Service X-Factor
Service X-Factor is the podcast where service operations stops being a cost center and starts being a competitive advantage. Hosted by Microsoft MVP Scott LeFante, this show reveals the secret ingredients behind service operations success and much, much more...transforming chaos into clarity and strategy into profits.
Service X-Factor
AI @ Work: From Buzz To Business Value
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
If your CRM could listen, learn, and log the work for you, how much time would your team get back? We sit down with longtime Microsoft MVP and AI futurist Chris Cognetta to unpack the shift from forms and clicks to agents, conversation intelligence, and contextual apps that actually serve the way people work. From hard‑won lessons on non‑determinism to real talk about governance and security, this is a practical blueprint for using AI without breaking your business.
Chris traces his journey from infrastructure and ADFS to dual MVP honors in Dynamics and AI, explaining why CRM got “smaller” as the stack expanded across Power Platform, Copilot Studio, Azure AI, and Microsoft 365. We dig into the skills that matter now—prompt engineering, human‑in‑the‑loop validation, and document classification—and why these fundamentals turn hype into repeatable outcomes. If you’ve ever heard “we don’t do AI,” you’ll learn how to build safe sandboxes with synthetic data, define connector‑level guardrails, and launch two‑week MVPs that prove value fast.
We also explore the next five years: agent‑driven sales processes that auto‑stage deals from call context, on‑site cues that prompt what matters, and domain LLMs shaped by your tenant data through Microsoft Graph. The result is a workday where the system already knows what happened, leaving humans to decide what should happen next. Whether you lead a services org, own a CRM roadmap, or just want to reclaim hours from busywork, you’ll get concrete steps to start small, scale smart, and keep costs predictable.
Ready to let the bots do the clicking so your team can do the thinking? Follow the show, share this episode with a teammate who needs a push to pilot AI, and leave a quick review with your top takeaway.
Venting And Team Dysfunction
SPEAKER_03Okay, how are you?
SPEAKER_00So is this a video or a voice? Just voice. Okay. Okay, so I took a shower for no reason. Well, you can go on video if you want. I got a vent, man. Oh, what's going on? Just yesterday's call? Oh, yeah. I I wasn't prepared for that either.
SPEAKER_01I was like, I what need you you heard the intent many times. What did you think the call was about?
SPEAKER_00I I thought it was going to be more leadership type focused. You know, what could what do we need to do? Not so much what I need to do, but you know, from an organizational perspective, whether it's you know, land, expand, delivery, whatever it may be. I wasn't prepared for like, hey, what's your big rocks for Q1? I was like, uh, it's like, I don't know. I was not prepared for that.
SPEAKER_01Yeah, and uh we we we really all had it out, man. So we have a lot of dysfunction in different areas of different pieces. A lot of it is because of, you know, and I'm talking to you off the record. This is Chris your friend, not Chris the business guy. Nope, but we have a lot of challenges with Avi, him not being around, not really being plugged into delivery. Um, it hurts us on all of our profits, so we don't make profit because the team's not managed. The team doesn't focus on a billability target. Have you heard the word since you've been here? No, not at all. Right. So as long as you're on that customer and that company's happy, Mike and Avi check off the box. But we don't get any bonus because those people, the 40 to 60 hours that those people should be putting in and putting in 25, the rest of the time they're off jerking around, they're not delivering AI, or who else has delivered extra stuff that you know of besides that core little group? Who else? Nobody, nobody, they don't follow the releases, they don't they don't follow it, hasn't and you know what? It's not just AI, Scott. It's been happening since two years ago. Yeah, they never even follow Power Apps. Oh, Power App's gonna be nothing. They're still doing 2011 CRM in their heads, yeah, and that is on Mike. He owns the people, he's supposed to manage that. Like we're supposed to you, you know, you've been in many of these calls as a leader and manager. Top top, who's the five percent that we're putting on plan? Remember the conversations? Yeah, we don't do that. When have you heard about it? Nothing. There's no pip, there's nothing like that. So there's some changes coming. Marty is one of the one of the changes that will be very positive for us because he's the sales guy and he knows what's happening. Um, and I think that'll help us get the sales engine moving faster so that we have more deals coming in and we can grow. Right now, we're strangled by what the team can do. Yeah, yeah. And and it's tough, right? But I can tell you that in five years, yesterday is the first day that I updated my resume.
SPEAKER_02Wow.
SPEAKER_01I'm not, I'm not, I don't want to scare you. I'm not telling you to go jump ship. Uh if I do decide that, you will know well in advance and time for you to go do what you need to do. Yeah, but that's not the plan. Yeah, I have too much invested here.
SPEAKER_00No, I I totally totally get it. Now that money, emotionally, everything. So I think I think Marty's gonna be good. I I even see it, for example, Peter. Like Peter's really sharp.
SPEAKER_01I I like and that will fix sales, and that's been a big stress. We'll fix Avi. We'll get someone that's gonna do finance and accounting and all the other stuff. You know what I mean? Yeah.
SPEAKER_00Yeah. So I think Peter's gonna be good in helping out. You know, like I was talking to him the other day when we were on a call with Tom and Tom was running it, and I said, Hey, look, if there's things that you hear that I should be doing better on some of these these calls to help out, I said, You need to let me know. I said, because that's the only way that if there's something I could do differently or better, that's the only way is if someone actually gives me that candid feedback. Period.
SPEAKER_01Were you on Foremost Forms?
SPEAKER_00I was.
SPEAKER_01We got a we got a escalation email from Chris Stedman on it. Oh, good lord. To the team's credit, they did go deeper into outcomes. They also went against my guidance and had a call that was not prepared for two days ago. I don't know what they were prep what you were not prepared for two days ago.
SPEAKER_00I think that that was a call that Chuck said I would hold off on it. And I said And Tom pushed it. Tom pushed it. It wasn't me. I mean it was Tom's call. I'm just there.
SPEAKER_02Tom pushed it and you got sucked in.
SPEAKER_00I got sucked in.
SPEAKER_02Yep. Yeah, he was feeling pressure from Microsoft.
Kicking Off The Podcast
SPEAKER_01They need more training, that's okay, it's new. But not having to prep for the call, that's that's bad because that's something that's completely different. But, anyways, let's get to our podcast because otherwise we never will. But I appreciate you letting me vent. I don't talk to anybody about it. Um, I I get it. I hold you in confidence. I don't want you to know that any decision I make, I make with you and your family in mind as well.
SPEAKER_00Oh, I appreciate that, Chris.
SPEAKER_01Okay. I do that for all the team. I have a lot of people here that I know, and that you know, I'm not gonna just do so, even if it means I'm gonna be unhappy to make everybody else happy, I'm going to do that for a period of time to make it happen. Just so you know. Yeah, no, I appreciate it, man. You got it. But I was really pissed off on that call and the way I was spoken to, and you know that if we were sitting in the room, I probably would have threw him out the window.
SPEAKER_00Yeah, I was like, Yeah, even even I was like internally, I was like, oh damn. I'm like, that's not gonna end well. That's not gonna end well. But but hey, but we're here, we finally get to. Okay, so what are we gonna do?
SPEAKER_01You're gonna record? What do you want to do?
SPEAKER_00I yeah, I'm gonna record. Actually, I have uh I'll hit the button on I got my loom note taker, so the loom needs to record. Um and so then I'll I'll just kick it off real quick. I'll introduce you. You can, you know, you'll introduce yourself, and then we'll just have I got some questions, like some softballs, like hey, you know, how did you, you know, you've been in tech long before AI, you know, all that kind of stuff. It's a conversation. I'm good. Yeah, a conversation, absolutely. You got it. All right, all right, so I'm gonna kick it off. Welcome everyone to the latest episode of the Service X Factor Podcast. I'm your host, Scott LaFonte. We are just a few days away from the holidays, and we have a special guest, a longtime Microsoft MVP, an AI futurist expert. I call him the expert of all things that uh I don't know, and then things that I do know. Mr. Chris Cognetta. Chris, how are you, man?
SPEAKER_01Hey Scott, thanks for having me, and hello to everybody on Next Factor. Happy to be here, and uh Merry Christmas and happy holiday to everybody here.
Chris’s Path Through Microsoft And MVP
SPEAKER_00Yeah, uh awesome. I know we were we were trying to get this scheduled uh for at least two weeks, and of course, scheduling conflicts always arise. And and so here we are, just uh it's a Christmas present for everybody that listens. Oh boy. Are you wanting to send out gift cards? I I am. I'm gonna send out gift cards to everyone that listens. All 10. Um, so Chris, you know, tell for those that somehow don't know who you are, um, you know, who is Chris Cognetta?
SPEAKER_01Yeah, well, uh so first of all, uh I am Chris Cognetta. I am the Chief Technology Innovation Officer and one of the founders at Congruent X. Uh, I have been with Congruent X over the last five years. Uh prior to that, uh, I have worked in many different industries, starting in uh telecom, financial services, e-commerce, automotive, um, healthcare, uh, pretty much uh higher education, um, pretty much all the different services and industries you could think about, but um held one of the guys that has held all the roles. So if you look at you know, Starbust Developer, uh, went from developer to an architect, went from an architect to uh to engineering, uh data warehousing, uh, which again, I always had the knack to figure out what was the next big thing coming. Um been an MVP now uh 13 years for dynamics, CRM, and started off as the infrastructure guy on that side, and uh very proud over the last two years to be able to get my AI MVP, um, which is not an easy task being that it's all new and it keeps changing, but uh you know, been doing that for longer than two years, obviously. Been uh been working with AI since conversational intelligence. And I know you follow along, Scott. You may have seen some of our dialogue crime, dialogue copilot stuff, which is you know, it's approaching four years old, but we were doing a lot of these things back then. So um the vision and and and what we're trying to do with that stuff is still real and uh continue to to do that. But that's just a little bit about me um and uh you know my background.
SPEAKER_00Awesome. No, that's that's great. I mean, I think that gives our listeners hopefully to say, like, wow, you know, like Chris has kind of been there, done that, not just the different roles, but in the different industries and different technologies. And so, you know, thinking back, Chris, I mean, you've been in technology for shoot probably as long as I have, if not a little longer. Uh, you know, I knew you know, I know we're going both on 25 plus years here. So we're we're no spring chickens. But take us, take us back to, you know, what got you into the Microsoft ecosystem in the first place? You know, was there something that just you know kind of a spark that lit, or was it just happened you know, by chance?
SPEAKER_01Yeah, um, it's a good question because I'm always asked, you know, are you a Microsoft guy? Like, do you only look at are you blinded by those solutions only? And the answer is no. I'm a I consider myself a technology evangelist. Um, I look at all the different solutions that are available in a build versus buy model type of thing, and then figure out what's the best that fits into that organization's infrastructure. They may not have one, then I can choose what I need to pick and what I need to do, right? Um, or or what the direction and strategy is for the company. But as far as how I got into the Microsoft space or how it became a focus, I'll say um I was uh working for myself and as a consultant, uh doing uh at the time was a Microsoft retail management system. Uh that's where you know you you do cash registers and stores, uh, had uh cameras for security, uh networking, cables. I mean, working for yourself, Scott, you know this. You'll do anything to any kind of job you can do to make it. So it was a little bit of my MCSE, my network background, my Nobel days, right? A little bit of uh architecture, uh, a little bit of e-commerce, um, a little bit of storefront management, inventory management. And uh I started doing that and met a company down in uh Delray Beach, Florida, who uh who was doing some of these things. And they were like, hey, what do you know about CRM? And I said, Well, I know a lot about uh at the time, you know, customer asset management systems and and these things, but um, you know, can you take a look at it for us? I'm like, yeah. So started looking at you know different options. Uh, you know, there was uh, I don't know if you recall back then, there was the outlook for CRM contact, yeah, all those things. And uh and at this time it was like 1.0 CRM. And Microsoft had uh it was a command line install. So if you want to know why I don't have an hair anymore, it's because I lived through those command line installs for uh for three years for three versions, right? And um we we we dug in and uh they basically paid me to learn CRM and implement CRM for the company. Uh and it was it was very successful. They uh continue to use it up until uh I think 2021 actually. Uh still stay in touch with them. And um, you know, there's always different reasons, you know, why people move from different CRMs. Um, our CEO would like to say that 50% of CRM implementations fail. So I consider that a success, right? Um and again, it was uh it was quite a different time frame back then, as you know, where CRM was being updated every 16 to 22 months, give or take. So as an MVP and and getting involved in the system and how I became an MVP was uh I went to a conference and uh there was a bunch of people getting together talking about it, and uh at the time it was you know infrastructure related, and you remember this word, Scott, ADFS.
SPEAKER_04Oh, yeah.
SPEAKER_01So uh so ADFS and IFD, the technologies to connect your CRM to the external world uh through a monkey wrench and everybody else's plan who didn't know about these things. Plus, I was a server guy, right? I did SQL servers, I did uh networking, I did Windows um AD and server setup. So I understood the background of how these things work. So how to make them work, and and little by little, uh that kind of coined me as the infrastructure MVP back in the day um and and started doing that work. Uh, did that through my career uh on my own. And then uh, as you know, I was at Tribridge for over 10 years, eight to 10 years. Um, and there uh not only running the practice, but also uh working towards uh both customer support, understanding their challenges and working with that, and also uh using my MVP skills to help uh as a servant leader get the rest of them team up to speed in those capabilities. We had an academy, helped grow those people, all that. So um it's been a whirlwind of stuff, and along the way, um CRM has gotten small. And people say, What? What do you mean by that? Right, Scott?
SPEAKER_00Yeah, oh yeah, no kidding. That is that is definitely not far from the truth.
SPEAKER_01And uh, you know, I mean by smaller is that what we used to do in 1.0 and 3.0 and 4.0 was all in CRM. Yeah, right now you have all these other applications that are interacting with CRM, whether it's Power Plat, Copilot Studio, to Azure AI and the functions in Azure, um, not to mention all the co-pilots and 365 stuff that's all integrated as well. You've gone from knowing, I mean, I can tell you, I knew everything about that system uh in 20 in four, in 2011, all the way even into the cloud aspect. And when it started to branch off, I think what's happened is that you got people who have stayed on the interface in the in the interface and know forms and fields and design and relationship. And you got others that have tried to work through all the other tool sets, whether they know flow, they know workflow. So it's it's no longer just a CRM consultant role. Yeah.
SPEAKER_00Yeah, no, you're you're spot on. I mean, we we talked about we talk about all the time at at some point in time, you know, we can see it, CRM sort of being the back end, right? Everything's being run with agents, right? And so you know, that I don't think that's too far one off from the truth and too far down downstream as well from a time perspective.
SPEAKER_02Right.
CRM Grew Smaller As The Stack Expanded
SPEAKER_00So um so it's another good question uh for you. So you know, you're you're now an AI MVP, you're I consider you a a an AI expert, futurist, whatever anyone wants to. Are you gonna tell them the name you gave me? Oh, what did I call you? Um I don't even remember Yoda. Yoda, that's right. Yoda. Um how how do you uh how do you balance that innovation with the reality of let's just say, you know, timelines for customers and and budgets? I mean, what what's your advice and you know how do you go about doing that? Because I think it's the biggest challenge right now that we see is you know customers trying to bite off more than they can chew, and then sort of scaling them back to get them ramped up. And also, of course, looking at it from not in just an implementation budget perspective, but you know, AI uses consumption, then you know calculating that consumption is not an easy thing to do up front.
SPEAKER_01Yeah, um, it's a really good question, and it all starts with the person themselves and their intellectual curiosity on how they're gonna not only understand what's going on, but their passion to keep moving and keep doing these things. Um, that's the best way I can explain it to you. There's a lot of people who just do their job, they they're good at what they do, type of scenario. In an evolving world, um, I can be on a call and I'm gonna be answering questions around whatever the the, you know, whatever it may be from a customer, but rest assured that that customer has AI and they're asking the AI that question before they ask me. If you're from the South Florida uh area, there used to be a commercial for uh a company called Sins Suits. You remember that? The suits warehouse? Yeah, oh yeah. We said the educated customer is our best customer because they wanted you to go out and look at all of these things about suits and all of that, and then come ask the questions. Yeah, right? Because then they could prove to you why their product was better. Well, AI is kind of doing the same thing. We had a good and and people say, Well, wait a second, it's not just AI, and you're right. I mean, people could go out to Google or a search engine and do the research and do that. It's just AI has made it much easier to get those answers and understand those answers and get. Viewpoints and all that, but um where I'm going with this, Scott, is that you know everyone has career goals and passions and and what they want to do, but this stuff is moving so fast. Um, you you actually have to set aside time to make that happen, and that that comes with personal sacrifice, right? Uh, and you know this very well because you're an MVP. Yeah, so uh it's no different than what we do as MVPs. How do you know about the latest release? It's because you got a copy in your environment and you're playing with it on your own time to do that in between calls, in between holidays, in between weekends. Um, that's the passion that remains to be there.
SPEAKER_00Like, wait, you get you get weekends off? No, you're you're spying. You didn't get the memo. I didn't get the memo. I did not.
SPEAKER_01Yeah, um, I it so it's very hard to balance work personal life right now, um, especially when you're you know being a founder of a custom company, trying to do things like that. You're an MVP, you're trying to stay on top of that. Uh, and I'm I'm an MVP in both CRM and AI. So trying to keep a foot in both of those is another whole struggle, right? Um, and then as you know, there people think that you just get MVP and you're done. It's an award you get and that's it. It's not, it's evaluated every year. You've got to speak, you've got to do engagements, you've got to do things that show that you do belong to be this industry expert. And um people forget that's another whole level expectation. But someone who's hiring you as an MVP also needs to give you time to be able to do that. And that's, you know, if you're an MVP listening, that's critical in in a company and in working with them and being able to use that. And in my role, because I do play the innovation officer, um, that time that I do spend to get the new technology and help bring it into our products and our offering for our customers can go hand in hand with MVP time.
Balancing Innovation, Budgets, And Consumption
SPEAKER_00Yeah, which is great. And and I that's what I love about some of the things that I get to do in terms of, you know, not even just on the delivery side, but on that sort of pre-sale side. I I actually have the time right to then take a look at what's going on and and build some of those cool things based on use cases, right? Not just going to build something for the sake of building it, which of course is is always fun, but building it to solve a real world problem, right? And and maybe not solve world hunger or peace, but solve the business problems and the business challenges. Um so I I agree with you wholeheartedly, one, to have that time. And then two, to your point, I think you said earlier is you know, staying on top of what's coming out, right? It's it's not easy. So to your to an earlier point you said, hey, you know, 16, 18, 22 months. I mean, I remember back in the SIBL days before you know everything was on-prem, there was no on cloud. And the releases were you know, you'd get one, you know, a big one maybe once a year. Right. In between, there's patches, sure, there's patches to fix things, but you know, and and back in the day, you would actually get a CD to install it on the multiple CDs. Oh yeah. So it it it was uh a different time back then, and now the it's just rapid fire innovation. Um so I think that's always uh a challenge for us to stay on top of everything. So here's a here's a a good one. So if if you know thinking back to you know early day AI and you getting started in that, what's one lesson that you may have learned the hard way when you know working with or deploying AI uh for any sort of solutions or production or what have you?
SPEAKER_01Uh the fact that it's not deterministic. And for those of you who don't know that terminology, what it means is that if you asked AI a question multiple times, you may get multiple different answers, and they all may be right, but they may not be the same. Uh, for example, if you asked AI, if you climbed a tree and you picked a fruit, you may get back the answer that you got an apple. If you run it again, you may get back the answer you got a pear or a banana, right? So that's a very, very simple, very easy thing to see. But when you're doing that against the system or an API or something like that, um, it's very difficult to get that process to be the same and understanding. And we've uh noticed this through our AI onboarding, right? Um, what we do in AI and how we use the tool set to help us get to that consistent result and do some of the heavy lifting. Uh, that is, you know, that deterministic behavior is something that you're not gonna program in. It's it's not going to work that way. So it's requiring you to change. Um, and it's also the second part about this is which is human in the loop. AI is gonna produce documents, it's gonna produce information, but if you don't absorb it, I like to use this word, I got a new word for it. Um, and I'm calling it valid date. Uh, good friend of ours, Mike Houk, mentioned this the other day, and I'm like, you know what? I think he's 100% on. And a good example, Scott, would be you get a pseudo design that's built out from an AI capability, right? It's a solution architect that built out different solutions, a lot like what we see in planner today. Does that mean it's right? No. So, as a solution architect, you're no longer worried about the relationships and how the data is and all of that. You are at a high level, but the system is mapping all those things for you. It's going to be up for you to validate that solution it provided using your experiences, using your Scott bot, your brain, to help you.
SPEAKER_00I'm like, I don't know if anyone wants a Scott bot.
Non‑Determinism, Validation, And Prompts
SPEAKER_01But you know what I mean? Yeah. And that is how um, you know, the with AI and lessons learned on, you know, you can't trust everything it provides to you. Uh, I I talk about this a lot in some of the classes I give. Another good example of this, I always hear people complaining about uh man, AI, I as a piece of garbage, I never got back what I wanted, yada, yada, yada. And I'm like, listen, guys, you need to be a prompt engineer. You need to understand how to interact and how to write your prompts in order to get the right response. If you don't, it's a science, right? You know, treat it like a three-year-old, where if you ask that three-year-old to go to the refrigerator and get you a drink, it could be anything. Yeah, right. Oh, yeah. Orange juice milk, chocolate syrup, tomato sauce, it could be anything, right? But if you said, hey, go to the refrigerator, open up the door, go to the shelf, the third shelf down, look for the red box and give me the red soda can. Chances are you're gonna get that Coca-Cola that you want, right? And that's the difference. And we all have this new skill that's developing, and it's been around now for a few years. But prompt engineering is a skill, so add it to your resume and learn about it. Learn about the different types of prompts and how to make the AI work for you and what you can do to make your prompts better. You can actually use AI to help rewrite your prompts and help you make faster, better prompts that the AI likes better, and you get more deterministic results. But those are the core challenges. But in these classes, people are like, oh, don't work, it doesn't do this and it doesn't do that. I'm like, well, you know, it used to be uh the example I was looking at, Scott, was uh Queen Queen Elizabeth had died. And Queen Elizabeth um had died, but the model was trained up to like 2022. And they're asking in you know, 2024, uh, is Queen Elizabeth still alive? And the model would say, Yeah, Queen Elizabeth is still alive, yeah. So it doesn't know anything. Well, and and that caused some confusion. So they fixed that in the AI model, where not only do they go off and use their LLM and the data that they use to train the model, but they also try to do a query of current events. So you may see a response like, well, as of this date, I don't have that information, but based on current events, this is what I found, right? You may see a response like that because that's the model bridging between the two. And now this is back in the 4041 GPT timeframe. As you look at the newer models, they're automatically accomplishing this, right? It's already been put into that. But this did cause a lot of confusion and a lot of skepticism on using uh AI and its responses. And you know, again, you're always gonna have the naysayers and what it can't do. Um, but uh I personally, what I have seen and what we have started on to do and what it's gonna be able to do for us, um you really need to get on board.
SPEAKER_00Yeah. Oh, yeah, absolutely. Well, and it's it's interesting because I'm trying to teach my kids, and I'm sure you're doing the same, uh, you know, how to use AI to get the information that you need. Right. And and I still see people, not just you know, the kids, but you're prompting it as if you're writing a text message and you're keeping it short and succinct, and to your point, you know, you need to be you need to be precise. Right. Because if you're too broad or too generic, you're gonna get those like, hey, I'm gonna bring back tomato uh tomato sauce instead of you know the the coke that you wanted. Um you know, so I I see it all the time. And you know, when I first started, I did a lot of the same thing, right? Because you you don't know. And and as you start using it more, you start saying, okay, you know, here's the instructions I want to provide, here's the prompts that I want to use. And even then, there's still times you're gonna make mistakes and then AI is gonna make mistakes, but it doesn't mean that it doesn't work, it just means maybe you didn't prompt it right. And that is what I'm saying. So you know, turning now to um you know, things that can grow in access, you know, how do you as the CTIO fall foster a culture of say AI experimentation without creating uh you know chaos? Because we know that uh you know AI can be unwieldy at times. So I mean, you know, how do you go about it with with your teams?
SPEAKER_01Yeah, and I think it's not just that congruent X because I'm being asked to do this for other customers, right? Um it would go without saying that you're talking about AI in one hand, the other hand, you're gonna be talking about governance and security and everything else, right? They go hand in hand. Um, but fostering a team is many ways to do that. Um, there could be incentives to how can I use AI to gain efficiencies in my role, whether it's building documentation, preparing calls, summaries, uh presentation work, all those kind of things are ways to chop about at a bit to get the team incentivized to use AI, but it's also in how do I use AI to get better? If you're still Googling, that's fine, but I don't use Google anymore, man. It's 100% AI, right? Everything I'm doing and I'm asking, I'm getting from those sessions and those chats. And then, you know, I'm using different AI models to help me provide those answers. But when you think about it, yeah, I mean, it's not current context, it's things I'm looking at are in stone. They've been there for many, many years. Why wouldn't I use them to be better? So it's bringing that mindset and challenging people to do that. Another creative way we've done is um creating a hackathon. And um hackathon is a giant investment from a company to get people to do that. And we did it for a few weeks. We we we had some success in delivering some agents and all that. But what I started to find out, it's always the same people, and this is where the word comes from, that are intellectually curious, curious curious, right? Those are the ones that are gonna take that ball and run with it. So you're gonna have to balance that with the people who are just doing the job, right? And how do I look at those people and understand why they are the way they are and why they don't jump into the technology and how do I bridge that gap? In some aspects, uh, I will tell you that we are gonna go through a very big emergence in the field itself, um, just from the ability of what these tools can do. Uh, I've got guys that have come out of college that are UX UAAI designers, know nothing about databases, know nothing about dynamics, know nothing about power platform. And they are very successful co-pilot studio agent builders and understand business models and know how to manipulate that information without any background in computer sciences, engineering, any of the things that we have been raised with. And it's alarming and refreshing at the same time. So I would say that if you're not using AI, are you behind? I think you are. Uh, I think you are, you should at least be uh using it uh and finding out what its capabilities are. It doesn't mean you're deploying it in production or using it in a product, but that would be the minimum. Um, I can tell you that there are many, many stories of customers that uh did not want AI. Meaning you have been on many of a call with customers that start off the call with, sorry, we don't do AI at this company. And I laugh, right? Scott, I laugh because I'm like, what what do you mean you don't do AI? So you're telling me you locked it all down from all your browsers, everything? So all that means is that somebody went home and ran into Chat GPT. Yeah. Right? That's all it means. You you're not stopping it, right? Yeah, you're not gonna stop it. It's an evolution that's here. You you really need to get on that wave, put in your governance, put in your security models to help you get there and maintain it. There's gonna be a lot of lessons learned along the way. Um, it is the fastest moving thing I have seen in my career, I'll have to say my lifetime, right? Um, based on what we've seen. Adoption of this thing is insane. Uh, you know, when you look at the older CRM cycles and then how fast they were moving, this thing was already providing value in you know, less than 30 days. Um, if you look at Moore's law on how fast the return of results with computers and all of that, that the adoption of this is going to shatter everything that we've known before. Um, so you may be an old school company, you may be somewhere, some industry that it doesn't apply. I get it, but I can tell you right now that your competitors are using these capabilities. They are looking at ways to do that. And if you're not, then you're behind.
Governance, Security, And Culture Change
SPEAKER_00That's some that's great advice. Uh definitely. And so I think that goes into uh sort of a question about that is you know, what's been the hardest part do you think about convincing customers to to really embrace AI instead of seeing it as something that they don't need or you know, whatever, whatever the the reasoning or rationale may be. I mean, what do you think is the hardest the hardest part?
SPEAKER_01That's like uh I forget the movie, but I'm gonna answer that question in 47 pieces. Yeah. Uh right, with Sam Kinnington, right? Back to school. Back to school, right? Yeah. I only have one question for you in 47 pieces. Um it's a combination of things, right? Um I think you know, looking at what's hard and why people are avoiding it, there's there's there's it depends to me on the role that you're talking to, right? If you're talking to the CTIO and the security team and all of that, they're 1000% scared that their data, their corporate IP information is going to get out into one of these models and they're gonna lose their competitive advantage. So their response is shut it all down. We don't know, we haven't figured it out yet. Close it all off. No one needs AI right now. We're gonna take a slow methodology methodological approach to get there, and that's it, right? Um, and then you got the business who's chomping at the bit to say, wow, uh, I don't have to look at spreadsheets anymore, I don't have to look at these files anymore, I don't have to, they're seeing all the other things that all the roadshow demos, all the you know, all the power platform conference videos, all those things say, man, I could really use this. Everyone in this world is asked to do more with less. And these tools are the key to doing that. So you can reduce your costs. It I have to say something else because a lot of people you haven't asked me this question, but I want to insert it here because I think it goes in alignment with what we're talking about. But the real question that you get is is AI gonna take my job. Yeah, that was right. And it's it's not necessarily that it's gonna take your job, but if you're the guy that's importing Express shed sheets and dealing with columns and formattings to get them in a down system, yeah, that job's gonna be gone. Right? Yeah, my advice to you is to learn how to use the AI to build that so that you're still the knowledge expert in those sheets and the data and program the AI to help get the sheets in there instead of spending 60 hours a week moving Excel files. Right?
SPEAKER_04Yeah, yeah.
SPEAKER_01And remember, I say program, but this is low code, no code. Somebody with no experience is doing this stuff. So, but the hurdles back to the you know, the story there. So you have that scareness on one side, you got the business wanting to move and the CEO wanting to move on the other. So it kind of falls somewhere in the middle of how a company is gonna adopt it, how they're gonna achieve it. Some companies' business models can't support it. Look at professional services, right? Professional services has been built off of how many hours a consultant's billing to do it to do a job. Well, if you take all those hours away from a consultant, the consultant's gonna say, Well, where's my job?
SPEAKER_02Instead of saying, Well, that means we can do five more customers. Right?
SPEAKER_01We can scale more, we can do more with less, which means we're gonna make more profit margin, which means people can get paid more, but people don't see it that way. So you have both, you know, pessimistic people that are understanding, you know, well, I don't know, I don't see. You got companies that really haven't seen the value. And I love the fact that, you know, at least that can grow an X with our AI onboarding process through that process, and you know, people telling us they don't want AI, but we use the tools every single day. Every conversation. So by using those tools and showing them the outputs of those tools, what happens, Scott?
SPEAKER_00What doesn't happen? That's that's really the question. I mean, they want AI then. Yeah.
SPEAKER_01They'll be like, man, I can't believe that you just got off this call and you gave me all these details of this workshop and my percentage and gap analysis. It took me three and a half weeks to get it from the last provider. And the answer is yes, we know it. Right? That's why what could you do with that in your business? Oh, so and you can take bits and pieces, make uh we call them MVP, right? You don't need an agent, it's it has to be built end to end. Build an agent that does the one part, execute the one part, then work on the other part. It they don't have to be completed projects that take six months to deliver, that kind of stuff. These things are happening in two weeks, right? It's a big, big difference.
SPEAKER_00Oh, definitely. I mean, they're not uh they don't necessarily have to be these massive projects, and that's where I I have customers that are giving me, say, 10, 10, 15 use cases for AI. And I said, okay, well, why don't we start smaller? Right? Let's let's let's start with maybe your top three. Right, and then and then go from there. Right? Let's not bite off more than we can we can absorb and chew and and overwhelm everybody with one testing and how to use it. And let's go up with three and and and then expand the footprint. You know, come up with the roadmap. And I think that's where I see a lot of organizations struggling as well is you know, where to start. Like, hey, I got all these ideas, but where should I start? I keep saying, Chris, start somewhere. It doesn't matter. Pick something, pick your top two or three issues, and let's start there. But stop talking about it and thinking about it because your competitors are already doing it and stop blaming the things that you should have been doing in the past.
SPEAKER_01And a great example of this is governance.
SPEAKER_02Governance did not exist prior to AI.
SPEAKER_03People are like, what? What? What did you just say?
Why “We Don’t Do AI” Won’t Hold
SPEAKER_01Yes, governance. We knew about it, we've been told about it, but now it's glaringly obvious that we should have been doing it and we haven't been doing it. Your SharePoint directories are a great example of data governance. Have you classified all those documents inside of there? No? Well, guess what? You should have been doing that 10 years ago, right? Classification of your documents. We're all driven on a security role. Oh, Bob has this security role, he don't belong to that team, he can't see that. But it's really at the document level so that all the tools across the platform work. If you don't have access to that document, your email's not going to send it, your thing's not gonna send it, and your AI is not going to expose it. But for some reason, this has AI has reawokened that crowd of what's needed there. And I get it, we don't want things to go into the wrong hands and all of that. And yes, we need governance. I'm not saying we don't, I'm just saying that it's not if it was such a critical path item, Scott, why did it drop off the board? Right? Yeah, and if you're trying to use it to delay using AI, then that's great. Congratulations. But you're really hurting your business. Business made a decision not to classify those documents and do what they're doing. It doesn't mean you can't have governance, you can still have governance in many different ways. It's not just file level, it could be connector level, right? If you got a connector and you talk to CRM, that's great, but it's still using Scott's credentials. And what Scott had access to in his browser and what rights he had in CRM still are at that connector, aren't they? Yeah, absolutely. Well, so does that does is that a type of governance? It absolutely is. It's just that again, pick a point, start as an organization to achieve those goals. Don't put that one thing out there and say if we don't solve this, we can't use AI because you won't solve it.
SPEAKER_00Yeah, that's I think some great advice for everyone that's listening. So here, you know, I know we have uh a couple more minutes because I know we're one, we're all pretty busy, and plus we're getting ready for the holidays. So I want to ask some uh a couple more questions. Like, so for example, you know, where do you see in the next five years the role of humans like us in CRM? I mean, what what are we going to be doing and using CRM for or what will customers be using CRM for?
SPEAKER_01That that is something that may be a shocking revelation to everybody on this call. Um, but uh I'll tell you what I what I feel um about and and uh if bear with me a second, I'll tell you a little bit of the story of why I feel that way. So probably eight years ago, I was having a conversation and I looked and I told the person in that conversation that you see your phone, you see all those apps that you have on your phone. That context of an app and having five different apps for something is going to be what business is like. And fast forward to today, we have apps that are contextual based. And what I mean by this is you have an app that customer services uses, you have an app that your bit your marketing team uses, you have an app that another person uses. That lens or that context of how you access and you do your job is what that contextual change has come, right? And that is what I see continuing to progress in the later years here of how AI is going to stop that. At the end of the day, when you do CRM, even ERP, right? When you do those systems and you talk to those users about their complaints, most of their usability complaints are about just accessing the system, clicking buttons. Uh, it's too much to fill out, it requires so much information.
SPEAKER_00Click here, click there. This is all required. Yeah.
Start Small: Use Cases, Roadmaps, Governance
SPEAKER_01Yep. All you gotta do is listen to one adoption session on that stuff, and you'll realize that people really don't like interacting with that system. There's many reasons why. Some of it's because the system doesn't give them value to use it. Right? It's more big brother. Put your information in here so your manager can see that you did that. But what if your systems work like your Alexa, where you're on site, it knows you're on site automatically, puts in that Scott was on site at XYZ Company, and this is what he did. Can it do that? If your system's not doing that today, then you're behind because it can do that, right? We we have many apps and things that are location aware, can use all the sensors on your phone, can understand that you're on site at a client. How about searching for records? If you're on site at a client, why can't it just pop up and say where you are? You want to try this technology? Go ahead and tell your whatever her name is. I don't want to say it so everybody's uh personal assistant goes off, right? Uh go ahead and tell Bob that next time you're at Publix, remind you to get a milk. When you walk through that public store, your phone's gonna say, Hey, you're supposed to get milk. If you don't know that, try it. Or, hey, next time I go to Home Depot, I need uh some nails and some glue. Next time you walk across Home Depot, it's gonna do that. Why can't it do that for your account manager walking into that deal and saying, Here's everything about this customer? By the way, they got a support ticket that's been open for two weeks, and Scott's pissed. Right? Yeah, you can give that advantage, and that's just one example. But so it's not only changing the context of how you work, but how the data gets there. Forms over data, people using data to get there and enter it, it's gonna be there for a period of time because it's just the way we work, right? And you know, me and you have ran into systems. Uh, I ran into a lotus notes system yesterday that's still been working for 25 years. So you can't tell me that within 25 years, these systems and those things won't exist because they'll still be there, right? Yeah, but I will tell you that we're moving and moving towards this ecosystem where those bits of information are pulled up from everything that you do, and you're no longer building systems to be a 360-degree view to try to capture what's happening because it's being captured in all the other avenues. Yeah. So and and you know, interacting with an agent where it sees your calendar, it sees the notes that you took, it marked in your visit, then there's no reason for me to go do that in CRM. Our dialogue co-pilot, for example, has the ability to identify what type of meeting that you're on. So you think about a sales process where you have um you have your you have your initial call, kickoff meeting, you've got your um qualification of that lead, you've got your discovery, your solution, your proposal, and your clothes. If you identify those calls and record them, the system can then move you through that process without you touching CRM because it knows that you've provided the context of what type of call is. Yeah, and I'm sure all your salespeople are on a call going, Oh, Chris, I want that right now. But that's that's the kind of capabilities that are there.
SPEAKER_00Yeah, the feedback that it gives is amazing. Plus, to your point, right, if it's putting in all the details for you, what does that mean for your salespeople? They can actually spend more time selling versus more time in the system.
SPEAKER_01Yeah, it's not that it's big brother saying, Well, I'm I'm I'm hijacking your phone and I'm grabbing all this stuff, it's so that you don't have to enter those notes. Yeah, right. And and I I we talk about AI and summaries and notes and everything, but that may be new to some people. There's still some people blocking those from calls. We've been doing that for four years. We've been in conversational intelligence and understanding that the context was going to change to that for a long time, right? And you're seeing it on how all these devices pick up and grab the conversations and do things with them. And and and why can't it do a lot of my business things for me? The answer is it can't. And it's not just sales. What about a design session? What about listening to a reverse demo? What about testing? All of those things have ways that you can use these AI concepts and contextual changes to make them better. And I think that's the road we are on. But from an AI specific Scott, to bring it back down to our lens, I think we're gonna move from we still have LLMs, which is the giant models that you're all aware of. But I believe, um, and I hope uh Microsoft doesn't take this the wrong way. Uh, I don't think they will, but I believe that they have won the game when it comes to enterprise, meaning that they hold all the data, right? The capabilities, the tool sets that are there, which is a great thing. And I think the fact that they have that, I think they're going to deliver something like a customer LLM or a domain specific LLM. So when you turn on your company and you start up your Office 365 and your CRMs and all that, this thing's just working behind the scenes. And it's building this LLM that is of Scott LaFonte Inc. of everything that you do, and now it's all available to you, and you're not, there's no infrastructure, there's no person that put that in there. It's all on the cloud, and it's all seeing of everything that you have. By the way, this is what Microsoft Graph does programming if you don't catch the connection, right? It's already there, we just don't see it, right? This would expose it and change how we work forever, and I think we are on that trajectory.
SPEAKER_00That's awesome. Yeah, I I would agree. And hopefully everyone's sitting there saying, Wow, okay, if I'm not on board, I better get on board now.
SPEAKER_01All right, so I gotta your point. Take that, go find that little POC that you could do to help solve something. Even if you do it at your own computer at night and work through it and then say, hey guys, I want to show you what I did with this. Be careful, don't use corporate data. Uh, you need examples, tell AI to make you a random file with the same kind of data. It's not your client data, but it can be the same columns and same rows. How many times do you do that, Scott? We're loading death data. Oh, all the time. Every single time. Right? So it knows it's smart enough to make the relationships between accounts and these contacts and these opportunities, can do a bunch of those things for you. So why not do that and and help uh make an example where it's not corporate data, but it is the format, it is the place, and then that way they can see, you know, you can you can see that. So that's one way to help uh influence change. But go ahead, Scott. You had another uh, I think you're to the closing question, maybe.
SPEAKER_00Yeah, I yeah, I got a couple rapid fire questions. Here you go. Um favorite Microsoft product of all time.
SPEAKER_02Favorite Microsoft product of all time? Wow.
SPEAKER_01I know it's a tough one. That's that's a hard one, dude. You're talking about like everything that I've so I'll I'll I'll go on the record as that I loved C. That was one of my favorites, um, from a developer tool, that kind of thing. But um I'll have to say that uh I'm pretty impressed with AI Foundry right now and and what its capabilities are, so I'm gonna use that.
SPEAKER_00You're gonna use that one? All right, I'm locking that one in. All right. One book or podcast that changed how you think about technology.
SPEAKER_01So the book is by the name of Bob Egger, and it was about uh it was in it was back in the telecom days, but it was basically you've you've all heard the story, it was about a killer app and creating this killer app that could do X, and it really helped change my thinking on how I deliver products, infrastructure to support the goal, keeping the business in mind. I think it's a great book.
SPEAKER_00Awesome. All right, I got two more rapid fire questions for you. If you could have dinner with any tech innovator, past or present, who would it be?
SPEAKER_02Past or present, huh?
SPEAKER_00Yeah, that's a good one too.
The Next Five Years Of CRM: Agents
SPEAKER_01Can I pick, can I pick two? Sure. Uh so obviously if the current one would be Elon Musk. And uh his his uh just like like can I follow him for work one one week? Right. Yeah, just uh just just watch it. I would think uh his concepts, the things that he does. He may not be the most polished person, which we see that a lot in the world today, um, and may not have um that, but you have to give it off for you know five different businesses, successful in every single one of them, um passionate, sleeping on couches in buildings that you know our ages is uh pretty crazy. And then um, you know, the other part that I mentioned that's why I said two people is because Elon Musk owns a company. You know the name of that company? That's Tesla. But but but Big Eye Tesla is the other one that you know um would love to the things that that man put together and drew um and talked about that are just becoming reality today. It's kind of like you know, you think about Einstein, you think about uh Thomas Edison, that type of brain, that type of person, um, would love to be able to have a conversation with either of them.
SPEAKER_00That's awesome. All right, and last question. If you weren't in technology, what do you think you'd be doing today?
SPEAKER_01Oh, that's the easiest one. You know the answer. I know the answer. So uh my father told me to work with my brain, but not my hands. Uh, but my passion is in automobile restoration, uh, cars, working on them, whether it's restoring them, racing them, uh, or just driving them. That continues to be a passion. And it's actually it's funny because it's become my stress reliever. So people be like, wait a second, you do this all day long, and you go outside in your garage at night and work and bang and hammer. And I'm like, yeah, that's stress relief. Yeah.
SPEAKER_00So you get to hammer away at things, absolutely.
SPEAKER_01Yeah, you know, take a fender out back, beat it up, right? Uh obviously you can't do that at night, but you can't do that on the weekend. But that's always been my uh I love cars from a kid, and uh I think there's a lot of synergies in between building cars and building products, right? You you've got designs, you've got envisioning of what you want it to look like. Uh, there's a mechanical aspect, which is the architecture, right? You can see the parodies of both of them. Um, and I think both sides uh bring me off. As a matter of fact, uh I just put a voice-activated GPT in my garage so that when I'm working on something, I can say, hey, what's the bolt size of a 7-8? This and my GPT, who is trained in auto mechanics, can reference that information and tell me. That's awesome. So I'm not going back and forth to the uh yeah, it won't be long before we have AI garages, right? I I need to have AI for my dogs.
SPEAKER_00Oh can you scan my car? Yeah, exactly. Well, awesome, Chris. I I think hopefully this gives everyone some really good uh information to digest and chew on. Plus, they got to actually get some insight into the brain of Mr. Chris Cogneta, which you know is either you see that as a good thing or a scary thing. I don't know. You you pick. But I I do appreciate your your time today. Thank you so much. I know you're you're busy, uh, so hopefully our audience uh gets some nuggets out of this conversation. And uh I wish you and your family a happy holiday.
SPEAKER_01Hey Scott, thank you very much for having us. Thanks everybody on the call for listening. Um, it's been a pleasure, honor to be on the show. Hope to come back, uh, do some more in the future with you, Scott. And uh again, Merry Christmas, happy new year, everybody, you and your families. Let's have a safe and good 2026. Thank you, everybody.
SPEAKER_00Awesome. Thanks, Chris. Thanks everyone for listening. Until next time.
