How AI Is Rewriting the Mortgage Business with Gino Fronti

Most businesses don’t fail because of bad products — they fail because they never build a scalable monetization and execution strategy.

In this episode of The Way, Dino sits down with mortgage industry veteran and tech founder Gino Fronti for a deep, no-hype conversation on how AI, agentic automation, and operating systems are fundamentally reshaping lending, sales, and modern businesses.

This isn’t about “using AI tools.”
 It’s about learning how to operate inside an AI-powered business model — before the market leaves you behind.

Gino breaks down why traditional CRMs, SaaS tools, and bloated tech stacks are failing loan officers… and how a new generation of sales-centric operating systems is eliminating bottlenecks, reducing human error, and giving professionals their time back.

If you’re a mortgage advisor, entrepreneur, or business leader facing shrinking margins and increasing complexity, this episode is required listening.

What You’ll Learn in This Episode

• Why most businesses fail by focusing on products instead of monetization
• The difference between using AI and operating within AI
• What “agentic AI” really means — and why it changes everything
• How eliminating human bottlenecks increases volume without sacrificing customer experience
• Why applications are “where loans go to die” — and how conversation-first workflows convert better
• The future of CRMs, loan origination, and operating systems
• Why EQ will matter more than IQ in an AI-driven economy
• How automation can enhance human connection, not replace it
• What margins, subscriptions, and volume economics teach us about the future of business
• How mortgage professionals can stay relevant as knowledge becomes commoditized

Thanks for listening to "The Way With Dino Katsiametis"
For full show notes, links, and extra episode resources, visit dinokatsiametis.com.
Follow Dino for weekly clips and mortgage leadership insights on
instagram.com/dinokatsiametis | linkedin.com/in/dinokatsiametis
Learn more about Ethos Lending at ethoslending.com.

Transcript

EP033 THE WAY GINO FRONTI EXPORT_1

Speaker: [00:00:00] The reason most businesses fail is they're so focused on the product that they're not photo focused on the monetization strategy. 

Speaker 2: If you're not getting the results you want, ask yourself if you're doing what it takes or only if you are doing what you think it takes. It's these kind of things that this episode is gonna be talking about.

I love that statement because so many of us are just doing what we think it is that is needed to succeed, and yet it's something totally different. And right now AI is really where everything is being shaped from. And if you don't start digging in and learning how to operate within it, I'm not saying learning how to use it, I'm saying learning how to operate within it.

There's a big difference. So let's talk with Gino. He's gonna tell you about DEOs, the company that he's put together and how it's going to change the mortgage industry. Something I truly believe after hearing [00:01:00] what it is that, that they're doing. And we're gonna talk about how to use AI and all the different platforms that are out right now and which one can best suit you.

So I hope you enjoy this show. So my man, Gino, I don't just like you 'cause your name rhymes with my name, but I like you because you've been around for a really long time and you're not like some of these other old guys that are like status quo. They are just jumping right into what the future holds and you're innovating and you're exciting and I love watching everything you're doing.

So I'm super excited to have you here. I wanna know about everything you're doing and I wanna start with who are you today, both personally and professionally? Oh, wow. 

Speaker: I'll tell you one thing that's interesting. I'm as disciplined as I've ever been in my life, so that is probably the key factor as we go through everything in my history and the things that.

I've achieved it's only gonna get better because I've never had the level of discipline and execution that I have today. So [00:02:00] today I am a regional manager at PRMG. I manage a team that's doing about 300 million, which is a far cry from what we were doing at Finance of America during our heyday.

Don Gotley and I we're running a team doing about 350 loans a month, 1.4 billion two years in a row. And that was a lot of fun. Really scaled the big business under Bill Dallas at Finance of America. Worked with Bill Dallas for about 12 years and that's how we grew our mortgage practice. In addition to that, I've been a real estate investor my whole life and a tech nerd my whole life.

And so today I'm also the founder and CEO of Dexo say I which I know we'll talk a little bit more about today, and I will never stop being a real estate investor. So serial entrepreneur I guess you could say. 

Speaker 2: Yeah. That's awesome. And you brought up Bill Dallas. He's a good dude. 

Speaker: He really has a great mentor to us.

And 

Speaker 2: yeah and I, it's weird when I think of him, I think of the word significant and for many different ways, right? He's [00:03:00] successfully has entered and exited the mortgage business financially, which a lot of people don't exit really, successfully. But he's done that. And he is, I think who he is as an individual.

I appreciate and I enjoy seeing, and I just, I want to get to know him better. I've spoken to him once and he was gracious and kind and and mentorship, like even without knowing me, so kudos to him. If he's listening right now I want you on the show, man. A hundred percent but let's go back to you now.

So let's talk I want to, I wanna fulfill the show. Way of doing things where we go back to day one then and then walk through your journey. But I do want to just start with, since it's who are you today? I wanna start with your new company. Tell us about that then I want to come back to it again.

Speaker: Yeah. Dexos is a operating system, not a tool. And the reason that we ended up with a platform like that is today the average age of a loan officer [00:04:00] is approximately 55 years old. There are tools and tools and tools being deployed. And in order to originate a mortgage, we have to have 55 different tabs open, where 22 different hats.

And not only is it inefficient and frustrating, it's doesn't have to be that way anymore. And the one thing that we noticed, and this is maybe I started saying this about a year or two ago, is that I have a choice in life. I am either gonna wait for technology to replace me or create the technology to replace.

So that was the birth and the innovation of Dex Os. And of course, like any startup, it has many levels of evolution. And what we're doing is we're creating a sales centric operating system for a loan officer to operate in one platform to do everything they need to do so that a comp compressed world, they can originate more loans in less amount of time and pro provide a better consumer experience.

Speaker 2: So what does that mean? I struggle with [00:05:00] that, right? Because I've also, and maybe 'cause I failed at this miserably, but trying to do everything in the one in one system seems to never work. But what does that mean? What, what do, what does a user do in your 

Speaker: system? Let me back up a little bit.

It's not your fault that you failed. Bill Dallas failed, encompass failed, all these other teams have failed historically in Mortgage tech. The reason for that is that it was built on, even the last 10 to 15 years have been built on a traditional SaaS platform. And what a traditional SaaS platform is a giant database with a user interface.

And the user interface creates a natural occurrence that the bottleneck is the user. And so no matter what, we are standing in the way of parallel execution. And so what's happening today is we're going from assembly line where you had to go step one, step two, step three, step [00:06:00] four, with you at the center of the decision making process with you at the, at every, the moment of a task.

And the way we solved it when we were at Finance of America is I threw bodies at it. I had 160 employees to do 350 loans a month. That is not the future, right? I was able to close every loan in 15 days. Frictionless, wonderful, but very expensive. So in a margin less world, that's not possible anymore.

So with the introduction of Ag agentic ai. Now I have specialized workforces that operate these tasks on your behalf. So now you're just the quarterback, but there's specialized teams that are doing all the data entry for you so that you no longer have to be the bottleneck. And there will be some very strategic human inputs that are required.

We do not wanna automate everything. We have a human in the loop ai, because if you automate everything like some of our competitors are doing, it's not a good consumer experience. It's not a good loan officer experience. The technology's not there to automate everything. So what we wanna do is automate what should be automated and [00:07:00] let the human do what they do best and make the impact where they need to.

Speaker 2: Oh man, I can't wait to demo that. 

Speaker: Is it live? Are you guys functioning right now? It's demo. The demo's live in its infancy stages. You could actually see a demo. You can go to our website dsos.ai for a already outdated video because every time we do a video within 10 days, it's outdated.

But today. Basically in its current functionality, you have a human to human phone call from a soft phone technology on your phone while you're driving. It records that phone call, it transcribes that form phone call, and it knows the context in which we're supposed to operate. So it takes all the data from that phone call and starts pre-populating your loan application.

With the data from that phone call, you're gonna tee up to the consumer that the next phone call is gonna be from an our AI assistant. So the second phone call comes from the AI assistant. So that one's an AI to human phone call. And because it already knows the context of the prior phone call, it won't ask any of the questions that you've already asked, and it'll only be [00:08:00] filling in the missing fields.

Then what it'll say is usually what will happen is the consumer will say, I don't have that. I'm not ready for that right now. Hey, I gotta go. And it'll say, no problem, we'll follow up with an email and now it'll take the context and the data from those two phone calls, populate it, all the application automatically.

Then it'll send out an email with just a few bits that are missing and ask for document upload. And so what we've done is we've eliminated the need to send the consumer an application link. And the reason for that is when we were at Finance of America, or with any of the POS platforms today, if you send an application link to a consumer, that's where applications go to die.

My team has gone back to taking phone applications to increase conversion rate, right? But the high trust conversation is really the only one that needs to happen. The rest can keep going. So to the consumer, they don't even know they're filling out an application. All they're doing is having conversations.

So we've eliminated the [00:09:00] application from the process, and I think that's one of the distinguishing things. There's a lot more to it, but that's the, in its infancy stage today, it does that. Then it runs the product through a product and pricing engine. Today we're gonna be with Optimal Blue and Loan Sifter is gonna be the tool that we're gonna integrate with to use that.

As we start ting on additional pieces, it takes the application, does income and asset verification, has the application completed, and then scrubs it against all the guidelines that are pre-populated in the system and makes a selection of product for you so that you take seven phone calls while you're in the car and when you get back, you see what the status of every application is.

You see exactly where they're at, what's still missing, deploy either another AI phone call, deploy AI email, or if it's completed, it'll say, we've selected these three products for you. And then it'll tell you where, got the guidelines, let's say for example, a home, a HomeReady loan, right? It already calculated income, it already checked the A MI and it says, this guy's eligible for [00:10:00] HomeReady is gonna price best.

So here you go. Here's HomeReady as an option for you. And then because we have to build trust with machines, because today AI still hallucinates and does all these things that are not reliable. What we've done is we've broken it out so that it shows you exactly where it got the guideline that it's utilizing to select the product for you and allows you to drill down and see exactly where it got the guideline.

It'll say page two of 57 of this PDF, and it'll open the PDF for you so you can validate. That's where we still have the human in the loop for the validation process. Maybe a year from now or two years from now, that's not necessary anymore. But for today where the technology is, we still believe the humans should be validating all the information before we give those options to the consumer.

Speaker 2: What 

Speaker: causes hallucination? There's a couple of things. So AI is non-deterministic by its nature. And one of the problems with non-deterministic lang large language models is that you can ask the same question 55 times and get a different answer. And just like humans, context is actually the most important thing.

And so [00:11:00] we've talked about prompt engineering quite a bit over the last couple of years. How you ask the question determines how you get the answer. And so in fact, asking great questions is one of the tools that each user has to educate the system. The problem is memory in systems has been poor. I know I tell it to remember everything I talked about and then halfway through the conversation I'm like, don't you remember what we talked about this?

Oh yeah, I'm sorry, I forgot. Or I say gimme a contrarian view because it's been programmed by the large language models to be nice to me. I don't want it to be nice to me. I want it to be accurate. So it has to do with where it's there, the technology stands today and where the programming has come on each one of these large language models.

So also by its nature, like you may say, I like Anthropic, and somebody else might say, I like rock. And somebody else might say, I like Claude. Those are some, if you think about the way that these la large language models have been trained. They've been trained, I'm gonna just say it this way, with bias.

With its [00:12:00] master's bias, and I just watched Frankenstein this weekend, so I've, I'm very much into the whole like, master creator type of thing. 'cause that's where we're at in the world. If you really think about it, it's like each one of these LLMs are built with an ideology in mind and they're following that ideology.

So within our system, we use different LLMs to do different things because some of them are better at some things than other. And as those things change, like for example, the best copywriter is Claude. So if you're gonna, if you're gonna have one, write your emails, Claude's probably the best one to write your emails, right?

But Claude's not the best one for research. Perplexity is the best one for research. And Perplexity does something similar to what we're doing in our system where it utilizes multiple LLMs and writes code for you with Python to be able to search them. It's very complex, but added space, what we're trying to do is build guardrails around the system so that we get reliable answers on a consistent basis.

Just give the human the opportunity to validate those answers before [00:13:00] it goes forward. Or if it questions something like, it's just like an underwriter, you could say, Hey are you sure the income is X, y, Z? I saw that there was an employment gap. Did you take that into consideration? And it could say, yes, we took the employment gap into consideration.

Here's how we calculated your income. And it'll show you how it cap the income. So it really is doing the branch manager, the LOA, the underwriter's job, like all these tasks that would normally be very time a time suck are all very efficient and quick today. 

Speaker 2: Let me ask you this question on anthropic.

Gmail, right? Google for Business and Gemini is integrated in, which makes it nice and easy. Is it worth going to philanthropic to figure out how to write an email if it's one paragraph? No. Or are you better off just doing it in Gemini? Is it that big of a difference? No I think 

Speaker: for low level tasks, use whatever you like.

I think it's just a matter of, I like blue jeans and somebody else, like [00:14:00] khakis. It's just a personal taste thing. Like in my household I use perplexity. My wife uses copilot and my son uses chat, GPT. And so I think it just, it's almost like what we say about CRMs, which one's the best large language model, the one you know, how to use.

And so I, today I say use that and unless you're drafting something super important, like responding to an email just make sure it has your voice. What about writing a blog? Oof. Okay. That's a little different, right? 'cause now you're starting to get a little more specific. I would say that if you're writing a blog I would probably use Chad GPT to write a blog because I think it still has the widest set of knowledge out of anything.

And, not to get too crazy, but it is capturing everything from everybody and it has the largest market share. So I self-learning at a faster pace than anything else. And but then again, if you have a Tesla, then I would say use grok. 'cause Tesla just released rock in your car and you could [00:15:00] vomit to grok while you're driving all these ideas that you're having.

And then when you get outta your car, you could have it all completely summarized. Wow. I mean there's, that's cool. The acceleration of the speed that things are changing at right now is gonna force us to be choosing a winner on a week by week or month by month basis, depending on what tasks we're gonna do.

So whatever advice you just heard from me right now is probably old. 

Speaker 2: Okay. Yeah. Let me ask you this question then, 'cause this is advice, this is one thing that I've had to make a big decision on and it's which 1:00 AM I going to spend? Most hours dumping into. And I ended up choosing chat because it is the biggest one and I figured if I'm gonna do anything, I might as well be in there.

Overall it seems to have the overall best of everything. And quite honestly, it's gonna write better than I can anyway. So writing any better won't even sound like me. You know what I mean? [00:16:00] But what do you say about that? Like one of the things that pro that prevents me from switching to go to Grok, for example, to write something is that it doesn't know anything about me.

Is it as simple as taking something outta chat GPT and downloading it and sending it over? Yeah. And then now it does. 

Speaker: I would say that for the majority of the population, this should just use chat GPT if you're not trying to use a specialized function. The reason I chose perplexity is 'cause I'm a research junk.

So for me, research is the primary function. It's not marketing, it's not writing emails, it's not doing any of those things. I am constantly researching things and I find it to be the most reliable. So when you use chat, GPT, almost 60% of his data sources are Reddit, which is not a variable reliable source.

So I would say when I'm using it for research, I even tell my ai, I say, please use X, Y, Z as your sources. And so instructing it on where to source this information is something that's [00:17:00] very important depending on the task that you do. And I'll give you an example. I personally, this is maybe not best practice, so I'm not saying do this, but I don't care what happens with my data.

Like I'm just that guy. I really don't care. It's all out there anyway, the moment we carrying this thing around, George Orwell was right. He just know that we were gonna opt in on purpose. He thought Big Brother was gonna get forced on us. And so I decided that I'm gonna use it for everything and I'm not gonna protect any of my data.

So I actually upload all of my medical records. And one of the things that I've done is I've said, identify first, first step, identified the top players in x, y, Z space. And then it tells me who they are and what their credentials are. And then once I did that, I said, okay. And once I identified each one of these people, a cardiovascular surgeon, a neurologist and I went through and I identified like 10 medical professionals and said, great, now create a medical team with these 10 medical professionals that said, I'm gonna upload my blood tests and I'd like them to each opine [00:18:00] independently.

And then I'd like to talk, I'd like them to talk amongst each other and gimme a summary of what my current health status is based on my blood tests. That's the kind research that I like to do. That's the kind of reasons that I don't use chat. GPTI just feel like I could operate faster in perplexity due to those things.

I get impatient with chat. GPT, it feels like it takes too long for me. 

Speaker 2: So perplexity is the better one for that. Yeah, for research. So would you say for the average person who does Google like that's what they've been doing for research and everything, that now perplexity would be the best new Google?

Speaker: I haven't used Google in three or four months and I've taken the next space. I actually don't use Chrome anymore. I use Comet, which is perplexity. Yeah. Two. And I've taken the next step this weekend, which is I now allowed perplexity assistant and pro and Perplexity Max to access my Gmail. [00:19:00] And so now it's sorting my email for me and tagging all of my email.

And so this is, again, this is constantly changing, but once you get married to something, it's very hard to change. So for me to change the chat right now or for you to switch, I it's like a commitment situation, right? Once you commit to one, unless it's something doing something you don't like or you need a special feature that's something else is better at.

Like chat. GPT projects is amazing, really is.

Speaker 2: What about A GPT? How would you is there anything else that has something similar to 

Speaker: A GPT? Not really. I still think that OpenAI really dominates the space as it relates to custom GPTs. So if I wanted to upload if I wanted to upload exactly what to say by Phil Jones and I wanted to drop my email that I personally hand wrote in, and I say, Phil Jones this, right?

And then it gives it back to me and then I deploy it out. That's a good practice. That's a good use of a custom GPT. You can let's say you're a loan officer. And you [00:20:00] wanna put 5, 6, 7 different books and voices on scripting, and you wanna put the whole Todd Duncan library of books in there.

And then you tell, watch this, and then you tell your assistant every time you respond to an email, drop the email that you were had received, drop it in this custom GBT and then respond in my voice with the Todd Duncan school. And now you're not even responding to your own emails. 

Speaker 2: Okay, hold on. So I got some questions.

So I'm in the middle of creating GPTs. I'm getting through all that. I understand the concept of it. I understand a project. If in for a second there, I was like why would you create a GPT? Why don't, if you just tell it, if you just go to any open chat and say, Hey, this is the email I'm gonna respond to do it.

Phil Jones it it'll do it. But then you said, tell your assistant. And then I'm like, all right, so maybe then I [00:21:00] create a GPT for my entire company and really dial it in and say, okay guys, when you're trying to sell or when you have issues, dump it in here and let it rephrase. That's what A GPT would be as opposed 

Speaker: to a project.

Yeah, I think I think there's some overlap. If you did a Venn diagram when they released these two things, I think there's definitely some overlap between the two. I would think that projects is probably a little more modern in its approach. And I think that pro what projects does, it allows you to really segregate different things.

So you could say, realtor communications. And then, so you can opt, you could, in realtor communications, you could set up Buffini Fairy and throw all those in there. And then you could say consumer communications and you could put in Gitr and Phil Jones. And so you could start putting different types of communication styles.

If you really wanna get crazy, take disc profiles and say, I wanna respond to a driver, I wanna respon, and you can start doing these different things. And I think projects just works better than for [00:22:00] that. So I would say maybe custom GPT is more general, more macro, and then projects for more specific.

It's wild, man. It's pretty endless actually. If you think about what we're doing and how fast it's moving, I don't know where this ends. I don't know where this, but I could tell you that the basis is that knowledge has been commoditized and for all of a human history, those that had a higher iq.

And combine that with some eq we're able to dominate their space and get compensated for it. I believe that the world that we're moving into is an EQ only world where EQ is more important than IQ because IQ is readily accessible for anybody. 

Speaker 3: Yeah. So 

Speaker: What YouTube did for the world. This is 1000 fold that.

So I think that, think about knowledge as a commodity. Improvely improve your human skills. Human skills empathy, sympathy, those things that machines don't have will be the [00:23:00] most valued skills on the planet. 

Speaker 2: Okay, so earlier I said I was gonna take you back to day one, but I'm gonna just blow that out 'cause none of it even matters anymore.

Nothing. About how we used to do things or that basic stuff even matters. It's all about what we're going to do moving forward. So I'm just gonna blow it out completely 'cause I want to stick on this conversation. There's so much good stuff here. Earlier you said lemme see if I can find it. You said that a margin less world.

What does that 

Speaker: mean? I think what happens is you're looking at revenue compression, not just in the mortgage industry globally. You're looking at revenue compression from a macro economic standpoint. Just different industries at different times for different reasons. But the reality is we've moved from a.

Being highly compensated per transaction world and maybe Walmart and Costco and all these large aggregators. Think about when they took over all the small mom and pop [00:24:00] stores that began the trend, maybe at Sears was maybe one of the first innovators of this. But now where we're at today, this is accelerating.

And we are in a high volume, low margin world where marketing is really what's driving sales for large companies. And they're okay because their volume is so great that they're okay with making a penny on each product. If you look at the whole entire Costco model, all the profitability comes from subscription.

They basically, they make a very small margin on product, which is what drives people into the store, right? The 1 99 hotdog drives people into the store, and then you walk out with a $700 receipt and you don't even know how that happened. But most of their profitability comes from subscription. So just how we monetize things is different and volume has become more important than price.

And the question is, as we move towards the future, do luxury goods even really have a space, right? Or does that widen where you just have [00:25:00] discount retailers and luxury retailers and nothing in the middle? So if you were to 

Speaker 2: be young again and you were thinking about getting to the mortgage business, would you do it?

Yes, 

Speaker: I would. But I would take a different approach. I would probably take a different approach. I would've set up much more marketing and much less knowledge. The knowledge is readily accessible, like we've already discussed. I spent the last 20 years being a mortgage guru, knowing every single guideline, knowing how to structure a deal, knowing how to take someone that doesn't qualify and find a legal and compliant way to make them qualify.

That's my super skill, right? I could take a pair of borrowers and find a way to put the square peg in a round hole. I probably wouldn't do that if I started over again. I would probably streamline my model. I would not work on complex pro [00:26:00] projects, and I would just do as much as possible volume as I possibly could.

Through traditional Fannie and Freddie FHA VA products, as fast as I could, do as much volume as I could, put away $10 million, put it all into the stock market and never work again if I was 20 years old today. However, now I'm gonna speak out of both sides of my mouth. If I was already an experienced Sloan officer like I am today.

That's no longer an option. It's very hard to start that today. If I already have the knowledge base, 'cause I spent the last 20 years gaining it, I would go in the complete opposite direction. And I believe that the advisor of the future needs to understand life insurance, health insurance, all the other financial product, all the other financial products.

I'd get licensed in all of those products because now I don't, the product matching is not difficult. In fact, with Dexo, we're building product matching and our future roadmap is to go into all these industries so that I could take people, give 'em multiple licenses and they could sell multiple products and create a holistic plan.

Speaker 2: Man. [00:27:00] Okay not very many people know what my bigger vision is at Ethos, right? I'm going for it. And right before you decided to go for it I thought I was in the back nine. I had decided I was gonna be semi-retired and all I meant by that was I'm no longer gonna work 60 hour weeks, I'm gonna work 30.

And I ended up averaging 18 point a half, 18.75 to be exact for the next two years straight. And, but I also got really good and I, and it took some work, but I got really good at forcing myself to keep a calendar and be dominant in my hours that I was working. And what I mean by that is, I forget the theory of whatever it's called or the law of, it's when you go on vacation and you know you're gonna be gone for a week, somehow you do all of the work that you were behind on all of the work that you're supposed to be doing next week that you're gonna miss out on.

And you cleaned your desk somehow some way, right? And boom. That's how I was every time. I learned how to put myself in that [00:28:00] mindset every single day. And I got really good at accomplishing stuff, right? So that was my goal and I did it. And then I realized I'm not ready to be on the back nine.

In fact, I haven't golfed in over a year unfortunately, because I came in with such a vengeance to go ethos right now, as far as long term, I've taken out all of the fat in the middle. There's no more branch managers and, regional managers and everybody in between. It's just, it's us to the loan officer, which we call mortgage advisors.

And you run your own business, right? Like it's all yours. I'm not looking for employees, I'm looking for partners. And the goal and the vision is I'm not gonna make a fortune on every single loan. I gotta keep it tight. 'cause that is the future, like you're saying, right? Like marginalized world. But then.

We're branching out. We already opened up the Ethos Real Estate Group. Next is the Ethos Capital Fund, and that's gonna be the Hard Money Lending company. I gotta figure out how to take out [00:29:00] Ethos Life Insurance 'cause they already own my name and but insurance we already dish out so much insurance to somebody else.

We're making somebody else so much money, it's time to keep that in-house. And then I think we probably partner with a financial planning company. And then finally the last company is gonna be personal loans. 'cause I, it, when I was producing, it's how I functioned a lot, right? Like I it was the circle of cash flow.

Yeah, I was dishing out all the time. But personal loans, I was doing it constantly and there's not a lot of money in it, but might as well keep it in house, right? Somebody has $10,000 in credit card debt and has a six 80 credit score because of it. They're maxed out. Let's get 'em, let's get 'em a consumer loan instead.

Pay off all the credit cards. Now they're up to 7 20, 7 40. Refinance, do whatever. And I kept thinking about this circle of cash flow, but instead of dishing it out to everybody, now I'm just gonna keep it in house and I'm gonna bring in every single one of these mortgage [00:30:00] advisors somehow, some way to have a piece of it, right?

Like the Ethos Capital Fund, the hard money lending. You don't, you're not gonna just get anything handed to you. If you close a deal, sure you're gonna get your commission, but put money into it. Put some into the fund, right? Be an owner and personal loans. Put some into the fund. Be an owner insurance.

Open up your own insurance, company your branch. You own half, we own half. We'll manage it all for you. And it's, I do feel like that is the future of where it's at. And it all starts from one of these companies and then it comes into the mix and just keeps going round and round. So I love what you said about that, and that's, that is the vision that, that I have.

Yeah, 

Speaker: I think you're at the right time. I think we've all tried to do that for the last 20 years. We've seen multiple companies and people fail at doing that. And again, it's because the amount of humans that had the knowledge necessary was extremely limited. And so now that we could [00:31:00] tap into that, and like I said, for me, the product matching piece that I do with Dexo, right?

Where I take a human and guidelines and find a way to pair those two together without the advisor being the person doing that, utilizing technology to do that. That's what's gonna enable us to be salespeople and get us out of the manufacturing process and check this out one level further. So now let's say I have all these people in my ecosystem.

I'm managing all their data across all of the different things that they have. And think of Fin Locker, as a soft idea of this but magnify that a thousand times and now New York Life releases a new life insurance product and you as an advisor come in this morning and says, New York Life released a new life insurance product.

And these t these 10 clients of yours will benefit from it compared to the policy they have today. You don't have to read all the guidelines. Think about who in your database could benefit from that. Like you'll know [00:32:00] with certainty that decision's already been made for you and that it's accurate.

And again, this is fast forward maybe two years, but at the speed that we're moving, that's happened. 

Speaker 2: Yeah, I. When I recruit and I'm talking to loan officers out there I tell 'em all the time, what I did is I took how I did business and I dissected it, and then we broke it out, right?

Because me, you, a few other people, like real anomalies, right? First of all, I don't tire. I can work all day and I just don't get tired and I'm fresh in my mind. I can wake up early. It doesn't matter. Other people can't function like that, and I don't want them to because it isn't the right way to do it, right?

There's a better way to do it. And what I tell everybody is, look, first, get really good at your trade. Know what the heck you're doing, right? Know how a loan works, know all of the positions, and then forget it. Become a master marketer. 'cause if your phone doesn't ring right, like old school saying [00:33:00] a phone doesn't ring.

But if somebody's not contacting you to get a loan, then you're not making any money. So it doesn't matter how good you are at doing a loan. So first, learn your trade and then become a master marketer. And with AI and where we're at right now, it's virtually impossible. I did it and closed over 150 million in production.

But man, I had systems in place, in place. I had a lot of people working for me. I had things going and I worked like a dog. But now I think we can do all of that with less hours and focus our attention on people. And you were alluding to that in the beginning. You guys are going back old school and taking phone conversation apps, and I think, and this is my big, piece that I just wrote a blog on it actually.

I believe. AI is not Satan. It's actually God. And when I say that, so many people think Satan is behind, behind ai, [00:34:00] and I think it's God's way of saying, are you kidding me? I'm trying to give you back your time. You can actually go and have relationships with people again. You can actually get home on time and go watch your kid play a sport or have dinner at the table with everybody because you didn't have to sit and do all of this stuff.

And oh, by the way, all of that stuff that you do, you're not even doing it really effectively and efficiently. Not nearly as close as what AI is doing for you. So I can't write, I'm a terrible writer. I'm a shitty writer. That's what I was gonna say. But all of a sudden, I'm actually putting together these great blogs and I don't care.

I don't feel like a fraud. I'm telling it what to do. I'm downloading everything, I think, and then I actually proofread it. I don't sit there and just say, okay, great. And run with it. And then I debate with it. I spar with it and I produce something that I think is really great. So here at Ethos, that's what we're trying to do.

We're trying to create this platform so that you don't have to [00:35:00] realize that one of your clients somewhere out there, can refinance right now or needs something from you. Ai, your personal assistant should be able to notify you of it right out of the gate, right? That's where we're headed. And man, I am excited about that.

So I love. Talking with you about all this stuff. I love that you're actually creating the company that's doing that for our industry. Because no matter, the one thing I've also realized is no matter how much we get into it and think it's not that easy to go create a bunch of stuff, right? Like it's expensive.

It's time consuming and it's something that for it to work properly without hallucination requires a lot more than just somebody doing a little vibe code. Yeah, 

Speaker: no it's absolutely, it's daunting is the word. And just to go back a little bit, based on everything you said, our tagline is let Dex os do the task so you can make [00:36:00] the impact.

Yeah. And that's really what the basis of our company is. It's we just, we have specialized teams that are better at those things than we are. Those AgTech teams have been programmed and trained. One of the unique things about Dex OS is being built by me and my fellow loan officer community, which is through sales mastery.

You know that I have the biggest loan officer community of absolute top professionals in the nation. And so utilizing that network to help build this product is what's gonna be the differentiator in this product. It's not some tech nerd at a VC firm that's creating this product and gives us a bunch of tools we don't want, don't need, and don't work.

It's us. So it'll have less features, but the ones we want and need. Because most of these things and I'll just use CRMs, I'll pick on CRMs 'cause they're the worst. They're built by CMOs, not by loan officers. And then they're built by these CMOs. And then loan officers come in and say, why does it do that?

How come it doesn't do [00:37:00] this? Why does it do that? How come it doesn't do this? So we've actually carved out a piece of equity for our strategic deployment team that's gonna help us build it. So that they could be part of this creation process and make sure that we take all of the best ideas from the top players in our space that do have the 20 plus years experience.

'cause the other thing we're looking to do is bring youth into the marketplace by giving 'em this tool. Yeah, man, it's pretty wild. I will tell you that I advise everybody to continue to practice, jump rope, get in your reps, become familiar with ai, what works, what doesn't, try the different systems, find your favorite and then start to build specialized individual things for specialized tasks that you need done.

Eventually, I'm gonna bring the enterprise solution to you, but in the interim, that doesn't mean wait. That means do the best you can with the tools that are available today. 

Speaker 2: The one thing I will say though, just to try and not make anybody feel like they're behind or too overwhelmed to start it.

It's just not too late. I, [00:38:00] in fact, if you had started a year ago. Everything you learned, everything you did wonderful. But it's, you can start today and in a matter of one week have learned as much as it took you one year to learn prior 'cause it's all changed so much and it's gotten so, so much better that it can actually guide you as to how to learn now as well.

Yeah, 

Speaker: I totally agree with that. If you look at where we are today is, and I'm trying to find a parallel, but let's just say let's go backwards a little bit. I was never a Jeff Bezos fan until now, and I'm starting to understand Jeff Bezos better and the tree and the way that he's done things and the mistakes he's made in a, as an entrepreneur, which are typical mistakes that entrepreneurs make.

And he was saying the other day, one of his key hires came to him and said, you have so many ideas that the speed at which you come up with ideas can [00:39:00] bankrupt Amazon. So what he said was, that was a really great sound piece of advice that he got from his COO. And he said, so what he did is he started holding back these ideas and that forced him to put 'em all down on papers instead of just vomiting them out of his head.

And then he said he had to perfect the ideas while they were manufacturing idea number 22. He was working on idea number 446, and he had to perfect them and prioritize them so that he could deploy them at the time that the manufacturing side of Amazon could catch up with the ideas that he was doing.

And so I was like, wow, man, as an entrepreneur, that's what we are, we're idea machines. And so that, that prioritization process is what I think every one of us should do is say, what do I wanna fix in my business? What will that do for me? Let me just flush this idea out, use AI as my assistant to flush this idea out and then determine which of these 10 ideas are most impactful to my business.

And then just deploy the one. [00:40:00] It's no different than anything we talk about at Sales Mastery, right? You can't do 10 things at once, but you could do one thing at a time, and I talk about it in my speech. That's just Habit Stack. So I would say Skill Stack, right? Learn a skill, perfect the idea.

Drop the idea in, master it, perfect it, implement it, move on to the next one. And if you think about at a very macro scale, the only thing that Amazon did for the world is build a better FedEx and UPS. It's it's all it really is a logistics company. It's not, it's really not an online sales platform.

It's actually just a logistic company that destroyed UPS and FedEx. I want to hit the button and get it at my door. Yeah. Yeah. What about Notion? Do you use Notion? No, I don't use Notion at all. I opened it once a long time ago, didn't like it, and I got off of it. Same thing with jva. Opened it once a long time ago, didn't like it, and I haven't given them another chance.

It doesn't mean it's [00:41:00] not good. What I would suggest to everybody, if you wanna learn something new notebook, LM Notebook. LM is my AI programmer calls that a meta skill. Okay, let's teach you a meta skill. The other things we've been talking about are all micro skills. A meta skill is notebook, lm.

'cause you can drop into Notebook, lm all the education from every YouTube video you want from every book that you wanna read. And then ask it to create a synopsis. Ask it to create a PowerPoint presentation. Ask it to teach you now with all the knowledge you gave it, how to implement that knowledge.

So I think Notebook LM is actually at the forefront of really educating humans about anything. How how is that any different than chat BTI just think it's a Google based product. And they're adding. It's easier to, the way that they're adding things into it, you could just, it's easier to drop in PDFs.

It's just functionally, it's just better. And I think, again, you're starting to hear about it. In, in [00:42:00] our circles we're using it my AI programmer is using it to teach his staff new things. When you watch the recording and they say, chat, GPT 6.0 is coming out and we're doing this, that, and the other thing he'll download that recording into Notebook LM and send it to his sales force and say, here, read the notes.

It just, I think it's the way the interface is better. 

Speaker 2: Okay. So quick question then. If it's a Google product, then how is it any different than Gemini? Can't you do the same thing in Gemini? And why would Google be competing with each other there? I 

Speaker: don't know if they ac if they acquired it. I think they may have been through acquisition.

And me personally, I hate Gemini. I find it to be the least reliable of all the LLMs now. They're coming fast, they're coming better quickly. But six months ago it was trash. Like it was completely reliable. And it's getting better very quickly. So I, I don't want that to say that it's not a good product.

But again, it's almost saying I'm building a house and [00:43:00] I use the Skillsoft for some stuff. I use a jigsaw for other things. I use a circular saw for other things. I use a hammer for other things. The different tools, do different things. Find the ones that work for you and utilize those on a regular basis.

But I think singular is not the answer today. Chat, GPT is the best singular answer. But if you want specialty, then find all these specialized things. Oh, I want learn something new. I'm gonna use note notebook, lm. Oh, I wanna write a blog. I'm gonna use chat. GPT. Oh, I wanna do research. I'm gonna use perplexity.

And look, I'm guilty of just default into perplexity 'cause I know it's so well. But in, in a perfect state, you would use the different tools for different things. You wanted to go back to the beginning. I'm a contractor by trade. That's how I got into real estate investment. So that's why I'm using these tool analogies 'cause I relate to that.

Fair enough. What is your tech stack the traditional Mortgage Tech stack today or by tech stack? For ai? Yeah, let's start with the Mortgage Tech stack. Wow. We're all stuck in the same stuff, man. I'm, I'm [00:44:00] using Blend up front. I've got I've got Encompass in the middle and I've got the total expert on the back, but that's gonna change at every mortgage company in the next two years.

P RMG is just about to release a new proprietary technology in house that gets us off Encompass and also has its own point of sale. So they're building a proprietary operating system as well. But it's still it's it's not really built proprietary. It's a skin that accesses all these things on the backside.

So that's one step in the direction. Of where we're going. It's cheaper and faster and can be deployed to everybody very quickly rather than actually building something from the ground up. But in the meantime, while companies are deploying that then you're gonna have people like myself building stuff from the ground up that's fully agent from day one, and it's not just a patchwork of different tools working together.

Speaker 2: What about mortgage tools that are out there right now for presentations let's call it [00:45:00] Mortgage Coach for example, stuff like that. What's happening

Speaker: with those kind of companies? Yeah, I think that they're challenged, right? They're challenged right now because unless you're an end, like I'm, I've been a mortgage coach user, by the way, just to clarify I've been a mortgage coach user since it was just a little better than a spreadsheet.

Don's been a mortgage. I ha I have two, over 15 years now, right? I've been using it. So for me, I have no reason to replace it. I think it brings tremendous amount of value and I have no reason to replace it. But if you're not using it today, I, there's a lot of people that are just dropping stuff in chat and saying, build me a comparison, put together an image.

There's a lot of people that are using the MBS highway tool 'cause it's a little more simple. Then the Mortgage Coach tool. And so they're saying, my, my customers are not the that savvy. They're not engineers. Mortgage Coach is too complex for them. I hate to even say it this way, but I guess what you really need to do is identify who your consumer is and then identify how you should present to them.

And yeah, so really like it, it can be as simple. [00:46:00] I have one loan officer does 65 million this year. Guess what? He uses an email, but three below. Option number one, option number two, option number three. Yeah. And his clients are uber affluent, like Uber affluent, and you're like, I can't believe that works.

He's it just works. I don't wanna change it. So do what works for you? Yeah, but I think all of those tools, I think every SaaS platform needs to find a new identity, reinvent themselves and come to the market with a fresh look and feel that maybe integrations with CHATT is where they're going.

If you start thinking about, now, which introduces, and I don't wanna get way too nerdy on this because we go down this rabbit hole, but what you're gonna be seeing a lot of is, you've all heard the term, A-P-I-A-P-I-A-P-I-A-P-I, APIs are being rep replaced or enhanced by MCP servers and MCP clients.

And that's the next level of integration. So think about what Zapier [00:47:00] did for APIs. CPS will allow companies to connect easier, faster, and more reliably. So I think the time of integrations is coming. That's how those systems can modernize themselves by tagging into LLMs and becoming a portion of that. I don't know if you guys saw, but Zillow struck a deal with chat GPT.

That's that's a game changer. Can you imagine you're gonna be searching for properties and chat, GPT, like Yeah. Pretty wild. 

Speaker 2: It's wild. So let's talk about CRMs for a minute. I'm struggling. I've always struggled with CRMs and I'm a user a big user, but I can never get anybody else to be a user 

Speaker: So I can do this 'cause I'm with you man. Ham. I'm beat head against the wall. It is not an easy adoption process. 

Speaker 2: No, it's not. And I spent all this money customizing [00:48:00] Salesforce and other than my personal team that was forced to do it, nothing. And I don't and I don't even blame them.

'cause it, unless you really spend big money customizing it a hundred percent, where it doesn't even look like Salesforce. It's just it. I, yeah. I finally had it when I had a younger loan officer come in and he just nailed it. He is, it's Salesforce and he's super techy and he is man, it just looks like something you used when you were a kid.

I was like, oof. Okay. It does, it looks old, it looks tired, right? Just like on LO's encompass lending pad, all, calyx. And then you gotta arrive and arrive. Looks new and fresh and easy and it's easy. CRMs, let's go back to this question now. Is there a world? 'cause I'm in the middle of go high level right now.

We're trying to put that together and make it all work. And I can't help but think [00:49:00] of can't we just have it be super easy? You look somebody up, it shows their contact information, it shows a couple little things and then chat GPT on the other side and you're just talking to it saying, Hey, here's, I'm gonna jump in the initial consultation I just did, or hopefully already integrated.

You already got their kids' names. 'cause I asked those questions. You already know where they work, how much money they make, all that good stuff. Later when we get the full loan application, I want you to update it all, but for right now, add this in and then later when you open it up and look, gimme a summary.

And then instead of going to have to go scroll through all these tabs and find just a quick summary on everything. Or you might even just say, Hey, when are all the birthday? Which you don't even need to do because it's gonna note it should notify you now. I thought, man, I just nailed it. I'm gonna stop spending all this money on go high level and I'm gonna do some vibe [00:50:00] coding and I'm gonna just create a super simplistic CRM and then attach it to some sort of, chat PT or something and just talk to it and have it stored.

But then just yesterday, I I'd been creating my clone in chat pt. I was pretty proud of it, knows a lot of stuff. But then I, last night I got home and my 11-year-old saw me working on it and I'm like, Hey, I got you wanna have some fun with me? I said, ask Chachi, pt, anything you want about me.

And of course he starts thinking of things that have nothing to do with anything I've told it, right? He says. Does my dad have any tattoos? And I have none. I got none. And he is ha. She says, haha. Yeah, that's a good question, Landon. Yes, your dad has tattoos. It's like, where did you get that from?

I don't have any tattoos. That's like, all right, that's where 

Speaker: the guardrails come in. And so that's part of the [00:51:00] orchestration layer that we call it is trying to put those guardrails around it so that, that's just plain lies really. It's just plain lies and it lies with confidence, which actually is worse 'cause you're actually, wow.

It might be true. And so what I would say is as I'm hearing you talk about this, it's something that happened by mistake on DSOs. So when I was building DSOs, I never set out to have it be A-P-O-S-I never set out to have it be a CRM, but it became those things by mistake. Which is the beauty of it is just so now.

Through an integration that we have that we're building together with Model Match. The minute I speak to a client and I enter a new borrower into DSOs, it goes to the Model Match database and it grabs all the information that I get that I can get from Model Match on that consumer on their property.

It's gonna integrate with title reports. I can pull all their title information, it's gonna automatically send follow requests to that consumer's Instagram, their Facebook, their LinkedIn and communicate with those things and [00:52:00] start to scrape all those things and bring all that data in. And you just had a five minute conversation in the car and now when you get back to your desk, you know everything about that person.

In fact, dude, fast forward, this is incredible. Yeah. Fast forward a little further and we're gonna have geofencing in the system and so you will know what coffee shop they frequent and so now instead of just sending 'em a Starbucks card, you could send 'em one to Pete's. 'cause you know that's what they like.

It's gonna know what sports teams they like. 'cause those are the ones they follow on their social media. So now when Ohio State wins a game, you could send them a special thing that says, congratulations on your big win. Hope you had a great weekend. And now you're really sending out automated, customized connections utilizing technology at your fingerprints at your fingertips.

And you'll be able to deploy all those things. It won't just do it automatically, but it'll say, would you like to send, would you like to send, would you like to send? Would you like to send? So you're just gonna be the person making the decisions as we, that's called hydrated data, right? So through model, match [00:53:00] and ingenious, I can hydrate almost any data that I want.

All the data's available. Why? 'cause you carry this thing. So everything you do here and hit accept, you know how we all joke around? I may have adopted like six kids by hitting accept here. So every time you hit accept is a data point that I can purchase and integrate into Dexo. Dude, what is this gonna cost?

Oh man, I don't think we have time for this.

Speaker 2: What are you rolling it out for? For everybody listening right now up. 'cause I'm not, I'm gonna roll it 

Speaker: out to wholesale first. It's the easiest place to revenue. And this is more business chat, right? 

Speaker 2: Wait, what do you mean wholesale? Yeah. Who are you gonna, who, who is that for you?

The 

Speaker: brokers.

Speaker 2: So my, my, so you're ta you're talking about ethos. My client. I have two 

Speaker: clients and now we're getting into my pitch deck. I told you in the beginning we might get into my pitch deck. How would you like to get it for free? Yes. I'm in. Yeah. Yes. Yeah. And so I, our go-to-market strategy is to give it to every broker in the nation for free.

Speaker 2: Ain't that some [00:54:00] shit.

Speaker: And we know who's gonna pay for it. So that's how we deploy it quickly, get it into the hands of the market and get IBS to beg us for it. Yeah. I worked for an IB and I've always worked for ibs. I was a broker when I started. Then I became an into an IMB and I wouldn't go back to the broker world for a million myriad of reasons for my own personal taste.

That's, it's a personal decision. I don't think there's any good or bad. Yeah. 

Speaker 2: I'm just going to go ahead and tell you that table that thought, because I can see an ethos shirt on you here any day now, and you never know what's gonna happen because that way your loans can just go and happen.

We'll take care of 'em. You're gonna go focus on dexo and you're gonna keep making things better for all of us. 

Speaker: Yeah. No, it's it's actually quite interesting. You and I could jump on a private call and I'll give you more of the roadmap. But the reality of the matter is we need, the reason most businesses fail is they're so focused on the product that they're not fo focused on the monetization [00:55:00] strategy.

And then what they do is they chase investors for more capital. Dilute everybody, chase investors for more capital, dilute everybody, add more features. So we have a go-to-market strategy that puts it in your hands for free, gets somebody else to pay for it. And if I give it to you for free, you'll use it right away.

I say it's the equivalent of Bo the broker market is the equivalent of bodybuilders. I could walk up with a little bag and say, man, I took this and I got huge. You won't ask any questions, you'll just use it. 

Speaker 2: I, this entire show, I've been formulating my own pitch deck on how I will become your ambassador if you give it to me for free.

And I, that was 

Speaker: time doing it anyway. No better way to get free. Many brokers are in the nation. 28,000 or so. Okay. Yeah. So I think I can get 20,000 to use 'em for free in year one. Yeah, 

Speaker 2: agreed. Agreed. And I will absolutely pitch it to everybody. 

Speaker: If it works half as good as you say. Yeah, there's [00:56:00] 20,000 subscribers.

It's worth quite a bit of money, whether depending what the irrelevant of the monetization screen what it is today. Yeah. The subscriber base is important. 

Speaker 2: Years ago, gosh, I don't know, maybe 10 years ago now, I went to a Tony Robbins seminar called Business Mastery and I actually will say it was probably one of the best seminars I've been to, of course, next to Sales Mastery just a few months ago.

So he said something that stuck with me and I've taken it to heart, and now all of a sudden I see it. He said, in the old days, you could start a company and it could just last for a hundred years. If it was a good company nowadays, and this was over 10 years ago, he said, if you don't sit back every five years, you gotta step outta your company and ask yourself this one question.

If I were to open a company like this again, what would it look like? [00:57:00] And if the answer is, nothing's changed, then consider yourself out of business in a short period of time. The world is changing. He said it's moving and it's moving faster and faster. Now, I don't think fi, I think five years is like probably five decades.

I think we gotta step back every year at this point and look at our company and ask ourselves, what can we do to make it better? Because if we don't, we're gonna, what's the term you used early on? We're gonna, we're gonna work ourselves right out of a business. 

Speaker: Yeah. No I I will tell you that being part of a AI tech startup forces us to do that almost weekly right now.

And there's we started with a roadmap that had 26 integrations, and I think we've now got it down to under 10. Because we integrate with less players to create more reliability. And there are aggregators of data and aggregators of services so that I can partner [00:58:00] stronger with these aggregators than if I was to go to each individual.

Like originally I was gonna have poly and lender price and Optimal Blue and loan sifter. And then I realized, what am I really looking for? I'm just looking for it to work. So do I really need all these different platforms or can I just select the one that's top tier in service and get access to everything I need and just integrate with them?

And and that was a function of something that happened when Steven came to Sales Mastery, or Steven's, my Bulgarian developer, my co-founder when Steven came, we spent, oh gosh, in the background of Sales Mastery, Steven and I are having Dexo Masterminds in the room for three hours at I'm already totally spent, just came out dinner, like we've been on stage all day managing the event.

And then I go upstairs and Steven's okay, and this and this. 'cause we hadn't had time to spend time together. And I can't tell you how much evolution happened for the product in just five, four hour sessions of us brainstorming together. [00:59:00] It's just amazing. You have to step back and reevaluate your business.

We've always heard the term work on your business and in your business and which one of the two are you doing? Yeah. And all of us are guilty of working too much in our business and not enough on our business. So I would say, depending on what space you're in, even weekly look backs or monthly look backs or probably a good idea.

And I thought about this concept the other day. Who takes the time to build a weekly business plan? 

Forget annual business plans. That's not often enough. Quarterly business plans, that's good. Monthly business plans, probably better. Weekly business plan. That's where it's at. Could you write 50 now with the help and use of ai, could you write 52 business plans a week that make you 1% better every week to give you exponential growth by the end of the year?

Speaker 2: That's a prompt. If I've ever heard one. I'm gonna write that down 'cause it's true. Yeah, it's wild man. 52 business. So 52 business [01:00:00] plans, call it 50. The other, this should be vacations. Okay? I can make you 1% better 

Speaker: every week. Exponentially better year over year. And then what I would say is these are the five subjects we're working on.

The five subjects are customer service, customer experience, marketing whatever it is, those things are for you, right? Every business is different. I'm not gonna tell you what those categories are, but I would break those categories up and say take these 24 things into consideration and let's, what can I do this week in my business to move forward on each one of these items?

Or you could do 'em, you could even focus them weekly. This week I'd like to work on marketing. Please help me. And you could, Matt, this is where it gets crazy. You could upload your calendar and start to have it create time blocks for you, right? So learning time [01:01:00] blocks and execution time blocks for the weekly business plan that you built.

Okay. The possibilities are endless, man. 

Speaker 2: They are. I gotta wrap this up, but I'm gonna ask a couple more questions 'cause I just can't. So first question is true integration, right? Like true integration that shaves hours off of your day. For example, I heard you say earlier, before you write an email, drop it in here.

Ask it to Phil Jones. It to me, that just sounds like so much time. I'm gonna, I'm gonna switch, I'm gonna go to this platform, I'm gonna tell it what I want it to say, and then it's gonna write it. And then I'm gonna go back into this platform and do it. Is there an integration that I don't have to switch around like that?

I can just say, Hey, write this email for me and put in my draft. Yes. My draft, yes. It was 

Speaker: very expensive. We're not there yet. We're not there yet. Okay. But we're in the infancy stage of email management with ai. So right now, perplexity that I [01:02:00] signed up on Friday, so I'm like three days into this thing.

And it's pre-writing some emails for me, but I haven't given it any training. It's using, its out of the box training. I don't even know how to train it to do that yet. 

Speaker 2: So Perplexity has its own like whatever you wanna call it for 

Speaker: specifically for writing emails. It's it's an email integration they just launched.

So I have Comet and I have assistant and email is part of the assistant package. But I haven't been able to train it. I can, I haven't been able to tell it Phil Jones it yet 'cause I've only been in it for 72 hours. I don't know if you can do that yet. That's a really good question to ask and then see if you can note Notebook, LM that and try to see if you could find a way to do it.

So I just, one, one thing I will caution you. Don't spend so much time building these systems that you don't use your job. 'cause your production will fall. Your team should let you do it [01:03:00] 

Speaker 2: and then deploy it to them. Yeah. And that's what I'm all about right now is I'm trying to figure out how to effectively use it for me and then fine tune it for a loan officer and then teach them and slightly force them, because sometimes you have to, but I don't wanna force anybody until I know it's actually worthy of.

There's so many little pieces that drive me crazy. Chachi pt, I told you my kid I told to ask a question, right? After that it asked how old is my dad? Does my dad look. I'm like, it doesn't know how old I look, Landon. And she laughed and said, ah, your dad, born in 1970, he's 53 years old.

I'm like, first of all, he asked, how old do I look? Not how old I am. And second of all, I'm 55, not 53. 'cause it. And I'm like and I was like, my birthday, how do you not know how [01:04:00] old I am? Like, that's not even a, that's not even anything big. That's just simple math. And she got it wrong. 

Speaker: Couldn't believe it.

You were gonna follow up that question with please take into consideration today's date. Then it would fix it. And again, so these, so how does it not know today's date? It, okay. So what we take for granted as humans is context. We take that for granted. We have a belief system. We've been trained, we've been coached, we've learned, we have, you have 53 years of it.

I have 47 years of it. We have innate context. Every time you start a new GPT, it has no context. It's zero baby. It's a newborn. Yeah. You gotta teach it. It's gotta learn. It has to learn what you want it to do. The one thing I'll say is if you have pro, you could share your GPT with other people and they can use, so once you got it, close enough, and also just, let's just be cognizant.

This is not meant to be a magic wand. Let's just say it's 80%, right? 90%. Is [01:05:00] that good enough for what you're doing today? Sure. For mine on enterprise level, I need it to be like 96 to 98%, right? Is our service level agreement. That's what we think we can achieve today with today's technology. On an enterprise level, with SOC two compliance, ISO standards true guidelines for enterprise systems, we think we can achieve 95 to number 96%.

Pretty easily getting to a hundred percent accuracy is not gonna happen, at least for the next couple of years. 

Speaker 2: Last question for you, and then I'll, I will let you go. Yeah. This one has nothing to do with technology unless of course you take it there. I love asking this question, and it goes back to just being a standard loan officer slash mortgage advisor.

If I were to pick you up and drop you off, in some city in the United States where you don't know anybody, you have all your knowledge, you have a little bit of money, you don't have a lot, you can't go deploy and pay for and do all this stuff, right? You have to go figure out how you're going to make a million dollars [01:06:00] doing loans as a loan officer, and you cannot go back and see your family until after you make a million bucks.

How are you going to do that? Probably gonna surprise you with this 

Speaker: question 

Speaker 2: or 

Speaker: Maybe not. Open houses and networking. I'm a very social person. I'm a very amicable person, and I can bridge the gap sooner and build trust sooner, belly to belly than I can in any other way, shape or form. So the people that you're seeing succeed today on social media, it takes 6, 8, 10 years to be a Neil Dinga, to be Deb.

Deb says she reached critical mass in seven years. So if you drop me in the middle of nowhere and I need a loan tomorrow, I'm going to the chamber, I'm going to open houses, I'm gonna go hit the realtor associations, I'm gonna sponsor the Women's Council of Realtors, all the old school stuff we did today, if you need to get money fast, that's still the best way to do it.

Speaker 2: I am with you, brother. I, it's funny because I love asking that question 'cause pretty much every single [01:07:00] person goes back to the grassroots of what got 'em to where they're at. And I think it is, it's the fastest way. And I'm gonna go back to that comment I made earlier about, I believe AI is going to allow us the flexibility and the time to actually be able to go back and do these things again.

Because we won't have to get so hung up on all the BS that we have to do all day, every day to do loans. And I love that answer. Yeah. So anyways, man, I want to thank you so much. I I didn't even get a chance to really ask you about Sales Mastery, but but I love what you guys did. I had a great time over there.

I really actually enjoyed myself and I wish you guys ton of success in Sales Mastery and DSOs and in everything 

Speaker: you guys are doing. Thank you brother. Really appreciate you having me on. I have so much fun and you could tell I'm not passionate about this technology thing at all. I call it the entrepreneur's disease, right?

I just, it's the people say, what's your hobby? And I said, tech. Yeah. 

Speaker 2: I love it. My, I don't know about your wife. My, my wife is she got a little upset the other day and [01:08:00] she goes, I don't know. Why don't you go ask your girlfriend? 

Speaker: No, because she's equally guilty. She uses copilot all the time and we're both like, ask your ai.

I ask my ai, I ask my ai. Yeah, we both, and we, because we both use different ones, like we have different opinions sometimes about the same thing. So it's a lot of fun. Great. Thank you. Max's great. Great. I love it. 

Speaker 2: Guys, we are here to empower, educate, and equip you the way lending should be done. And I'm always bringing the best in the business here for you.

So I hope you enjoyed it. Hope you got a lot out of it. And if you did, please don't just like it, but leave some comments so we know about it, and then share it with your friends. We love you and we'll see you next time.