Marketero · Entrevistas
Why ROAS Sucks: The New Attribution Metric Everyone Should Use...But Aren’t Part 1 | PT EP 576
23 ideas de marketing sacadas de «Why ROAS Sucks: The New Attribution Metric Everyone Should Use...But Aren’t Part 1 | PT EP 576», de Perpetual Traffic…
- 23 ideas despiezadas
- Fuente Perpetual Traffic
- Dura 46 min
- Publicada en 2024
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.
Medición y experimentosLas métricas reportadas por plataformas publicitarias o herramientas de atribución pueden mostrar un ROAS positivo mientras el negocio real pierde dinero; la única métrica definitiva del éxito publicitario es el impacto financiero neto en la cuenta bancaria del negocio.
don't believe those guys like you have to believe like your bank account that's that is the source of truth that's it that person where I'm like man 440 roaz and we gained $62,000 this last like three months and they're like well then why did I lose 10% of my business huh like when I said hey here's the excellent metrics that we get to show and actually hurt that business then they measure the only thing that matters
Escúchalo en el minuto 31:28
Riesgo de acoplamiento y sobrecoste al combinar Meta Ads con Google Performance Max
Canales de adquisiciónSi no se controla la adquisición de tráfico frío, Google Performance Max tiende a hacer retargeting pasivo del tráfico procedente de Meta Ads en lugar de captar usuarios nuevos. Esto encarece el coste de adquisición de nuevos clientes (nCAC) y genera una dependencia artificial donde un mal desempeño semanal en Meta arrastra automáticamente los resultados en Google.
Google ads and we see that performance Max was running I think like 10 campaigns or 12 campaigns whatever is here there's like 12 performance Max campaigns that said uhoh that performance Max campaign is going to remarket all of that kind of semi poor meta um meta traffic now meta wasn't doing poor it was just overspent but then Google's going to subsequently overspend what that what is that going to do it's going to drop your endc drive it up the new cost requireed first- time customer is going to be way more expensive than rather than half meta traffic and half Google traffic which is what…
Escúchalo en el minuto 37:05
Medición y experimentosUn ROAS muy alto reportado dentro de plataformas como Google o Meta (ej. 500% a 800%) suele ser señal de que los anuncios están captando audiencias tibias o compradores previos, en lugar de generar demanda nueva. Las campañas de adquisición de tráfico frío suelen mostrar un ROAS bajo dentro de la plataforma, pero son las que realmente impulsan el crecimiento global del negocio.
they going to measure my campaigns typically have really poor Roa as an appp and I want that because it means it's cold traffic and it's growth High you know five six 800% roas means this warm audiences that probably already bought from you or already been to your site three times from meta now Google's going to scoop it up
Escúchalo en el minuto 18:14
Regla del 70% al 80% de inversión en prospección para escalar
Canales de adquisiciónPara escalar un negocio de forma rentable y sostenible, entre el 70% y el 80% del presupuesto publicitario total entre canales debe dedicarse a prospección (tráfico frío nuevo); asignar menos del 60% satura las audiencias de retargeting y deteriora progresivamente el Marketing Efficiency Ratio (MER).
audiences Omni Channel marketers think about what sort of spend do I have in these three main buckets it's it's prospecting remarketing and acquisition you want to have 70 to 75 sometimes 80% of your adment into prospecting now is how much my Aspen is going towards prospecting and when you look at those channels we had about only maybe 50% maybe about 55% of our total budget across all channels going towards prospecting which isn't scalable because now every dollar you put in you only get 55 cents back that's not that's not row as metrics that's myr Med the efficiency ratio if I keep…
Escúchalo en el minuto 38:23
Alinear la medición en métricas de negocio globales en lugar de ROAS por canal aislado
Medición y experimentosOptimizar y medir canales publicitarios de forma aislada mediante el ROAS de plataforma crea una falsa ilusión de éxito que puede dañar la rentabilidad del negocio. El rendimiento del marketing debe medirse con métricas holísticas como nMER (New Media Efficiency Ratio), CAC de nuevos clientes frente a LTV y margen de beneficio real, alineando todos los canales hacia un único objetivo común.
that's not Raz that's that's nmerb New Media efficiency ratio new customer acquisition cost new you know versus lifetime value of it there now you're have a business and you're not going to bidding strategy your way to success you have to take a step back further and say how do I actually grow my company with my unique product or service with my very specific profit margin and lifetime value of my customers and there's there's everything is so specific to a company that you can't measure in Channel row as because your company will die because but the agency will win and then you keep paying…
Escúchalo en el minuto 25:07
Las campañas de prospección no de marca deben evaluarse por métricas globales y no por ROAS atribuido directo
Medición y experimentosLas campañas de prospección (no-brand) capturan el primer clic pero rara vez cierran la conversión directa en plataforma, ya que los usuarios suelen finalizar la compra buscando el término de marca; por ello, medir únicamente el ROAS intraplataforma hace parecer fallidas campañas que en realidad están aumentando los ingresos globales del negocio.
this is bad why because we're measuring inapp Ras but what actually did happen is we went after new cold traffic on campaigns that are only designed to get the first click they're not designed to get the second third fourth fifth click because they're non-brand standard shopping campaigns so when you look at the standard shopping campaign it's only going to attribute the first click because think about how people buy if pen I'm going to Google best office pen and then I'm going to be introduced to a company called stainless pens.com and here I'm gonna buy one what is the next thing you're…
Escúchalo en el minuto 43:35
Medición y experimentosUn ROAS alto reportado dentro de las plataformas publicitarias no equivale a crecimiento real del negocio. Las agencias o gestores de tráfico pueden inflar artificialmente el ROAS concentrando el presupuesto en remarketing o búsquedas de marca propia (branded keywords), atribuyéndose ventas que habrían ocurrido de todos modos mientras los ingresos netos del negocio se estancan o caen.
that they had their worst months in in over a year and their agency reported back to them it's like hey everything's great we're at 400% I forget the it was like some sort of ridiculous number 400% row ass on response was no you guys are crushing us because our bank account is down and all you're doing is you're just either the agency itself Mis reading the data or trying to take the platform level data and extrapolate that onto the overall health of the business which is roas like roas is a false god you can manipulate that however you possibly it like here it it's funny I said Raz isn't…
Escúchalo en el minuto 20:51
Evaluar el escalado a audiencias frías mediante el MER incremental y no por el ROAS de plataforma
Medición y experimentosAl inyectar mayor presupuesto publicitario hacia audiencias frías, las métricas atribuidas dentro de las plataformas (como el ROAS reportado o el volumen de conversiones directas) pueden desplomarse fuertemente aparentando un fracaso, a pesar de que el retorno real a nivel de negocio sea altamente rentable cuando se mide a través del ratio de eficiencia de medios incremental (incremental MER / nMER).
$67,000 and more $117,000 more spend so divide that by 17245 that is a 3.93 incremental myr nmer which means my 400% myrr not Ras but media efficiency ratio so that's that incremental growth scenario my Raz is down 40% my conversions are gone 80% yeah this looks like a complete agency were measuring in app but all we did was just put more ad spend to a colder audience of good products that we the metric itself is showing this massive loss but at the end of the day it's really it's about new customer
Escúchalo en el minuto 45:25
Estrategia High-Low basada en Pareto para presupuestos en Google Shopping
Canales de adquisiciónPara maximizar el crecimiento en Google Shopping, se recomienda concentrar la mayor parte del presupuesto de prospección (alrededor de dos tercios) en el 20% de productos estrella que generan el 80% de las ventas (estrategia 'High'), asignando un presupuesto residual ('Low' o catch-all) al 80% restante del catálogo.
shopping campaigns and the standard shopping campaigns took a high low uh strategy and what high low strategy means you take your top products your peretto rule the 20% of your products and make 80% of your sales and you put a large amount of AD spend there this is $2,100 per day then we take a what they call the low strategy which is a catchall so let's take the other products that are the 80% of our products that only make 20% of our revenue and we stick a lower amount of AD spend there that is $1,000 per day so what we're doing is we're forcing 23ds of our budget on prospecting into the…
Escúchalo en el minuto 40:45
Auditar el rendimiento desde el backend del negocio antes que desde las plataformas publicitarias
Medición y experimentosPara tener una visión real del crecimiento de un negocio, la evaluación publicitaria debe comenzar en los datos del backend (Shopify, CRM o herramientas de atribución externa como Northbeam o Wicked Reports) en lugar de limitarse a las cuentas publicitarias individuales como Google Ads o Meta Ads.
you would never start like yeah you will do your screen shares inside you know the the MCC inside Google of course level of course but you would always start with like the back end either you know with Shopify or the CRM or even looking at something like a North beam or a wicked reports like that's where you get that bigger view of where are where am I growing the business and I an agency and you're thinking channels you are looking at the wrong freaking thing and if you only control one channel you only control that one aspect
Escúchalo en el minuto 10:34
Medir el éxito del paid media en la salud global del negocio y no en el ROAS de plataforma
Medición y experimentosPara evaluar con precisión el impacto del marketing pagado, se debe medir su efecto en el backend del negocio (nuevos clientes vs. recurrentes, nCAC, nMER, LTV y rentabilidad global) en lugar de guiarse únicamente por las métricas de ROAS reportadas dentro de cada canal.
say move the needle I don't mean just like make the row as go up but actually grow the client's bottom line uh and is we measured our Google ads success based on the business health so when I ads I will hop into the back end of their Shopify measure all customers versus new customers develop an endak identify what the end mirror was the New Media efficiency ratio you know compare that against LTV look at profitability
Escúchalo en el minuto 5:15
Performance Max sobre-atribuye conversiones al reciclar audiencias y retargetear clientes existentes
Canales de adquisiciónConfiar ciegamente en campañas automatizadas como Performance Max de Google puede inflar artificialmente las métricas, ya que tienden a sobre-atribuirse resultados haciendo retargeting al tráfico generado por otros canales (como Meta) o a clientes que ya compraron, lo cual es especialmente ineficiente en productos sin valor de vida recurrente (LTV).
say should I just slap performance Max on and cross my fingers no that's that's being a bad marketer that's being a good measurer because performance Max over attributes what's running in Google and it was like eight different performance Max campaigns I was like uh oh so the meta that might be driving kind of poor traffic we know that performance Max is going to remarket that and it was a product that doesn't have LTV so like okay well then performance uh performance Max is going to not only remarket the poor traffic coming from meta but also the return or the the people that already bought…
Escúchalo en el minuto 16:48
Medición y experimentosPara evaluar la salud y eficiencia real del marketing sin depender de las métricas infladas de las plataformas, se deben contrastar interanualmente tres datos esenciales: la diferencia en gasto publicitario, la diferencia en ingresos totales y la variación entre clientes nuevos versus recurrentes.
what is the actual Topline metrics year over year that's the first thing that you want to look at and you want look at spend differences then you want to look at Revenue differences and then new versus returning customer differences those are those are going to tell you 80% of what you need to know and what we
Escúchalo en el minuto 36:35
Medición y experimentosPara evaluar la salud y eficiencia real del marketing sin depender de las métricas infladas de las plataformas, se deben contrastar interanualmente tres datos esenciales: la diferencia en gasto publicitario, la diferencia en ingresos totales y la variación entre clientes nuevos versus recurrentes.
what is the actual Topline metrics year over year that's the first thing that you want to look at and you want look at spend differences then you want to look at Revenue differences and then new versus returning customer differences those are those are going to tell you 80% of what you need to know and what we
Escúchalo en el minuto 36:35
Medición y experimentosEvaluar el impacto de campañas basándose exclusivamente en el ROAS de cada canal individual es una métrica obsoleta; ignora el impacto de la visualización sin clic (view-through) en redes sociales o YouTube y el comportamiento multidispositivo, provocando que los canales de captación se subestimen mientras Google se atribuye injustamente las ventas de marca.
agency's fault as to how how the measurement has not progressed with the problem is we have not kept up with social media um dilution of that attribution or the multiple device days now it's social media has non- click attribution and then YouTube has non- click attribution which means yeah you've seen a Facebook ad and then you've Googled something or you've seen a YouTube ad and then you've Googled attribution and then you have a tablet a phone and sometimes like two laptops so now you have four devices that are also when you look at attribution are going to get you know you're standing in…
Escúchalo en el minuto 12:54
Insuficiencia del ROAS tradicional ante la fragmentación de atribución
Medición y experimentosConfiar exclusivamente en el ROAS como métrica principal resulta obsoleto porque no contempla la dilución de la atribución provocada por las redes sociales y el uso de múltiples dispositivos en el viaje del cliente.
it's mostly the agency's fault as to how the measurement has not progressed with real life what I mean by that is we've been using Raz for the least 10 years the problem is we have not kept up with social media dilution of that attribution or the multiple device dilution of that attribution everyone has a disconnect in their mind of well I don't know what to do so let's just measure row
Escúchalo en el minuto 0:00
Reducción del 50% en presupuesto de Meta genera solo una caída del 4% en ingresos reales
Medición y experimentosAl recortar el presupuesto de Meta a la mitad en una cuenta, la facturación total del negocio solo cayó un 4%, demostrando que la plataforma estaba sobre-atribuyéndose ventas que habrían ocurrido de todos modos o impactando audiencias que no generaban crecimiento incremental.
year-over year and we saw that when meta cut its budget in half we only saw like a 4% drop in actual Revenue so okay So Meta was way over spent or meta is going after the wrong Market or meta is going after War a es or meta just taking credit for a viewr conversion that was going to happen anyway right and we okay well meta was cut in half because the company was not growing and that didn't change the bottom line and then
Escúchalo en el minuto 18:44
El riesgo de gestionar canales publicitarios en silos independientes
Canales de adquisiciónGestionar canales publicitarios como islas separadas fomenta incentivos desalineados; por ejemplo, plataformas que inflan su ROAS invirtiendo excesivo presupuesto en retargeting de clientes que ya no volverán a comprar, canibalizando el presupuesto necesario para la adquisición real.
it's all product based it's all like individual Silo based hey I'll sell you Facebook hey I'll sell you Google hey I'll sell you SMS hey I'll sell you email well who's the one who's bringing the problem was is you'd have an agency that was managing meta that is you know measuring inao as and making you know probably not smart decisions each day like meta doing $5,000 per day out of a $10,000 per day budget in remarketing of a person that's already bought that's never going to come back again and now
Escúchalo en el minuto 7:03
Retención y relación con el clienteAnte la cancelación de un cliente, la agencia debe asumir la responsabilidad total del fallo en vez de culpar al cliente o escudarse en métricas intermedias de canal.
have their you know post getting fired water cooler meaning where it's like that client was crazy right yeah they then go and report on another internal Raz metric exactly and that's that's the uh you know jao willings extreme ownership it's like if a client leaves like it's up to it's it's our fault it's not because they're crazy you know we point is is like uh you have to take ownership of it uh and I I think
Escúchalo en el minuto 33:50
Las conversiones por visualización pueden inflar hasta un 70% los datos de Facebook Ads
Medición y experimentosEn auditorías de cuentas publicitarias en Facebook, hasta un 70% de las conversiones reportadas pueden corresponder a conversiones post-impresión (view-through), acreditándose compras de usuarios que ya tenían intención de compra y habrían convertido sin necesidad del anuncio.
yeah you know we have a a client that's like just started with us and we know 70% of their conversions on Facebook are coming from these view through conversions which basically means these are people that we're going to buy anyway like some of it yeah but you know
Escúchalo en el minuto 23:59
Límites del enfoque monocanal frente a la visión estratégica integral
Canales de adquisiciónOptimizar plataformas de forma aislada e hiperespecializada (como solo Google Ads) limita el crecimiento real del negocio si no se integran de manera holística el resto de canales pagados, CRO, email y SEO desde una perspectiva estratégica global.
think has always kind of hamstrung us um in really being able to move the full needle for a business not just necessarily the Google portion is because we were just a Google ads agency media but we couldn't ever really move the needle outside of Google ad because paid as one and when we're looking a sergus CMO um and so that's where I think that when you can take control of a company's narrative and traffic and kind of think you know five layers above what maybe an individual agency might be looking at in app you actually can see the the forest for the trees and so tier they do they just look…
Escúchalo en el minuto 2:30
Divergencia entre métricas de Google Ads y resultados globales al escalar tráfico frío
Medición y experimentosAl escalar la prospección en Shopping invirtiendo 17.000 $ adicionales en 5 días, el panel de Google Ads reportó una caída del 80% en conversiones atribuidas, pero las métricas globales del negocio crecieron: +24% en sesiones y nuevos usuarios, y un aumento del 9% en ventas totales (de 14.000 a 15.288 pedidos).
says we spent $177,000 more in this last 5 days and we lost uh 80% of our conversions so we're down 80% of our conversions and our conversion value is so it looked like we lost 80% of our conversions in Google most profitable $20,000 Google says we're down 20381 $177,000 more and we lost 20 grand in day over day we have 24% more sessions what the business growth measures 24% more new users that are 7.7% % more engaged good what are the sales we have 9% more conversions 14,000 sales up to 15, 288 sales and our Revenue went from 215,000 to
Escúchalo en el minuto 42:26
Tamaño y tasa de crecimiento del mercado global de agencias
Investigación de mercado y clienteSe contabilizan unas 143.000 agencias a nivel mundial con una tasa de crecimiento anual del 5% al 6%, impulsada por las bajas barreras de entrada del sector.
we just bought another report that shows there's 143,000 agencies in the world growing at a 5 to 6% clip because what do you need you need a laptop and some subject matter expertise and that's pretty much it there's no barrier to entry and then you undercut people in
Escúchalo en el minuto 34:46
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.