Michael Hartmann: 0:24

Hello everyone, welcome to another episode of OpsCast, brought to you by MarketingOps.com and powered by all the MoPros out there. I am your host, Michael Hartman. Flying solo as a host this week because we are recording this during Mopsapalooza 2025. So if you're there and listening to this later, we hope you had a good time there. Uh, Mike and Yumi, I'm sure we'll be back soon. In fact, we've got planned a recap, so watch for that when coming into your inbox. So today I am joined by Tracy Fudge. She is an AI operations architect and agentic workflow designer, and uh, she's doing that on her own with her consultancy AI by Thrive. Tracy has spent the last few years deep in the AI trenches experimenting with language models, automation frameworks, and what she calls agentic workflows. She helps marketing and operations teams not just understand technology, but actually put it to work. So today we're going to be talking about how to move beyond the AI buzzwords, not that those are out there, to build real usable systems that save time, spark creativity, and drive better business outcomes. So, Tracy, welcome to the show.

Tracy Fudge: 1:29

Thank you so much, Michael. I'm really excited, and I'm I am, I'm I'm missing MAPSA, but uh I'm there virtually and in my heart, a lot of my friends are there.

Michael Hartmann: 1:39

Me too. I have serious FOMO going on right now. Well, I'm glad, but I am I'm grateful that we've been able to stitch this together. I was going back to some of the conversations that you and I had to try to get this started. It went all the way back to it is now late October. Uh, and I think it's we started in February sometime. So um good things come to those who wait, right?

Tracy Fudge: 2:02

Yes, exactly.

Michael Hartmann: 2:04

All right. Well, why don't we let's start with this? Like, let's start with maybe a little bit of like a short version of your journey. Um in particular, like the last couple of years, like what led you from marketing and operations, uh, kind of the traditional one we were just chatting about this before we started recording about how so many people are still focused on platforms, um, and getting you into AI systems, agentic workflows, things like that.

Tracy Fudge: 2:29

Yeah, so it was it was an easy transition. Well, one, I've been in tech my whole career. Um, you know, I started in software sales uh before I really understood what software was. You know, my explanation was, well, what's software? And I was like, I don't know, I think it goes on hardware. So that's how far back it goes.

Michael Hartmann: 2:46

You're not gonna start talking about floppy disk drives, are you?

Tracy Fudge: 2:49

You know, there was a day when we when software was a new word. Um, and you know, so then fell into mops. Um, you know, I was managing uh social media platform for somebody, and they're like, hey, can you go manage this email platform we have? And that's sort of, and most people fall into mops by accident. I'll start. You know, there's no college courses or anything like that.

Michael Hartmann: 3:11

Not yet.

Tracy Fudge: 3:12

Not yet, yeah. But the um the the AI part of it was just something that was just intriguing to me. And you know, systems are connected via API. So the agentic part of it was already being done. We just weren't calling it that, right? So when I um decided it to go deep and really learn, you know, on what agentic platforms were, you know, I had known Zapier and Zapier, you know, nobody really knows quite it.

Michael Hartmann: 3:41

They told us that we've had people on from there. It is Zapier, like happier obviously fingers. Not that we're like, we're not this is not a promotional thing, but I think I now have that that rhyme in my head.

Tracy Fudge: 3:54

And I don't know how far back you've been using it, but I was using it before people did correct me on what to call it. Um, and uh so that's been gosh, and I mean well over 10 years. But um anyway, the the whole AI part of it, the agentic part of it was sort of an easier transition to sort of understand what this all meant. And obviously I was curious about well, what's an LLM? And oh, there's other types of LLMs. Oh, what do these frontier models mean? And then putting them all together and making the intelligence layer over the connections. Um, you know, and that's just that's much more complex than people realize because you go into the conversations around data. And when you're inside of an organization, sometimes marketing doesn't have all the data or the systems. You know, I it systems sit over there. So that's a big uh gap that's still there. And teams are starting to come together and realize that AI kind of goes over the whole thing. Um, and AI needs to work on the same set of data, uh, not the same fields per se, but let's all talk about the same data. Let's make sure that the pipeline reflects in the CRM, et cetera, et cetera. Uh, so that's really where where I got into it. And it's it's still a new, you know, people uh frankly don't understand it. They think it's just workflows. Oh, just flow build six workflows, and that makes it, you know, and that's not it either. Um it's part of it, but certainly when you're looking at a business, you've got to connect it all together. Uh and finance needs to have the same initiatives as RevOps, as Mops. I mean, it it needs to all flow together.

Michael Hartmann: 5:42

Yeah, it's I think that's the that's the uh the goal, right? The only thing you you said something that uh marketing sometimes doesn't have all the data. And I I I would push back a bit on that and say that I think most marketing teams have more data than they know what to do with, um, and probably lack skills and knowledge to be able to do any much with it is is like to me that's the bigger gap. It's not data itself, but the ability to understand it.

Tracy Fudge: 6:10

Yeah, um, and I think you I would say that you are correct in that aspect. When I think of data, I mean the the ability to action on it, to add some predictive analytics, right? Uh to be able to hold the data in a sandbox and just kind of play around with it, apply some machine learning, some prediction. And that way your lead scoring's a little bit better. But if you're still saying because person A watched Webinar X, let's give them five points. And it doesn't take into account all the other things that said prospect or lead did along the way.

Michael Hartmann: 6:44

Sure. Yeah. Yeah. That's interesting. Yeah. Um well, so you you kind of hinted at this when you were talking about this, but yeah, something you said to me, which I've been thinking about, is that you said AI isn't really new, that it's yeah, automation and data taken to another level, I think is the way you put it. Like, what do you like? What do you mean by that? Break that down a little bit.

Tracy Fudge: 7:05

So hooking up, let's say HubSpot to Clay or HubSpot to Salesforce, that happens for the most part, out of the box, right? API, API, field match, field match. Um so that's already sort of an ingenic portion connecting APIs. I think uh, you know, when you add in the intelligence layer, for instance, if you were to drop the data that's moving over to the marketing automation platform, drop that into a knowledge play, uh, an LLM node to do something with the data, make decisions, apply tags, do whatever. And maybe that when you drop it into say N8M, and I'm just I'm making I'm making reference to things that we all know or to a Zapier. But the point is that connection doesn't have to be direct anymore. You pause it, you do things with it, and then you map it.

Michael Hartmann: 8:00

Right.

Tracy Fudge: 8:00

Um, and then you iterate, and that mapping can change. So it's that part of it that is new, but the actual technical API stuff, even installing a platform like an N8M, things things are so easy, uh easier to understand once you've developed on one platform, you can do it on another. One CRM largely looks like another, you know, minus some capabilities, but the concepts are the same.

Michael Hartmann: 8:33

Yeah. It's it's interesting because you um it feels like well, one thing is like some of these terms get thrown around as if they're the same, right? And I like in my head, I try I've started to try to break down like um AI does not necessarily mean automation, right? Uh AI also I think generally gets conflated with LLMs, where it feels like there's a whole lot more that's under the AI umbrella. LLM is sort of one type of AI um modeling kind of thing, right? Where you've got you've mentioned machine learning. I think predictive analytics could be a part of it, or it's a component. But it feels like people kind of mix those up. And to your point, I think some people are just still have maybe not blinders on, but like their their focus is on like small relatively like let's do the replace this small thing that we're doing with something that has AI as in quotes about it. But um when they start to open up the the aperture a little bit, they start to get overwhelmed, right? So for those people who are like, hey, do you feel free to disagree with my my assertion there about like the mixing of terminology, which is I think very common in a scenario like this, but also like any recommendations for those people out there in marketing, marketing ops who are maybe feeling a little overwhelmed about like where to even start with this AI journey, if you will?

Tracy Fudge: 11:05

Uh well, most people that I know that did it, let's say within the past three years, because that's really where when I went in.

Michael Hartmann: 11:14

We it's crazy that it's been three years for stuff, Rick. It feels like it was just like the other day.

Tracy Fudge: 11:20

I um I actually started, there was somebody something kept coming in my feed, right? It was this no code, low code. And I was like, okay, this might be part of it. And I think um that is easier to start because I'm not a software developer. I can I've done Python, you know, I did a serious analytics course where I had to write Python code, but then there's stuff that does it for you. So, you know, that sort of skill set goes away. But the the whole idea of um AI just being connections, it could be um, you know, an LLM play. But in terms of starting, I don't start with the start with the low-code, no-code platforms. And I personally started on Maven and I stayed there for the whole time. The the courses uh people then uh did more courses and more courses. And plus you're with a cohort.

Michael Hartmann: 12:17

And I think um Maven is a like a training.

Tracy Fudge: 12:22

Yeah, uh not the platform. Okay, uh I'm just saying Maven was where I started.

Michael Hartmann: 12:28

Sure.

Tracy Fudge: 12:29

Um and they do uh, or at the time were really the only ones giving true how to start with AI classes. Okay. Um I think that that's one way to start, just real basic. What is AI? And now, fast forward two and a half, three years, you could get the same thing on a YouTube video. I mean, honestly, just break it down. I think the hard part is this. Our background is marketing and marketing apps. And so we're flooded with platform. Oh, Salesforce has AI. Look at these Salesforce agents, right? Spot has AI. Oh, everything's got AI. So I think there's a false sense of being AI when it's not AI. It is just the platform's version of AI. You have no control over the prompt, uh, little control over the knowledge base, um, and it's context, and you don't know what model is in there. Um, a true AI person that wants to add the knowledge layer has all the levers to temperature and knowledge base and prompting and system instructions and prompt chaining, and that's the limitation. So you might, and then another part of it is some of the AI in these things, it's just filling out fields, right? It's um it's just doing, oh, that field needs this data or that field needs this data. And you can do that now with a lot of platforms, you know, Airtable off is their uh AI component, Zapier's got one, they've all got one, but it's kind of filling out the architecture for you, which is part of the part of part of what AI is. But once the scaffolding is built, where's the knowledge? You still have to create that. And that cannot be done yet without a human that knows what the hell they're doing. Yeah. I think unless you've you understand data, and I I always I've always said this is a data problem. Um, understand not just where the data is, but what the data is, uh, what data matters, uh if the data changes what's the story it's telling.

Michael Hartmann: 14:47

Yeah.

Tracy Fudge: 14:47

Exactly. And we're not having enough data conversations.

Michael Hartmann: 14:51

Um well, they this goes back to like one of my absolute strong beliefs right now is that we just lack in marketing in general, in ops in particular, um the skill set to really do much with data, right? I think there's just not a deep understanding about how to do it. Um I want to go like something that's just sort of clicked with me that I think maybe is part of the uh the challenge for some people in this, in that we're so used to having what I would call, I guess, deterministic kinds of solutions. So, you know, I get a bunch of job titles, I'm gonna put something in there that I define the rules to say this is the job level, right? So that which is sort of inferred from the titles. And yeah, I think an a very obvious use case for probably LLM in this case, right, as part of a flow is to use that to categorize jobs, people with job titles into categories that make sense and use some intelligence on that, maybe with a little bit of human oversight. But I think, yeah, what you described, right, in some of these other ones, there's a little bit of this trade-off, right? You could either do that yourself and you have a little more control over what's happening inside the black box, or you take advantage of some of these tools that are being uh embedded within other platforms. You mentioned a few, right? But it becomes even less apparent, like what's happening within that black box. And so there's this lack of, I think there's this like, do I can I trust it? Can I not trust it? Yeah, how can I tweak it? Can I tweak it enough? And it feels like maybe that's part of the like the resistance for people who are struggling with it, where they, yeah, you've got less like, oh, I want to take advantage of the stuff that's being provided in the platforms that I'm used to working in every day, but it doesn't really give me, like, I don't know enough about it. So somebody starts questioning it, I can't answer it, which puts them in a position of not feeling comfortable about it. And then at the other part is like the other struct struggle is like I still have to, like, how do I fit in time to go beyond that? And that's I think it feels like that's maybe the rub that's the or the the the the things that are making it difficult for some people to go, like, how can I step in and take a step forward in this? So does that sound right to you?

Tracy Fudge: 17:06

Yeah, it's a gap. And I have friends at very large organizations, and um, they don't have time to go learn this. And I think when you're a W-2, uh you don't have a lot of time. We're all overworked. I had, I chose to take time off. Well, I have a side business, and that's what's been supporting me, to be honest. Um, but I've taken uh a lot of time and taken very uh a few projects, but it's not, I'm not working 40 hours a week for someone else. Uh, it's probably 20, and then I work the other, you know, even weekends to learn this stuff. And I think it is it's a real um miss of an opportunity uh because you've got really smart people that are working really hard in these huge organizations that are not getting upskilled in this. Because you and it's and here, here, here's the other element to this. When you go in and design a workflow, it's not something you touch today for an hour, you come back next Tuesday and you finish it. No, it's deep work, it's 12 hours sitting down, getting up just to go pee or eat and doing it end-to-end. And there's no, it's heavy, deep can't outsource that part to AI, right? You can't. And it is you you've got to remember where you are and you're deep into that. You're living inside the workflow. I mean, I dream in spreadsheets and workflows because you know, and not everybody has that ability to say, okay, you wake up at seven and you realize that it's you're gonna work on this until seven at night, and you don't stop. You don't answer the phone, you don't, you don't get on social, nothing. You're just inside that workflow. Um, and hopefully in that time you you have enough good data, good parameters to test it and it worked out. But some of that can just be, you know, beating your head against the wall to try to make it work. And um, you've still got to get it reviewed for the client, you know, for the client or for yourself or whoever to see your art because you're the one that designed it.

Michael Hartmann: 19:24

And here's the that's the main right, and that's hard to to test.

Tracy Fudge: 19:31

Nobody's done this before.

Michael Hartmann: 19:32

Well, and because if you're using so in my head, one of the distinctions I've started making is there's the AI comp AI can be uh there are AI components could be used in some sort of automation or workflow or agentic model, right? So pick the tool. But there are other like there are other more deterministic ones. And I think what again I get back to like when you're trying to validate, right? It's it becomes less of a um did it do what it was supposed to in in an exact way, or did it produce an output that I is expected even if I don't understand how it came to that? And that's a very different validation. And and then if you're not comfortable, if you're the kind of person who's not as comfortable with these sort of um what's wrong, right? Like less, like unless like, oh, this is this directionally correct, right? If you're comfortable with that as a quality assurance um level of quality or completeness, then you're gonna really struggle with this stuff, I think, right? Because a lot of times what I've seen, even when I'm just using, say, uh an LLM, Chat GPT, Claude, whatever, right? The output I get from my back and forth in terms of prompts, right? Sometimes I go, I could I scratch my head, I go, like, where did it come up with that? Right. It we've all heard about the hallucination thing, but also sometimes it just doesn't freaking follow the directions, right? I remember working on something where I was like, I want you to draft a document. I had we'd gone back and forth, I had a lot of stuff, dropped this document and came up with something. I was like, I like that, but I don't like this one piece of it, right? It was like I think I had done a header image or something. I said, Yeah, just go fix that. And it completely redid the whole fucking thing.

Tracy Fudge: 21:19

So yeah, and I was like, no, like that cost us down, right? Like that that you can't put the whole prompt in one place.

Michael Hartmann: 21:28

Well, and I hadn't, like I had gone, but like I had produced something that was probably 80% of what I wanted. It was in line. I'd gone through like as many steps. It was just like this last one, like the formatting, the layout of this document wasn't what I quite wanted. And so I asked to make one simple change and it completely like it redid the whole thing. And I was just couldn't understand why.

Tracy Fudge: 21:51

Yeah.

Michael Hartmann: 21:52

And that's the kind of stuff. So if you take that, right, where I'm like going back and forth and it took me time to do it, and you say, like, I'm gonna do something like that, maybe not exactly, and I'm gonna be make that as a part of an overall flow, then like, how do you test that and evaluate it? That's like that's where the uncomfortable like people get uncomfortable with how do I fit this into something where people need to trust the overall output of that, not just one piece of it.

Tracy Fudge: 22:20

Yeah. I was introduced to a platform called Cassidy. Um, it is it's a pretty cool platform. I use it mainly for uh doc gen. So creating um a workflow to create an RFP process or a workflow to create uh a presentation or talking point or or something like that. It's to do one thing is eight different agents or assistants, right? One of which is the JSON, and the JSON is gonna be how that document is formatted, and you tack that on to the end and you go, okay, I want the header to be this, da-da-da-da-da-da-da-da. And then and then it goes through that node, which is a code node, um, and then it's formatting. And that uh had I not gone through Cassidy workflow on a simple doc generation thing, yeah, I wouldn't have realized that no, no, we we first have to have a research agent. Oh, and then we oh, first we need to have the brand voice agent, you know, or wherever you put that in, how do I sound? So you've got eight different assistants to do one eight-page output. And I know that sounds possibly gratuitous, but it's how you do it.

Michael Hartmann: 23:44

Yeah. Well, I think so going back to my original question, to you, like, where should people start? It feels like the way to start is because you can't do all this in one sitting. If you're a kind of person who's busy with their day job and you can't do it all in one sitting, you could start to experiment with pieces of what might be a bigger process, right? And you like figure that out. And then because I it I I spent a lot of time trying to learn how to just use I have a bias towards ChatGPT, not for any good reason other than it's just the one like was one of the first ones. I've spent more time with it, and I'm open to the other ones. But I've done lots of individual things. It wasn't until I'd been doing that for a while where I sort of lifted my head up and said, okay, I keep hearing about this agentic right in quote stuff, you know, what does that really mean? And I played around with different tools. And I don't think that if I I think if I'd started with the N8Ns of the world, or like I I I've had pretty good luck with relay.app, right? There are others, I know there are plenty of others, but if I had started with those as opposed to learning how the LLMs work, I think I would have struggled even more because I would have tried to do what you said, like I would have tried to do like a whole bunch of stuff in one uh AI node of a process as opposed to breaking it down into smaller chunks, which does feel on the surface to be inefficient, but it feels like like now I understand that uh in order to maintain a level of quality or consistency, that's kind of what you have to do.

Tracy Fudge: 25:21

Yeah.

Michael Hartmann: 25:21

Which means then you have to understand like which it should be a strength of ops people, right? What's the process that I go through? That what's the well, how can I break down this overall process that I want to automate, have agents do into smaller steps so that I can trust the output at the end. Yeah.

Tracy Fudge: 25:40

Yeah, it's it's and you evolve to that point. We all start, we're all different learners. And for me, I like being part of a cohort, a learning cohort. And I didn't want to, I kind of paused in my mops world intentionally because I didn't, I felt like mops wasn't where I wanted to go. So there were a few mops people in the community where I where I ended up, uh, and I still am, but not many. It's CEOs are in there. You know, it's just different people having different questions. And I personally would be exposed to more use cases than just hooking up a CRM to a marketing automation plan. You know what I mean? Like I it and that's where that's where um so I went, you know, that's where the decision I made to get out of the mops mentality and the mops world because it's all mops. And while mops is very relevant, take the blinders off, right? Yeah, you just need to go and understand it. And it is it's overwhelming. It's overwhelming. I cannot tell you the it's just it's it's it changes the way you see everything.

Michael Hartmann: 26:57

Yeah.

Tracy Fudge: 26:58

It's not just AI, it's it changes everything. I mean, I look at things so differently now uh in the world, and um it it changes everything. And I don't know where to tell people to start that. I don't, I for okay, so fast forward today. If I were somebody getting started, I would say go to N8N, do one of their starter templates, and just build it. Just at least get the the click and the link and all that stuff, just understand it. Yeah, and mops people will understand that because most mops people, I mean, actually, I wouldn't say most mops people. There's a set of mops people that don't hook systems up. You know, they're not on the API side and they're not on the data side. But the ones that are will get that part. Um and I would start there. N8N's gotten a lot better. I was an NADN, I left. I didn't, I was not a fan of the community. Um, I didn't feel like there was a lot of help there. I thought it was a lot of bravado and not things that were true. Uh, and it turned out that was the case. A lot of people were saying, well, we'll come in and do all this magic, and and they couldn't, they could just build the thing inside of NADN and it was visually appealing, but it didn't do anything. So anyway, um, but now I think it's a lot better, and they do have things for people to go learn quickly. And I don't, and I think you just do these P these bytes of things, maybe an hour each, and just build and then and then move into the data layer. And that's where Airtable and Super Base uh or Superbase, you know, that's another one.

Michael Hartmann: 28:37

Do you know how to say I have no idea?

Tracy Fudge: 28:41

But they're very good. Uh and when I started to step into data and believe it or not, uh into knowledge graphs. So, and we could go into this, but AI has to know context and it has to know how things are tagged so that it can go and retreat it. Um, this is different than Google's algorithm looking for keywords. So AI has to have context. So you've got to have a defined ontology. And that means, you know, this is a rotter word for saying tagging, you know, you've got to know what things mean and then how they're related. Uh, and that's where triples come in. So AI can't move beyond that layer. So if you don't have the knowledge graph and the nodes and ontology, it can't work. So when that light bulb went off in my head, I was like, damn.

Michael Hartmann: 29:37

Yeah. So that's like con that's like con providing context to a whole nother level, right?

Tracy Fudge: 29:44

Yeah, but it's getting it right. It's so it doesn't hallucinate, right? Right. So you'll see, and they may not be called knowledge graphs, but it is.

Michael Hartmann: 29:54

It's gonna No, I it it's a good, I think like me, it's a good mental model, like in my I I can already kind of envision it what you mean, but ontology is a big word that I think most people listening or watching may not really fully understand.

Tracy Fudge: 30:06

Um it's getting it is giving the relation of how things are related.

Michael Hartmann: 30:13

Yeah, yeah. Um it's like librarian, right? Virtual librarian. Um not that anybody goes actually goes to libraries anymore, but um that's a whole other topic. Uh so I'm curious, like what are what are some of the like can you give us some examples? Maybe uh going back to early on, like things that you've done where you've kind of that were helpful in your learning and still were maybe productive, maybe weren't productive in terms of an actual thing you can apply in the work, but it was like a good learning experience. And maybe something more recent that you've done as you've kind of learned and evolved and and what you've seen as possible.

Tracy Fudge: 30:55

Well, I was lucky enough to have my own business to sort of act as a playground, right?

Michael Hartmann: 31:00

Sure.

Tracy Fudge: 31:01

So I automated um a lot of the front office, a lot of the back office.

Michael Hartmann: 31:07

Um for your business, you mean?

Tracy Fudge: 31:10

That's where I practiced a lot of it, um, where I learned um real life things that that were impactful. So um that's how I was able to. To do things live and play around with it. Not everybody does that. They're still so if you're in an organization, you've got their platforms, you know, you've got their stuff, and it may not, they may not have NADN, you know, they may not want you playing on an agentic platform to understand it all. But those the so the efficiency, so that's how I learned a lot of it. I just threw my own business, my own use case, I built my own knowledge graph. Um, I built my second brain, and that's all knowledge graph work. Um, and I'm actually building the second brain for someone as a client right now. And that's actually all encompassing because uh a person's second brain that's an evolved person, say a diplomat, um, does a lot of different things. So you have to connect a lot of different dots for that person. Um, but from a workflow standpoint, I've done lead routing, uh, which was honestly, it was easy. It was an easy problem for when they started talking, I knew right away, okay, that's the problem. We need the lead routing to be fixed. And all it was on the sales side. They couldn't keep their territories and their data and Salesforce up to date. And it was honestly, I created an SOP, uh, I created an ideal framework for it. This is how we want it. And they built it. And then we cut lead routing down from 48 hours to four minutes. Um, and then I did some predictive lead scoring uh for somebody, but that's on the mop side. And I've other done other boring things um that people might find are boring, but it is um just workflows and filing. Uh, you can build an ontology with a workflow now. Like to do it right, you'd have to look go say you had a thousand documents. Well, someone would have to physically go through all thousand and apply an ontology, relate it, build the triples and stuff like that. So there's workflows.

Michael Hartmann: 33:23

So sorry, you've used the word triples now multiple times, and I I was just gonna assume that it's a no I don't know what you mean. What does that mean?

Tracy Fudge: 33:31

So it is meaning that um Johnny kicked the ball in the street. You got Johnny ball street.

Michael Hartmann: 33:39

Okay.

Tracy Fudge: 33:40

Well, someone else could have kicked the same ball into the ocean. So it's it's the it's how it's related in triples, three things. And so long as you can match the three things, then everything you can see how everything is related. Uh and I'm not doing a great job of explaining it, but you'd have to see it. It's just applying relationships by three.

Michael Hartmann: 34:04

Okay. Interesting. Okay. I mean, that makes sense. I mean, triangulation on stuff kind of makes sense. Okay. I'm gonna be uh you're gonna see my heads like you're gonna see my eyes going back and forth because I'm still thinking about this. Keep going. I'm sorry.

Tracy Fudge: 34:19

No, that was it. It is, and I'm not an ontologist, so there are people, you know, as I stepped into this world, I'm in the other cohorts and taking more classes. Um, you know, I am I'm I'm kind of addicted to taking classes, um, to be honest, because I don't want to do it myself. I want to sign up for a class that I've invested in and it's got beginning and an end date, and it keeps me accountable. Plus, you meet people along the way, right? Um, which is a great way. And and community and networking is gonna matter now more than ever. Um it's gonna matter now more than ever because it's gonna be who you know.

Michael Hartmann: 34:53

Sure. Yeah.

Tracy Fudge: 34:55

And at my point you are how many degrees you have. Um, so I don't know if I answered that question.

Michael Hartmann: 35:02

No, I think I think I think if you did, I think I'm trying to think about like if I was someone listening, watching whatever, and um I didn't have my own side business, what could I do? And I think about some of the things that I've done, because most of mine has been uh more on a personal level. So interesting. So I've done automation actually related to the podcast, where um I have now a link for people to schedule this recording, right? And part of what I have done, this doesn't involve AI at all, which is interesting. But I did play with an automation platform that goes this looks for when uh that new thing is added to my calendar, it goes looks for new ones and it does a couple of things with it, right? Adds a couple of people to it. Um I've been thinking about like how could I take that even a step further? So I still do a manual effort of prepping. So I've taught uh in this case GPT, uh how I like to do the preparation document. So for people listening, right, we actually do a fair amount of preparation for this. It's not a total wing-it thing as much as we try to make it feel organic. But you know, that's still a fairly manual process based on some. You've everybody knows we we have initial discussions about these things, and and uh you know, I've got recordings. So I would love to get the process if I could set set it up where like it can go back and go find my notes and transcripts from our planning discussions to generate the document, right? That would be an automation that would would benefit for me. This still doesn't have taken me as much that much time manually, so it hasn't been a top of mind. But I have done other things um that are more focused on still manual but recurring things that I'm doing, or you kind of hinted at this, like doing research now. Um, I find it really, really valuable in doing research on stuff. Um to put like I literally had today, had people here, service repair people. We have an old uh cooktop having problems with it. And so I did a bunch of research on is this stuff I could fix myself, right? Did it I did it with uh we had a deep freezer that went bad over the summer. Um, and I did a bunch of research like what are the possible causes and is other things that as someone who's not super handy could could fix. And it's a it it made gave me a solution that was a $20 solution as opposed to a replace it for a few hundred dollars, right? Which is beneficial. And I probably could have done that research on my own, but having the LLM do a bunch of that for me, come back with stuff, and then I could refine it. Um I've even done stuff if you're a kind of person who likes to cook at home. We took photos of all the stuff from the pantry and said, come up with some recipes. Take a picture. That's what we did. Yeah, it's really actually quite good at that. I found it's not so good at generating photos, but it's or pictures, but it's really good at deciphering what's in photos. I I've been sort of blown away. I didn't believe it would work well, and it's been really good. Yeah, so we actually did that and it generated a number of recipes, and we tried one and it was terrific, right? No, no, will everything be that way? Right?

Tracy Fudge: 38:11

I doubt it, but um so I think walking into your kitchen and all you have to do is talk to your kitchen, yeah.

Michael Hartmann: 38:22

So I think I think I wouldn't yeah, go ahead. I think if people think about like how could I try to learn some of the stuff even if I don't have a site, like it doesn't have to be a business context. I think you can learn a lot from doing things, and there are like I've played with connecting like my email to one of these automation platforms using an AI kind of engine to automate drafting replies and things like that, or categorizing things as important or not important. Um honestly, I'm a little bit of a like I'm paranoid and like just like the amount of access that those tools currently have to have to things like your inbox, not comfortable with personally. So, but I know it's possible. Like, and it's definitely uh like that's a very common use case that most of these automation platforms will give you as a starting point. Like open up your email, classify the emails, draft the reply to those ones that you think are important, and then you can go in and and then send you a summary on a daily basis. Pretty common thing you could do. And um, like I think there are things that people, if they thought about it, even if they're not doing a work-related one, because there are concerns about you know uh security and proprietary, you know, confidential information, things like that, that would probably make it harder. Um, but there are ways you can learn. So I think that would be high encouragement is to do that kind of thing. So I oh I get so one of the things you have I think encouraged me, um, and I have yet to to really do that, and I've already mentioned right that I have a bias towards Chat PT, GPT because I'm most familiar with that, and then I particularly like um Grok because you can put on the uh what do they call it, a persona, I think? You can do the unhinged comedian one, and it's like I just that's more of an entertainment thing than anything else, just to see what it comes up with. But um how do you like what what could you share about what you know about some of the let's keep into the LLMs with this, I guess at this point, like what like what do you see from them? What do you think that they're like what are their strengths and weaknesses? Because it feels like they all have sort of various things, and I know it's evolving, so like by the time this gets out, it's probably gonna change a little bit. But in general, what's your take on those?

Tracy Fudge: 40:42

I um so I I don't really use grok. I have in the past, but not to the extent. So my main ones are ChatGPT. I just pay the 20, and then Claude. I I think I did bump it up to the higher one because Claude um has some memory and context issues, and I hate having to start over in the middle.

Michael Hartmann: 41:01

Uh yeah, yeah.

Tracy Fudge: 41:02

So, which is very irritating. Um, and then Claude's limitation was not being able to hold the context inside of folders or projects. You know, you can create folders. So I have a I have a huge folder structure, and I was actually arguing with ChatGPT. Because it used to be able to drop into the folder and ask a question and uh about anything in that folder. Well, today I found uh that it didn't it didn't have the context of a conversation that I had just had in that same project.

Michael Hartmann: 41:34

Yeah, I've found that I've I've had to give it explicit direction, go to this other chat that's in the same folder.

Tracy Fudge: 41:41

Yeah, and that was not the case. So I don't know what's going on. Maybe they realize that they let too much out with chat GPT that this is a lot of power requirement to make your, you know, the return of your request happen. Um but I use both. I actually um I intentionally did not do Claude Desktop until a month ago or two months ago because I didn't want to get too advanced because I knew that that would then uh I would build MCP servers. So I've evolved into Claude Desktop, MCP servers on my on my desktop to go into these systems uh for me. But that's not for everybody. You still have to know what the hell you're doing. Or what it's doing, you know.

Michael Hartmann: 42:31

Yeah.

Tracy Fudge: 42:32

Build, you know, build the JSON to be, you know, uploaded into, you know, uh N8N. I mean, you've still got to know uh a few things about that. But that that's what uh and I use Claude for that. So but I still go back to chat GVT for creative. I've I've started to write more often. Um I love to write and I've started to write more often on LinkedIn and I've done it.

Michael Hartmann: 42:56

Do you find it do you find it better at drafting initial stuff with a lot of context or you drafting stuff and now asking it to um give you feedback on it?

Tracy Fudge: 43:07

I've been training it for a while. Um, and assuming it's still got the same project context. That was one of my things I was frustrated with uh because it has my brand voice, it has my last post, it knows sort of where I left off last week. Right because I get one loud. I don't, I I I took last week off and I was like, I'm not going on LinkedIn. So I took a break. Um, but it's still so that for that I use it. Um uh I use Chat GPT for that. But Claude, I use for the technical stuff. But Claude's gotten better uh with the creative and the writing um as well. And my you know, there's so much noise, and I know everyone says this, there's so much AI noise, and now with all these layoffs and things, and I just I want to post something that matters. It's just not more of an echo of what's already and sometimes that's hard.

Michael Hartmann: 44:02

Yeah.

Tracy Fudge: 44:03

Because you want to call out the noise for what it is, right or wrong, and then you know, have your point of view. And it's it's hard to get a point of view sometimes when when it's so noisy out there, you know. You don't want to, you know, it's hard to not be afraid, you know, these types of things.

Michael Hartmann: 44:19

So yeah. It's interesting because like my biggest, my latest challenge, even though I've got some folder structures inside ChatGPT, because I'm also at a paid level, I think that's a you have to have at least the basic level folders. Yeah. Um is that even within that, I've got like I'll I I I actually did something with my wife too, for some she does in her work, which there's ongoing doing a lot of the same things, and so having context is really good. What I found is when I it goes on for a long time, it um chat PT GPT tends to lock up, right? If I if it's if the chat thread gets too long, and so that's my latest one is like how do I retain what is like I did a lot of work training it to voice and stuff like that. Um but the performance on new asks is still there, and that that's my latest challenge. But um it's what's a screened this week.

Tracy Fudge: 45:17

I don't know what it is.

Michael Hartmann: 45:18

Well, I've been noticing it for a few weeks, so I don't think it's anything specific to this week um for me, but um and and I think it's just just that, right? The the threads are just so long now that uh it may just be browser, right? The browser rendering it takes too long and it's just locking up. So uh but again, like this is like part of the learning process that we go through, and it it would help me knowing like, hey, if I'm building some sort of automation for a work context and um I wanted the context to be there and available over time, right? I need to think about how to structure it so that I don't run into the same thing, an automation step where I'm like, hey, asking a whole bunch of stuff to a say uh uh LLM engine, and uh because it's like if it's building it out over time, is that gonna cause that to slow down and not run correctly? So things like that, right? So I've kind of tied tying that together. Um let's let's let's wrap up here because you kind of alluded to this. Like, how how how can people think about what's the right way to start to incorporate AI, LLM, you know, automation slash agentic stuff into their, let's keep it to Mops, RevOps, whatever, uh, into their world where it's it's a complement uh or a um I like the term I heard somebody use, like it provides a time dividend, right? To like into their their day-to-day um work effort, right? How did like what are some ways they can do think about doing that?

Tracy Fudge: 46:56

So are you asking from the perspective of the mops individual or mops?

Michael Hartmann: 47:02

I think I think mops people as individuals, right, but as part of a team, even if it's a team of one or two people.

Tracy Fudge: 47:10

Mm-hmm. Content is a big thing. Um, and I think content's really relevant. Uh, but you can't just put it into Chat GPT and say, write an article on this topic. I mean, you've got to iterate it, you've got to give it examples, you have to understand. But I think getting content out for the team um for um, you know, email is another one. I have not ventured into email for exactly what you said. I want to go in and read my emails and things like that. And I'm scared to, and I I use uh Google Workspace and I'm scared to turn on Gemini and my email. I don't know what it's gonna do. Right. Um, but I do think that, you know, find a few workflows that are for you and build those out. Uh for the business, if it's if it's easy to show a return, then do it. And you know you could show it because you want to show ROI. And that's gonna elevate you to leadership as well. Look, I I shortened your lead routing. We could do this. Look, let's um, and do some predictive analytics, pull some data out of sales if you can, you know, the last four campaigns. Uh, what else did those people do that converted to a sale? Did they visit uh a certain and besides the webinar, did they do these other things?

Michael Hartmann: 48:37

Um looking for pattern, like looking for patterns tied to exactly uh the desired outcomes.

Tracy Fudge: 48:44

Most companies aren't capturing those events, right? So you've got to figure out, okay, can I pull in data on the white paper they read, not just the webinar they signed on? Can I pull uh and we already know what emails are opened?

Michael Hartmann: 48:57

Um those you could provide it, you could provide it not only the data, but also the content, right? The examples of the content. This here's this content, it's tied to this data point or this kind of activity that's in the data versus this other piece of content. And um you could then tell it to look for what are patterns in the content that uh are tied to um better conversions, more pipeline, like whatever that is that you're trying to optimize for.

Tracy Fudge: 49:31

Possibly look at the last thing they did, like they looked at an ROI white paper or something. Let's just make that up. So then you have the AI note that, right? But then craft a draft email around RLI. And it's something they just did, so it's fresh in their brain. It may not be the thing that pushes them over the to be a uh cli uh a paying client, but it's something that's relevant. Um, that's something that a lot of people uh could easily do.

Michael Hartmann: 50:01

Yeah. I mean, I think I think what I would do, like if I was in an organization that had like product marketing and or just content marketing teams, and um we had a sales team that had some sort of call recording platform, I would be like, that sounds like such great context if you could capture the transcripts from those sales conversations, then tie it back to which of those and ended up turning into deals. Mm-hmm. And capture the language that's in there, both client side and sales side, to feed back into how are you thinking about the content you're publishing and the material you're providing to the sales teams, right? Because you because like I think there's so much crap out there on B2B websites in particular that is like just gobbledygook nonsense that doesn't really tell anybody anything about what you actually do. Like if you could drive, like if you could do that, take that back, feed that back into your content engine, um, which then hopefully is gonna drive more traffic and awareness, right? Like you could build a pretty strong flywheel, um, I think in terms of that that's just simply based on what are the actual terms and uh things that uh customers, prospects actually care about, and make sure that you've got that really well honed on your public-facing website and the content that goes with it. Like to me, that's like it's it's not trivial, I get it, but like it talks about like uh an ROI over the long run, both short term and long term. I think that that one to me is a use case that I would love to see more people trying out.

Tracy Fudge: 51:38

And then you could retarget people to go to that new dynamically created blog post that was created, you know. So these things are dynamically um done because there's APIs in there that are doing it. Yeah. So there's the API work, but absolutely.

Michael Hartmann: 51:53

Yeah, so anyway, well, hey, I it feels like as so many of these are that we could go on and on and on. Uh, but unfortunately, we do have to we do need to wrap up. So, Tracy, first off, thank you for sharing. It's been a lot of fun. I enjoyed the conversation. I'm so glad we were able to make this work uh while everyone else is out there having a fun time at Mopspalooza. But if uh if if folks want to connect with you or learn more about what you're doing or go deeper on this topic with you, what's the best way for them to do that?

Tracy Fudge: 52:24

Uh find me on LinkedIn. Um, my website's AI by Thrive, and I'm Tracy with an EY at AI by Thrive, if they won't email me.

Michael Hartmann: 52:33

Perfect. Well, thank you again, Tracy. Appreciate it. Thanks always to our listeners and now watchers now that we're uh going live with videos. Well, we appreciate your support. And as always, if you have ideas for topics or guests or want to be a guest, you can reach out to Mike, Naomi, or me. We'd be happy to talk to you about it. Until next time. Bye, everybody.

Tracy Fudge: 52:54

Thank you, bye bye.