Hey everyone, welcome to the webinar. This is about what if this is the last website you'll ever build? Very controversial. So AI led digital experience for the future of manufacturing. I have Aswin and George with me. If you guys want to introduce yourself, guys,

George, go ahead first. Okay, hi everyone. George Trang, I'm the Senior Director of visual experience and technology at hexagon. So our teams are responsible for our global digital platforms and making sure that our web experiences are modern and help our customers get what they need to find quickly. For those of you who aren't familiar with hex gun as a company, we're a global leader in measurement technology, so we help a lot of industries. Help build and innovate with confidence. A lot of our solutions work around productivity and sustainability across a whole bunch of different industries, especially around manufacturing, around construction, aerospace and the like, and so at the end of the day, our solutions and the solutions that we manufacture are really to help bridge this gap between the physical and digital worlds. So excited to be here, really looking forward to having this conversation about these AI digital experiences and kind of what it means for our industries today.

Excellent. So thank you George. And first, thank you Akshay for involving me here. What I do in my day to day is I'm the principal industry lead for manufacturing at Sitecore. And in addition to that, I'm a researcher. So I spend my time at the University of Pennsylvania working on what is next for manufacturing and industrials? And I love the provocative nature of the question, what if this is the last website you'll ever build? Because if it is, what are the things you should be thinking about, and what are the things you shouldn't be thinking about, as in, what are the things you need to untrain yourself to think through? And I can't wait to get into the topic.

Yeah, I know it's been exciting thinking about this, because we're all trying to figure out better ways always, right? So let me ask you guys the first question, and then from there, let's see where it goes. So the question is, you know, where does the traditional manufacturing website model breakdown for, like the modern buying groups, right? Is it with the long, linear sales and what signals tell the current leaders it's no longer to fit for the purpose it was built?

Go ahead to chime in afterwards,yeah, I'll give you a little bit of kind of our experience, kind of work, having being in this position, right? Is, you know, we are, as a company, we have a lot of experience with long, much longer, buying cycles, right, and basically interpreting how we work around the situation in order to provide the best experience for visitors to our website, right? And I think what we've kind of come to conclude, after some research and just some you know, hits and misses, right, is to is that the modern buying group really is just that which is a buying group, right? You have a whole bunch of different people from a whole bunch of different roles coming in, whether it's whether it's engineering or technology or procurement or finance or, you know, executives that are coming in, all with different intentions, all with different perspectives, and all looking for different things. And they're all kind of coming to this, to your website, try, all trying to find different things. So this idea of kind of this, I guess we'll call it legacy idea of this kind of linear buyers journey, of someone coming to a website and going to your homepage and navigating to this, and then going, clicking on that, and drawing going to this, CTA, it's starting to fall apart a little bit. We see that a lot more customers are, you know, they're, they're using search engines. They're using AI chat right, to to do their searches for specific needs. And part of that, it really is for us, kind of the result of that is for us to really focus on the content and how we structure the content on the website, right, and so that it is AEO and geo friendly, right? It is, how do we? How do we, you know, make this, make the content relevant for multiple different parts of the part, multiple different parts of this buying group, but also be able to provide all the information without overwhelming whoever is coming to our website and taking a look at it. I mean, that's that's been the biggest challenge, right is kind of flipping this linear buyers journey on up on its head, and how do we provide all the content that these buying groups are looking for, but also not overwhelming, overwhelming people with information that they're not looking for from.My perspective, I think the past always informs, like, where companies are headed, right, which is when we think about manufacturing marketing, right? What is it usually like? The end of the day, it comes down to, let's go to a trade show. Let's have some flyers out there. We'll give out the flyers. And so that DNA, I think, informed a lot of manufacturing websites, as in, like, how can we recreate those brochures? The websites are very functional in that sense, and experience became secondary. And one of the things that is very interesting about manufacturing is, at least in my experience, given the DNA and cultural kind of background of people that get into manufacturing, experience was originally seen as something as being like we don't really need to care about that. We just want to make sure we're just telling just the facts and just giving the top line information. And it will be a very kind of like a knowledge based sale, as opposed to emotion. Because experience was almost seen as a dirty word. And one of the interesting things that I think is causing people to change is one the proofs in the pudding. When we manufacture, started seeing other manufacturers who said, like, okay, like, there's more we can learn from say, how a CPG company is able to architect their experience. Like, if we take those learnings, it becomes much more important to start thinking about websites that are more experience latent than just those more brochures. So I think that was mark one of this evolution. Mark Two of that evolution is like, let's think about how manufacturing buys your main right. These are huge capital investments that will take many years to pay off. And so in it, like, if you're just going Amazon and clicking and buying something on, like the cart and so on, that's a very like almost instantaneous sort of thing. So you wrote, you don't really need to care that much. But when you have a complex capital buying group that is associated with people that have Signing Authority, people that are dealing with procurement, all of the constellation of people that you're bringing on board, they have fundamentally different needs. And when you start thinking about those fundamentally different needs, linearity, by nature, just breaks down because like, different people are coming in. So if you're architecting an experience that is as robust and dealing with all of those things, linearity by just as I mentioned, breaks down. So now we're in a brave new world. We have, yes, let's go and let's start investing in experience. We have to start thinking about non linearity, and we have to start thinking about the buying group and disaggregating that group instead of being a giant blob. So when you're there, I mean, I just had a recent conversation with an HVAC company, and they were like, I don't know how to navigate this world. Was the question that was, like, it was just a very honest and vulnerable moment that they had, which was like, I don't know how to do this, because all I know is how to go to these, one of these giant trade shows, like, Yes or something, and just have all of those leads come in. It's very much an open question. And like, I hope that one of the way things that we can arrive at the end of this conversation is, how do we think about that, like, next stage now that you're in that era of uncertainty. So I'll pause here for a second, actually, if you have any thoughts or anything else youwant to add. No like, I think you, you both kind of covered it, right? It's with all of this and with AI added to the mix on every single thing everyone does, including, like a common person who doesn't even have anything to do with it, I think a lot has changed, and a lot is in flux. Where will come out of this. It's very interesting, because if you're able to build stuff on your phone by sitting in a coffee shop, right? Like, think about that for a second. Like, where does this go in terms of adding additional problems from, like a governing and legal perspective, and at the same time making things easier for proof of concepts or internally for you to do things. I am actually very curious to see six months down the line what happens. Either it's going to be a big bust, as in, something horrible is going to happen, or we will finally start seeing things settle down and companies starting to say, okay, AI isn't going to go away. This is how we're going to do it. So coming to that, I did have a question for you guys. So how does like as of now? How do you guys see AI changing digital experiences specific to manufacturing? So like, from creating pages, the ideation, product journeys, intent driven interactions across like roles and channels. Like, have you seen something change? And if so, what?

So I can offer some thoughts. And I think the phenomenon that's happening with AI is this shift left, idea right, as in that people earlier. On in the value chain, are starting to design more prototype things. And what I find really exciting about that is there always used to be this. When we think about the content supply chain for manufacturing, person one used to do their tasks. They used to kind of package it off, send to person two, and so on and so forth. While it had its virtues, the problem with that approach is it introduces a lot of latency in the system. And I was just an Italian manufacturer, and it's interesting. You wouldn't even think of them as like, having this sort of DNA, but they were thinking about like, how do we see the latest trends in like, social media, and how do we build marketing campaigns around those trends? And if you want to stay at the speed of, say, social media culture in general, one of the things that the shifting left is really going to play a role, as in that the marketer can start saying that, hey, this is what I'd like to design. Then they go to the technologist. And so that becomes a much more faster loop than we have seen in the past. And so it might be really interesting to think about what the downstream implications of that are. George, I don't know from your perspective, like, how is your organization seizing on AI?

Yeah, I mean, I think we're, you know, like a lot of organizations are, you know, I think we're still feeling our way around kind of how to best leverage AI for our use cases, right? I think there's a lot of experimentation or anything, but I feel like, you know, we're not alone in this, in the fact that a lot of organizations don't necessarily know how to use AI, almost like, I'm gonna coin it like, almost like next gen kind of AI marketing, right? We have a lot of content creation, content generation, you know, prompting, you know, getting us to first draft state pretty quickly, but then it's still, you know, a lot of you still need the manual tweaking and everything, and we still wind up with web pages, and kind of, the end result is still kind of what we've always had, right? It's just getting to that end result a little bit faster. Also there, you know, we've seen even, even in the industry, right? The the content management industry, right? There's a lot of push for, well, we can do, we can take your one piece of content and we can create, you know, 18 different variations of this to to tailor to, you know, either ABM or for a specific industry or specific target target market, right? And so then it's you, you still kind of wind up with that same kind of result of like now, but more of you have more kind of targeted campaigns and everything, but you're creating but faster. And I think to me, this is kind of almost kind of gen one of AI based marketing, right? Like, how do we get to where the traditional marketing, digital marketing pipeline, but faster building in the efficiencies? But I think what we really need to start thinking about, or even just kind of wondering about, is, what does that next version, next generation look like? Is it, you know, are we still, is it still web pages? Do we, you know, kind of back to this idea of the topic of this webinar, right? Are we even? Are we even just trying to get to web pages faster, or do we need to look at a whole different other format together? How do we engage with these buying groups. And, you know, I kind of what you were talking about earlier, Ashley, about, you know, people buying on Amazon and and these longer buying cycles for B to B, right? I very much subscribe to the theory of buyer groups are even the, the biggest buying group, the biggest company with the longest sales cycle is still made up, made up of people, at least for now, right? And because of that, right, these same people on their weekends, in their evenings, they are the Amazon shoppers they've been over time, they've been led to develop, to expect a certain kind of user experience based off their personal lives. And that's not You're not going to be able to divorce that away from how they approach buying, you know, even subconsciously in a, b to b type of cycle, right? And so it's as even as the B to C universe changes, right, how to it's almost like they kind of, the B to C world kind of takes the lead in how this, how this translates, right? Eventually, no, maybe no one goes on Amazon, and all our shopping happens in a chat box, the road, GPT, right? That becomes the new normal. That means, from a B to B perspective, those That's that same expectation is going to be put on, you know, manufacturers, or any, any other kind of B to B industry, because that's kind of what they've been you know, people have been trained to to learn and to use and to expect, and so I think this very much so are how we kind of approach this next generation is very much going to be. How do we what are the interfaces that we're going to be going through? What kind of content are people going to be looking for? What's. Really important to these people, and how do we deliver that content through these new channels that may we may not even be fully aware of or fully prepared for yet?

So George, there's two things that you said that really got my mind going, which was, number one, is like, not yet these buying groups are going to be old. So it's like, it is conceivable in the future that you have a procurement agent or something else like that, right? So how do we make sure that that procurement agent is able to go to your website and get the information that it needs? I mean, does all of your content need to be in Markdown so it can kind of go through that agent like make sense of that data? So there's a set of considerations that are here that we need to start thinking about. And I think this is the perfect shove moment for us to think about that. And the second thing that you said is, like these are like engineers selling to engineers in many contexts, in manufacturing, right? And the thing, and actually, I think you did this earlier, which is that like this content just can't be like cowboy content, as in, it just can't be stuff that's just created. So for example, if you have, like a sub assembly for a break, or like a caliper, it has to be exactly right. It can't be just like, oh, this is some hallucination or this part, because you're selling the engineers and the I wouldn't contend that manufacturing has perhaps the highest threshold, because it has to, you know, people's lives depend on this. And, you know, it has to match the Bill of Material almost one to one. It just can't be anything fake, so that that's an interesting kind of side aspect to all of this, as well. As in, we're kind of pressing the accelerator, but we just have to make sure that this crazy out there.

Yeah, yep, we have a question. So we could probably take that before we get to the next one, because I had a juicy question for both of you. But so Michael asked, what are some practical ways manufacturers can start using AI to improve digital engagement today without launching a massive transformation project? I think most of the marketing leaders would want to know this as an answer to so Aswin or George, you guys can jump in

George after you, please? Yeah. I mean, I would say, you know, kind of what we were talking about earlier for this kind of current generation of AI marketing is, you know, how do we, I don't think we've, we're fully transformed into the fact that, like we have to, you know, everyone's shopping through gpts or anything like that today, everything is still kind of living in a very kind of traditional sense. But I think it is back to, how do we use AI to make what we do more efficient and make what we do more research driven and knowledgeable, right? Instead of, instead of guessing, and I think this is where kind of AI tools in the present today, really are able to help us. Right is, you know, being able to go out and scour the web or scour knowledge sources for information on, you know, on your on target demographics or buying groups or specific industries, or, you know, even specific companies. Right, being able to do that, do that quickly, and spin that up quickly. We all know this stuff changes so quickly, being able to kind of get up to date type of things, even on competitors. Right, being, understanding competitors, creating battle cards. I was speaking with someone from a company who they're like, oh yeah, we have our battle cards with our competitors. Are automatically generated every evening, right? Like they're always up to date. It's kind of crazy, and it's all through AI and everything, right? And so it's basically, I think, today, really where, where we can really take AI into practical in a practical sense, is to accelerate and make more efficient some of the processes that we're already doing, whether you know that's if you're on kind of the content marketing side, right, creating that content and being able to make that content more relevant, preparing that content to be more kind of AEO and geo friendly, right? Or even on more of, kind of the sales marketing side, right? It's, you know, finding out, doing research and find out about your competitors, and all these types of things, the things that you're already kind of doing your day to day, or you would like to do more of in your day to day, be able to plug a little bit of AI into that, to be able to help you do a lot of that work faster. I think that's something that can start today

I'm going to answer with an anti answer on this one. Just imagine I'm a hammer salesman, and I show up at your office and I give you 50 hammers, and you have no nails, but you have 50 hammers, and I tell you to start smashing away. You can solve any problem under the sun. So what ends up usually happening is, if I give you 50 hammers, you're going to smash up your entire office, all the desks, all the furniture and so on and so forth. And this is the I say it half cheekily as the end answer, because I still believe that the foundations of it are strategy. Where do you want to get to? What are the things that you. Need to do in order to get there. Like, it could be AI, it could be a bunch of squirrels running in one of those things powering your kind of experience. It doesn't matter. I think AI is a tool that is not a master, but a servant in service of something. So my anti answer is, I think it starts with strategy. And you said something that is really, really useful to kind of think about it is like, what is stuff you're already doing? Where are the wins? Like, for example, battle cards, or whatever else the case may be. But it starts with stuff that can actually add value in making those interventions. So that's why it's strategy is not a dirty word to me. I think it's absolutely essential before we start. You know, just launching AI willy nilly everywhere, yep.

So my my question answer to that, knowing your marketing leaders, right? Well, our competitor is doing it. Why can't we use AI for everything? Because in my mind, if I'm the marketing leader and you're telling me, let's strategize, let's think about this. To me, all I'm hearing is slow, slow. It'll take time. And the answers I'm hearing from leaders are, well, if a competitor is competitor is doing it, why can't we do it? If they can generate with AI, why can't we incorporate AI everywhere, right? So they're not being afraid to be bold. My retort to them is great if your throughput for marketing campaigns or content or code is increasing 50 fold, you also need a human checkpoint at each one of those, which also scales with the throughput. So yeah, don't have people building it, but have people reviewing it. That's something that you shouldn't ever get away from having a human checkpoint, whether it's code or content, right? That's what comes to mind when I'm hearing you guys speak. Is governance, governance, governance and how impatient leaders are in how to get work done, is the scariest point to me. So actually,just to kind of think about this on this site, which is that I don't think strategy and speed are antagonistic. We tend to think of those things as being like those. But a great strategy can make you 10x in terms of your execution, because it helps you think through all of the kind of issues around the corner. So it's almost like a measure twice cut one sort of thing. So if you have a great strategy, you can be faster to market, instead of just going wild with all of these experiments, but having to scale them back, that's the only thing that I want to make sure that strategy and speed are not mutually exclusive.

Yeah. Yeah. I mean to me, like strategy is very much like you're defining, you're defining where the road, like you're drawing the map of where you want to go, and then the tools are just to help you get there faster, right? And without knowing kind of where to go, how to run that navigation, you know the even go as fast as you want, you'll never get to where you actually want to get to. I think the other point of this is, you know, how, I guess, how, how we look at AI tooling, right? Is, you know, definitely, from a from a governance perspective, AI is, I guess, to get to, kind of the the nerdy part about this, and kind of the underlying, underpinnings of AI, like, AI, the underpinnings of how AI works, and, like, how this magic of things like chat GPD works, is it? It takes stuff it already knows, and it pre, you know, it kind of regurgitates it back, right? It's not, it's not coming up with new ideas. It's not kind of synthesizing new things. At the end of the day, it really is taking stuff that you already have, or whatever is on the internet or data sources, right? And then kind of resynthesizing repackaging up into something nice. That's what AI is really good at. But coming up with new ideas or new strategies, probably not so much. And so, you know, if we're saying, like, I don't know, I'm maybe a little kind of pessimistic on this, but in my opinion, right? If you have four companies that all Do you know for competitors that all do kind of similar things, and they're like, they're all for them. We're like, All right, we're just going to run everything through AI, chances are pretty good. They're all going to wind up pretty close to the same thing, right? Like, they're the to me, the the innovation just has to come from external to AI. You can't, almost, like, depend on AI for that type of innovation. And I think, to ashram's point, like this is where strategy really comes into play, because I feel like the strategy is what defines that innovation. That's the roadmap to get you to whatever you want to get to from a destination perspective, if everyone's like, Oh, well, why don't all that AI define my strategy, chances are pretty good they'll point up in a similar spot. But if you're have a very defined strategy, and you know your market, whatever there is that, you know, human touch to it, then AI can help accelerate you to get you to that destination. And so I mean, like, I just feel like, in situations where everyone's like, Oh, our competitors. Doing this and that it's like, great. Do you want to be like your competitors? You know, I don't. We're not at a point where AI is at a, you know, is so novel and so innovative that we're, you know, we can kind of depend on them, to depend on AI to generate these brand new ideas. Rather, I think it's a it's we're still at a place where AI is still tool, right, acceleration, tooling, in my opinion.

George, to that thing. The meme that I think about is like, you know, when people go to Claude and, like, make me a billion dollars, there can't be any competitors. Don't make any mistakes, and you get any sort of reaction you might get, yeah.So a couple of days ago, so there's like so many thoughts flowing in my head, so I'm just going to spit it out and let you guys handle the thing from it. Google filed a patent a few days ago for an AI generated page specifically to do with how companies would no longer need to create a page. It'll create it for you. And again, the interesting part is all of the hallucination, factual data that needs to go in for depending on the sector, right? So that was kind of interesting. There was another post which came by, I think, on LinkedIn, a couple of them, really, where they were talking about they were able to hook up to Google ads, they were able to hook up to Tiktok to automatically generate this ad content and have hundreds and 1000s of variations of these videos. And again, the first thing which came to my brain is, whoa, who's looking at all of this to make sure that it is representative of your brand, right? Because they were talking about a brand. It wasn't like a personal thing, and that, to me, was a bit scary. And again, like the title of webinar is, what if this is the last website you've ever built? So going into that, if manufacturers want to stop rebuilding their site every few years, how should they think about designing a like a foundation that can evolve continuously, maybe as as AI capabilities mature, but as we mature, I guess. But how do you guys think they would have to build these sites? Aswin, you have to go first.

You are responsible for the title. I'm originally Canadian, so I always ask other people. So my thinking is as follows, which is, what, what is the web like? I mean, like, I don't mean to be glib about this, but I just want to ask this question, like, what is the web? Which is, it's a set of information that is being transmitted from point to point. I guess that's that's the kind of like a just a very crude definition of it. And when Netscape came along, we thought, oh, wow, the browser was going to be the way that we access this information. So you evolve from, like ARPANET to Netscape, and you start to go down this path, right? And then, for some reason, we thought that this was the natural evolutionary end state, as in, that if you're able to have like the way that the web is looking like, and we built some great applications on top of it. I'm not denying the great amount of innovation that has happened. So what does the future look like? I think there are few trends that are going to drive the the that are going to necessitate a different requirement of digital experience going forward. Number one, most of the people that are most of the entities that are going to be surfing the web, or whatever else this construct is going to be called, are going to be agents. I think it's going to be few, less, fewer involvement, more agents that are going to be as part of the process. So if you want to build a web that is congruent to their needs, so that they don't get lost, they don't get kind of trapped in different doors, you don't have to start thinking about, you know, I offered some minor stuff in the past, right? Markdown as a as a bare minimum, being able to think about how this can be easily scraped for AEO, all of these other kinds of things, I think, are going to necessitate the kind of spec sheet for what the future is going to look like a future website. But we are living in like, I guess Dario Ahmadi call this like the Centaur age, right, which is half human, half horse. So we shouldn't just build a digital experience just for, you know, AI agents and forget about humans. So it looks like a bunch of binary but it should also be something that humans can use. So the the way that I'm trying to make sense of it in my mind is like, you almost have, like, a sensor at the top, like, you know, when you know when you can choose different languages, it's almost as in, like, what kind of experience do you want? Do you want the agent driven experience, or the agent based experience? Or do you want the human experience? The human experience all of the things that we've told you for the past 10 years still hold, I'm not telling you like we should just. Burn everything down. But we're hitting that transition point where the web will become increasingly a playground for agents, and when you start thinking about what those requirements are, and over time, it will be increasingly agent driven with some human kind of things. So that's how I see the evolution of what we're seeing go on. And I know it's not a definitive answer, but I think this can start to kind of lay a foundation towards what the future might look like. George, I'd love to know how you're thinking,

yeah, no, I was nodding the whole time, because I fully agree with you. That's kind of exactly where my thoughts are at. I was when you're I was thinking about, like, when you were talking. There was that, I don't it was a meme or video or something that was, like, you know, you had, yeah, like, two AI thing, like, agents in phone call, like, with phones next to each other, and they're, they're like, talking to each other. They realize each other, AI. And so they, like, start, you know, beeping and, you know, going, you know, using a more efficient language to talk to each other, whether it's fake or not. You know, that's whatever. But the fact is, I think that's, that's kind of what it's going to become, right? Is we're going to fly, we it's, we're almost, we're using agents to write for other agents, in a sense, right? And so to consume. And so I like, I like, the how you kind of use that phrase, like the like, we're in, like, this Centaur kind of mode right now, where it's kind of like half and half. So obviously we still have to provide for the human experience. But I mean practically, how we go about this is, I think if we were to start a scrap everything start over today, right? I would put a lot more focus on content structure, right. And so, how do we structure our content? Like, Oh, regardless of what the website is going to look like, a what kind of content we we have be like, how do we structure that content and organize our content and see, make sure that we have that we fill in all the gaps of this universe of content that we want to have to be available, right? And then we can take that, and then it's, you know, how do we format that out, whether it's we make it a little bit more, you know, instant markdown, or whatever the case, to make it more, you know, Agent consumable, AI consumable, or we plug it into a CMS system that to make it more human consumable that, right? But almost separating that idea of content away from your delivery channel, right? Which, I mean, if you look at the history, you know, the past 15 years of digital experience, there's been kind of trying this trend already. It's like this whole idea of, like, multi channel, reusable content. We're just looking at agents as another humongous channel that's going to be coming in, right? And so I feel like now more than ever, this kind of decoupling of your content, not only just the decoupling of your content, but the amount of effort and focus put into the structuring of that content to make it organized and understandable by machines, right? Is, is going to be super important. And then, you know, I feel like, you know, the traditional kind of new website process is, oh, we'll just take our new website, we'll come copy and paste some text from the old website to the new website. We'll give it some new colors and make it a little prettier, right? And I think that isn't going to be sufficient moving forward, right? It really needs to be a rethink of how your all the content that's on the site, how we restructure that content, how we reformat that content, and how we organize that content, even from like a taxonomy type of perspective to be used in these various channels. Again, web is one of them, and now agents is another.

I do have one question that I want to ask both of you, right personalization is something that I think a lot about. Right personalization, it's assumes person right, that's it's talking to humans as the ecosystem becomes increasingly agentic. If we're to believe that claim, what does personalization mean for the agentic Web? Does that matter as a tool like is that something that's not going to be as important? Is that just an artifact of humans visiting these websites? So I keep thinking about that. I don't know if I don't know if I have a definitive answer to that, but I'd love to hear Akshay, what you and George, you're on to something there. What if, like, one of the channels George was talking about, is GE right. I want it to be crawled by all of the agents. If someone searches for something that I provide as a manufacturer, I want to provide a different coupon code to them, as opposed to if they come to the site directly. So sure, I would like to personalize it, even if a bot was trying to read my site and give them like a 25% discount, but if you come to my site, I might give you a 30% discount. So but that's also scary to me. Again, I keep saying scary, because unchecked personalization with financial, financial implications is is also a bit weird. But I mean, there's all kinds of personalization, right? We can do offers. We could do here's it could be free, things that we give out, like a white paper or whatever it is, but I do see personalization in the realm of both bots and human beings in your analogy, Aswin, that's my take on it. Anyways. Yeah, I'm kind of interested in whether we get to a future where we do get additional context from things like agents or chat, you know, through chat interfaces, right? You know, I was the couple of weeks ago. I was like, I finally gave up on chat GPT and kind of move fully into Claude, right? But I'd use chat GPT for so long, like, I know so much stuff about me or whatever. And so, you know, had this whole process where there was like, oh, you know, you know, you can transfer it over. You just ask it for it was kind of shocking how much information you knew about me, like when I had it spit out, right? And or do we get to a point where, you know, if we're through these interfaces, right, like that con, we can get to a point where that context can be passed on to whatever the target to our content, right? Hey, this is George, and he works in this industry, right? And he's this level of, you know, he has this job title, right? And, etc, etc, how it's almost like the dream kind of scenario for marketers to have that much amount of context to be able to personalize against right? And so I don't know if it'll ever get to a point where all that context gets kind of passed along right, to get to a point like that, but I mean that that would be kind of almost like a dream scenario where, like, the information that you give back from a content back to back to an agent is fully contingent on who's asking that question, right? Five people ask the same question, but based on the context of what this agent or this this platform, AI platform, knows about this person, right? To be able to personalize based on that, I feel like, is a really interesting thing that we can know because, I mean, that's always been the dream, right? How do we do kind of contextual signal based personalization. And we have that now, we have this opportunity, potentially, of having all the signals based off of a whole history of things that have nothing to do with your before they ever hit your site, right? Ever, it's because they're like, Oh, I'm, you know, putting in a chat, you know, some guys putting in the chat GPT of like, how do I get a promotion for my job? This is the job role I have, right? Like, and now it knows who you are and what job role you have, and being able to get some of that information for that contextual personalization, I feel like that's the dream, right, to get very high intent, high signal type of personalization. So I don't know. Maybe it'll never happen, but that would be incredibly cool to see from a personalization perspective, right? Like the first time someone ever visits your site you already know so much information, give them a personalized experience from minute one, which would be incredible.

Yeah, no, it makes total sense. Also, all of the privacy, incredibly scary. Yes, that word, bring that word back. You'll be surprised you're saying that in some cases. So say I'm looking for a car, or whatever that is, and I get to say, You know what? Use everything that you know about me to go find me, which I've which we've talked about, like I actually feel, no matter how much privacy stuff is there to protect us, which makes sense from the way I look at it, is it's more of don't sell my stuff to someone else. You know, I'm going to you Amazon for a specific reason. So I don't mind if I open the floodgates for Amazon and say, okay, you know, I want to buy this. Go find me the best thing of it. I'm willing to let go of my privacy barrier in certain cases, and I think people will be willing to do that too if you give them a chance, because it makes their lives easier, like giving the full context without you having to. There is one additional kind of aspect that I want to highlight here, which is that, who owns the data right? We're assuming implicitly, George, if I'm understanding you right, like a lot of that personalization is within the LLM that you're using, right? Is that fair to say? Would it? Would a brand want to own that is, that's, that's my question to you, yeah, yeah, no, I think, I think that's a very valid question, right? Like, does the brand want to own that, or is it just a matter of, you know, someone, the brand provides a whole bunch of different, basically, types, you know, variations of content. It right, depending on how something, the context of how something is asked, isn't necessarily. Do I need to know who that job role is, not necessarily. But I can say, like for this, these types of roles, have this kind of content go out. For this types of roles, have this kind of content go out right? And then, you know, for, for, for for an agent to come by and be like, Hey, I'm this kind of user, this kind of content comes back out, right? Like, I don't necessarily need to know about that user, but that that information is kind of owned by the agent, in my opinion, right? And it's just, we're providing that amount of content now. But I think the difference now is the interface, right? It's not that I have to have 18 variations of this content on my web page, like I have 18 variations of the content through an API endpoint. No one will ever see it until someone gets to it, right? And so from a user experience perspective, you don't feel like you're inundating your visitor, because I'm like, Oh, I have to cover all these different things for 18 different ways that almost gets kind of disguised and hidden behind, behind the technology a little bit, which is interesting. Yeah, maybe there's like a data exchange. You own all of the things. And when you go to a site or want to do something, because they Okay, here you can have a, b and c of my info from my exchange, and it goes and gets, I don't know, it'll be a very interesting there's a product idea right there. Yeah, you want to build it actually, I need a cut. But saying that though, like, so what are like from your perspective, what are the potential mistakes manufacturers are going to make when they're going to build these digital experiences, right, like, if they're using the power of AI if they want to get to a site that they never have to rebuild, what common pitfalls or mistakes that they should avoid?

So I can start with this, which is one of the things that became very clear in the chat that we had, is the, what is it called the usual wisdom of eat your vegetables and get 10,000 steps in I think that is absolutely true, more true than ever with the age of AI, because taxonomy, making sure you have a good command of that, how that looks like, mapping those out, doing the hard yards with the unglamorous stuff, believe it or not, sets you up for success. So I think that's the number one. Like, there's a big shiny object out there, there's 50 hammers that I'd love to sell you, but you have to do these foundational stuff in order to be able to think there. So that's number one. And number two for me is because, as a believer in strategy, I do believe that thinking strategically through what you want to achieve, the how just kind of sorts itself out, but it's the what that needs to be answered first into what do you want to be when you grow up as a brand? Or those are the kinds of two big things that I ask manufacturers to think about as they start planning their digital transformation strategies forward. George, yeah, I mean, I think it's a lot of the things that we've already talked about right is almost making sure that you're, you're thinking forward of kind of what this potential next generation of marketing is going to look like, and being prepared for, I mean, it's already kind of half here, you know, it's, we're already kind of half horse at this point, or half human, depending on which way you want to look at it. And so I think it's almost like ignoring that is at your own peril. You know, you want to just kind of go with the traditional route of of being able to, you know, of like, Oh, I'm just gonna pretty at my website, and I'm gonna do some new visuals, and I'm gonna, you know, even rewrite some of the content, right? Like, I think that'll get you, that'll get you kind of halfway there. But, I mean, I think Aswin makes a really good point of, like, going back to the basics, to the foundational stuff, right? And investing your time there, because that's going to benefit, I feel like is going to benefit everything that foundation is being built on. Really need to work on that foundation. And I think ignoring that foundation is what will get people is we were looking for speed, right, even if it's using AI, right? Like, oh, I want a new website. I'll have, you know, cloud code, develop me a new pretty website, and use, you know, use, use other tools, and use figment MCP and whatever, right, like you can get the prettiness out of it, and that's fine, but I think so much of it is going to be dependent on the foundation and really focusing on that foundational stuff In order to get you to that next level. Because, frankly, like, things are moving so fast that we can't really predict the future. What we think right now, today, at, you know, on what is it, march 12, you know, and 2026, it's going to, by June, it'll be something could be something completely different, right? And. So the foundations is, what's going to save you is, if you have the foundation, right? You know that top layer, we can always adapt and change and everything, and so, yeah, I mean, I agree with that. I'm like, it's, it's very much, you know, I think the, probably the biggest mistake, is ignoring the foundations and just worrying about the pretty top layer, right? Really putting that, putting the effort and investment into some of that foundational strategy and content work, just to kind of add, sorry, one second, I think, foundationally, it's a mix of like two things that, I think two emotional traits or personality traits that are going to define a frontier marketer in manufacturing, which is trust in guts. It's like you have to have a trust that your system is able to kind of do all of these great things, like we talked about George, like, all of the foundational elements. But guts can't be under emphasized, because even though there's all of this great stuff that's happening, there's got to be a sense of imagination and a sense of bravado in order to say, like, I want to be differentiated from the pack, so in order to kind of seize on that, I think that's very important.

Yeah, totally agree. Cool. So to circle back, and it makes total sense, right? So I think it's even with or without AI strategizing and restrategizing regularly doesn't hurt, because you are more you can course correct as you strategize, so that will never go away, and it should be there. The second thing that you guys are saying is, if you build the base right, which you're calling as Foundation, which is how you store the data, how you are able to differentiate between if it's an agent, I'll give you this way. If you're a human, I'll give you this way. If we have the base right, all the governance checks and all balances are right, then you could put as many pretty faces, or you could make it look 15 different ways on 15 different sites, but the course stays strong in terms of how the data is structured. Tomorrow, you need to adapt to give a subset of the data to another different channel. Let's call it comes along because of AI innovation, and so be it. It's easier for us to do that because all the metadata, taxonomy, everything is set in a way that we can extract the data that we want, and that's what matters the most, is what you guys are saying. So if I if a manufacturer does that, then they would no longer have to rebuild their site or restructure their site every few years. That's the promise George and Aswin are giving you guys. Yeah, that promise is valid for a week, by the way. Short Fuse on this you're building us that data exchange, right? That's going to be the other third part, right? That's right, the data exchange. It can even be a physical thing. You plug it into the USB and then transfers to being, you know, like a security key. But seeing all of that, though, like we covered a lot of things, I do want to be sensitive of your time and stuff. Do you guys have any other questions you want to go over? I haven't seen any questions come through the chat yet. It's totally up to you guys. I want to say that like, Thank you for the opportunity to have me on. If anyone listening now or in the future wants to get in touch with one of us or multiple of us, please feel free to do so, because, like, this is a topic that I think all of us are incredibly passionate about, and I'd be more than happy to jump on a quick chat with any anyone that's interesting.

Yeah, I totally agree. Goes for, goes for, definitely goes for me as well. And of course, you know, thank you for for this opportunity. I mean, I thought it was a great conversation. Learned, learned a lot. I hope you all everyone watching and listening Did, did as well, at least, at a minimum, get your gears turning a little bit of kind of what, what's coming in the future. But, yeah, I mean, I think it's a, it's a super interesting topic. It's something that, you know, we're, we could be somewhat right about or completely wrong about. Guess we'll find out, you know, in a week. Thank you so much. Guys. Have a good rest of your day. Perfect. 

Take care. Thanks a lot. Bye.