Florin Armasu lleva 15 años construyendo infraestructura de email marketing para marcas como Nestlé y Brown-Forman. Es el CEO y fundador de Data Innovation, con sede en Barcelona, y el creador de Sendability, una plataforma agentic de email marketing y CRM construida sobre software open source. En esta conversación con Sophie Steffen para The Open Rate Club, Armasu recorre pieza a pieza el stack que utiliza en producción: mail transfer agents open source que llevan funcionando desde 1998, un sistema de monitorización de entregabilidad que consolida datos de siete años, y las lecciones concretas que solo se aprenden cuando una marca de gran consumo se bloquea accidentalmente en Hotmail y contempla un agujero potencial de cinco millones de euros.
Entrevista Expert Insight


Lo que sigue es un recorrido por la maquinaria del email marketing tal como se practica realmente: qué herramientas sobreviven al contacto con la producción, por qué la limpieza de listas no es la cruzada moral que algunos pretenden, dónde se esconden los fallos de entregabilidad a plena vista, y qué métricas merecen atención frente a las que simplemente son Apple Mail Privacy Protection mintiéndote con total naturalidad.
Conclusiones clave
Conclusiones clave
- El stack open source: Mautic para orquestación de campañas y KumoMTA para la entrega, combinados con Tableau para analítica cross-ESP, forman una infraestructura de nivel producción que compite con plataformas enterprise a una fracción del coste.
- Entregabilidad rentable: Tratar la entregabilidad como puro cumplimiento normativo lleva a enviar cada vez menos emails hasta que el canal muere. El objetivo práctico es maximizar ingresos gestionando la reputación, no eliminar todo riesgo.
- Fallos silenciosos: Los problemas de entregabilidad son a menudo invisibles hasta que resultan catastróficos, porque a diferencia del posicionamiento en buscadores, no hay un momento obvio de “página cinco”: simplemente dejas de llegar al 35% de tu base de datos.
- La métrica que importa: La tasa de crecimiento de base de datos activa – el porcentaje de tu lista total que interactúa con tu contenido a lo largo del tiempo – es el indicador más revelador de la salud de un programa de email.
El stack de email open source que impulsa campañas reales
La recomendación habitual en email marketing asume que utilizas uno de seis ESPs comerciales y que tu única decisión técnica consiste en elegir entre ellos. El stack de Armasu funciona de otra manera. El front end ejecuta Mautic, una plataforma de automatización de marketing open source comparable a Iterable o Braze, que gestiona la orquestación de campañas y los recorridos del cliente. El pipeline de entrega ejecuta KumoMTA, un MTA open source construido por antiguos ingenieros de SparkPost, junto con SparkPost, Amazon SES y Qmail, un software creado en 1998 que Armasu describe como “un dinosaurio en plataformas de software”. Sigue en producción. Sigue entregando emails.
La capa de analítica ejecuta Tableau, porque Armasu lleva siete años extrayendo datos de todos los ESPs que sus clientes han utilizado y consolidándolos en una vista única que permite comparar entregabilidad en Gmail, Hotmail y Yahoo a nivel de newsletter, durante años, muy por encima de la política de retención de datos de dos años que la mayoría de plataformas aplican. La capa de verificación agrega resultados de Webula, BrightVerify, IP Quality Score y Email List Verify, más siete años de datos históricos propios. Y la creación de contenido pasa por Brand Expand, su propia herramienta para automatizar la distribución de contenido en email, web y redes sociales.
“Intenta integrar tus herramientas, las seis, siete, ocho, cuatro herramientas, las que sean, en una sola. Es cada vez más fácil hacerlo con agentes de IA porque te puede dar más control y más flexibilidad si tienes todo incorporado en un solo dashboard.”
Su recomendación experta para herramientas de entregabilidad es Inbox Analyst, un sistema de monitorización basado en paneles utilizado por Cheetah Digital, Oracle y Adobe que rastrea dónde aterrizan realmente los emails usando usuarios reales en lugar de cuentas semilla. La mayoría de herramientas de entregabilidad dependen de seeds, que no pueden replicar las condiciones de un usuario real ignorando un email, moviéndolo a spam o simplemente sin importarle. Inbox Analyst utiliza un panel de usuarios reales, lo que hace sus datos considerablemente más valiosos.
Lee nuestro análisis completo sobre el stack de email open source
Higiene de listas de email y el dilema de la lista gris
La recomendación estándar para la construcción de listas es el doble opt-in. Es seguro, es conforme, y tiene un modo de fallo del que casi nadie habla: si tu reputación de remitente ya está dañada, el propio email de confirmación aterriza en spam, y pierdes suscriptores antes de que confirmen. La alternativa – opt-in simple con verificación de email y campos honeypot para atrapar bots – funciona, pero requiere un nivel de pragmatismo que incomoda visiblemente a los puristas del cumplimiento.
La posición de Armasu sobre la limpieza de listas es clara. La mayoría de herramientas de verificación clasifican las direcciones en tres categorías: válidas, inválidas y una zona gris de no-respuestas y dominios catch-all. La recomendación estándar es purgar agresivamente. El argumento de Armasu: la adquisición de usuarios es cara, y descartar el 80% de tu base de datos porque una herramienta de verificación devolvió resultados ambiguos no es buena práctica. Es desperdicio. El enfoque práctico consiste en separar lo claramente inválido de lo incierto, probar lo incierto con envíos reales, y dejar que los proveedores de correo te digan qué direcciones rebotan. Un estudio de Boston Consulting Group identifica la mala higiene de listas como uno de los riesgos de entregabilidad más significativos, pero la solución es limpieza reflexiva, no tierra quemada.
“No te obsesiones con ello. No elimines a todo el mundo. Date una oportunidad porque invertiste en adquirir a ese usuario.”
El problema de la lista gris es particularmente agudo con los dominios catch-all, direcciones donde el servidor acepta correo independientemente del valor de la parte izquierda. Verificas sophie@yahoo, sophiesteffen@yahoo y sophiesteffen2048@yahoo, y las tres devuelven “okay for all”. La herramienta de verificación no te ha dicho nada útil. La única forma de resolver estas direcciones es enviarles y monitorizar los rebotes, lo que requiere una monitorización a nivel de infraestructura que la mayoría de marketers no tienen.
Lee nuestro análisis completo sobre higiene de listas de email
Fallos de entregabilidad silenciosos que cuestan millones
Los problemas de entregabilidad no son como las caídas de posicionamiento en buscadores. Cuando tu página cae de la posición uno a la cinco, lo notas. Cuando el 35% de tu base de datos deja silenciosamente de recibir tus mensajes porque estás en la lista negra de Hotmail, puedes no darte cuenta durante meses, porque tus números de Gmail están bien, tus números de Yahoo están bien, y nadie pensó en comprobar Hotmail por separado.
Una marca de gran consumo con la que trabaja Armasu en Barcelona lanzó una campaña de reactivación a través de Ogilvy. El email era bueno. La estrategia no: lo enviaron a toda la base de datos, incluido un segmento masivo de destinatarios no comprometidos. El resultado fue bloqueo en Hotmail, que representaba el 35% de su base de datos. Si el bloqueo hubiera persistido un año, Armasu estima el impacto en ingresos en cinco millones de euros. Fueron contratados para arreglarlo. Llevan seis años trabajando juntos.
El problema de las IPs compartidas es igualmente insidioso. Una pequeña empresa – una plataforma de psicólogos – estaba compartiendo sus IPs de envío con 2.000 remitentes. Si alguno de esos remitentes se comporta mal – y con 2.000, siempre hay alguno – el problema del vecino malo arrastra la reputación de todos.
“La entregabilidad a veces se trata como una responsabilidad de – trabajas con Salesforce, por tanto tienes muy buena entregabilidad. Y te puedo decir que no es así.”
Apple Mail Privacy Protection, desde octubre de 2022, pre-cachea cada email abierto en un dispositivo Apple, generando un evento de apertura fantasma independientemente de si el destinatario realmente miró el mensaje. Las empresas informan tasas de apertura medias del 60% y celebran, sin ser conscientes de que un porcentaje significativo de esas “aperturas” son los servidores de Apple realizando el equivalente digital de marcar cada carta como “leída” antes de que el cartero haya llamado.
Lee nuestro análisis completo sobre los fallos silenciosos de entregabilidad
Métricas de email más allá de las tasas de apertura
Las tasas de apertura, para quien todavía las siga con cara seria, son ahora tan fiables como un pronóstico del tiempo escrito por alguien que nunca ha salido a la calle. Apple Mail Privacy Protection las infló. El pre-fetching de bots las infló. Las tasas de click-through se resintieron a su vez, porque el denominador – las aperturas – es ahora ficción, lo que convierte cualquier ratio construido sobre él en igualmente ficticio.
Los ingresos por email enviado es la primera métrica que Armasu realmente valora. Muchas empresas rastrean el lado de costes – pagas por contacto almacenado o por volumen enviado – pero nunca conectan envíos individuales con eventos de ingresos. El problema de atribución, reconoce, nunca se ha resuelto correctamente. Pero no rastrearlo en absoluto es peor que rastrearlo de forma imperfecta.
La métrica que Armasu considera la más importante es la tasa de crecimiento de base de datos activa: ¿el porcentaje de tu base de datos total que interactúa con tu contenido está creciendo o reduciéndose a lo largo del tiempo? Si el círculo de consumidores que interactúan con tus emails se contrae, tienes un problema de entregabilidad o un problema de relevancia, posiblemente ambos. Si se expande, tu contenido está llegando a nuevos segmentos y tu reputación es lo bastante sana para soportar la expansión.
“Si tu base de datos está decreciendo, eso significa que eres demasiado conservador. No intentas reactivar a tantos usuarios, y eres muy preciso. Entregas perfectamente una cantidad muy pequeña de emails, pero eventualmente tu base de datos va a llegar a cero.”
La tasa de quejas necesita medirse por proveedor de correo, no en agregado, porque Gmail solo informa de quejas a nivel de campaña: sabes que la campaña generó quejas, pero no puedes identificar qué usuarios individuales se quejaron. Yahoo y Hotmail te dicen quién se quejó, así que puedes suprimirlos inmediatamente. Gmail no, lo que significa que tus opciones de supresión se limitan a ajustes estratégicos a nivel de campaña.
Lee nuestro análisis completo sobre métricas de email más allá de las tasas de apertura
IA, trampas de precios y la cuestión de la infraestructura
Armasu utiliza la IA como la mayoría de profesionales experimentados: para generación de contenido, variaciones de líneas de asunto y como sparring estratégico. Construye una estrategia, se la da a un LLM y le pide que la cuestione. La respuesta es útil, pero tiene límites claros. Cuando le pidió que generara una estrategia de entregabilidad integral a partir de 750.000 páginas de documentación, la respuesta volvió como un manual de cumplimiento: no hagas esto, no hagas aquello, ciñete a los usuarios más activos. El punto final lógico del cumplimiento perfecto es enviar cero emails. Cero emails significan cero quejas. También significan cero ingresos.
El lado negativo de la IA en email es la avalancha de cold email. El spam generado por IA está aumentando de forma pronunciada, señalado y condenado por M3AAWG (el Messaging, Malware and Mobile Anti-Abuse Working Group) pero sin signos de ralentización. Armasu recibe estos emails a diario. La mayoría son gramaticalmente perfectos, temáticamente irrelevantes, y creados por alguien que confundió “personalización a escala” con “acoso automatizado”.
En cuanto a precios, la trampa oculta es la facturación por contacto. HubSpot, señala Armasu, es excelente con 1.000 contactos. Con 50.000, cuesta 3.000 dolares al mes, y estás pagando por cada suscriptor no comprometido que permanece en tu base de datos: contactos a los que no envías emails pero que no puedes eliminar porque podrías resubscribir accidentalmente a alguien que se dio de baja. Acabas pagando una prima mensual por el privilegio de almacenar un activo que no estás utilizando.
Y luego está la geo-patriación, un concepto de un evento reciente de Gartner en Barcelona. El 70% de las organizaciones de EMEA están diseñando activamente para acercar sus stacks tecnológicos a sus oficinas centrales, impulsadas por la incertidumbre geopolítica y las preocupaciones sobre soberanía de datos. Para la infraestructura de email específicamente, esto significa alojar tu MTA, tus datos y tus IPs de envío en una jurisdicción que controles. Armasu ofrece esto como servicio: dile dónde quieres la infraestructura y él la instala allí.
Transcripción completa de la entrevista
Leer la transcripción completa de la entrevista
Sophie Steffen: In today’s Let’s Marketing talk, we’re going to talk about the real behind-the-scenes workflow of email marketing, from the tools and messy execution to deliverability challenges and metrics that actually matter, and lessons learned the hard way.
Sophie Steffen: Welcome, Florin, to this chat. Would you like to give us a quick intro about yourself and also what do you do?
Florin Armasu: Hi, Sophie. Nice to reconnect with you. I’m Florin Armasu. I’m the CEO and founder of Data Innovation. We’re based in Barcelona, Spain. We have built Sendability, an agentic email marketing and CRM platform.
Florin Armasu: I was managing that team at emailing network in the past, part of Reworld Media, an email marketing agency. We’re agentic in this space. It’s an overused word today, I suppose. Everybody does AI agentic and stuff, but we really mean it in the sense of we’re preparing for this change in technology, in software especially, that has two categories of users.
Florin Armasu: One is real human, and another one that consumes the platform is an agent. So we’ve seen many different infrastructures for sending emails and sending commercial and content emails and managing these customer journeys and everything, and we wanted to build our own stack, especially because we think that nowadays it’s more and more possible with the open source platforms that are available.
Florin Armasu: And there’s so many tendencies on doing so. Like, we’re gonna speak later about a concept that I just learned last week from a Gartner event called geo-patriation. That means bringing your own stack closer to where the company is, and it makes sense in this context of the world.
Sophie Steffen: Yeah. We’ll talk later about this, which is very interesting. I know we had a chat before this recording about it, so I’m excited to move to that question. But before, let’s start with the stack reality. So again, thanks, Florin, for joining this chat. And let’s skip the marketing theory and dive directly into your stack, the tools that you’re using.
Sophie Steffen: So could you tell me what specific email service provider, ESP, and other supporting tools you had open yesterday on your screen when you were getting a campaign out of the door?
Florin Armasu: Yeah. So yesterday, mostly nowadays I’m using our stack, but in the past we have – I’ve been working with Salesforce Marketing Cloud, I’ve been working with Klaviyo, I’ve been working with Cheetah Digital, and I manage clients that have been in many different environments.
Florin Armasu: So we’ve been working with Nestle, we’re working with Brown-Forman, we’re working with agencies that are spread among very multiple tools, and they are also using our stack to manage the email marketing and deliverability, and to balance between what they send and how much profit is this making for their company.
Florin Armasu: So, a concept that we are calling profitable deliverability. But speaking about the stack, we’re mostly using Mautic. It’s an open source marketing automation as a front that would be comparable to Iterable, to Braze, to the front, the campaign orchestrators that many people know and use. On the back side, on the delivery pipeline, the mail transfer agent, we’re using KumoMTA, which is again an open source MTA that was delivered by some, a team that was part of SparkPost many years ago.
Florin Armasu: We also use SparkPost. We also use Amazon SES and Qmail, which is the oldest MTA that was created in ‘98. It’s like a dinosaur in software platforms. We’re having another platform called Volume Deliverability Manager Suite that actually does the orchestration of how to – what’s the safe volume to send on each mailbox provider.
Florin Armasu: Depending on your deliverability, on your reputation at Gmail, at Hotmail, you cannot always impact the totality of the database that you have. So you need to select from the users that you have, the ones that are more active or less active or more engaged or less engaged.
Florin Armasu: It’s simply impossible, as it is in social networks or in some other channels, to reach 100% of the number of subscribers if you don’t have a good reputation.
Florin Armasu: That’s the work of deliverability. Yeah. And we have more layers. We use Tableau as a tool to do our best analytics.
Florin Armasu: Tableau is an application part of Salesforce, but it’s the leader of visualization and business intelligence. We have been extracting for all our clients the data from their ESPs or from their infrastructures with the same idea in mind. So if your data gets locked in a vendor and you have a contract that has a data retention policy or it’s subject to some restrictions, you are basically left incapacitated to work with your own data, with your own sense.
Florin Armasu: Or maybe you want to see how the sends were working three, four, five, seven years ago. The normal data retention policy is two years. Maybe you want to see it top to bottom or at a newsletter level by mailbox provider, but you also want to see multiple ESPs. That’s why we use a different – we use Tableau.
Florin Armasu: We use email verification services. We have been checking all of them, like from Webula, which is Oracle, BrightVerify, which is Validity, and some other smaller, IP Quality Score, Email List Verify. But then we developed our own stack, which would be an aggregator, and all the data that we gathered in the last seven years.
Florin Armasu: I’m using also an application that we have created to automate the content creation and distribution, which is called Brand Expand.
Florin Armasu: And that’s about it. I think what – if I could make a recommendation here to the listeners or to the viewers – try to integrate your tools, the six, seven, eight, four tools, whatever number the tools you are using, into one.
Florin Armasu: And it’s easier and easier to do with AI agents because it could give you more control and more flexibility if you have all everything incorporated into one dashboard.
Sophie Steffen: Well, that’s a good recommendation, and I mean, you mentioned quite a lot of tools here. Do you have any favorite tool, like an expert pick that you would recommend to our listeners?
Florin Armasu: Well, for deliverability specifically, I think one of the best tools is Inbox Analyst, and it’s also incorporated, like Cheetah Digital uses it. I think even Oracle and Adobe now have it. The other day, I was seeing a situation analysis on Massimo Dutti, and they were also having Inbox Analyst.
Florin Armasu: So it’s a big tool. It’s a big panel of users. Yeah. And it basically allows you to see where your emails are going, and not only with seeds, because most of the tools that are out there are based on seeds, like Email Console, GlockApps. Most of the deliverability tools that are available are based on seeds, and a seed cannot replicate the real conditions.
Florin Armasu: Like a user going to the email, ignoring one email, moving one email to spam. It’s not really easy to simulate it. And Inbox Analyst has a panel of real users that allow them to see where the email lands, so that’s valuable insights.
Sophie Steffen: Yeah. Okay, thanks for sharing that. And let’s talk about what happens before sending out emails.
Sophie Steffen: So you have to build lists and, what is the actual tool workflow for cleaning and verifying email data so that you don’t destroy your sender reputation?
Florin Armasu: Well, that’s a little funny because we’re moving – I mean, the standard recommendation is to do double opt-in, and it’s fine. I think it’s a safe bet to do it.
Florin Armasu: It’s been a standard recommendation for a long time now, but it has like a point where you can fail, you know? So if you have a bad reputation, the confirmation email will gonna land into spam. So you’re gonna lose a lot of subscribers at the beginning when you are doing the confirmation of the registration. So you need to be aware of that step ‘cause you could lose a lot of initial subscribers in that step. And there’s also the alternative to just do a single opt-in and verify the email that the user is inputting, verify that it’s not a bot, because that’s very common nowadays.
Florin Armasu: And that can be done with, it’s called a honeypot. So it’s some lines in the form that are not visible to a human, but they are visible to a bot. So the bot fills this, and you know that it’s not a consumer. It’s a bot, so you just discard this subscription.
Florin Armasu: And it’s also good if you are having consumers in your database that were not sent in a long time, like more than six months. Before starting to send to them again, it’s a good practice to clean them again, to clean the list using a data verification tool ‘cause there might be a lot of spam traps and other kind of unuseful or doubtful addresses inside.
Florin Armasu: Do you have any case that you remember of data validation practices that went wrong, Sophie?
Sophie Steffen: Well, a lot actually. So usually what I feel is the biggest pain point is that data cleaning so that the actual database that you have is managed. So that’s something, I mean, multiple clients manage it different ways.
Sophie Steffen: Do you have any recommendation here what’s the best practice for this cleanup, you know, for cleaning out that the actual database is cleaned? Or do you have like an internal tool or external tools? You mentioned that you use this data cleaning tool but, you know, overall to prepare lists, is there a go-to tool that you would recommend?
Florin Armasu: I mean, you always, in my opinion, you need to apply a practicality to this because there’s – and we’re gonna go into that a little bit later. I think if we speak about more in-depth deliverability, I think there’s a trend to consider deliverability as overlapping 100% with compliance and it’s a tendency to think of it like only compliance. And the problem with this idea is that you don’t think in practical terms. Now, if I invest a certain amount of cents in, or dollars in acquiring a new user, how much dollars will I expect from this user to bring me into the business?
Florin Armasu: And then we can be very lax about all these processes and say, “Okay, just discard 80% of the users, it’s fine.”
Florin Armasu: It’s not fine ‘cause user acquisition is very costly. So it’s better to – of course, you should try. Most of the tools classify the cleaning in three segments, which is valid, invalid, and in between called gray list. So in between is practically a non-response. There are domains that are configured to accept emails from any type of – I mean, with any type of left part, so it’s called okay for all. So even if I would choose to check sophie@yahoo, for example, or sophiesteffen@yahoo or sophiesteffen2048@yahoo. In all three cases, I will get the same answer, okay for all. So that’s not really a useful answer.
Florin Armasu: You cannot do anything with that. But then when you send the first time, you differentiate between the bounces. So the ESPs, the mailbox providers are practical, and they’ll tell you, “This recipient does not exist,” so you need to eliminate it fast from your list. What I’m trying to say, there’s cleaning solutions.
Florin Armasu: There are many. Bouncers, Eurobounce, Webula, BrightVerify, whatever. You can use ours. That’s an aggregator of different tools and plus our historical data. Sendability verification. And the important thing is don’t be mad about it. Don’t eliminate everybody. You just give yourself a chance ‘cause you invested in acquiring that user.
Sophie Steffen: Yeah. It’s a fine line with – as you’re saying, it’s how to walk these gray zones between deliverability and the legal aspect of sending out emails. So, yeah.
Florin Armasu: Yeah. And maybe – I don’t want to sound that I’m encouraging people to buy lists or whatever, which is a problem in this space.
Florin Armasu: This is not at all what I mean. I mean, if you have a – but there’s a lot of technical infrastructure problems that could make you lose database, and you cannot lose what you have acquired, what’s your most important asset, very fast without even trying to activate them and to get some revenue.
Sophie Steffen: Yeah. So we’re moving from the list building and list cleaning into the actual process, so the execution and also the failures. So could you walk me through the last deployment pipeline of your latest email campaign, so from the beginning, which tools you touched, and also design or copy, until you hit the send button?
Sophie Steffen: So walk me through this process.
Florin Armasu: All right. Before I jump into this pipeline, I just wanted to say that I remember I read a Boston Consulting Group study recently, and not cleaning the database is considered one of the most important problems of deliverability, because spam traps create a real problem. Like, it’s a risk that you cannot really afford, and that comes from especially not having good user acquisition practices.
Sophie Steffen: Yeah, it starts before actually, right? So actually the performance and the success of an email campaign depends on how you acquire the user.
Sophie Steffen: So once you have acquired the user, yeah, what happens then?
Florin Armasu: So yeah, most of our clients, and I would use an example of a very well-known fast-moving consumer goods company that we work with here in Barcelona. The first thing, the strategy around sending emails has basically three main types of journeys.
Florin Armasu: One is activation. Activation means that you are sending different types of content from a welcome email to different types of interactions with the consumer, where you try to get the interest, to get them interested into your CRM program or email program. And so this is a tier of consumers that you don’t really know about them.
Florin Armasu: You just know that they are new to your program, and you don’t know what they – especially what they like. It’s not really easy to personalize the content except based on geo-demographic data. So you need to think really strategically about this welcome or activation series, how you construct it, how much time you leave between impacts.
Florin Armasu: It is all about less about instinct, more about observable data ‘cause there are series that work better, and you would consider them not so good at first sight. But you’re not the average of your list, so you need to judge based on what the consumers tell you.
Florin Armasu: The second type of customer journey is about users that are very engaged with you. So let’s say the clickers, the openers, the ones that are really interacting with what you are sending. And that should in theory amount for a high percentage of the daily volume that you are sending.
Florin Armasu: And you also in these cases, in these tiers, you need to think about relevance. What are you sending – I mean, the biggest error most companies do is just communicate what they want to sell. And sometimes you need to think on the other end of the line there’s a person, and they are interested in different stuff.
Florin Armasu: You can inform them, you can entertain them, you can engage in so many different ways with them. So it’s very similar to real life. You cannot go on the street to say to somebody, “You know, you want to buy my product?” Then nobody will buy you. You need to first get to know you, why they need the product, how the product is different than other products or services that are available on the market, and that. So this is the phase where it’s the exploitation phase. You have an asset, the active part of the database, and you’re sending emails or doing other types of messages, SMS, WhatsApp, whatever is in your CRM, with the objective of making money.
Florin Armasu: But you don’t need to forget that these consumers also need something in return. So that’s the relevance part. And there’s another part where the interest starts to fade, and that’s the third part of the journey, which is called reactivation or sunset policy, and that applies to the less engaged tiers, the ones that you have tried to activate, you have engaged with at some point maybe, or you have never engaged with, and now you need to try to see if you can recapture their attention, or decide if you want to still send them emails or not. The trick with this part is that, and that’s another deliverability part that nobody is really spoken about, is that the modern spam filters have two layers also.
Florin Armasu: So first layer is do we allow this sender or this IP to send emails to our consumers? And that’s the most common one, the one that we know for everybody from many years ago. There’s a machine learning application, but it’s kind of simpler. And the second layer is, does this email that we are seeing good for Sophie, good for Florin right now, or shall we deliver it into inbox or into spam?
Florin Armasu: And that’s the part where it gets trickier because these consumers that are unengaged, are they really uninterested in your content, or they are not receiving in their inbox? Because if somebody receives in spam, it’s like – it’s the same story, you know. If a tree falls in a forest and nobody hears it, does it make a sound?
Sophie Steffen: So basically coming back to this pipeline, you would first categorize the type of user or the type of campaign into either this activation campaign, the engagement campaign, or reactivation campaign. And could you just circle back for the listeners, you know, like which tools you would touch, from – for example, you can pick one of the segments, say in reactivation campaign, because I feel that is something that a lot of people when they have users in their database, they haven’t been touched for a while, so they want to reactivate them. So what would be the process here? Would you – once you have that segment, until you hit the send button, what needs to happen so that you get to that point of sending that email out to reactivate users?
Florin Armasu: So an important step is, once you know to whom you want to send, the question is what you want to send.
Florin Armasu: One of the common mistakes in content creation is sending the same as always. Because people react very well to novelty, so it’s trying at least to say the same thing in a different form, in a different format.
Florin Armasu: You cannot just send the same newsletters. There’s a catch though, because it’s easier to send emails along with the active users, and it’s more complicated to say, “Look, I’m gonna physically now need to send to actives and inactives at the same time so that the inactive cohort doesn’t get sent directly to spam, and the message should be different.”
Florin Armasu: So it requires a kind of orchestration there that we have developed the platform in order to allow for that. But that’s not always very common. Then, after you have content that is really good for reactivation – it can be maybe even a text-based message.
Florin Armasu: It needs to include persuasion tactics. It needs to include – it needs to be different than your common newsletters that you send. You send it, and you measure. So a very important thing is you monitor and measure. So you monitor deliverability in Google Postmaster Tools, Microsoft SNDS, at least the big two.
Florin Armasu: Which a month here in Spain accounts for 80% of the consumers. In other countries it’s less, but it’s still a significant quantity, more than 50%. So you need to at least measure this. And you need to see if you get complaints. You need to see if you get bounces.
Florin Armasu: You need to see if you get responses. And responses – it’s like the platinum of all user interactions. It’s the most important one.
Sophie Steffen: Unless it’s a complaint, right?
Florin Armasu: Yes. Yeah. Well, I mean, a complaint, touching the spam button is kind of strange.
Florin Armasu: Because Gmail doesn’t tell you that a user has clicked on the spam button unless you have a good reputation, and it comes with spam and unsubscribe, so you get the unsubscription back. Yahoo and Hotmail, they tell you, they inform you. The user has made a complaint, so you need to take it off the list.
Florin Armasu: And Gmail, you cannot take them off the list. You just see a campaign has had a higher percentage of complaints, but you don’t know which were the consumers that complained.
Sophie Steffen: And within this workflow from identifying the segment you want to send an email to, creating the content, being persuasive and creative about the actual copy and design you’re using, until you hit the send button, do you find yourself specific bottlenecks that keep coming back at the same point in the funnel?
Florin Armasu: Yeah. Well, my biggest struggle in the 15 years that I’ve been doing email is the content creation, to be honest. I mean, infrastructures, the platforms are really built to help at any stage of this funnel, of this pipeline. If you do a little bit of integration, you have all the data where you want it.
Florin Armasu: But content is still hard. It’s becoming easier now with all these AI tools and all these AI content creation platforms. Still email – it doesn’t have a unified language. So email is HTML, but the HTML is displayed differently in different providers.
Florin Armasu: We used to have a very big provider called Litmus that was guaranteeing that the email is more or less visualized the same in all types of browser plus device that they use. And also it came out now with the dark mode, there’s so many different possibilities that the logo doesn’t appear well on top of a dark background or because it doesn’t have an outline and stuff like that, and it gets the same color as the background.
Florin Armasu: So for me, the content creation is where companies need more help. They usually work with external providers. But the good thing is that it’s becoming easier and easier to use agents and to create a quality HTML that gets delivered well in all parts of infrastructures. And we have our tool called Brand Expand that creates automatic contents that can be published on the web, on the social networks, LinkedIn or Instagram, or inside the sent by emails.
Sophie Steffen: Yeah. Okay, perfect. And moving into the failure section, do you have any story about an email automation migration or campaign that completely went wrong? And in case so, what broke, and how did you fix it?
Florin Armasu: I mean, I have a real story that affected a lot of consumers, again, from the fast-moving consumer goods company that we work with here in Spain.
Florin Armasu: And they were – they did a reactivation campaign through their creative agency, Ogilvy, and they just sent it to everybody ‘cause they wanted to – the email was good, so they thought it’s gonna be good for everybody to receive it. That’s not a very good practice because there’s a guardian between the brand and the consumer, which is the spam filter.
Florin Armasu: So sending to too many inactive or unengaged recipients at once will only guarantee or damage your reputation, and that’s what they did. So they basically got blocklisted at Hotmail, and that represented 35% of their database. And if they would have continued to be blocklisted after a while, it could have been a big problem for them.
Florin Armasu: Like a year of blocklist at Hotmail would have meant at least 5 million euros in terms of money. So we’re talking a lot of money. And that’s actually what I’m talking always about deliverability. There are sometimes these failures are silent, so you don’t really realize that you have a problem at a mailbox provider level, and you only if you look in comparison with the other ones. So compare what’s the open rate, what’s the click rate, what’s the bounce rate, what’s all the main KPIs at Gmail, at Hotmail, at Yahoo, at all the other ESPs, you can realize that you have real problems in one of them. So we started working with this company because of this problem, and we’ve been working for the last six years.
Florin Armasu: Then there’s more problems to solve. There’s always more problems to solve. One other common problem that I see all over, and that’s really sad sometimes when it happens. There’s a lot of platforms that offer shared IPs, which means that the sending IPs are shared between many senders.
Florin Armasu: I saw a company, small company that was a psychologist’s platform. So they had been sharing their sending IPs with 2,000 other senders. If some of these senders are misbehaving, which is called a bad neighbor problem, you can do everything perfectly and you have bad deliverability because your sender reputation is good, but the sender reputation is constructed with the sending domain and the IP.
Florin Armasu: So if the IPs have bad reputation, part of your emails go to spam. So that’s a little bit my story. If you want another one, the Apple Mail Privacy Protection, MPP – that was launched in October ‘22 – created a lot of problems for a lot of marketers because it inhibits the possibility of doing send time optimization and stuff like that.
Florin Armasu: Basically because every email you send, and the user has an Apple device, an iPhone or iOS, and they consult the email on that, the email is pre-cached by Apple. So when you send an email, you automatically get back an open. That drives the open rate in iCloud or some other domains to 100%, but it does not only affect iCloud.
Florin Armasu: It affects Gmail, Hotmail, all the other, if you are reading them on an Apple device. So that creates a problem. Open rates have raised by a lot. Like, there are companies that now are at 60% open rate on average, and it’s not real.
Sophie Steffen: Yeah. I mean, metrics obviously depend a lot on how you read them and what’s behind the metrics.
Sophie Steffen: So moving into that part and also deliverability, I mean, that’s like the big topic, I feel, of this conversation, which keeps getting harder. Also looking into the recent updates of Google and Yahoo of inbox protections. So do you have any obscure settings, specific settings that you can monitor, or that you ensure, like tools to make sure that they actually hit the inbox and not end up in spam folder?
Florin Armasu: I mean, I already recommended the Google Postmaster and Microsoft SNDS, and I think that you – it’s very important that you use these two tools that are official from Microsoft and Hotmail to get an idea about where you are.
Florin Armasu: And then you can use any other tool, from Sendability to Inbox Analyst to whatever other tools are there, Lookups and whatever. But in the end, delivering emails is like a multi-variable problem. So how much can you influence the positive signals, how many positive signals you can give a mailbox provider, and how few negative ones you show.
Florin Armasu: So that’s where you have some degree of control. If you are diminishing the negative KPIs, the negative signals, you’re gonna improve your reputation, and you’re gonna do more business with your emails. And I wanted to touch just a little bit on the signals.
Florin Armasu: For many years, I was focusing on clicks, which is a good level of engagement. So if a user is engaged, they click in your emails. But recently I realized that I’d somehow forgotten – I mean, that was really done a lot in the past, this asking the user to add you to their address book.
Florin Armasu: This is such an old thing in a way but that sometimes you say, “No, look, this is unnecessary.” But it’s a positive signal. So you have a consumer that has you, a brand, sending them emails in their address book, that for the spam filter is a positive signal.
Florin Armasu: And the most positive signal of all is the reply. And imagine – I mean, probably you’ve seen so many brands that have no-reply. There’s many brands that still have no-reply at something.
Sophie Steffen: Yeah, yeah, yeah.
Florin Armasu: And that’s encouraging consumers not to reply to you and losing this advantage of having a positive signal.
Sophie Steffen: Okay, so that would be the two main metrics that you actually care about also within this Apple Mail Privacy Protection, where you don’t have so much reliability on open rates. Would those two metrics be the ones that you care the most about?
Florin Armasu: Yeah. Yes, these are two metrics that I care a lot.
Florin Armasu: They are not so much of a vanity metric like open rates. I also – we also do care about some metrics that we can measure, but – the metric would be getting out of spam. It’s not easy to measure how many people just took the message out of spam, but you can infer it from having a consumer not interacting with your emails and then suddenly they start engaging again. There’s other metrics that are really important, and I was thinking about this a lot, like revenue per email sent, you know?
Florin Armasu: There’s many companies that don’t track that. So they track the cost ‘cause you sometimes pay per contacts stored or sometimes pay by volume. But you don’t track the revenue that each email that you send is bringing to your company. I would also consider a very important KPI, and it’s actually part of our health score, is the active database growth rate.
Florin Armasu: So very similar to monthly active users. Did your overall reach increase or decrease by sending this campaign?
Florin Armasu: If your database is decreasing, that means that you are overly conservative. You don’t try to reactivate so many users, and you are very precise. You send, you deliver perfectly a very small quantity of emails, but eventually your database will gonna go to zero.
Florin Armasu: So it’s not a case. Complaint rate, I told you, you may need to measure it by email provider ‘cause Gmail only communicates that at campaign level. So you have in Postmaster Tools this KPI is shown there. And it’s only showing the total sender complaint rate and by campaign. And then, click-through rate sometimes – it also has decreased lately because the denominator increased. So you think about open rates, open rates are highly skewed right now. It’s not – they are not consistent.
Florin Armasu: It’s a lot of non-human interactions, like fake interactions, pre-cached interactions. So the click-through rates decreased. It’s better to look at the click rate by itself.
Florin Armasu: And we have created a health score, and some platforms also do have it, like SparkPost, and that’s an aggregate of – you give a certain score to click rate, a certain score to open rate, a certain score to bounce rate, a certain score to all these KPIs, and then you have an aggregated value that shows you in one number where you are in terms of reputation.
Sophie Steffen: Okay, you mentioned now quite a lot of metrics.
Sophie Steffen: Is there one that you would highlight out of these? I mean, they’re all important, but is there one that you would say, “Okay, this is the one that marketers should really obsess about?”
Florin Armasu: Let’s do an exercise. You tell me what’s your preferred metric and I’ll tell you –
Sophie Steffen: Well, I mean, you’re the email expert.
Sophie Steffen: I mean, I’m a performance marketer, so for me, obviously, it would be interesting to see depending on the goal of the campaign, what would be my goal if it’s informing, also depending on the segment, if it’s activation, engagement, reactivation, probably it would be a different KPI. Usually from email campaigns, it would be kind of the ratio of people opening, so understanding out of all the emails that you send, how many opens did you get and how many clicks or engagements, so looking at this ratio, so you can compare it within your email campaigns.
Sophie Steffen: But obviously, I mean, if you can tell me the revenue of an email, that would be super interesting for me to understand how to measure that.
Florin Armasu: It’s true. You know, we never solved correctly the attribution problem, so we sometimes between channels, like your performance expert, and performance basically deals with search, with social, and with display.
Florin Armasu: And not that much with email and other channels. But actually, there’s a degree of interaction between channels, and there was, many years ago, the position called integral marketing communication. Like omni-channel. But since we didn’t solve quite well the attribution problem, sometimes we forget about that.
Florin Armasu: But to answer your question ‘cause I’m not going to another territory, for me, one metric that is very important, and we never think about it on a daily basis, is this activity expansion. So is my email, are my emails touching and have been engaged by a higher percentage of the database during the month?
Florin Armasu: So what you ideally want to see is that the brand, the communications that you send are eventually touching all the database that you have. So you can’t impact all the inactive users. Sometimes the inactive percentage of a database is big, it’s like 50% or more.
Florin Armasu: But ideally what you want to see is that your emails are being read and consumed by a higher percentage of your database in time, as the time passes. Because that means that – I mean, you work a lot to create the content and to send it. If the circle of the consumers that engage with your contents is shrinking, that means either you have a deliverability problem or you’re not doing a sufficient enough good strategy of expanding your database.
Sophie Steffen: I think – okay, so it would be basically the percentage of users engaging with your content throughout your whole database. So the more people engage with your content over time, the better.
Florin Armasu: Exactly. That’s exactly the one.
Sophie Steffen: Okay. Perfect.
Florin Armasu: That’s the golden phrase.
Sophie Steffen: I’ll take that one.
Sophie Steffen: Just to wrap the part of deliverability up and then move to the last chapter, do you have any story about deliverability? We talked a lot about it, but also in the realm of AI, you know, how these two interact and how they affect maybe the audience.
Florin Armasu: Well, yeah, I do. Thank you for the question.
Florin Armasu: It’s really interesting. You know, the deliverability with AI, people that are working in deliverability are kind of scared of the newest changes, like the Gmail resuming the content of the email, so the consumer doesn’t go to read the email because they just have a two, three lines description of what’s inside.
Florin Armasu: It’s like a more advanced preview. But what I actually think that the real danger is, and it happened – I tested it by uploading all the deliverability documents that I could find from big senders like Salesforce, to Mailgun, SparkPost and all the other ones.
Florin Armasu: Then from agencies, then from deliverability experts, and finally even researchers, if you can believe it. There are PhDs that study in their doctorate how spam filters work and what’s relevant or not for the consumers. And the answer came back from – and I tried it with Gemini, I tried it with different LLMs, with ChatGPT and also with Claude, and the deliverability rules that they deliver by analyzing – I think the total was 750,000 pages of deliverability knowledge, and it looks like compliance.
Florin Armasu: So don’t do that, don’t do that, don’t do the other. Stick to the very most active pieces of your database. And I think that overlapping deliverability with compliance is not the way. It’s not the way because if you don’t produce revenue for the business, I mean, you can be as strict as you want.
Florin Armasu: And then there’s an asymmetry between email. In email we are very righteous, but the same asymmetry doesn’t apply to social media or to other channels. So in social media you can say that I didn’t like this content and I don’t want to see it again, but nobody guarantees you that you’ll not see that again, or you’re not gonna be punished by 4% of the company revenue because you showed it again.
Florin Armasu: That doesn’t happen, so yeah.
Sophie Steffen: So basically, the AI output was to follow the legal rules to the thumb? Is that the outcome?
Florin Armasu: I mean, the recommendation is – when will you have zero spam?
Florin Armasu: When you send zero emails.
Sophie Steffen: Yeah.
Florin Armasu: If you don’t send any emails, you don’t have any complaint.
Florin Armasu: Then, for example, the spam filter guidelines – Gmail is, dangerous is above 0.3%. And then you have some ESP telling you, “Oh, you have a very high complaint rate. It’s 0.15.”
Florin Armasu: And I mean, even Gmail doesn’t seem to be bothered by that, and you are bothered by that. It’s strange.
Sophie Steffen: It’s kind of like the – we’re training – the analogy is like, okay, if you don’t wanna make mistakes, don’t work, or don’t do an email campaign, right?
Florin Armasu: Yeah, yeah, yeah. I totally agree with that. It’s like the one that doesn’t do mistakes probably doesn’t do any work.
Sophie Steffen: Yeah. Well, that’s interesting. Thanks for sharing that. And yeah, also moving into a little bit the industry reality and trends, we talked a little bit about the beginning on your insights with the Gartner study, but first let’s talk about your personal experience.
Sophie Steffen: Looking back at sweat and maybe tears of your 15 years of career so far, is there a single most important, maybe painful lesson that you’ve learned about doing email marketing the right way?
Florin Armasu: I mean, yes, there is. And it’s kind of sad because most deliverability problems are sometimes invisible.
Florin Armasu: I spoke to many companies that were not aware that at Outlook they could do way better, or at Gmail they are not doing it really well, and stuff like that. Because it’s not like in Google Search where you can say, “Yesterday I was on page one, or position three, very high position, and today I’m not appearing in search,” or, “I’m on page five or seven.”
Florin Armasu: And so deliverability is sometimes treated like a responsibility of – you work with Salesforce, thus you have very good deliverability, and I can tell you that it’s not like that. So you can work with the biggest, best infrastructure there is. It also depends on your strategy and on your knowledge.
Florin Armasu: I had the – I have spoken several times with the deliverability architect at Salesforce, and he was kind of surprised that I know a lot of things. And the painful lesson is this, you are having silent problems until one day where you can’t deliver anymore, and changing IPs, changing domain, changing ESP is hard.
Florin Armasu: It is three months. It takes a lot of money, and you can even go to the – like the saddest lesson that I learned is I had sometimes heard, “We’re not doing email anymore. It’s too hard, and we’re not getting so much money out of it.”
Florin Armasu: Email is a very old channel. It’s an interesting channel.
Florin Armasu: It’s 50 years old. The first email was – would be probably considered today as spam because there was a guy selling computers, and he sent an email to 14,000 other users, and he sold, I don’t know, 90,000 in hardware, in PCs. But it’s still a good channel, and it’s at the middle of the other channels, so it’s a good identifier for social media.
Florin Armasu: You create social media with an email or with a phone number. You can use it as SEO purposes, so if you have a very good keyword strategy, you can also apply it in email and insert the same keywords in your copyright or in your copy of the emails. If you have an e-commerce, let’s say you sell on Amazon or sell in some other parts, or you have a YouTube channel.
Florin Armasu: Sometimes when you’re launching a new product or new content, email is really good in those very first moments of the new product, the new campaign. If you have a group of users that have really engaged with that category or product, and you send them an email, they could really help you in other channels.
Florin Armasu: So you don’t measure the revenue in email, but you get quite a start. Like jump-starting a car engine, email is very good at that.
Sophie Steffen: Yeah. Do it smart, as the other channels as well. I think it’s how you use it. So yeah. Thanks for sharing that. And we spoke a little bit about your AI experience, but also looking into trends, we see AI is rising everywhere, everybody is using it in copy.
Sophie Steffen: It’s changing how the industry operates, also how we build email marketing systems. So how do you adapt to this changing of how AI gets into the industry framework, how are you changing your work? How are you adapting to this AI trend?
Florin Armasu: I will answer, but I will start, if you allow me, with the – so we can establish a baseline. How do you use AI for emails today? Are you using it?
Sophie Steffen: How do I use it?
Florin Armasu: Yes.
Sophie Steffen: Well, I think the typical spellcheck is very useful. Content ideation, you know, subject line A/B testing variations. So as a brainstormer, I think it’s good for copy, let’s say, for email copy.
Florin Armasu: You just – I mean, you mentioned two of the ideas and the use cases that I see, and I think they are the most important ones, so content generation. There was a company that I think still exists, and it’s a big one called Phrasee, that was doing email subject lines with AI, that was –
Florin Armasu: I mean, AI, it’s different types of algorithms now. They were doing machine learning years ago, like 10 years ago, and now probably they are doing much more advanced stuff. And I think now you can use AI in all parts of content generation, from subject to inserting the right keywords for the right category of users.
Florin Armasu: For personalizing, it’s way more advanced nowadays. Without doing so much self-promotion, our tool, Brand Expand, is doing a lot of AI content creation for different phases of the funnel of the customer journey. And I think everybody should use – there’s Jasper, there’s Contents AI.
Florin Armasu: There’s a lot of platforms that can create content that you can use in email, and they are all with AI. Then, a level that you – and the strategic co-creation, let’s say. So the idea to think, “Okay, I have this objective in mind. What would be the strategy to – how would it be to execute it in email?”
Florin Armasu: I think AI is very useful at this stage, and you mentioned this use case. There’s a funny thing. You don’t need to outsource responsibility of the strategy onto the AI, because some – you’ll get a very average strategic plan.
Florin Armasu: They are getting better and better at it, but you still need to think, and you could think better. So one of the good examples would be do a strategic plan, explain it to AI, and ask it to challenge it to tell you what are the things that you are not seeing, what are the problems and the issues that you are not observing.
Florin Armasu: And then there’s a level of operational intelligence, like I told you, selecting which are the users that are more probable to interact with your emails tomorrow. It’s a problem that you can solve better with AI, you know, because it’s very – you could do it with other tools and with – even if you can set it up with a logic, with a segment logic.
Florin Armasu: But to execute it in practice in production, let’s say, it’s complicated without AI. And there’s – I wanted to make a mention in this point about how AI is changing email marketing. I think there’s a new avalanche of spam, of cold email.
Florin Armasu: I’m getting every day a lot of emails about services I haven’t requested, and most of them are created using AI. This is a negative usage of AI, and actually it’s pretty much not well seen by the community, by M3AAWG. That’s the Messaging, Malware and Mobile Anti-Abuse Working Group.
Florin Armasu: It’s not well seen, and it’s rising.
Sophie Steffen: Yeah. There’s definitely rising AI content and emails within the ecosystem. And also moving into the industry pricing, is there any hidden costs of email marketing that you think is out there and that people don’t see until it’s too late?
Sophie Steffen: Especially looking at the per-contact pricing model that oftentimes punishes growth.
Florin Armasu: Yeah. That’s such a great question. Yes. Basically, many years ago, email was charged by 1,000 emails sent, and there was a volume discount, a volume rebate when you are reaching certain volumes, and that’s changed five, six years ago or maybe even more, when a lot of tools started charging by the number of contacts that you have stored there.
Florin Armasu: Maybe the first one was even HubSpot. So HubSpot is really, really good if you have 1,000 emails. It’s really – you have a suite that is perfect at a very low cost. But when you scale, you get really, really taxed on that growth. So imagine 50,000 emails at HubSpot can cost you 3,000 a month, which is quite a high price.
Florin Armasu: So why is it hidden? Because you don’t think. You say, “Oh, that’s really nice.” For a small business, having 50,000 contacts is not from one day to another.
Florin Armasu: And you don’t think that you’ll have users that are unengaged. You’re not gonna send 12 or how many emails this type of pricing allows you to send per month.
Florin Armasu: You’re not. So you’re paying for an asset, you’re paying too much for an asset that you’re not exploiting, and you want to keep it there so that you don’t accidentally resubscribe an unsubscriber or stuff like that. But you can’t, because you’ll be paying every month for somebody that you’re not even contacting.
Florin Armasu: So yeah, I think there’s a pricing trap that you should avoid. So if you’re being charged by contact, switch to another solution that doesn’t charge that.
Sophie Steffen: It’s a good signal, right?
Florin Armasu: Yeah. And then there’s many other companies that – I don’t know, I’ve seen so many use cases that you have a Ferrari and you are using it like it’s a bicycle.
Florin Armasu: So you’re not using all the features, so why pay hundreds of thousands of dollars just to have an infrastructure that you don’t use? So it’s more common now to think about, “Am I going to use this infrastructure? Am I gonna pay per usage or per –”
Sophie Steffen: And moving into the question that ties to the very beginning, this – Gartner identified that the current geopolitical situation, context, is pushing companies to repatriate their CRM and email structure into regions where they have a more secure relationship.
Sophie Steffen: Does it matter really where your email infrastructure is hosted, and should it matter? What’s your point here?
Florin Armasu: Well, my point is that I was not really aware of this trend. I was thinking it in my own terms. So we live in troubled times, and we’re not gonna enter into politics of it.
Florin Armasu: But ideally you would like to have your data and the consumer data in a region that is more friendly to where the company is located in terms of legislation. So without saying big stuff, but imagine you someday have taxes on importing your data from a country that’s been putting a lot of taxes nowadays.
Florin Armasu: It’s better to have it in a data center near you. So even you cannot have it in your own country, it’s better to have it closer to you or in a more geo-friendly geography.
Florin Armasu: And that refers to any kind of infrastructure, not only to email.
Florin Armasu: That refers to any type of platform that you are having. So there’s also in this Gartner event that was last week in Barcelona – they are saying that in EMEA, Europe, Middle East and Asia, 70% of the organizations are designing actively to geo-patriate, that means bring closer to their home offices, their technological stacks.
Florin Armasu: So why not do it with email too? What we can do for our clients is install the infrastructure where they want it. So if they want to install it in, if the company’s in Canada and they want to install it in Canada, I will be more than glad to set up the infrastructure in a server there.
Florin Armasu: If they are in Spain, the same. If they are in Germany, the same. Would it be a risk to have it in the US? Well, it’s not a risk in any way. Gmail and Hotmail, the consumers, to be honest, are the ones that have their inboxes all over the world. So if you have a Gmail address, it’s more likely that your data is being stored in the US or processed by the US. Although there’s so much concern about data transfer, like personal data transfer, you still accept these terms because there’s no other solution.
Florin Armasu: And that’s it. But yeah, I think it’s a good idea, and you need to consider it. I mean, if you could set up a lean independent email infrastructure close to you, and even owned by you in a server, owned by you, would it be smarter or not so smart?
Sophie Steffen: Then go for it. Yeah. Florin, we’re reaching the end of this conversation. Is there anything else you would like to share, where can people find you and find more about your work?
Florin Armasu: Yeah. Thank you very much. I feel that I didn’t ask you enough questions, so –
Sophie Steffen: Well, it was the interview about your field, not my field, so that’s okay.
Florin Armasu: Well, we need to do the performance interview.
Sophie Steffen: The reverse – challenge accepted.
Florin Armasu: The reverse roast. Well, you can find us at sendability.com, which is our platform that we have been launching this year, although we have been speaking about it for a long time.
Florin Armasu: You can find us at the Deliverability Summit in Barcelona. It’s in a month, less than a month. We’re gonna – I’m gonna give a speech there. It’s at La Pedrera in Barcelona, and all deliverability experts of the world are coming. There’s actually a very nice series of events organized by Andrew Bonner and Nelly Bonner, called Email Festival, and it has a sender symposium.
Florin Armasu: It has many other events there. I’m gonna be at Web Summit in Vancouver in May and in Lisbon in November. We’re a beta startup. We’re not actually so much of a startup anymore because we have – we’re a specialized technology infrastructure provider, and we have some years of experience.
Florin Armasu: But yeah, we are gonna be there. And also, you know, many people contact me on LinkedIn. Please do go ahead. If you have a domain or an IP that has a reputation problem or you just want to know how you are doing, it doesn’t – it’s not so complicated for me to take five minutes away and answer a request like that.
Florin Armasu: So no strings, no cost. Just shoot me a question and I’ll answer.
Sophie Steffen: Perfect. We’ll tag everything in the description of this video, so people can find you. And yeah, thank you, Florin, for this chat full of insights and chat about deliverability, and hopefully a lot of users will have a better idea of what’s going on in the world of email marketing and email infrastructure and where we’re heading. So that’s it for today. Thank you to our listeners.
Florin Armasu: Thank you very much, Sophie.
Sophie Steffen: Thank you. Yes, thank you. And yeah, if you found this useful, please feel free to share it, tag it, and see you next time.
Herramientas mencionadas en la entrevista
Las siguientes herramientas y plataformas fueron mencionadas durante esta conversación.

