Metadata Demo

Julia Nimchinski:
And up next, the future of marketing today, metadata. Welcome to the show, Lisa Sharapata. Pleasure having you here. It’s good to be special title. I need PO by AI and GTM strategy. What a treat. How are you doing? How was last week? And yeah, excited to dive in.
Lisa Sharapata:
Yeah, well, it’s great to be back, thanks for having me, and, you know, joking a little bit about my title, because the panel I was on earlier, we were talking about it, but, no, last week we hosted an Agendic go-to-market three-day event that, you know, just brought in a lot of, like, really showing of what people are doing, like, really peeling back the layers on, really, the marketing side, very specifically. So, it was great. Honestly, saw… it was… it was everything I hoped it would be and more, so it was great. Yeah, thanks for asking.
Julia Nimchinski:
Amazing. What’s your biggest takeaway, and what excites you the most, transitioning to the next year?
Lisa Sharapata:
Yeah, well, I actually… I thought I was putting together predictions for this week, so I’d been working on that, and in talking with a lot of people, and just everything that I saw, I did put together predictions, and I’ll kind of go through that as I’m showing what we’re doing at Metadata, but… really, the biggest thing I’m seeing is that people are still trying to figure this out, how to build, what to build, what do you need agents for, and then everything changes again, right? And so, I think that there’s that piece, and then there’s a really big balancing out with what we were talking about on the panel, brand, and events, and that need for, I think, real… real hand-holding, and people to have the confidence that they’re making the right decision, and it’s, you know, an invest… like, what they’re buying is the right investment for their organization, and I think that there’s… the struggle’s real, right? Because so much is changing so fast. you know, I feel like people are a little bit, paralyzed by knowing that in 6 months, everything that they’re doing could be obsolete. So, yeah, that… those are the kinds of themes that were coming up quite a bit.
Julia Nimchinski:
Definitely. Well, let’s dive in.

Lisa Sharapata:
Alright, well, great. I’m gonna try sharing my screen and see if I can do it as a… a, like, a regular slideshow. Otherwise, I’m gonna have to… whoa, whoa, whoa, hold on. Is your brochure, right.
Julia Nimchinski:
Yep.
Lisa Sharapata:
Yeah, let me just see if I can… Yeah, it’s not gonna let me, it’s blocking the whole thing out. So, alright, we’re good. I’m just gonna go… So I got all my tabs open. So, alright, well, like I said, I started off with some predictions, and… I really am big right now into the show, don’t tell, even though I’m gonna tell, but I’m gonna show at the same time, and I really want to bring some of these things to life. So, starting with this idea that the MQLs are out, this whole top of funnel notion. And we’re really moving into a place where marketers can be getting to pipeline and revenue. And then moving away from, I would say, even a reactive or predictive models into agentic workflows, and then I… my big thing is, I think 2026 is really going to shape the market. I think that things will start to kind of… Settle a bit, and we’ll see… what’s gonna happen. So, with that, this notion of prompts to pipeline, I mean, first of all, because if you’ve got all of your data sources unified into a central system, you’re now able to see everything coming through from You know, a touch out in the market, an impression. all the way through to what happens to that account at the end, and so… with that, I’m gonna… I’m gonna actually dive into Metadata Platform and show some things with the MCP server. So to get started with that, and just to try to help bring this, to life a little bit more, I’ve already built these out, because I only have 15 minutes, and I won’t have enough time to go through it all, but The first thing I just want to show is how this is set up, because, you know, still, I think people are trying to figure out, well, I’ve got these agents, how do they work together? How does this become an Adventix system? And there’s different ways you can build this stuff out, but in metadata, what we’ve done is we’ve built an MCP server connector. And what you do is you just come in, manage your connectors, add a connector, and we have it turned on here. And then there’s, like, 70 different agents here, 71 agents here built out, and they’re all… put into this connector, right? And then they’re connecting back to the metadata platform. So, now you’ve got the power of your LLM, plus the power of that platform, and all these agents that can talk to each other and work together, baked in. So. The first thing I just want to start with is this idea of we’ve got… I’m going to start with the prompt, I’m going to analyze. I want to see what was happening last quarter, so that I can get my board slides ready, but also start thinking about what I’m going to do next quarter. So, in this case, I’ve got my prompt in, I’m asking for performance analysis. And what is happening is my agents are actually going in to my metadata platform, they’re looking for all the different performance stats. and then… and the metrics, and they’re actually starting to… we’ve got, like, these skills created now within Claude, which was kind of a new thing, like, a month ago, and I feel like, it’s like an agent… version of a Claude piece that’s, like, similar to an agent that you’d build in ChatGPT. So, it’s pulling in… that information, and then it’s starting to put together this analysis for me. And this will give me direct links right back into my metadata platform. So, I can dig in deeper. I can come in here and see what actually happened. I can look at this campaign closer, you know, all of that is at my fingertips if I want to. But, I don’t need to, because, you know, all of that information was analyzed and synthesized for me, but if I need to fact check it, it’s here. So, I can start to see all of this information. And I’m like, okay, great. Now, I want to take this a step further. Take everything that you just analyzed, and I want you to now put on your creative digital advertising pat, and I want you to look at the latest advertising trends. and generate 3 new brand creatives for me. So, since the last time we talked, Julia, we now have our creative agent working, and it’s… it’s really starting to create some good stuff. I mean, I think… That’s been one of the biggest challenges for most people out there, is on the creative generation side, so… The first thing is we’ve got, you know, it’s going out into the market, it’s looking at what are those top trends, it’s pulling in that information. Now it’s pulling in the specific metadata brand guidelines and brand kit. To create new metadata ads based on what performed well, what’s trending, and our brand. So… You know, it’s going in, it’s… You know, doing all its beautiful work. It’s starting to pull in images, from the internet. It’s, you know, it’s going to town here. And, so now it’s generating these 3 different creatives that I asked it to create. Including the copy, the headline, everything that I need. It’s sending me a link to my platform of… because it’s storing it in the asset library for me, but if I just want to peek at the… the asset that it created, it’s all generated in here, or I can go right into the platform to look at this stuff, so, I mean, that app was beautiful, I would use that. Like, I’m, like, so excited about it. And they’re not always, but then you can iterate, right? And so, you know, I’m just gonna… Pause here and… and kind of go into my next prediction, but you can start to see how I’m now agnically able to analyze, prompt, and get to a place where I can see how much pipeline something’s created, how much it predicts it will create, and create and deploy this new campaign. I could do that right from Baud. So… So cool. But I’m gonna go another step further into this, because… so that’s kind of a big picture thing, but I also want to start breaking it down into more of, like, this… how are we shifting from predictive to agentic, and how are we getting away from being reactive to, again, using agendic workflows and a, agency, basically, to… to take off some of this heavy lift for marketers as well. So… With that, I just first want to talk about, like, so here’s an example. Day-to-day, a digital marketer, digital advertising specialist, they’re going to be required to be looking at their bids. Are they performing? Are they, you know, getting outbid by a new competitor coming in? They might also be tasked with doing some A-B testing and managing all those campaigns and trying to figure out, like, what’s working and what’s not working. It’s very time-consuming, and, you know, it’s kind of soul-sucking, right? And then. Another piece of that, and we talked… we just looked at campaign creation, but… There’s no way that a human can Test and monitor all these A-B tests and change their bids at the speed at which things are happening, at the speed at which an algorithm’s changing, or, you know, going in and dropping things down and picking new audiences and setting all these things up. There’s… you just… you can’t keep up today with how cumbersome the technology is, so… You know, this shift to Agentic is really getting more human-focused and allowing people to be more creative and set the vision. Well, you’ve got these agents going in, and instead of doing these one-to-one-to-one tests, you can run multivariant testing with AI and machine learning all at the same time. So you can pull in multiple offers, multiple audiences in multiple different, channels, and so… Oh, here’s my creative that came up. But here’s an example of what that would look like. So… In this case, we’ve got these… Variety of different audiences created in each of these different channels, with different graphics, different campaigns put in, and different offers. And what, again, is happening is the agents behind the scene, they’re figuring out, okay, I’m gonna test this audience with this ad, and this offer, and this channel. And it’s running all these different experiments at once, figuring out which works the best, and then it’s making the adjustments for you to optimize for what is working the best in each channel, and it starts to turn off the things that aren’t working, turn on the things that are working. And then, the last thing I’ll show is that bid agent. So, that’s kind of the big picture. This gets granular. And so, we have a bid agent turned on, and you can pick which things you want it turned on for. And so, in this case, I’m just showing Like, here’s some example from today, where we have the agent turned on, and it’s going in And it’s… changing the bid. you know, in this case, 30 cents, in this case, doing the math really quick, but, like, 11 cents, or whatever. So, you know, these are variations that really no human would even consider doing, because you don’t have the time to mess around with that. But it’s doing that all day long, figuring out what is the most optimum number to get the outcome you want, and you set the outcome. So you could say, oh, I care about cost per lead, but you could say, I really care about seeing lookalike accounts that typically turn into revenue for me. And that’s what I want to see this optimized towards. And so, again, just this ability at the fingertips of a marketer to turn all this stuff on, like, directly from Claude or ChatGPT or in your metadata platform. Is, you know, a game changer. And then, the last thing I want to leave you with is… a little teaser of what I’m seeing coming into the market. So, I mean, I do think, again, this next year is gonna be… kind of where everything… the dust settles. But, from a marketing standpoint, the thing that we’re starting to work on here is basically what I call vibe marketing right now, and this is a prototype of it, but you can come in and you can just start. Clicking on, like, what I’m trying to do. And it will help you, you know, get… All of this information. You know, ad creatives, The next step of things generated for you right in your flow of work. without having to… to really… to know all of those different things. So… you know, I think as we start to see how things shift out, or maybe new companies emerge, and they don’t have a full marketing staff yet, or whatever, but, this, I think, is the future of how, you know, things are gonna start to build out of the art of the possible. So, that’s what we’re working on next. I’m so excited.

Julia Nimchinski:
Phenomenal demo, Lisa. For somebody who spent, like, a decade in marketing, this is the dream, so… really love what you’re building. I’m curious, before we close up. What are you excited the most? What feature? Anything on the roadmap for metadata? Just as a teaser for 2026?
Lisa Sharapata:
Yeah, well, what I just showed, the Vibe marketing piece, I think is… is really cool, and I’m the most excited for that, for sure. But in general, what I’d say is… I’m getting a chance to be more strategic and more creative than I’ve been in a long time, and I think it’s because that ing part of marketing is… getting off of my plate, and, you know, even as, like, a CMO, I was still having to really… to get into the weeds a lot, and my team was too, and it took away kind of that soul-sucking part of, like, being in spreadsheets and stuff really took away from the creativity side, so I’m starting to see that emerge, which is what I’m the most excited for.
Julia Nimchinski:
Amazing. Thank you so much.