Marketero · Entrevistas
PPC Live Online ft. Ayisha Yousef, Paul Salame, Jérémy Courty
47 ideas de marketing sacadas de «PPC Live Online ft. Ayisha Yousef, Paul Salame, Jérémy Courty», de Jyll Saskin Gales | Google Ads Coach. Sobre todo de…
- 47 ideas despiezadas
- Fuente Jyll Saskin Gales | Google Ads Coach
- Dura 1 h 21 min
- Publicada en 2025
Ver la entrevista entera en YouTube
De qué habla
Las 47 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 47 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.
Como montar POAS sin ser un genio
Medición y experimentosPara tener POAS hay que calcular el beneficio bruto: ingresos menos coste de la mercancia, y en ese coste meter todo lo que se pueda calcular, directo e indirecto, promociones, materiales, mano de obra, envio, comisiones de pago. Eso se carga al feed con el atributo de coste de la mercancia, preferiblemente por feed suplementario desde el gestor de feeds y no por reglas de feed, que es la peor via. Al ser un…
But um pas is not that hard to calculate. So in order to get pas all we need to do is calculate the gross profit. Um and we can do that by taking revenue which most of us already have available to us and remove cost of goods sold. And the cost of goods sold is basically as both direct and indirect. literally everything. So whether it's the cost of the promotions, the materials, the labor, the shipping cost, payments, get as many things in there as possible, um and then we can just apply it um to the feed um with the attribute cost of goods sold. And then we can do this three ways, three…
Escúchalo en el minuto 14:43
Puja por margen: objetivos de ROAS al reves de lo que parece
Canales de adquisiciónSe coge el dato de Google Shopping, se anade el margen bruto en una columna y con eso se calcula la rentabilidad. Despues se agrupan los productos por tramo de margen con etiquetas personalizadas y se reparten en campanas distintas: objetivos de ROAS mas exigentes a los productos de margen bajo y objetivos mas relajados a los de margen alto, jugando con esos objetivos hasta encontrar el punto dulce, con los numeros…
Uh margin bidding. So again like grabbing your Google Shopping data, adding your gross margin in a column and then from that you can work out profit um and obviously you can use your margin groupings to make custom labels. all these products in different campaigns. Have higher RAS targets on your lower margin products and have lower RA targets on your higher margin products. Play around with those RAS targets. See what's a sweet sort of spot and then obviously you've got the profitability numbers right there. So, you know, it's quite simple, right? when once you have the reporting there for…
Escúchalo en el minuto 50:04
Los tres cubos de lo basico antes de tocar la estructura
Canales de adquisiciónCuando las cosas van mal, primero se cubren los basicos, agrupados en tres cubos. Uno, feed y datos: no enchufar el feed de Shopify tal cual al Merchant Center, sino auditarlo, ver que atributos faltan y anadir los no obligatorios propios del sector, porque eso es lo que empareja tu producto con la consulta; entrar en un CSS; meter el dato propio desde la herramienta que sea, tipo Klaviyo, bien segmentado; y los…
we chop and change everything, I think we need to cover uh our basics. So when things are not going our way, um I've categorized it into three uh different buckets. So the first one is feed in data and that's effectively everything going on the right direction. Feed optimization is a big one. Don't simply like put your grab your feed from Shopify and plug it into uh your merchant center. Um look at ways of doing audit. Look at whether you're missing some u specific attributes, whether you've got non mandatory attributes from specific vertical that you can add into into the feed. Ultimately…
Escúchalo en el minuto 57:07
Canales de adquisiciónNo hay ninguna casilla en la plataforma donde decirle a Google que quieres mas beneficio o mas clientes nuevos. Lo unico que entiende es un objetivo de ROAS o de CPA. Por eso hay que guiarlo con datos de negocio y con estructura: separar campanas y montarlas de forma que reflejen el objetivo, priorizando los productos que lo cumplen y despriorizando los que no.
lower margin. Um because at the end of the day obviously Google doesn't actually know what your business goals are. Like there's no way to input into the ads platform. Okay, I want to drive more profit. I want to drive more new customers. Doesn't know. All it's got is a rare target or a target CPA. So, you need to guide it through your business data and tell it okay look I want to achieve this obviously through control through splitting out your campaigns and through having that structure which complements your business goals. So,
Escúchalo en el minuto 47:33
Un penique de valor al cliente nuevo: guiar a Google sin ensuciar el informe
Canales de adquisiciónAl usar el objetivo de captacion de clientes nuevos en Google Ads, el ponente pone siempre un valor de un penique en todas sus cuentas. Cualquier valor mas alto se suma al valor de conversion de la cuenta y arruina el reporte, mientras que un penique basta para que Google entienda que quieres priorizar a esa gente. Es una forma sencilla de mejorar la incrementalidad y dejar de pagar tanto por clientes recurrentes.
>> This probably looks familiar to a lot of people now, but obviously you can optimize towards new customers on Google as well. So from an incrementality perspective, again, like super useful, especially for branded bidding. Uh really simple. I always set a 1 P LTV across all my accounts just because if you add any more than that, it will add that to your actual conversion value within the Google Ads account, which will obviously, you know, just ruin your reporting. Um and 1 Pence will guide Google anyway, so it will know that you're trying to prioritize these people. So this is a really good…
Escúchalo en el minuto 44:28
Customer match bajo de 1.000 a 100 usuarios: ya no hay excusa
Canales de adquisiciónGoogle redujo este ano el minimo de las audiencias de customer match de 1.000 a 100 usuarios, con lo que pasan a estar al alcance de practicamente cualquier comercio. Lo primero de la jerarquia de audiencias es el dato propio, asi que hay que insistirle al cliente para que suelte sus listas de correo; y si no las da, se pueden montar audiencias de compradores anteriores desde la seccion de administracion de Google…
But anyway, hopefully we can all agree that the top of our hierarchy of audiences is any sort of first party audiences, first party data, sometimes customer match audiences, there was a big change earlier this year from Google. I can see some nodding heads. Hopefully you heard about it where the minimum requirement for customer match audiences dropped from 1,000 to 100. I can see you nodding. Yeah, this is massive. This means that it's now available to pretty much every retailer, every client out there. So, do bug them for their, you know, their email lists. If you can't do that, I just…
Escúchalo en el minuto 20:42
Canales de adquisiciónPMax es un gran culpable de irse a por la fruta madura, sea por busqueda de marca o por shopping de marca. Excluyendo la marca completamente de PMax se controla cuanto gasto va a generico y que resultados da, y se deja aparte una campana pequena de shopping de marca que gasta poco pero mantiene la visibilidad para no regalarsela a los competidores.
What I like to do with PMAX is actually excluding brand completely because what you can do with that like Pax is just a massive culprit of going after low low hanging fruit whether it's through branded search, branded shopping, whatever it may be. So this is a really good thing. This is a really good strategy to try and have more control over how much spend is going in generic versus brand through Pmax. Uh, and obviously that just helps with incrementality as well because then you can control how much you're spending on generic PMAX, how much what results you're seeing from it, and also…
Escúchalo en el minuto 46:00
Medición y experimentosPara la busqueda de marca siempre activa, mira cuota de impresiones y cuota de clics contra numero de clientes nuevos en tu plataforma de ecommerce y contra el CAC, y ve probando: si estas al 80% de impresiones y 85% de clics, bajalo y observa si el CAC mejora o empeora y cuantos clientes nuevos pierdes. La excepcion es el periodo de rebajas, cuando los competidores empiezan a pujar por tu marca con mensajes de…
So, when it comes to like always on branded search activity, what I tend to do is look at impression and click share and then I look at it versus like new customer OS, like new customers, Shopify, whatever ecom platform you're using, and basically just incrementally test. So, make some tweaks. Say you're on like 80% impression share and 85% click share through a branding campaign. Okay, let's try and lower it. What impact does that have on new customer numbers, what impact does it have on your CAC? Does it get healthier? Does it get weaker? Um based off of that, obviously, you know, fine-tune…
Escúchalo en el minuto 43:24
Canales de adquisiciónComo el presupuesto se concentra en los mas vendidos, que se agoten es un problema grave. Y no basta con mirar la referencia principal: si se han agotado las tallas o colores populares, el producto se sigue sirviendo y gastando mientras caen la conversion y el ROAS. Con diez, veinte o cincuenta referencias criticas, se habla antes con el cliente, se comprueba el stock, se piensan productos sustitutos y, si no hay…
changes. So one of the biggest things that we can do here is monitor our stock. especially the monitoring as the stock of the best sellers because um as Jeremy nicely pointed out, we put a lot of money behind our bestsellers. So, when these run out of stock, that can be a really big issue for us. And an additional thing to look out for is not just the bestseller skew itself. If it comes in multiple different colors, sizes, is it that the popular sizes have still have already sold out? Um, so that product might still be getting served, spending money, and then we'll start to see a row dropping…
Escúchalo en el minuto 27:11
Escala de incrementalidad por tipo de campana
Canales de adquisiciónOrdenando los tipos de campana de menos a mas incremental: busqueda de marca abajo del todo, porque esa gente ya te conoce y ya te iba a comprar, normalmente porque la educaste en Meta o TikTok; despues shopping de marca, mas incremental que la busqueda de marca porque el espacio de shopping tiene mucha mas competencia; luego generica, gente consciente del problema y en mercado pero que aun no sabe a quien…
going after. So this is just like pretty much every single campaign type on Google ads. So branded search shopping least. Um, so you see at the bottom here, you've got an increment incrementality scale from least incremental to most incremental. So in my opinion, this is how it sort of ranks. So at the bottom you've got branded search. So these people, you know, brand aware, um, they're ready to buy from you pretty much. Um, so they're the least incremental. Um, you've probably educated them through like Meta, Tik Tok, why they should buy into your brand. Um, I put brand shopping next. So I…
Escúchalo en el minuto 37:05
Canales de adquisiciónRegla estricta con sus equipos y sus clientes: pasado Halloween, ni se toca la estructura de campanas ni se cambia lo que se mide como conversion. Si cambias la estructura o el seguimiento entrando en noviembre, el aprendizaje automatico no tiene tiempo de asentarse con esa informacion y te estas preparando el fracaso.
throughout that. So my general rule of thumb as well and I'm thinking about that timing. So even though we have take action across October as well, I really have quite a strict rule with the teams that I run and even my clients. My gosh, aren't I terrible. But no major changes to campaign structure or conversions past Halloween. And that's why we've got the scary ghost. So if you're changing your campaign structure vastly moving into November or you're changing what you're tracking, you're really setting yourselves up to for a fail. So think feeding on machine learning if you're only giving…
Escúchalo en el minuto 11:17
Revisa el Merchant Center a diario, fines de semana incluidos, y automatiza la alarma
Canales de adquisiciónDurante Q4 la revision de incidencias del Google Merchant Center deberia ser diaria, tambien sabado y domingo. Como no se pueden crear alertas desde el Merchant Center, se montan del lado de Google Ads: una regla que cada manana mire las campanas de shopping y avise por correo de las que no estan recibiendo impresiones.
doing during this time? Number one checking the GMC. So check your Google Merchant Center. Check your issues. see what's going on in there. And I would really recommend this should be at least a daily check, including the weekends. Sorry guys. Um, and as well, we can create alerts. Like, of course, it's kind of impossible to create alert from the GMC, but we can do it Google ad side. And what I just tend to set up is well lots of different alerts, but every day in the morning um choose the campaigns um any shopping campaigns and any um campaigns that have received less than zero impressions…
Escúchalo en el minuto 23:55
No dejes las promociones solo en las extensiones
Mensaje y copywritingMeter la oferta unicamente en las extensiones es un error: hay que tomarse el trabajo de llevarla a los titulares y las descripciones, jugar con ellos y usar cuenta atras donde se pueda en las campanas de busqueda para crear urgencia. Es la parte que mas tiempo consume y la que te hace destacar en una pagina de resultados muy competida, asi que se planifica con antelacion y se revisan las hojas de cambios, porque…
honestly just a job in itself so I really love that everyone in the room does not just rely on extensions for your promotions and discounts. Tutt toot. Don't do that. Take the time to get your um your offers in your actual headlines, in your descriptions. Play around. Use countdown where possible. Obviously in our search campaigns, create that urgency and also plan this in because this is the part that takes a lot of time and it can make you stand out from the competition in that SER which is a very competitive space. So, plan your creative changes, plan out your um sheets ahead of time and…
Escúchalo en el minuto 18:10
Canales de adquisiciónClasificacion de productos por rendimiento con formulas ajustables: zombi es el que no llega a mil impresiones en noventa dias, heroe el que gasta mucho, factura bien y es eficiente, escudero el que gasta menos que un heroe pero tiene margen para escalar, y villano el que va mal. Cada grupo va a su campana con su propio objetivo de ROAS. Con todo en una sola campana y un ROAS unico, Google gasta hasta cumplir esa…
particular instance. So this is an example of like again like some performance um categories I put together based on just formulas. Um so for example like a zombie product would be a product that's got less than a thousand impressions last 90 days let's say. Obviously it's not one size fits all like you can tweak these definitions. Uh hero products are there products that you know have higher spend. They drive great revenue. Their efficiency is really strong and you want to keep on just pushing those as much as possible. Psychics are their products that don't get as much spend as heroes, but…
Escúchalo en el minuto 51:02
Que pasa si no pujas por tu propia marca: el peligro no es perder la venta, es el intermediario
Canales de adquisiciónSi no pujas por tus terminos de marca, casi todos te compran igual por el resultado organico, unos pocos se van a un competidor (normalmente un distribuidor, porque robar a un cliente de marca es dificil) y alguno no compra. El problema real es el distribuidor: el usuario entra ahi buscando tu marca y se lleva otra, como quien busca Nike Tenis y acaba comprando Adidas en la tienda multimarca. Paul Smith desaparecido…
So, what actually happens if we don't impact this. But for the most part, if you're not bidding on your branded terms, or at least if you're not bidding aggressively on your branded terms, um most people who search your brand and don't click on the PPC ad because you're not visible, um they'll just buy through an organic listing. Like they're visible there. There will be a few people who go to a competitor. For the most part, it's probably retailers because again, they probably still want your brand. It's very tough for a But obviously that comes with like context as well. So SER context is…
Escúchalo en el minuto 40:46
Informe de valor de vida por referencia: un producto de 24 libras que vale 45
Medición y experimentosCon los datos de cliente por pedido de Shopify se monta un informe por referencia que compara el valor medio del primer pedido con el valor de vida de esos clientes. En su ejemplo, un producto con 26 libras de primer pedido acaba en 27 de valor de vida, es decir una libra de diferencia y poco recorrido; otro con 24 libras de primer pedido llega a 45. Ese segundo puede parecer poco prioritario por su margen o su…
some examples as I mentioned like gross margin um lifetime value propensity for second purchase from a particular skew. So by that I mean okay these are all the products that were bought by these customers for the as part of their first order. how much LTV do we see from those orders? And then I've actually built a slides, but um your report basically tells you, okay, which products are driving the most the highest LTV customers and therefore the most profitable customers because you might have a product that you know it's it's low in gross margin or it tends to see a low AOV first order…
Escúchalo en el minuto 47:58
El ROAS de la plataforma sube mientras el de cliente nuevo baja
Medición y experimentosEl escenario no incremental es exactamente ese: aumentas la inversion, el ROAS que ves dentro de Google sube, pero el ROAS de cliente nuevo (ingresos de clientes nuevos divididos entre la inversion) baja. Eso significa que el dinero extra se esta yendo a clientes recurrentes o a fruta madura, gente que iba a convertir pujaras o no. Lo que quieres ver es que los dos suben a la vez. El panel de Google no debe decidir…
so I would never let the Google ads data fool you so never let the dashboard let's not look at in platform rowass. I take it with a pinch of salt and he still said that. So again like I just take Google ads in platform whereas a pinch of salt so obviously use it to like help you guide help guide you make decisions like in platform actionable decisions but don't let it guide you on whether a platform is actually working Everything else is constant. So in a non-incremental spend scenario, you'll spend. So in the first scenario, Google rowass is going up, but you'll see that new customer rowass…
Escúchalo en el minuto 34:34
Medición y experimentosSi subes la inversion en Google mientras escalas Meta o TikTok, esos otros canales tambien empujan tu margen de contribucion y tus ingresos de cliente nuevo, y ya no sabes que produjo que. Al probar escenarios de inversion hay que dejar quietos los demas canales; si no, no es un experimento controlado y la conclusion no vale.
is really important as well like this sort of data to make sure that there's no external factors which are impacting this. So by that I mean if you're increasing Google ad spend but you're also scaling Meta ad spend or Tik Tok ad spend whatever it may be obviously those other channels are also going to have an impact on your contribution margin potentially increasing your new customer revenue potentially increasing so you always want to have like control variables in place when you are like incrementally testing different spend scenarios and different advertising channels otherwise you know…
Escúchalo en el minuto 34:03
El dia que quitaron el horario de anuncios en Reebok y se apagaron los anuncios
Canales de adquisiciónEn Black Friday la gente busca a horas raras, a las dos de la manana, asi que hay que desactivar cualquier ajuste a la baja o apagado por franja horaria. Ella pidio a su equipo que lo quitara, el mensaje paso del responsable al directivo, y lo que hicieron fue eliminar la programacion entera: los anuncios no se sirvieron en absoluto y hubo que darle explicaciones al responsable de la marca. Lo correcto es dejar la…
couple of bits and actually I was major in my team when I worked on Reebok. So during um Black Friday, people just search at random times. I'm sure a lot of us have got ad schedules in place. Um, so I told someone in my team or I told the manager, then the manager told the exec and they told them to remove the ad schedule. So what they actually did is just they they our ads weren't live at all. What we need to do is obviously you just add like 0% throughout all of the day and all of the time, but they just removed it and that was a really big issue that I had to um I had to explain to like…
Escúchalo en el minuto 22:23
Antes de culpar a la IA, mira si tienes lo basico puesto
Medición y experimentosSe ven demasiadas cuentas donde no funciona nada y resulta que no tienen la lista de CRM cargada, siguen en el CSS de Google y les faltan atributos en el feed. No se puede echarle la culpa a la IA: el trabajo es alimentar al aprendizaje automatico con los datos correctos. Y no sigas haciendo lo que siempre hiciste, porque lo que probaste hace dos semanas puede estar completamente desfasado la semana que viene; hoy…
So my key takeaway feel like it was in 5 minutes, I'm not sure. Um but when I think when things get tough, we need to cover all our basics. So too often I see accounts where oh it's not working, but we don't have the right CRM list in it. we probably still on the Google CSS. We don't have we're missing attributes in a feed optimization. So we can't just blame it on AI. Ultimately, again, we need to fuel the data with with the right um fuel machine learning with the right data. Don't keep doing what you've always done. What you tested two weeks ago might actually be completely different next…
Escúchalo en el minuto 1:10:21
Por encima de 10.000 libras semanales, planifica el presupuesto dia a dia
Canales de adquisiciónLos dias de Q4 son muy irregulares y sobreinvertir en un dia de conversion baja te estropea el resultado global. Con cuentas que gastan mas de unas 10.000 libras semanales en Google Ads, la ponente recomienda una planilla de presupuesto diario planificado: cuanto hay que gastar cada dia, cuando relajar los objetivos de ROAS y cuando subir presupuestos, sin confiar en que el crecimiento organico del volumen lo haga…
Okay, we're getting into the gritty part of it. So, budget planning, I just think this is so important. So, spending, you know, the days throughout Q4 can be very up and down. And if you overinvest in a day with a low conversion rate or something, that can really really mess up your overall um outcome. So, I really recommend that if you have an account that's spending say more than about £10,000 per week on Google Ads to do a daily budget laid down. And this is going to be in the um by the newsletter as well. So, always plan everything out. have an idea of what spends that you need to get to…
Escúchalo en el minuto 16:24
Un CSS baja un 20% el coste por clic en shopping
Canales de adquisiciónEntrar en un servicio de comparacion de precios distinto del de Google supone un 20% de descuento sobre los CPC de shopping. Para el ponente es una decision obvia y sin embargo muchos grandes minoristas siguen en el CSS de Google y compiten peor que sus rivales solo por eso.
and in the UK. Advertisers choose to work with Product Hero for their shopping ads because they get a CSS discount of 20% to roll down their CPCs. CSS comparison shopping services no brainer for me. It's 20% reduction on your CPCs on shopping. Just do it. Um, we talked to so many like large retailers um out there. They're still they're still in the Google uh CSS. They're just not as competitive as they could be versus the competitors. First
Escúchalo en el minuto 3:34
Con la IA en los resultados, el feed importa mas que nunca
Canales de adquisiciónAunque la experiencia del usuario este cambiando mucho, la optimizacion principal sigue siendo la misma y gana peso: los metadatos que le das a Google, que en ecommerce llegan sobre todo por el feed. No basta con tener las columnas rellenas, tienen que tener sentido, porque es de ahi de donde Google tira para emparejarte con consultas mas amplias o de cola larga. Si el feed no era prioridad, este ano toca doblar la…
I think for me the primary optimizations that we still need to be making and it just gives it even more importance is the metadata that we're feeding and the main metadata that we're giving Google Ads is via our feeds when it comes to shopping and e-commerce. So, if it wasn't already a priority for you to be looking at your feed, checking your feed health, really making sure that all of your columns are not just completed, but really make sense because that is what Google's going to be drawing on to really match you to different queries um where they're just a little bit more either broader…
Escúchalo en el minuto 1:13:42
Tabla de precios de la competencia con ChatGPT, y mandarsela al cliente sin que la pida
Investigación de mercado y clienteTruco visto en LinkedIn: darle a ChatGPT los enlaces del mismo producto en varias webs de competidores y pedirle una tabla con los precios por referencia. Para clientes que solo venden en su propia web se puede hacer con cinco o diez productos comparables de la competencia. Lo valioso no es solo el dato: es mandarselo al cliente en pleno Black Friday sin que lo haya pedido, que es de esas cosas que se olvidan y…
So, um, if you guys aren't already avidly using LinkedIn like me, um, it's So, this post is just saying about how he's using chat GBT, which I don't actually really use that much, but he's just given it a really simple prompt. It says, "Please create a table with the prices for each of these SKs." and he's just checking his products on competitor websites. So, link the products from website one, website two, website three. Such an easy way and it gives you a table. Now, I'm thinking about doing this, but I've actually got some clients that they only sell on their own website. So, what I'm…
Escúchalo en el minuto 24:46
Google dice 15 conversiones al mes; los datos dicen entre 500 y 1.000
Canales de adquisiciónGoogle sugiere algo asi como quince conversiones al mes para que el shopping estandar funcione. Genie Shopping, un proveedor de CSS con cerca de mil minoristas, miro cientos de campanas y encontro que el punto dulce esta entre 500 y 1.000 conversiones por campana y mes: en ese tramo, el 70% de las campanas cumple o supera su objetivo. El ponente no dice que por debajo de 500 no se pueda, solo que esa es la distancia…
Um, sorry, no product here is in the house, but I'm going to plug in Genie Shopping, a CSS provider. Um, but these guys are as sister company. So they are a CSS provider and then you can and they've got about a thousand retailers and then they book. So a lot of uh a lot of campaigns. We know we need a lot of conversions data for the algorithm to work. Um Google is telling you something like 15 conversions a month I think on standard shopping. Now what we seeing here is that these guys looked at um hundreds of campaigns that they're running and effectively the sweet spot is that really between…
Escúchalo en el minuto 1:05:23
Canales de adquisiciónFuera del comercio nadie sabe que el Black Friday es el ultimo viernes de noviembre, asi que se ve un pico grande de trafico de busqueda el penultimo viernes, lo que en el sector llaman fake Friday, y otro subidon el domingo siguiente a ese viernes falso. Hay que tener la campana lista para esas fechas, no solo para el dia oficial.
So, first up, fake Friday. No one outside of e-commerce or retail actually knows when Black Friday is. Of course, we all know it's the last Friday in November. No one else knows that. My mom doesn't know that. I don't think my plumber knows that. So we actually see a big search surge in traffic um on the second to last Friday. So this is called fake Friday. And then of course we have the Sunday after fake Friday which also sees a surge. And then we've got our
Escúchalo en el minuto 8:23
El cliente que no compra hace un ano cuenta como captacion
Retención y relación con el clienteEn un cliente con ventanas de doce meses y de sesenta o sesenta y cinco dias, veian muchisimos recurrentes entrando por Shopify y ninguno por Meta, y resulto que los de Meta eran todos clientes dormidos. Si alguien no compra desde hace un ano y vuelve, eso es practicamente recuperar un cliente: aunque la plataforma de ecommerce lo cuente como recurrente, desde la optica de captacion tiene sentido tratarlo como…
I forgot lapse customers. So yeah, like again obviously it depends on the brand, but like for because I'm one of my clients like they've we've got like a um a last 12 months um a last 60 day 65 days. I going to Meta now. Uh we've got a last year's purchases pixel audience within our ASC campaign. Uh they're seeing loads of returning customers come through Shopify. We're not seeing any come through Meta. And they're actually all lapse customers. And from their perspective, they're like, "Okay, elapsed customer is basically customer acquisition because if you've not bought for a year and you…
Escúchalo en el minuto 45:01
Canales de adquisiciónDos criterios rapidos para despriorizar producto en shopping: los articulos con muchas devoluciones, donde no tiene ningun sentido gastar en publicidad, y los que estan muy bajos de stock y ademas tienen buena velocidad de venta organica, porque se van a agotar igual sin pagar por ellos.
profitability returns. So products that see high returns like what's the point in advertising them right? There's no point. Uh and then finally just based on stock. So if you got products which are really low on stock probably like high organic sales velocity as well so that at least they can sell out through your organic channels and then yeah like definitely worth dep prioritizing those products as well.
Escúchalo en el minuto 48:39
Canales de adquisiciónDos criterios rapidos para despriorizar producto en shopping: los articulos con muchas devoluciones, donde no tiene ningun sentido gastar en publicidad, y los que estan muy bajos de stock y ademas tienen buena velocidad de venta organica, porque se van a agotar igual sin pagar por ellos.
profitability returns. So products that see high returns like what's the point in advertising them right? There's no point. Uh and then finally just based on stock. So if you got products which are really low on stock probably like high organic sales velocity as well so that at least they can sell out through your organic channels and then yeah like definitely worth dep prioritizing those products as well.
Escúchalo en el minuto 48:39
Anadir al carrito como conversion primaria casi nunca deberia estar ahi
Medición y experimentosAuditando cuentas en el ultimo ano se encontro muchas con la accion anadir al carrito marcada como conversion primaria. Eso solo tiene sentido en un comercio de ticket muy alto y muy pocas conversiones, del tipo joyeria o diamantes; para el resto es darle a la maquina la senal equivocada, y mas cuando campanas como PMax solo ven el objetivo de CPA o de ROAS.
have to give the robot, the machine learning, the best um data that we can in terms of conversions. We've all spoke about PMAX doesn't have all of the data that we have. It only has the CPA goal, the ROS goal. So make sure there's been quite a few occasions that I've audited accounts literally in the last 12 months and even though they're only using them to optimize towards you know they're separate reports so many accounts have um like add to basket as a conversion action that's set to primary which unless you're a very um high value retailer with just a few conversions coming through which…
Escúchalo en el minuto 12:25
Modelo mixto: shopping estandar para alimentar, PMax para los best sellers
Canales de adquisiciónCuando los CPC suben, el anunciante pequeno se queda sin trafico, sin conversiones y entra en un bucle en el que la herramienta ya no puede pujar bien. Volver a shopping estandar les devolvio control del emplazamiento, visibilidad en busqueda y negativos, y sobre todo conversiones con las que alimentar al algoritmo. Y en vez de team PMax contra team estandar, un modelo mixto: cuando ya hay datos de conversion…
PMAX. And the reality is that it's all about feeding the tool with conversion data. And some of them are not meeting that threshold because when you've got that um increase in CPCs on their account, you just being priced out of the auction, you're no longer driving enough traffic and then you go into that loop of not driving enough conversions and then again um you're not able to kind of um the tool is not able to bid effectively. we've had with some of those clients. We wanted to have full control over the placement. We wanted to have greater search visibility and being able to add negative…
Escúchalo en el minuto 1:04:20
El dato de negocio es tuyo; el de la plataforma puedes perderlo
Medición y experimentosLo que cambia con el tiempo es el dato que entra en Google Ads, y eso complica optimizar y sacar aprendizajes. Por eso el ponente defiende apoyarse en el dato de negocio: ese siempre esta ahi y no lo pierdes, mientras que la vision que te da una plataforma de terceros puede desaparecer. En paralelo hay que asegurar lo que si depende de ti: seguimiento del lado del servidor, conversiones mejoradas y modo de…
mean it's always there. Um I think the only thing that really changes is the data that comes into Google ads and obviously you know that makes that makes it a bit more difficult when it comes to like optimizing um and you know having that data within the platform to actually use to get gather your learnings from whatever strategy you put in place. But again, like there's obviously things we can do like these days to help with like just making sure like the data that we get is 100% whether it's like server side tracking through, you know, um through enhanced conversions or um consent mode as…
Escúchalo en el minuto 1:15:13
Medición y experimentosDesde la optica de medios pagados, crecimiento real es la actividad de marketing que aporta ganancias significativas e incrementales, y varia por negocio: clientes nuevos, facturacion total, ingresos de cliente nuevo con un objetivo de CAC. Conviene mirar el negocio por arriba (CAC, clientes nuevos en Shopify, ingresos de cliente nuevo) y por abajo (cuenta de resultados, margen de contribucion, beneficio neto, que…
part of our definition. So, how I've defined it is through the lens of pay media, it means marketing activity that contributes to meaningful and incremental gains for your business. But one-sizefits-all. So for example, some businesses their goal might be to increase new customers. Some might be just increasing topline revenue. You growth goal. It might be increasing new customer revenue. It might be, you know, having a new customer revenue target, but also having a CAC target in place as So what does like causing growth business in two different ways. Topline performance, bottom line…
Escúchalo en el minuto 32:22
CPA, ROAS y POAS: bueno, mejor, optimo
Medición y experimentosLas tres metricas valen y se usa la que se tenga disponible, pero hay una escala. El ROAS ya tiene en cuenta el valor de la venta, que es mejor que el CPA, pero no el beneficio de esa venta, y ahi te tropiezas: acabas vendiendo productos de valor muy alto que ademas tienen mal margen, y sacrificando otras ventas por ellos. El POAS, que mira beneficio, es el paso siguiente segun madura el cliente.
So, the differences and the different metrics in terms of CPA, ROAS, and POAS. So, we've spoke about it a little bit Okay? And I like to look at them as good, better, best. Now, you use what is available to you. Of course, we can see that the different metrics have different things included in them. However, you know, this is just an evolution of your client and an evolution of what you've got available to you. So, if you are still using Rorowass, this is good. You know, we are still taking into consideration the value of the sales. However, we aren't, you know, taking into consideration the…
Escúchalo en el minuto 13:23
De toda la transparencia nueva de Google, lo unico accionable es el informe por recurso
Medición y experimentosLas ventanas que Google abrio este ano (desglose por canal en PMax y demand gen, terminos de busqueda en PMax, informe por recurso) se agradecen pero saben a poco: dan una pista, no informacion suficiente para cambiar la estrategia. El desglose por canal solo sugiere cosas como hacer mas video si ves clics de YouTube, cuando YouTube es mucho mas que emplazamientos de video. Lo unico realmente accionable hoy es el…
Now, you mentioned data and transparency. And so, some changes that have happened in Google Ads this year that I'll admit were not on my 2025 bingo card were more transparency coming to a lot of campaign types. Like, for example, in demand gen and performance max, we can now see that channel breakdown in a way we couldn't before. We have search term reporting in Pmax, which we did not have before. We're getting asset level reporting, which we did not have before. So, I'm curious for >> Yeah I can uh kick us off. I think that it's great that Google has given us a little bit more insight, a…
Escúchalo en el minuto 1:16:15
Rastreador de gasto por hora para los dias pico
Medición y experimentosElla usa un rastreador de gasto por hora todos los dias, y en Q4 se vuelve imprescindible: en los dias de picos grandes te deja comparar lo que llevas gastado con el mismo momento del dia anterior y decidir si hay que empujar o frenar, en vez de enterarte al cierre.
and this thing, this is a hourly spend tracker. Okay. then. So then you know how good they are and how insightful they are. Um so I use this actually every day but during Q4 during those periods where you know you can get really big spikes and you can see um what you're spending comparing to the last day. This is actually the one that Iceland's data. But yeah, it's a really good way to understand whether you need to push, pull, and you'll know where you're at.
Escúchalo en el minuto 26:11
El evento de ofertas de Amazon de octubre arrastra a todo el ecommerce, y luego viene el valle
Canales de adquisiciónAmazon coloca su evento de grandes ofertas al principio de octubre porque sabe que la gente ya esta buscando, y eso tiene un efecto domino: suben las tasas de conversion en Google y en tu propia web aunque no vendas en Amazon. Despues la ponente espera y vive un valle entre octubre y mediados de noviembre, y en ese tramo recorta inversion.
the first events that we have is right at the start of October. Um you know, Amazon are smart. They know themselves that um people are searching at the start of October. So, they've put an Amazon big deals event and that can often have a ripple effect across the e-commerce world where we see CVRs, even actually increase via Google and on your own websites. I've got to say, I do usually um expect a lull and experience a lull within my campaigns. Um so, I do tend to pull back during that period between October and then sort of like the start and mid of
Escúchalo en el minuto 7:33
Un cliente recurrente solo es incremental si esta dormido o a punto de irse
Retención y relación con el clienteEl unico caso en el que pagar por un cliente que ya te compro aporta algo es cuando es un cliente inactivo o uno a punto de abandonarte. Los recurrentes activos ya estan comprados, son fieles y van a convertir igual: no tiene sentido reinvertir en ellos por publicidad, que ese trabajo lo haga el marketing propio y el organico.
increasing. Um the only thing I like caveat of this like the for returning customers, the only time I see a returning customer actually being incremental is when it's a lapse customer or if it's someone about to churn. Otherwise I think active returning customers are going to convert. You know they're brand loyal. They've already bought into the brand. So there's no point in you know going back and reinvesting in them. Like let marketing and organic social do that job for you.
Escúchalo en el minuto 36:29
El 43% de los minoristas usa el descuento de hasta un X por ciento
Oferta y precioCuando el cliente pregunta que tipo de promocion hacer, el estudio de IMRG ya lo contesto: la gran mayoria de minoristas se queda con el descuento porcentual de hasta un tanto por ciento, un 43%, con mucha diferencia sobre el resto de formatos. Si todos lo usan, por algo sera. Ademas se puede mirar la biblioteca de anuncios cambiando las fechas para ver que corrio la competencia en anos anteriores.
Cool. Oh yes. And this was also in this IMRG study as well. So, if your client is coming to you and they're asking you, "What should we use? Should I do I don't know, like um from a percentage off or um maybe it's a set like 20 off or something like that." Well, they've done the research. The vast majority of retailers are going to stick with it up to percentage discount off. So, that research has been done for us. If everyone's using it, there must be a reason. It works well. So 43% and that was the um the by far the biggest margin. We can also use our um ad library tool as well and you can…
Escúchalo en el minuto 19:29
Canales de adquisiciónLas consultas de Black Friday, Cyber Monday, ofertas y descuentos arrancan ya en octubre, asi que hay que estar visible antes. Un estudio amplio de IMRG entre minoristas del Reino Unido encontro que los que alargan la campana ganan mas dinero, un 81% incremental segun la cifra que muestra la ponente, y eso ya descontando el coste de mantener las promociones mas tiempo.
into the mood for it. So, queries for Black Friday, Cyber Monday, deals, discounts, these start as early as October. So, therefore, it's really important for us to be getting ourselves ready ahead of this and to be there and to add to that um a really nice study done by IMRG. retailers in the UK and what they actually found is that those retailers that have a longer campaign. So this is actually making more money. So there's a reason why retailers are going early with their campaigns. It's because they make more money. And of course, they've taken into consideration the cost of So, this is…
Escúchalo en el minuto 6:04
Los CPC subieron un 20% con la demanda plana, y la conversacion con el cliente cambio
Canales de adquisiciónEn un minorista de electrodomesticos del Reino Unido que invierte unas 100.000 libras en PMax, desde marzo los CPC llegaron a estar un 20% mas caros mientras el precio, las estadisticas de subasta y todo lo demas seguia igual y nadie habia salido de la puja; con margenes tan estrechos, el golpe a la rentabilidad fue enorme. Los datos de Google para ese sector en el Reino Unido mostraban demanda plana o ligeramente a…
This, can you see that? Um it's basically a graph um year to date of CPCs for one of our electrical retailer in the UK. They spend about 100k uh on Google ads on PMAX. So quite a decent So quite quite decent, um uh player. Um this is what they saw. This is what we saw saw from March onwards. Now everything kind of from price point auction inside everything kind of was quite similar year on year and month after month. No one dropped down massively from the auction. um in some cases PMAX like those CDCs were 20% more expensive and because margins are so tight within that um within that vertical…
Escúchalo en el minuto 1:00:14
Reposicionar el buscador pagado en la era del cero clic
Canales de adquisiciónPPC siempre se ha medido por conversiones, fidelidad y retorno, y eso sigue valiendo, pero el recorrido del cliente es mas desordenado que nunca y estamos entrando en la era del cero clic, donde Google y las demas plataformas quieren retener al usuario en su casa en vez de mandarlo a tu web; ya se ve caer el trafico incluso en marca. Por eso el ponente propone reposicionar el papel del canal: ademas de convertir…
But PMAX, I'm I'm sure we've talked about that for so long, but Pmax alone isn't isn't the solution in my in my opinion. So, we need to drive conversions, we need to drive um sales, of course, and PPC is really good at that. This is where PPC historically has always sit driving conversions, loyalty, return on ad spend. I don't disagree with this. We're still able to drive those conversions, but now consumer journey is even messier than it used to be. We For me, it's about repositioning the role of the channel. Yes, it's a by driving conversion, but ultimately it's about the intent. It's about…
Escúchalo en el minuto 1:07:54
Puja por beneficio en un catalogo de 20.000 referencias
Canales de adquisiciónEn Travis Perkins, con unas 20.000 referencias cuyo margen iba del 10% al 85% y miles de libras al mes en PMax y shopping, el ponente estuvo tres anos pujando por beneficio: como el objetivo del negocio era ganar mas margen, priorizaron los productos de margen alto y despriorizaron los de margen bajo. Con esa dispersion de margenes era una decision evidente.
example, if you're trying to increase profitability, then like optimizing towards gross margin and bidding towards profit is a massive win. Um like I did this with Travis Perkins for three years. um they're got like I think 20,000 SKs in their inventory uh all ranging from like 10% to 85% margin. So it was a no-brainer for us. We're like yeah like business wants to increase profitability. Um we've got you know thousands of pounds to spend a month on PMAX and shopping. Let's do profit bidding and let's prioritize the product which have a higher margin and dep prioritize the product should have…
Escúchalo en el minuto 47:01
Si todos usamos la misma herramienta, la ventaja esta en los datos
Canales de adquisiciónSi todo el mundo usa la misma tecnologia, el mismo tipo de campana y hace lo mismo, al final nadie tiene ventaja competitiva. La IA cambia el trabajo, pero para eso: para dedicar mas tiempo a aportar valor estrategico al cliente. Hace diez anos el trabajo era montar grupos de anuncios y poco mas; ahora, usando todos la misma inteligencia de Google, lo que distingue es con que datos alimentas al algoritmo.
done right that's what we do um but the reality is that if we all using the same technology we all using the same campaign type we all doing the same thing at the end of the day we're left with no competitive advantage so that's kind of the way I look at it so um yes AI is revolutionizing the type of work that we do and how we do it but ultimately this is only so that we and spend more time adding strategic value to to our clients. I remember like when I started 10 years ago, it was all about building out groups, siting and that's literally all I did. No conversation best manipulate those AI…
Escúchalo en el minuto 54:30
Que el Black Friday caiga antes o despues del dia de pago mueve la venta
Canales de adquisiciónEn 2023 el Black Friday cayo antes del dia de pago y eso empujo a mucha gente a buscar y comprar el Cyber Monday. Este ano cae el 28, ya cobrada la nomina, lo que anticipa un ano fuerte y con una curva parecida a la del ano pasado, asi que los datos del ano anterior sirven de guia. Ademas el Cyber Monday se va a diciembre, y eso hay que tenerlo en cuenta al repartir el presupuesto.
headline acts of Black Friday. This year falling on the 28th which is after payday. And if anyone remembers 2023, Black Friday came after um before payday. So it pushed a lot of people to actually search and buy under Cyber Monday. This year we've got our Black Friday after everyone's been paid. So that really tells us that we're in for a bumper year, the same as what we saw last year. So we've got a very similar trend as last year. So that's really great for us. That means that we can use our data from last year to really help and guide us. And then we have got Cyber Monday being pushed into…
Escúchalo en el minuto 8:52
Google apunta al domingo posterior al Black Friday como el dia de mas volumen
Canales de adquisiciónGoogle estima que el dia con mas volumen de busqueda de todo el periodo no sera ni el Black Friday ni el Cyber Monday, sino el domingo intermedio. Pero estan hablando de volumen, no de ingresos: puede ser gente que ya compro y esta comprobando si consiguio el mejor precio, asi que hay que contar con una tasa de conversion distinta ese dia y hasta protegerse de el.
um our spends. Now, we've got a big star over the Sunday after Black Friday, and that is because Google themselves are um estimating that this is going to be the day with the most amount of search volume out of the whole of the periods. Not Black Friday, not Cyber Monday, the Sunday after. Now, of course, they're thinking about search volume. They're not thinking about revenue. So we still need to think about different CVRs during this time, but that is the day that they are pinpointing to have the most amount of volume. So it could be the case that we need to protect ourselves from that…
Escúchalo en el minuto 9:36
30 peniques de cada libra online en Reino Unido son de Amazon, pero el 70% compraria una marca nueva
Investigación de mercado y clienteDos cifras para cerrar el panorama: treinta de cada cien peniques que se gastan online en Reino Unido van a Amazon, asi que el rival real no son los otros anunciantes. Y en el otro lado, el 70% de los compradores esta dispuesto a comprar marcas que no conocia: la mayoria no busca una marca concreta, busca una cosa, asi que hay partido.
So couple of things to close with. 30p of every pound spent online in the UK goes to Amazon. It's a lot, right? That's even before we start. So, let's not make it easy for them. Let's give them a fight. And they're not even so much on Google anymore. So, let's do it. Um, but yeah, that's that's what we're up against. We're not even up against each other. We're up against Amazon. But hopefully Um, 70% of shoppers are open to purchasing brands that they've not heard of before. So, we've talked a lot about like branded search, branded Pax. Most people are not searching for a particular brand…
Escúchalo en el minuto 28:49
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.