, if you're going up to bat, you're actually just competing against five or six other of your competitors because that person has shown an interest or a propensity to want a specific app or want a specific product.

Escúchalo en el minuto 15:54

Al escalar agresivamente el presupuesto en Meta, el algoritmo prioriza la prospección sobre el retargeting

Canales de adquisición

Cuando se incrementa el presupuesto de pauta agresivamente en Meta, la proporción de retargeting disminuye porque el algoritmo sale a buscar nuevas audiencias para evitar saturar y encarecer el público de remarketing.

When you're pushing the budget larger, there's a smaller percentage of retargeting because you're pushing a larger net. You're capturing more conversions that are positive from new people. So that wider net brings in a lot more. So that's where we have seen when we have pushed budgets aggressively, we've been able And okay, because Meta's like, oh my gosh, there's all this new money. I'm going to actually go after new people because I don't want to scale the remarketing because then that becomes competitive in and of itself later.

Escúchalo en el minuto 30:23

Preparar Black Friday con anticipación para calibrar márgenes y descuentos

Oferta y precio

Planificar las promociones de Black Friday y Cyber Monday con suficiente antelación permite comprobar la viabilidad de aplicar descuentos agresivos en un entorno de márgenes reducidos y máxima competencia, donde gran parte del volumen de ventas procede de la base de clientes recurrentes.

And I think when you're able to prep Black Friday, Cyber Monday earlier, you can prove the stomach for an even steeper discount later because it's the most competitive time, smallest margins. It's really sort of your repeat customers mostly. But I mean, it is a customer grab and obviously it's revenue.

Escúchalo en el minuto 40:21

Riesgo de desarrollar plataformas de atribución y analítica internamente

Medición y experimentos

Intentar programar y mantener una solución propia de analítica y atribución publicitaria in-house suele acarrear costes elevados y desarrollo fallido en comparación con el uso de herramientas especializadas existentes.

Otherwise, none of this stuff really makes a whole lot of sense unless you can figure out a way or create your own wicked reports, which good luck trying to do that. We tried to do that three or four years ago. Don't do it. I'm telling you through 200 grand down the shitter on that one.

Escúchalo en el minuto 43:16

Diferencia de ticket promedio (AOV) entre clientes nuevos y recurrentes en caso analizado

Oferta y precio

En la cuenta analizada, el ticket promedio (AOV) de los clientes recurrentes alcanzó 198 USD en comparación con 159 USD para clientes nuevos, registrando un incremento general de AOV del 20% al 30%.

Repeat customer here is $198. $159 for a new. All these AOVs are up 20%, 30%. as the conversion rates hold, they've managed to get 20 more percent on AOV, which is going

Escúchalo en el minuto 32:15

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.