Marketing Academy podcast by GCMO, hosted by Yitzy Tannenbaum, with Alex Greenshpun.
00:00Alex One of the myths that I bust is this one: well, context is important, so let's just drop as much context as possible, and now I have a smarter AI. What actually happens is that you have a very, very confused bot in front of you, and it's just trying its best to keep you happy.
00:16Yitzy Welcome to the Marketing Academy podcast by GCMO, where in every episode we sit down with a marketing expert to deep dive into a hot topic shaping the industry. I'm Yitzy Tannenbaum. Let's get started.
00:30Yitzy Welcome to the Marketing Academy podcast by GCMO. I'm Yitzy, and today is a very exciting day for me. A lot of buildup, Alex, so no pressure. My guest today is Alex Greenshpun, who is the founder and leads the 10x Company and the 10x community. From everything I heard of Alex, she is at the cutting edge of everything AI. Everyone uses AI, and "hey, I made a picture with Nano Banana, so now I know AI." But then there are people whose computer looks like a developer screen, because they're writing all these complex connections and code, and you're really on the forefront of this whole thing. So thanks for taking the time.
01:13Alex Thank you, Yitzy. Thanks for inviting me. Excited to be here. And how did you take a look at my computer screen already?
01:29Yitzy I will say that Rotem [unclear]...
Alex: How did you figure that out?
Yitzy: I was in a webinar and you showed your screen. It was like...
Alex: Right, it's dark and...
Yitzy: It's very dark and Matrix-y.
Alex: Yeah, there you go.
01:42Yitzy So tell us a little bit about you, how you ended up in the AI marketing world.
01:47Alex It's kind of a fun story, but I don't know if it's a unique story. I've been in marketing for over 15 years now. And when AI came into the picture, I was already trying to do more with less, because obviously I'm a marketer. That's what we do. So I kept trying to see how I could use this cool little tool, this GPT bot thingy, and do more with my marketing, save time, maybe be more efficient. While most people at this stage were still playing around with it, writing funny Dr. Seuss-style poems and Donald Trump-style speeches, I was already trying to figure out: can it help me write my blogs? Can it help me write my email outreach?
02:31Alex So I kept playing with it. But the progress is amazingly fast, and as things were progressing, I felt that while I was in-house, there was just not enough time in the world to dive deep enough into this technology and really understand what it can do. I was actually VP Marketing in my last role, at an AI company, dealing with AI and writing about AI for the company blog. And I was thinking, wait, how can I learn this? How can I make this really part of how I work?
03:04Alex So basically I quit my job, and I said, you know what, I will use this time to learn as much as I can and build as much as I can. It was right around the time MCP came out, which was at the end of 2024, I think, so at the beginning of 2025 I quit my job. I started building my own MCP servers, my own automations, agents, systems. And of course, together with Noy and Hadar, we started running the 10x Marketers community.
03:39Alex What I was thinking was, at some point I'm going to get tired of learning all this new stuff, and I will eventually look for my next CMO job, and I'll just roll into this new job with a squad of agents behind me, everything ticking on its own. It didn't happen. What actually happened was that people started coming to me and saying, "Hey, Alex, how did you build this content engine system? How did you do this automation? Can you help us? Can you help us learn this? Can you build this for us?"
04:13Alex So I started working with companies doing this on my own, while running the 10x Marketers community, and Noy was also doing this on his own. We looked at each other and said, hey, why aren't we doing this together? So we opened 10x Company, and this is what we're now doing. We help marketing teams and GTM teams in organizations of different sizes implement AI in a really smart and scalable way. Not just as a bunch of tools that you connect, where you need to keep track of more and more tools, but as a system, an orchestration of knowledge, methodologies, and tools. And so here I am. [laughter]
05:02Yitzy It really is a crazy story. And the community is huge. I think you mentioned over 2,000 members, super active. Very hard to keep up with what's happening, but it really is amazing. And what you were saying about trying to keep up with the speed of AI very much resonates with me, because with my day job I don't have the time I would want to invest in playing around with things and building things on my own.
05:33Alex Yeah. I think it's quite an active community. We have about 2,000 members now, from lots of leading brands and companies, and we're super happy and grateful for everyone taking part, because it's the people who make it so amazing. And I do hear you, it's so hard to keep track. This is why we actually have an agent doing the summary.
06:00Yitzy Of course you do. [laughter]
06:01Alex We have an agent doing AI digest summaries of what's happening in the news that is actually relevant to marketers, not simply AI news. We're trying to help people let go of the FOMO. That was the original idea of the community, and it's still what we stand behind, because there's so much buzz right now. Through the short history of AI there's always been lots of FOMO and buzz. And now there's, "Oh, I built 50 agents, and they all work for me while I sleep, and I just drink coffee and click Enter all the time. That's all I do now. I don't even do any work." That's how people talk about it on YouTube and LinkedIn and Twitter, and it's really easy to feel like you're behind.
06:47Alex But the reality, and I'm telling you this from the perspective of somebody who talks to dozens of marketing teams every week, is so far from it. People are not yet building agents. People are just starting to understand how to work properly with context. And agents are not those scary robots watching you sleep while they're working. That's not what agents are. Agents are basically just a folder with the right context. That's the simplest way to look at an agent. So this is what we're trying to do. We're trying to bring it back to Earth for people, to make them feel, hey, it's cool, and if you have questions, we're here for you. I'm hoping this is something we've managed to do well. [laughter]
07:37Yitzy I will encourage our listeners to join the community if they want to learn more about AI and how to implement it in their day-to-day. But I will ask you: AI can do everything. I think there are things it does really well, and there are places where marketers use AI and it's not really hitting home. When you come into a company, what are the first places, the first low-hanging fruit when it comes to AI adoption? Where should we adopt AI first?
08:09Alex It's an interesting question, because it's kind of like asking where we should adopt electricity first, and that's how we look at AI. I think the right way to look at AI is not as a bunch of tools but as infrastructure. AI is the new electricity. So yes, you can work with a flashlight. You can bring something battery-powered into your workspace, turn it on and off, and say, "Oh my god, it's amazing. It was dark and now it's light, and I can work better and longer because I have this light." That's how most people still use AI today. It's an on-and-off button.
08:47Alex We need to change the mindset. The first thing you need to do is not start implementing tools, or start with "this is the agent we need to build here today." That's not the first step. The first step is changing the mindset, and the mindset is that we are working with something completely new here. Just think of the infrastructure we had to build historically, in companies and factories, to work with electricity. It takes some work, but you build this infrastructure and then you can take off. You cannot just bring flashlights into the factory and say, "Okay, we're using electricity now." That's ridiculous. And this is what's happening, not just in marketing. I think it's currently happening with most AI implementation.
09:40Alex So the mindset shift is the first thing. Once you have that mindset shift, the first thing you understand is that AI is all about context. So the first thing you want to do is start documenting, start building the right context layers of your marketing team. I talk about context in two ways, mostly. One is the knowledge: the what of your marketing, the what of your brand, your product, the company. This can be dynamic. Obviously we are dynamic, we change all the time, products change, features add up. That's one layer of context. The other layer is more stable. It still changes, it still grows, but it's more stable. It's our methodologies, the operational memory of how we work.
10:31Alex Right now, both of these are not well documented. And this is the biggest boost you can have in your company, not just in marketing: starting to document this, starting to build these context layers. On top of that, any agent and any system can really do a great job for you.
10:53Yitzy I feel like a lot of people will just take what exists: the brand book, the messaging doc, the ICP list. Would you build something dedicated, or can I just take, "Hey, here's my knowledge base, here's everything I have, let me just dump it in"?
11:14Alex It's a great question. Think of it like onboarding a new employee. Would you dump all of that on a new employee and expect them to just understand it all? Probably not. You still need to be very picky about what you add, because there is also a way of structuring those documents and explaining the value so it's easy to understand and not overwhelming. You check that there are no ambiguities and nothing repeating from different angles, which might cause confusion for people as well, not just for AI. So there is a method to it.
11:55Alex But the first step is just documenting. I'm talking about even documenting transcripts from your calls with your founders. Keep them documented. Make sure you have a folder with all of those transcripts. You have transcripts from calls with the AEs and salespeople. Keep these documented as well, because all of that is context that will make you and your marketing machine much, much smarter. So there is a way to build it properly, but you also need to start somewhere. I would say start with collecting all of these data points first, and then you can think about how to build it properly.
12:39Yitzy Interesting, because I would assume that when it comes to AI, it's quantity over quality, and it sounds from you that quality may be even more important here.
12:53Alex It's actually great. We have an AI accelerator where we help people who work in marketing use AI better, as infrastructure, and one of the things I like to do in the accelerator is bust myths. One of the myths I bust is this one: well, context is important, so let's just drop as much context as possible, and now I have a smarter AI. What actually happens is that you have a very, very confused bot in front of you, and it's just trying its best to keep you happy. So it's trying: "Okay, so we said our brand voice is funny and mature, but also that we are very formal. Okay, I'm trying to make it work. Here you go, this is some generic... sorry, [laughter] but this is probably going to work for you." And that's how people get slop all the time. It's one of the ways people get slop.
13:52Yitzy Interesting.
13:53Alex When you work with AI as infrastructure, you start thinking almost like a scientist sometimes. That's why concepts like context engineering have the word "engineering" in there, because you start thinking, wait, if I change that variable, what will it affect down the line? That's how you start working. You start thinking about the documents you upload: do I really need that little document and that paragraph for this particular task, or is it just going to cause more confusion and maybe not-so-good results? That's how you should think when you're working with AI.
14:35Yitzy I'm sure throughout this conversation you're going to bust many myths, both for me and for our listeners, so I'm excited to learn a lot. So first I give it the context. But you also mentioned the methodologies, how my company works, how it operates. How do I structure that? How do I feed AI that?
14:59Alex The operational context and the methodologies, you can think of them as skills. I'm sure you've heard of skills, agent skills. Anthropic started all of that, and now pretty much every agentic platform supports skills. The way we think about it is we look at the biggest workflows we have and break them down into several steps. We see which building blocks are supposed to be there to connect it to something that actually gives us value. Then we can focus on any one of these building blocks, or later, if we want, connect them together into a bigger methodology.
15:42Alex It's just like documentation in general. You say, "Okay, what's my process for writing and distributing SEO-aligned blog posts?" And you say, "First I'm doing keyword research." That's one of the things I need to do. I have a specific way I do keyword research. I have specific tools I use for that. So you want to document all of that. And you probably also have a lot of interesting things that are never documented, that swim in your mind all day long, because you are the expert. You say, "Wait, I know my competitor uses this keyword a lot, and it sounds like a keyword we should use as well, but it's actually the opposite of what we represent." That's nothing the AI can know on its own, and it might not be documented in any of the context files you uploaded.
16:35Alex So this methodology must include all these little shticks, and you have to teach it your invisible knowledge. I call this the messy middle. We have a lot of knowledge in our minds, and I think the biggest challenge right now in working with AI is how we bring it out, how we document that knowledge, how we make our AIs learn what we know.
17:03Yitzy That's so interesting. So it sounds like a big part of what 10x does is not even AI-based. It's helping companies understand what their processes are, how to articulate them, and then how to feed them into AI.
17:16Alex Yeah, definitely. It's one of the first and most important steps in AI implementation in general. It's not the only thing in working with AI, but it's fundamental. And by the way, it's not straightforward. Sometimes, while you build it, a couple of new processes are already being born. So it's alive, and it keeps getting updated, and you need to figure out: is this process a building block that can connect to this particular skill, or should we build a completely new system for it? So there are almost strategic decisions that suddenly come into play in your AI-related decisions as well.
18:11Yitzy Which is true with every process we build in marketing, because as the market evolves, so do our marketing machines, and so does AI.
18:20Alex Yeah, exactly. We've been experiencing something really amazing in terms of the pace of growth. It's kind of wild. Take something like video generation. Look at what was possible eight months ago, and then you have something like Seedance and it's mind-blowing. You had to connect five different tools to build something that sounds this way and works this way, and then you have Google Flow and Seedance, and you do it in minutes now. This is what we see in every aspect of working with AI.
19:07Alex What's important is our part, our role as humans. Where do we come in, if AI suddenly can do this and that, create content, create videos, do outreach? Where is my role as a marketer there? Actually, our roles become much more important, because we become the orchestrators. Marketers become the orchestrators. We're not just orchestrating agents. We're the ones deciding on the logic, deciding on the importance of particular steps in the systems we build, and of course the strategy.
19:42Alex And by the way, AI is leveling the field for everyone. This is something we need to talk about. Yes, you can do this now with agents, but other teams can do this now with agents too. So what's your differentiator now in the way you work? It's always the people, it's always the expertise. I actually believe that at some point we will see even bigger teams, [laughter] not smaller teams. At first you will see smaller teams working with AI, and of course in startups we'll see lots of small teams doing lots of big things, which is terrific. But as companies grow, we will actually see more people in the marketing team, because we'll need more of these orchestrators working with the agents and bringing in their unique point of view and their domain expertise.
20:36Yitzy I would say that's another myth you're busting. A lot of people think AI will replace us, and I very much agree with you, I'm a big advocate of this. I think it will replace tactical work. If you do the same thing again and again and again and don't use your brain a lot, those people should be worried. But if you're building strategy... When we look at AI, AI is trained to do more of the same. Whatever is the most common thing, that's what it does. As marketers, we're trained to do the exact opposite, to find the weird angle. How do we stand out? Today there's a crazy amount of noise. You mentioned videos. I was one of those people eight months ago building those videos with the five different tools, so I know that very well. Our role has always been: how do we stand out, how are we unique, how do we build a strategy and a process that is quick, efficient, and differentiated? We used to use freelancers, we used to do it ourselves. Now we're able to spend more time on the strategy and leave the tactical to the AI.
21:50Alex Exactly, and that's the beauty of it. One of the things we see at 10x Company: once we build the infrastructure I described, with the context, the operational context, and the knowledge, and the first system is running, suddenly you see a spark in the eyes of those marketers. They're like, "Oh wait, hold on, are you telling me I can also do this and this, and connect this and this system?" They get all these ideas, and they do what marketers do: they get creative about it and start thinking of lots of different use cases. "Maybe instead of this gated landing page we have right now, we can build this cool application, maybe even an MCP server that connects to the agent of the user."
22:44Alex We're seeing this every day. It's like all the gates, like chakras, are opening. That's how I think about it, because they see this amazing possibility, and they know it's right here. It's not somewhere on somebody's YouTube. It's right here and they can do it. And this is why I think we will see more people in marketing teams, because yes, you will do all these things, but you will want more. That's human nature. We will always want more.
23:15Yitzy And in marketing you have to, because our audience has a very limited attention span. They can only see a certain amount, and we're still going to fight for the same attention. As you said, the playing field is leveled, but it's leveled across the board. So we're still going to have to be creative, bring our innovation, bring something unique. But now we're not spending time writing eight-page white papers. We're spending time thinking about our ICP, what they're looking for, and how my message is going to resonate, which is what I believe marketing has always been. It's just getting easier.
23:58Alex Exactly. It's getting easier to get the data you need. For example, it's so easy to just go and do research with a bunch of scrapers, or just one scraper, into the LinkedIn posts of your ICP, their Reddit threads, the G2 reviews, whatever you have. It's so easy because it's right there. You can just go and grab it. But how do you orchestrate it? What's the logic? What do you ask? What are the questions you need to understand so it actually gives you value, and not just some slop like, "Hey, your users are really happy about this competitor." Okay, great, that's not really helpful. So we need to be the ones who bring in that judgment, that logic, the systems thinking and the strategic thinking. Those are going to be the real moats [unclear: captions read "modes"; "moats" fits the context] in this new... [cut off as Yitzy comes in]
24:50Yitzy Something I wanted to ask you. One of my pet peeves of AI today: I'll work really hard on something and send it to a colleague or a manager for review, or to get their perspective. They'll put it into AI and send it back to me. And I say, A, I did that. B, I could have done that myself. C, you're not providing me any value. So how do I make sure that what I'm getting from AI is the quality I'm looking for? Especially with those people who throw it into AI and throw it back, that's not really helping. How do I use AI as a tool, versus as just a...
25:45Alex Slop machine.
Yitzy: Exactly. A slop-ifier [unclear]. And now that I've asked the question, I think I kind of know the answer: it all comes down to how you train your AI. But I'm wondering what you think.
25:58Alex It's not literal training, because we don't really need to go into the models and fine-tune them or anything. But it is the foundational knowledge layers that you build for your AIs. And now, with skills and the current capabilities of agents like Codex and Claude Code and Cowork, it's become really easy. It's right there.
26:23Alex What you're describing, by the way, is workslop. That's the terminology for it, and it's very, very common, especially in bigger organizations. I know you write this on your own, but people are often just writing something with AI. They get a 20-page document, they send it for review to their manager, the manager reviews it with AI and sends it to somebody else on the team, who is now stuck, because maybe their neck is on the line, so they have to actually read it. [laughter] And they don't understand how it makes sense. It's very, very common.
26:59Alex And it comes back to the first thing I told you. The first thing you need to do when working with AI is not to try to implement the tools. It's the mindset shift: this is not an on-and-off flashlight, this is infrastructure. If you're just sending people batteries and a bunch of flashlights, you're not really getting any value out of it, and at the end of the day you're probably wasting their time. So people need to change their mindset about this, and also about their responsibility within that supply chain of doing work and using AI.
27:42Alex As I said before, we are the orchestrators. When you are the orchestrator, you know you are responsible for the output, and you are responsible for the inputs the AI gets. Whatever happens within that process, you are responsible for guiding it. You are responsible for saying, "Hey, you know what, stop right here." I do it all the time with my Claude Code and Codex. I say, "Hold on, hold on," hit Escape. "Stop. What are you doing right now? Why did you go there?" You have to do this sometimes. Of course, we later build tools that help us do that. We have skills that run and do the review for us. But still, I read whatever it is I'm writing. I write it together with the AI. It's AI-assisted. It's not a replacement of me.
28:23Alex This is an important thing to understand. We still need to be in the picture. Otherwise, what's the point? Why are you there? If you're not even reading this email, if you're not even reading the report, what are you doing? Where is your value? Where is the particular domain expertise that is yours? Otherwise anyone else can do the same thing, and your competitors are probably doing the exact same thing, and there's no differentiation anymore. That's the danger. The danger is offloading the thinking to AI, and we really need to take responsibility there. Because again, yes, I have electricity, but I know when to turn something on and off. I know I need the air conditioning on right now and not the heat. That's up to us.
29:20Yitzy I really like the electricity analogy, because electricity makes our lives substantially better, especially in these summer months, but it doesn't solve all problems. It solves many, but there are still things that are not solved by electricity. So it's understanding that AI is a tool, that it's going to solve many issues and make our lives substantially better, but there are issues that we as humans will still have to handle. Drawing the line, understanding what is our responsibility versus the AI's responsibility, is really important. My advice to our listeners: if you find that all you're doing is putting things into AI and sending them out, maybe relook at your responsibilities.
30:12Yitzy That brings me back, Alex, to the idea of enablement. You come into companies. Many marketing leaders have a variety of employees who adapt differently to technology. How do I bring the team in together? How do I enable them? How do I teach them to use this very strong tool?
30:37Alex Excellent question, because there is no single correct answer there. It's part of what we do at 10x Company. We come into teams and help them work together in a more standardized way. Instead of everybody building on their own in their own little silo, we have a standardized way. We build a marketing brain. We build the marketing systems and tools on top of that. And then we show everyone how to use it, when to use it, and when not to use it. When do you build your own thing and share it with your team, and when do you build your own thing and just keep it to yourself? Because there's a difference.
31:16Alex And that's part of the slop, by the way. The slop, the slopification, is happening within skills as well. People say, "It's so easy to create a skill. I'm just going to say, 'Hey, Claude, create me a skill to remove em dashes from my email.'" And now you have a skill just for em dashes. And then another skill that removes "it's about, not about." I'm just giving you examples. Obviously it's not the best way to use skills. But it's slopification. This skill is called the copy editing skill, and the other one is also called the copy editing skill. Maybe both of them are good, maybe both of them suck, and you don't know which to use. Then your team suddenly has 200 skills across them, and you don't really use most of them, because you don't know which to use and which is good.
32:10Alex So a standardized way of operating is the baseline we start from. But very importantly, we need to understand as marketers that not every marketer on our team is going to be at the same level of working with AI and agents, and that's fine. Everybody should work through the standardized baseline in the standardized way. But above that, you should encourage people to explore and grow on their own. Don't try to box them in with, "No, we don't build skills, we don't do that." On the contrary, those explorers are going to be the ones who take the rest of your team forward.
32:53Alex So what we do at 10x is select a couple of champions, especially in bigger teams. It's usually really easy to see them. You see the sparkle in their eyes when you talk about these things, and they come to you and say, "You know what, I actually built something that does this." We select these champions, work with them, and give them as much knowledge as we possibly can, so that they can later show their teammates: "Look, I built this thing. You're doing this manually here. Take this, I just built it, and it's running on its own." And the teammate says, "Oh wait, how did you do that?" And the champion teaches them. That's how knowledge transfers, and everybody benefits.
33:43Alex What companies usually do is bring in a workshop. We've done some of these too, people ask for workshops, and that's great. But they bring in a workshop and say everybody has to listen. What happens is that not everybody is at the same engagement and motivation level. If I were in-house right now and we had an AI workshop, I would probably say, "Oh my god, that's so awesome, I want to be there." Somebody next to me would say, "No, I have to pick up my kids, and now I have to watch this workshop instead." Different layers of motivation, different levels. So usually nothing changes after these workshops. The champions, the people who were already working with AI in a smarter, more unique way, continue doing that, and they benefit. But the rest of the team stays cut off. So it's a culture thing. It's less from the outside in, and more from the inside out.
34:47Yitzy I'm a big fan of appointing champions, getting the people who are passionate about it and love doing it involved, and getting them to teach and support others. I think it works really, really well. Now, when you look at marketing departments, you're seeing this AI evolution, roles changing, structures changing. Do you have any advice or guidance? How are you seeing this evolution happen?
35:25Alex It's a great question, and it depends on how far into the future you want to look. Boris Cherny, the creator of Claude Code at Anthropic, just posted a tweet saying there will be, I don't remember exactly right now, four roles remaining in the future. One of them will be growth, another will be somebody responsible for bringing ideas in, and so on. Archetypes, basically. It's nice to think in archetypes. I don't think that's going to be the exact reflection of what we'll see.
36:03Alex One thing we keep forgetting, and you mentioned it before, is that new technologies bring new problems. Before the industrial revolution, we could not have imagined what new jobs it would create. We could not in a million years have imagined that there would be somebody responsible for posting things on Twitter. [laughter] How could we think of that? We didn't even know what Twitter was. And we don't have to go as far back as the industrial revolution. Even in the '90s, you could not have imagined all the amazing things the internet would bring. You could not have imagined smartphones, or the different digital roles we have with the internet. This is what we will experience, but much, much faster, with AI.
37:02Alex As more and more companies adopt AI, we will see new roles being born. There are already GTM engineers, and there are already AI builders working within marketing teams only to build things, because again, it makes sense that not everybody is going to be at the same level. When you're in-house, you don't have enough time to go and learn how to build something or how to manage a particular agent, so you need some support. Right now we're seeing the beginnings of that. The roles will change. There will be new roles.
37:43Alex I don't think the question of whether or not somebody will be replaced by AI is the right perspective. We are marketers, and I think the fundamental thing about being marketers is that we keep on learning. "Always be learning," that's literally my motto [speaker unclear: the captions mark a speaker change before "always be learning" but no change back, and the thought runs on unbroken, so it is most likely still Alex], and I think every good marketer I've ever met has that motto too. As long as we keep up that almost beginner's-mindset perspective and keep on learning, we will be fine. Thirty years ago we didn't have a role for social media. We couldn't have imagined it. But maybe you were doing something else in marketing and you grew into this role.
38:34Yitzy Most of the roles in marketing departments didn't exist. Email marketers, growth marketers, social media marketers, influencer marketers: go back to the '90s and none of them existed. Was there an influencer manager ten years ago? I very much agree. Now we kind of take all the AI roles and lump them into this one go-to-market engineer, or whatever we call them. I think new roles will happen, but I like to use the word evolution. People will evolve, and as marketers we're very accustomed to that, because our role always evolves, no matter what. So this is our evolution. As marketing leaders, we need to think about how we enable both our department and the individuals within it to evolve. The champion, yes, that's the low-hanging fruit. But how do we make sure every single team member is evolving and using the tools? Because, as you mentioned, my competitor is doing it too. If my email marketer or my influencer marketer isn't evolving with the technology, it's putting me, my department, and my company behind.
39:55Alex Exactly. And I'd like to offer another perspective, from a different point of view. Within the team, yes, we will have new roles. But I think in the not-so-far-away future, marketers and people with our skills will be the leaders and the most important people in the company. I strongly believe that, because we have a unique combination of skill sets. I believe that about marketers and product leaders, by the way. I think we kind of converge on these. We will be the leaders of companies. I believe VCs will want to see more CEO-CMO leaders in the future, because we always keep on learning, we have the perspective of how to bring about growth, and we keep thinking about things creatively. Especially if we are able to think in systems, and we know how to combine our domain expertise and our ability to bring about growth with the AI systems we can build, then we have a unique perspective and a unique value.
41:00Alex Up until now it was usually the builders. They were builders, and they're still running companies right now: "We're building this product, this is what we think we should do." And that's great. But now, when everyone can basically build anything, the thing you're still missing is distribution. How do you make people use what you've built? How do you make anyone notice you? This is why marketers will become super important, literally because of AI. AI is leveling the playing field, and not just for marketers, for everyone. Features that used to take months can now be built in hours, sometimes less. So we need to understand where our differentiator is, beyond what our product has right now. And that's marketing.
41:57Yitzy I love that, and I very much agree. I think everyone is worried about this big unknown. But I also think marketing has always been a project management role. Not always, but definitely in recent years, we were usually orchestrators. Instead of AI, we had freelancers, we had agencies. Our LinkedIn agency would help us with that, or our content writing agency or video agency, and we would play with the various execution tools. But we were the brain, we were the innovation, we were the idea people. How deep you got into execution depended on your role, but for a long while, pre-AI, marketing was very strategic. So I think we're just evolving the execution piece into something different, more efficient, more precise. But the ideas are still what drives our departments, which is very cool.
43:07Alex Exactly. And it's very similar to project management. Orchestrating agents is very similar to project management, and you will be orchestrating a lot of those agents in the near future. You're probably already doing some of that: different skills doing different elements of your current work. Then you'll have things running autonomously, but you still need to be there to check in, because this is client-facing, world-facing. It's not just some email you're sending to a guy who's looking at it with an agent. It's for your audience. Another point, of course, is that the audience is also going to use agents to read or consume whatever we're delivering, but that's another point.
43:50Alex The world is going to evolve. The scale of everything we do is going to evolve. So the ability to connect different systems together and understand, wait, when I change this... It's like those magnetic blocks my son really loves. If I change this and do this, suddenly I have this huge tower, and the ball can roll over here and there. That's how we think, playfully almost, but with the systems we build. I think that will be a huge value point in the near future.
44:26Yitzy Let me ask you something. I feel like the world of marketing today is going to a very human place, where we look for human connection. There's so much noise and so much advancement and new technology out there. In a weird way, as a marketer, what I look for is how to connect to a person on a very personal level. When people see AI, when they see the em dash, when they see that sentence that says "we're not this, but that," they automatically go, "Just another AI post." On LinkedIn, sometimes you see those AI posts and you're like, ugh. I think we almost look for the mistakes, for the humanity, in everything in life, but specifically in marketing. Do you have guidance on how to be human in such an advanced, AI-driven world?
45:31Alex It's a fantastic question. I don't think I have a single correct answer for it. It's something I think all of us should be constantly thinking about. The first thing I would say is that communities will matter a lot more. Obviously I'm running a community with my partners, so it's an obvious thing for me to believe. But I also believe we can always bring our humanity, even if we use AI tools to assist us with, I don't know, some copy creation. We should always bring forward our humanity and be creative about it. Physical events will be more important. Meetups, roundtables, what have you. They're going to be more and more important, and more unique.
46:30Alex And right now we're definitely at the stage where everybody's allergic to obvious AI outputs. I have an instant, infinite eye-roll reaction when [laughter] I see a post like that.
46:43Yitzy We all do. Yeah.
46:44Alex Right, exactly, we all do. We have a sixth sense now for anything AI-ish. Why? It's not because of AI. That's the very important point to understand here. We don't hate this thing because it was written by AI. It's the white space that's missing. It's the human being within the picture that is missing for us there. Where is your idea? Where is your original thought in there? What's the value you are bringing?
47:20Alex By the way, you can check this yourself. Right now on LinkedIn, I think pretty much 90-plus percent of the posts are written with AI, assisted by AI, or fully generated by AI. Check it with the people you like to follow. Sometimes you read something, and it might not be too obvious, but it still feels AI-ish. Yet it's interesting, and you keep reading because it has a unique perspective. It gives you unique value. You don't have that allergic reaction to something like that. You might even see an em dash sometimes, but it gave you some actual value, something to think about, and you read it.
48:00Alex On the other hand, you see all these sloppy posts with lots and lots of em dashes, and obviously "it's not X, it's Y," and no substance. All of these structural, aesthetic things are irritating, but we can overlook them when there is actual value, actual substance. And this is how I think we should think about marketing in a world with AI. We always need to bring substance. We always need to bring value to what we do.
48:45Yitzy I love that: ideas, substance, and creating value. I think that's the hardest thing, and it's always been the hardest thing. Our customers have such short attention spans, looking at our website for a few seconds. How are we hitting home? We should use electricity, or in this case AI, as a tool to figure out what they're looking for, what they care about, what resonates with them. But it's our role to come up with the idea and be innovative. Alex, I don't know if you expected this, but I thought my big takeaway from this conversation would be how to configure my MCP server. [laughter] And it really is inspiring to understand what our role in all this can be. So, Alex, before we go: some cool examples of marketing departments using AI, some use cases, some fun ideas you could leave our listeners with.
49:54Alex There are lots of use cases. Of course, it depends on the actual audience and the actual ICP of these companies. But I see a lot of cool ideas around gated content, like I mentioned before. People are building games, mini computer games.
50:12Yitzy Love that.
50:14Alex Some people, as I mentioned, are building an MCP server with specific knowledge that's very unique, maybe data points it connects to that are very unique, and now their users, their audience, can use it to get this knowledge for free. It's super valuable.
50:33Alex Those are some examples. And of course the value comes even from the everyday things. Suddenly you have a skill that documents the tone of voice of your brand and the tone of voice of your founders, and you already have the brain with all of the previous ideas, all the things that worked and the things that didn't, and it gets smarter all the time. Suddenly even your brainstorming sessions become much more interesting. You get interesting, valuable perspectives even from brainstorming with AI, especially when you connect it to systems like competitive intelligence, which is also super easy to do now with AI.
51:17Alex For example, we have people running an automated competitive intelligence system that runs every couple of days. It checks competitors' social media, checks YouTube, checks their blogs, and, depending on the industry, maybe even publications like academic papers. Then it compares everything to what we have currently, our current messaging, our current publications and content, and it ships a couple of points of view and ideas to their Slack that the content team can pick up and start working on. It's not just content. It could be something for events: "Oh, this is an important angle, maybe we can do something fun at a physical event based on that."
52:04Alex So the response is much faster, the ideas are much richer, and of course, when the human comes into the picture, it bounces back more great ideas. Things that sometimes took weeks to figure out just appear on your Slack in minutes, every day. And now it becomes more a question of priority: when do I want to look at it, when do I actually want to react to this, and when is it okay for later?
52:33Yitzy Alex, I came to this episode to learn about AI, and in the weirdest way, and I don't know if you anticipated this, I learned more about people.
52:45Alex I think that's the biggest compliment you could give me. [laughter] That's amazing.
52:50Yitzy When we started out, I told you this could easily be a Joe Rogan podcast and we could probably be here for four hours. To me, this time went by very, very quickly. So I want to thank you for your time. I want to encourage our listeners to join the 10x community. There you'll hear about humans, but there you'll really hear about AI: how to configure it, how to get it to work. And invite 10x Company to your organization to help you go through this transformation. Alex makes it sound easy, but anyone who's leading a marketing team knows the challenges involved, and getting help can be very valuable.
53:35Yitzy I do want to thank Google for Startups. If you have a startup and you're looking for a place to get it off the ground, use AI, but also turn to Google for Startups. Google has a very strong AI itself, which is provided as part of the Google for Startups program, so check them out. I want to thank our lovely producers, who introduced me to Alex and very much looked forward to this conversation, Rotem [unclear: captions read "Rotam Shah"] and Rachel. Thanks for your help. Alex, thank you so, so much. I can't wait to do this again.
54:08Alex Thank you so much. It's been super fun. Thank you.
