Marketero · Entrevistas
FACEBOOK MARKETING FLITTER IN THE TWITTER: Trevor Longino & Charley T discuss targeting post iOS14
23 ideas de marketing sacadas de «FACEBOOK MARKETING FLITTER IN THE TWITTER: Trevor Longino & Charley T discuss targeting post iOS14», de Professor…
- 23 ideas despiezadas
- Fuente Professor Charley T
- Dura 56 min
- Publicada en 2022
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.
Canales de adquisiciónEs preferible probar creatividades en audiencias amplias y sin restricciones de intereses para descubrir qué ángulo de mensaje resuena mejor, dejando que el algoritmo identifique y agrupe naturalmente al público comprador sin limitar el potencial de escala.
I do the targeting I do the testing at Broad to figure out the messaging that I'm best at delivering and then letting that messaging collate its own audience and then deciding do I then Facebook can find the folks that are in one Circle or two or three or four and I'm not spending nearly as much and I'm not limiting the people that might want to actually see my content because I've decided to use some predetermined audience parameters from
Escúchalo en el minuto 26:07
ConversiónNo se debe invertir en publicidad de Facebook sin haber validado previamente el mercado y el producto; encender anuncios pagados sin tener una ruta de adquisición y conversión rentable configurada vuelve el costo de inversión ineficiente y perjudicial.
I think a lot of people go into Facebook advertising before they know who their Market is and before they know what product sells well yeah and generally speaking my advice to those people is don't because Facebook is really expensive the and what I mean by that is the exchange rate on your investment with Facebook versus other forms of growth and acquisition and Revenue as far the exchange rate is far worse figure out a way that your business works right get the profitable customer Journey acquisition machine figured out build your factory to build the Factory and then complain when it…
Escúchalo en el minuto 34:02
Canales de adquisiciónLa subasta de Facebook equilibra el presupuesto, la tasa de acción estimada (EAR) y el puntaje del anunciante; al aumentar el gasto de un anuncio, el algoritmo debe buscar usuarios fuera del núcleo de máxima afinidad, disminuyendo la tasa de acción y elevando el costo por conversión.
so the page score is part of the advertiser score that Facebook brings in ultimately the bid in the auction is budget so how much money is that ad actually going to spend today budget is every ad so the example here we have an ad that is in the middle of a circle a dot in the middle of a circle that ad is in the middle of the circle the further away you get from the center of that Circle the more the lower the match rate of that ad with having a high estimated action rate for that audience is you generally have to move further away from the middle of that Circle depending on how much more…
Escúchalo en el minuto 17:47
Medición y experimentosEs responsabilidad del profesional de marketing formular hipótesis, validarlas en múltiples canales y cuestionar las métricas reportadas por plataformas publicitarias (como Facebook) cuando difieren de fuentes de datos internas como Google Analytics o el CRM.
is and then as I've learned this I validate this on other channels and if they work if they don't work that is the business owner or the marketer working for the business owner to solve and say all right this was our philosophy this was our Theory this was the hypothesis we tested it we learned this now we're doing this instead and that whole funnel a good marketer always tests a good marketer always asks and so even if Facebook tells you this is what works if you're a Google analytics or your CRM tells you something different you begin to question right this has
Escúchalo en el minuto 55:07
Investigación de mercado y clientePara validar el lanzamiento de un producto y priorizar las funcionalidades del roadmap, se deben resolver tres incógnitas fundamentales (quién es la audiencia, cuál es su problema y qué solución se ofrece) testeando una matriz de combinaciones de audiencia, mensaje y propuesta de valor mediante experimentos pagados. Las combinaciones con mejor rendimiento estadístico indican qué características priorizar en el…
yet how to prioritize their roadmap and part of what this is telling them the experiments we've run for them is here is who wants what you're selling and here is the features they care about uh by now it's like 36 different ways of creative message audience because fundamentally you've got to solve three things to go to market right who's your audience what is their problem what do you offer to fix and by doing that with a quite complex Matrix of different solutions and different audiences we're able to find two or three that statistically outperform against and with that then you're able to…
Escúchalo en el minuto 46:53
Canales de adquisiciónEn lugar de depender de la segmentación manual por intereses o datos demográficos —que suele ser imprecisa y desactualizada—, se debe utilizar el mensaje del anuncio para que el algoritmo construya su propia audiencia. Cada creatividad encuentra a las personas que reaccionan positivamente y genera su propio público similar; por ello, la investigación de mercado en paid media consiste en probar qué mensajes capturan…
knowing that every ad effectively creates its own audience based around how people engage with it like uh every ad understands who responds positively negatively to it therefore every ad goes and finds other people the next thousand Impressions that audience is going to look a lot like the 500 people that loved you and that compounding interest of an ad building its own lookalike audience uh you can even identify the ads that are ultimately earning spend desirable to contributing to the delivery of a desirable CPA those are the ads that ultimately your messaging is most effective with and so…
Escúchalo en el minuto 5:14
Medición y experimentosFacebook Ads puede utilizarse como un canal económico y ágil a gran escala para validar qué audiencias responden a qué mensajes; los aprendizajes sobre interés y conversión obtenidos allí se pueden exportar directamente a otros canales como SEO o eventos presenciales.
the chance to measure both interest rate at the top of the ad and then once you come to my landing page How likely to fill it out and go further that right there once I've learned that I can export that knowledge over into any other channel I wish to export like because Facebook has now told me these people this message and what I've learned from these people this message applies to when I go to a live event and I want to think of who I want to meet at the event or when I'm learning about SEO who do I want to come to my website all these questions are informed by Facebook as a very easy to…
Escúchalo en el minuto 10:02
Canales de adquisiciónProbar creatividades contra múltiples grupos de interés en campañas digitales de Facebook permite descubrir exactamente qué perfiles responden mejor, generando datos que justifican y optimizan la compra de medios a gran escala en canales masivos como televisión o vía pública.
and then it was running against like 30 40 interest groups based on that Insight which I could do could then take that data here is the audience data buy interest grouper who responded really well so in that way we can sell 50 million dollars of AD time for the year based on five million dollars of Facebook spend want to optimize out of Facebook that interest targeting lets you learn and export data out and that's like the right you learn who cares and then you can go and use that information anywhere else that seems applicable yeah now let's find the people who are most receptive to this…
Escúchalo en el minuto 30:44
Canales de adquisiciónMantener una alta puntuación de página y una elevada tasa de acción estimada permite acceder a inventario publicitario de mayor calidad (posiciones tempranas en el feed) con costes por mil impresiones notablemente más bajos (10-15 $ CPM frente a 35-40 $ en ubicaciones tardías de navegación pasiva), facilitando captar usuarios de alta intención incluso mediante segmentación amplia.
uh very if the estimated action rate and cost your cpms is very low let's say you've got a really engaging ad that's working abroad and you've got a 4.8 page score odds are you might be the third piece of content when somebody opens up Instagram it's like your sister some brand you love boom here's an app expensive that app that's like the third thing in the feed might be 15 CPM might be 10 CPM the ad that's 20 minutes into Doom scroll might be 35 40 cpms bell curve if you were doing better at that page score than your competition odds are you are going to even at Broad retarget bottom of…
Escúchalo en el minuto 20:13
Medición y experimentosEn lugar de exigir una precisión algorítmica perfecta en Meta Ads, conviene tratar las audiencias como hipótesis de investigación de mercado, acumular millones de puntos de contacto a través de múltiples experimentos iterativos para diluir el margen de error muestral y contrastar los perfiles reales de los leads convertidos mediante herramientas de enriquecimiento de datos como Apollo o el onboarding.
honestly I care that it works more than I care that it's accurate right if I say here's my intersection of eight different audiences even if I assume two of them are wrong targeting one I'm gathering millions of points of data which automatically levels out a lot of the inaccuracy when I go through my full set of 18 different experiments for a client I'm averaging between two and five million individual data points at the end of that right which means sampling errors have to be really big testing it right 18 distinct times over the course of two or three months I'm my onboarding I can see who…
Escúchalo en el minuto 53:34
Medición y experimentosEjecutar experimentos rápidos mediante pauta pagada permite intercambiar presupuesto por tiempo, validando audiencias, ofertas y embudos en semanas con volumen estadísticamente significativo, reduciendo el riesgo de quemar meses de nómina sin tracción.
because you haven't learned your business yet and Facebook is when you when you're running Facebook ads to answer business questions around who's my audience what do I sell you're you're spending money instead of time and what you can do is in 90 days 18 different experiments around audience at the end of 90 days you've answered what's my funnel and then you're able to you do so you can spend six months which is going to be 200k in the salary and maybe answer it or you can spend a month and get that answer spend 20 30k in money but it is one of the most agile in terms of pivoting and it's one…
Escúchalo en el minuto 36:31
Canales de adquisiciónEn negocios estacionales, los meses de baja demanda son el momento óptimo para investigar audiencias y realizar experimentos publicitarios: la menor competencia abarata los costos de pauta y permite entrenar los algoritmos de las plataformas antes de escalar el presupuesto en temporada alta.
insights and market research and those 8K a day months and understand who are the people that are still holding on days got it but I always say when it's to school that's the best time to text because what works there is going to be a tremendous Advantage also though nobody else is bidding right so that's the time when you can learn for cheap likewise if you're finding the the in the cheaper traffic and the lower traffic months learning as much as you can and it's important you're also teaching Facebook's algorithm yeah right doing upfront targeting lets you also tell Facebook I want more…
Escúchalo en el minuto 12:46
Canales de adquisiciónLas variaciones en el puntaje de página de Facebook (Page Score/Advertiser Score) modifican significativamente los CPMs y condicionan la asignación de ubicaciones prémium en el feed.
and Advertiser score that paid score what I have found is it really matters in the hyperbole so if you're over uh if you were at like a 3.8 4.0 and then you immediate and then you move to like a 4.5 4.8 you will see a pretty big shift in the cpms of your advertising and it's not just the cpms that you're advertising it also has to do with the placements that you get also the same could be said for if it goes to like a 2.5 or something either way and one thing I think people don't people don't think about is that the placement of an ad on Facebook is tremendously important and premium…
Escúchalo en el minuto 19:07
Canales de adquisiciónComenzar con audiencias preseleccionadas por intereses para alimentar el píxel con eventos de conversión iniciales permite que el algoritmo aprenda con precisión quién es el cliente ideal; una vez validada la conversión, cambiar a segmentación amplia (broad) permite escalar el presupuesto con mayor rapidez y estabilidad.
um so I actually very much like broad once you know what the hell you're doing uh once you've in two ways once you've trained the algo with you said directed traffic to your page you got the conversion events you wanted everything worked now the way it should because you were pre-selecting your audience up now when you go broad Facebook already knows who you want and then I find the algo learns faster and spend right but when if you've already put 10K into pre-targeted audiences when you move over to a broad reach you can ramp up way faster now you know what works and Facebook does too so…
Escúchalo en el minuto 23:26
Canales de adquisiciónEl uso de la publicidad pagada en Meta cambia según la madurez del negocio: en etapas tempranas sirve como herramienta de investigación para descubrir quién es el cliente y qué resuena con él, mientras que en etapas más avanzadas funciona como motor de volumen para alimentar un embudo de adquisición ya rentable.
I would thriving a higher volume of uh quality or improving value quality traffic into profitable customer Journey acquisition growth and acquisition marketing as the acquisition of profitable customer Journeys and generally I use Facebook ads and other paid media to acquire a higher volume of people into those Journeys so I would consider myself a farmer more than a hunter more of a researcher than either if you're in your early stages of business is who I deal with mostly and a lot of the questions you end up asking are how do I find customers what is my customer who should I be selling to
Escúchalo en el minuto 2:25
Canales de adquisiciónTestear y validar ofertas y creatividades publicitarias en Meta antes de trasladarlas a plataformas como Pinterest (que pueden tener CPMs más bajos pero algoritmos de optimización menos avanzados) asegura saber qué mensaje funciona de antemano y optimiza el presupuesto.
and that moved over to Pinterest and got even cheaper because HR Tech where do you find 30 to 50 year old women Pinterest cpms there even cheaper than Facebook but we validated Market we validated the creative on Facebook and then Pinterest which is a less sophisticated advertising tool we knew what was going to work off we went to the races right like that can work very
Escúchalo en el minuto 40:09
Investigación de mercado y clienteSe pueden emplear campañas de anuncios de bajo presupuesto para responder rápidamente quién tiene interés real en un producto y qué argumentos resultan persuasivos. Al ejecutar anuncios con creatividades idénticas dirigidos a diferentes grupos demográficos (por ejemplo, 20 años vs. 40 años), se comparan las diferencias en tasas de activación para validar hipótesis de demanda antes de escalar la inversión hacia…
by using Facebook to relatively inexpensively answer the question of who cares about my product and what language persuades them so like broadly speaking change how you think about Facebook for a moment and if you don't have it as a system to drive Revenue right now but if instead you say I have three billion people on this platform and I can ask them all questions and every ad that I run once I segment it into an audience that I've controlled up if I run identical ads to 20 year olds and 40 year olds who have an interest in a particular space given I can see and can replicate a difference in…
Escúchalo en el minuto 8:11
Canales de adquisiciónEstar categorizado en un grupo de interés en plataformas publicitarias no garantiza intención de compra positiva: incluye a usuarios que interactúan críticamente o se quejan de un producto, sumado a errores técnicos de asignación de la plataforma, lo que reduce sustancialmente el tamaño real de la audiencia cualificada.
being in an interest group theoretically defines that you've liked the page you've talked about something we might feel negatively about something and talk about it have bad service odds are you're more likely to talk about bad service at a so if we know that let's say half of the people in that interest group theoretically are talking about but let's just throw it out there 50 of the available audience that is accurately being tracked might feel negatively about something but they're about one-third of the people in that audience were there basically just buying a mistake like just errors in…
Escúchalo en el minuto 50:20
Medición y experimentosAunque la segmentación totalmente abierta (broad targeting) en Meta optimiza conversiones algorítmicamente, impide extraer datos cualitativos precisos sobre qué nichos o perfiles respondieron mejor al mensaje para aplicar ese conocimiento en otros canales como eventos presenciales, Google Search o patrocinios.
and that if you let Facebook answer for you Facebook will tell you kind of blankly this ad made this much money works for but you really can't see if I want to export Knowledge from Facebook on what audience cared the most about what my ad had to say so that I know what events do I attend in the real uh on Google search how might I wish to write my ads to persuade a particular audience or how I might I want to do in audience testing or buy a newsletter subscription uh and subsidize somebody to run ads in a newsletter right all this information can't really extract from Facebook on Broad yeah…
Escúchalo en el minuto 3:47
Investigación de mercado y clienteUtilizar Meta Ads como una herramienta de investigación de mercado y creación de intención a escala permite validar la demanda de conceptos y ofertas antes de escalar, aumentando significativamente la tasa de éxito de nuevas iniciativas.
same spot is using Facebook as and I always say it's a market research and intense creation device at Scales probably the world's greatest market research and intent creation device yes because of the market research you're doing so you're investing in market research to get to instead of a 20 success rate 80 failing 20 being successful instead of getting a 20 success rate getting to a 30 or 40 or 50 600 or more but like I'm I'm at about 50 these days yeah that's fantastic so you have a two and a half X better success rate by doing this yes
Escúchalo en el minuto 40:56
Medición y experimentosAunque la segmentación abierta optimiza con alta eficiencia las conversiones dentro de la plataforma publicitaria, encierra los datos psicográficos y de intereses en una caja negra, impidiendo transferir con facilidad esos hallazgos de audiencia a otros canales como televisión o prensa.
the problem is then you can't export what you've learned right Facebook will let you optimize in network and will do a great job of optimizing even broad but it won't let you know what that optimization was and that's why if you why Facebook does this right if they lock in a super good audience and don't tell you who it is you're stuck spending money on Facebook good for them but if I want to learn an audience and then use it elsewhere in my Facebook whereas Facebook might be know to Target that but Facebook won't tell me unless I pre-select and then
Escúchalo en el minuto 27:18
Canales de adquisiciónEn un caso de HR Tech B2B con valor de contrato anual (ACV) de cinco cifras, la captación de leads en Facebook arrojó un CPL de alrededor de $80 USD frente a los $750 USD por lead registrados en LinkedIn.
yeah so uh that for them they were doing 10K a month they were doing I think 25 a month on LinkedIn and they're spending like 750 a lead on LinkedIn right and they were happy because their ACV was this was an HR tech company you don't better uh we were getting like 80 CPL
Escúchalo en el minuto 39:35
Medición y experimentosProbar conceptos creativos contra segmentaciones abiertas ('Broad') proporciona datos de rendimiento más fiables y extrapolables para escalar que testearlos en nichos de intereses específicos, los cuales añaden variables y sesgos difíciles de comprobar.
I would rather do wide scale concept based testing around creative specifically so that I can understand what Creative Concepts work I feel like the validity of the data that I'm getting from an interest group I know that if I put an ad at Broad and it works I can trust that it works if I put in at an interest group to say that that interest group is affected I've got a lot of caveats that I have to check off my list
Escúchalo en el minuto 42:45
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.