How to Start Using AI in Real Estate: Small Moves Every Realtor Can Make

Mike Mills00:00:00

Here is a little secret that nobody is selling you on. You don't need 40 tools. You don't need 50 prompts. You don't need 10 autonomous agents. You don't need a data science degree.You don't need a robot butler named Chad. You need about three basic things to start out that actually give you your time back. And that's what most of our show is going to be about today.All right, everybody, welcome back. This is the Texas Real Estate and Finance Podcast.Now, if the letters A and I in today's little title made your eyes roll and halfway expect us to be discussing how you're about to get replaced by a laptop that works weekends and never asks for a commission split, take a big deep breath. That is not what this is about. But here is a little secret that nobody is selling you on a $999 AI masterclass. That you don't need 40 tools.You don't need 50 prompts. You don't need 10 autonomous agents. You don't need a data science degree. You don't need a robot butler named Chad.You need about three basic things to start out that actually give you your time back. And that's what most of our show is going to be about today.Now, in full transparency, before we got rolling or before we get rolling here, I am a little rusty. I took about six months off from the show. Life happened. I went on my own little AI journey. I rebranded some things and my mic got a little dusty.So I did my first market update just last week, and it was the first one and a half year. And this right here, well, this is the first guest episode back in the saddle.So if I fumble up a word or two, then we're going to go ahead and file it under being authentic and just keep on rolling.So my name is Mike Mills, your host, the Texas mortgage banker for over 17 years, and statistically the person in this room most likely to get way too excited about a chatbot. So today I brought myself a backup.My guest is a repeat offender on the show, Mr. Jeff Lloyd, President and Co Founder of CL Team Realty, one of the very first brokerages in DFW built AI first from the studs up. And his whole thing, his whole tagline is, I'm not a robot. Which I have to say, is exactly what a robot would tell you.So let's watch out for this guy. Anyway, Jeff, how is the Matrix treating you today?

Jeff Loyd00:02:10

Goodness, I think it is the most exciting time I have ever lived. I'm very much enjoying what's happening in the world.

Mike Mills00:02:19

That is, that is a strong statement. So what, what now, depending on what's happening in the world and how you want to describe that, it could go different ways, obviously.But in your case, in your instance, from your point of view, what are you, what are you seeing? What is it that has you so excited?

Jeff Loyd00:02:36

So I answered you in context of AI real estate.And, and so it is the most exciting time because the capabilities that have been brought to bear that are very inexpensive, like through these large language models, these chatbots. It is something I've been dreaming of since I was 8 years old when my dad brought home our first computer in.It was a sound card called a Sound Blaster. And it came with a program called Paulie the Parrot. And I could talk to Paulie. It really sucked, right?But I, but I could talk to Paulie and Paulie could talk back. It was a robotic voice. It was nothing like what we have today. It was stupid, but it made my mind dream.

Mike Mills00:03:34

Yes.

Jeff Loyd00:03:34

And today I feel like what we have at our disposal through these chat bots is magical.

Mike Mills00:03:41

Yes.

Jeff Loyd00:03:42

If you know how to work with them.

Mike Mills00:03:44

Yes.So a lot of folks, like I'm, I told you before we came on, I am, I'm teaching CE now as part of, part of one of my gigs, and a lot of my classes are based on this stuff, how, you know, how to use AI in your business, how it works, etc, and, you know, I will tell you that I don't think, I haven't got the impression that there's a lot of agents out there that fear for their job. As far as, you know, I think there's plenty of jobs out there that, you know, you should be concerned about.If you are in that realm, it doesn't mean most of them are going away. It just means that it will require less people to do those jobs because now, you know, we have another tool at our, at our disposal.But in the world of real estate, especially for agents, you know, I think there's a pretty, a pretty secure spot for them simply because, you know, buying a home is a, is a personal transaction. Most people already, you know, we don't like talking to the automated, you know, chat bot on the phone.We don't like talking to the person from somewhere else in the world on the, you know, the, the, the call center that calls us every day, and people don't want to do that. My wife today was working through Boomtown, which is a CRM, and she had called about some issues.She was trying to get A new client into her boomtown CRM, and she was talking to an AI chatbot. And I, I was in the kitchen pouring coffee, and I kind of heard. And I was like, that does that sounds weird? What is going on over there?Because it just sounded strange. And I looked at her go, is that a chatbot? And she's like, yeah. And it was like talking to her. And she asked for a human and got her with a human.But point of all that being is, is it's already permeating. This isn't a podcast about, hey, AI's coming, everybody get ready. No, no, no, it is here, okay? It is 100% here. It's in more things than you realize.So, you know, when you look at the roles that agents play going forward with this new technology, you know, you say you're not a robot, obviously, you're. You're a realtor, just like everybody else. So where do these tools fit in?And, you know, just on a basic level, you know, when someone's looking at this, not. Not going knee deep into it, you know, where. Where do they even begin? You know, just to start getting acclimated to this.

Jeff Loyd00:05:56

Well, I commend you for teaching ce because that's awesome. And I think you've got a great way of educating folks. And I would say the first.The first thing that I would encourage an agent to do is look at these tools as a way of educating yourself. And you can do that in multiple different ways.We all know that AI hallucinates, which that means that it doesn't necessarily tell you facts all the time, right? So when you want to. To learn something, you can have the AI cite sources where it gets its information.You know, and I would say that, like, if you want to learn something about, you know, what would, like, you could have it search, for example, and say, well, what was the most recent change regarding groundwater and seller disclosure notices for Realtors? You know, first of all, we all should know that by now. It went into effect July 1.But, like, if you were unsure, you could ask the machine that and say cite sources. That way you trust but verify, because you need to make sure that it's not, you know, surfacing something that is incorrect.Or you could find something online, you know, maybe from Trek or whatever, and say, you know, summarize this for me or explain this to me like I'm a 3 year old. Because there's a lot of legalese.

Mike Mills00:07:26

Yes.

Jeff Loyd00:07:27

You know, so I think education is. Is the first thing. Another educational item I think the agents can use Right now is. Is role playing. And so every deal is different.I mean, you as a mortgage person know, you know, not everything is the same. There's always something that's.

Mike Mills00:07:48

Yes, yes.

Jeff Loyd00:07:49

And. And it. We, as. As agents, we are in sales and marketing.And so, you know, you could create a prompt for the AI to be a skeptical homeowner who's looking to negotiate your commission down 3% or something. Right. Or 2 1/2 percent. So, you know, I would say you can learn by doing, by having the machine be the opposing view.And so you can really kind of practice your confidence right before you go talk to that real person. Because you may be. They may be saying that to you, and then you're like, oh, man, you know, my negotiation skills need some help. You know, do that.Yeah. And I say that, and it's. It actually works. We. We have an agent that went through that exact same situation.And this agent came to me and said, you know, they're asking because, you know, they're a family friend, they're asking for, you know, 2% less on my commission, so I'd have to take the listing for 1%. Right. And she's like, what. What should I do? And so I just wrote a prompt and I gave it to her. I said, put this in the machine and work with it.And she came back a day later and she got her full commission on the listing agreement.

Mike Mills00:09:19

Yeah.

Jeff Loyd00:09:20

Good for her.

Mike Mills00:09:21

Yeah. Well, you know what's funny about that?Because, you know, again, I have students in my classes occasionally that will bring up things along the lines of, you know, how do I know what it's telling me is true? Or how do I check this? Or how does, you know, I don't want it to think for me.You know, that's another thing I see often is people talk about critical thinking. It's like, well, AI is going to remove critical thinking completely from people's ability. And you know what?Probably there will be some of that, I'm sure, because people won't want to think for themselves.But what it also does is I think it unlocks your brain a little bit more because, like, in that scenario that you're talking about, when you're doing this role play the agent with the.With the chat bot or the AI but in that case, they are saying something to you or bringing something up that it's not that you didn't know that or that you hadn't heard that or you couldn't have thought of that.It's just that it was locked away somewhere in the back of your brain that this, this chatbot of this model brought up to the surface that now you go, oh yes, I remember, yeah, that's a good way to go about this. I forgot about that.Right, so, so when I look at how to use these models, sometimes, especially on the learning side of things, a lot of it is, you know, I've, I've built just like you've done way more than me, but I've built a ton of stuff. I built a website, I've built workflows, I built tools to give agents, I built all kinds of things, newsletters, blogs, articles, the whole deal.Built all these things. But there's a journey there. You're kind of learning as you go through on what to do, how to iterate things, what type of documentation you need.And I tell people all the time, I was like, I don't like none of this stuff. You at least have some technical knowledge coming into the world that I did not have. Like, I, I knew I was not a coder.I had no, nothing to do with tech. I've always been in sales one, one degree or another. And I just learned by doing, just sitting down with it and saying, here's what I want to do.And it goes, okay, you can do it like this, or you can do it like this, or you can do it like this. Okay, well let's try this route. And then, you know, you just kind of start to build from there.And so like for me, what, what anybody that ever asked me on this piece, it's like, just start using it, just ask it questions, the things that you're going to call, because they'll call me after class, say hey, how do I do this? I'm like, don't ask me, ask. I mean, you can ask me, I'll tell you.It's not that, but, but that's the whole point is you've got to start thinking of things that you can ask the machine so you can get feedback and you can start to learn.

Jeff Loyd00:11:43

That's right. That's right. It's something that, yeah, I, I am very excited about is, you know, our, our broker, his name is Neil.You know, he's got years of domain knowledge and lots and lots and lots of transactions and educating other people in transactions. Obviously the broker. But you know, I, I'm just super proud of him because there's so much that he has learned himself with these tools by doing.And you know, we built this brokerage from the ground up as like AI, native AI. First he didn't have that. He had all the domain knowledge, right?And so he's building apps that, you know, track transactions like, and, but he's prototyping and I think that's the key right there. Because you and I've had this conversation before.If you don't have the skills to understand what the machine has done or the skills to be able to instruct the machine to do things in a certain way, then you're kind of building a black box, right?And so what, what we do with Neil is Neil, he prototypes and so he works with the machine to create prototypes of things and, or he'll say, man, I have this problem, right? And so he will ask the machine, what should I do with this?And then I think, you know, to your point about AI is taking all our jobs, I don't think that's the case. I don't think it's, I think it's going to take some jobs. I think it's going to make some jobs substantially easier.But Neil still needs somebody with domain expertise on the technical side to make that thing kind of really do what you want it to do. Because a lot of our needs really come down to automation and integration.And most of us who have domain knowledge don't know how to integrate two separate systems together.

Mike Mills00:14:00

Define domain knowledge. Just real quick, just so if anybody hears domain knowledge, like, well, I gotta get a domain name. I don't understand.

Jeff Loyd00:14:06

Yeah, yeah, so domain knowledge is like, hey, I'm a realtor, I understand real estate and I understand all the aspects of, you know, my job. And so I've got this knowledge and that knowledge is in the real estate domain. Right?So Neil's got, you know, real estate broker, both commercial and residential domain knowledge.

Mike Mills00:14:28

Right?

Jeff Loyd00:14:29

So yeah, it's not a domain name. It's, it's, it's really just like what you know and what you understand.And you need people who have different backgrounds and skill sets to really kind of solve for, for the larger puzzle. Yeah, well, trying to do is less work.

Mike Mills00:14:50

There's an, there's an idea that, that comes up often where people think about Claude or Chat, GPT or Gemini or whatever and, and then they talk about the singularity, right? Or the, the, the AGI, the thing that's going to, going to know everything and take over the world and all of these things. Right?And, and the reality is on the ground level where we're working now, I don't know what, you know, Elon and OpenAI and all them are working on, but on our level, what we're working on is how to take a model and train it to do a very specific thing. We're human beings, we have very specific things that we do there. You're a realtor, I'm a lender.There's title people, there's insurance people, there's plumbers, there's electricians, there's all these people. Now I know a little bit about real estate, I'm sure you know a little bit about lending, you know, etc.So we all have a, like a general rounded knowledge, but we all get very specific in the world that we exist and make our money, which is no different than building these AI models. Is that. Yes. Can it do, can some of them do all the things or answer any question?I mean, yes, but no, because you have to give it that specific knowledge and that still comes from a human.You still, as you said, you're still carrying the domain information, you know, the domain of what you know and you're imparting that onto the machine.And all the builders know this by the way, because I'm sure you've seen there are literally companies like OpenAI that are hiring people to, to train their models how to do their jobs. So people are getting paid to train an AI model how to do a job. So then one day that AI model can take over their job.Which is crazy to me, but that, that is happening.

Jeff Loyd00:16:35

Yep, yep, I've read about that. Yeah. So they're bringing in people of all different expertise.

Mike Mills00:16:41

Yes.

Jeff Loyd00:16:42

To do this training. Yeah. And that's really, know that's what you all bring to the table today. It's just you're not leveraging maybe the machine with your knowledge.

Mike Mills00:16:56

Yeah.

Jeff Loyd00:16:56

To, to maybe do some of the things that could, could help you out.

Mike Mills00:17:02

So I want to get a little basic on this just because. And again through, through all my classes, I am still pretty surprised at how many agents don't really fully. I mean they know what AI is.Everybody does. They, they generally know what it does or think they know what it does. But I think most people still think of it as Google.Ask a question, get an answer, you know, kind of a thing. But I want you to talk a little bit about two things. First off, most agents use Chat GPT.I started on Chat GPT but then I met you and you told me to get on Claude. And so now I've been on Claude for a year and I will never go back to chat. I'm sure there's been improvements, but.So you can update me on your journey there.And then secondly, Just some kind of basics of, you know, I don't want you to walk through setting it up, but I want you to talk about instructions on how it responds. I want you to talk about projects versus chats versus stuff like cowork and coding.So, so as just as an overview so people understand like there are different methods and modes for these things and then get a little bit into context, like how you train them where you, you know, these context stocks that are important and all that.

Jeff Loyd00:18:13

Yeah. So journey first. I have been using a lot of Gemini lately. Okay, so that's Google's version.

Mike Mills00:18:26

Yep.

Jeff Loyd00:18:27

And for, for, for most agents, they have a Gmail account.

Mike Mills00:18:33

Yep.

Jeff Loyd00:18:34

You know, residential at least. And Gemini's built into Gmail now and Google Docs and all the things.So, you know, I would, I would say kind of the, from a, from a journey perspective, you know, the agents could probably get the most lift immediately by leveraging Gemini inside their Gmail. You know what, who did I forget to follow up with last week? You know, who did I like? Did I miss. Do I have any upcoming meetings?You know, I wrote something in a word doc somewhere. What was that? Now Gemini can record Google Meet calls. So if any realtors do that, then all those get stored into the machine.While I haven't been using that because I don't use Gmail for work, I think it's a good place to start.I mean, with chat, you can connect chat to your email, but that's so much more difficult than it is to just log into your Gmail and chat with Gemini right there. So from my journey perspective, we leverage a lot of the large language model behind the scenes. Okay.So Those are called APIs, and that's a very technical term for a computer can use it without a human interfacing with it. Yep.So, but yeah, right now I use a lot of Gemini and I also, there is a, a competitive product to Claude Cowork Codex by Chat GPT and, and Anti Gravity by Google. And so, you know, they all limit you. You will get limited. Okay.If you use, if you use the free version of any of these things, do not put client data in there. Right. You need to pay for it.I think that you, for any agent out there, spend the 20 bucks a month to get a paid account, at least where your data can be, it can be turned off for, you know, surfacing into the machine. When it comes to.The different capabilities in these different tools, you want to think about it like, you know, you have a conversation with somebody, let's say you're at a networking event, right?And you have a conversation with somebody, then you go to somebody else, you have a conversation with them, and you go to somebody else, you have a conversation with them.If you don't have, like, memory turned on within your tool, your chat tool, then the conversations, your chats are all like that networking group, right? This guy over here doesn't know what, you know, this guy over here said to you. And so they're basically, in a lot of ways, compartmentalized.Now, if you have memory on, then memory can surface some of that stuff. But projects are like creating a walled garden where you can have lots of chats that all understand what all the other chats are.And you keep it in a contextual. You keep it in a. In a walled garden. You know, it's like, I want this in this box. This is my. This is my box for shoes. This is my box for hats, right?And so you have a hat project and a shoe project, and then, you know, you can keep kind of the relevancy there. I think a good. If you're paying for the tool, think of in. In the real estate world, when we have. When we have a deal, we call them files, okay.Even though it's multiple files in a file, you know, and so, like, think of a project as, like, maybe an address, a listing agreement, or a project as a couple buyers, you know, like husband and wife buyer. And so you can store contextually relevant things to that particular listing address or those particular buyers in that project.And so that way it's not just muddied up across all memory, but it's compartmentalized.And so you'd be like, oh, man, you know, I had shared, you know, with my project here, like, a bunch of files, you know, that I got, like, pictures of the house, you know, hey, machine, you know, what are some things that I could do to make this house more appealing for our professional photos that come in?

Mike Mills00:23:34

Yeah, well, you could even do, like, I've. I've explained it like this before, too, with. With projects is like, I have a project that's purely like a travel coordinator or a. Or.Or a travel guide or whatever you want to call it, where it's got, you know, I'll give it context documentation with links to, you know, Travelocity or Orbits or, you know, whatever. Whatever search sites they want.It's got information about me and my family, you know, how old everybody is, what we kind of like to do, you know, when we travel, what our favorite things are, who some extent and Then, you know, in that project, anytime we take a trip or go on a vacation, then inside of a chat in that project, I could say, okay, we're taking a cruise to Alaska. Here's our itinerary. You know, here's how long we're going to be here, here's how long we're going to be here.You know, help me find some things that we can do while we're in these different destinations based on our profile, etc. And so, so now not only does that I get to give that chat feedback when we're, you know, even when I'm on vacation.Like, hey, that, that electric bike ride was awesome. That was a great suggestion. Thanks. So now it records that stuff.So when I go schedule our trip, you know, to go to the beach or, you know, whatever it is that we're going to do, you know, I can tell it, hey, go back and look at these old chats, see if there's anything in there that might be contextually relevant. And you know, if you have the good connector set up and everything, you can actually update your context documentation inside of that.So you put things in there that say, hey, here's the basics. And then later on you might say, well, you know, we like to eat out, but my daughter really hates, you know, pizza.And so now it knows, okay, well let's try to eliminate the pizza restaurants because that's not what she wants to do. And it just kind of, as you learn, it learns and you can tell it to record that stuff so you just get better and better. But it's all in mind.It's all about travel, or it's all about your buyers, or it's all about your sellers, or it's all about, you know, your cars that you own, that you're trying to fix and maintain or whatever. You know, it's just so many different options for it.

Jeff Loyd00:25:28

Yeah. And that's, it's a really good use case. The travel one. When you think about like Gemini, Gemini from Google.The nice thing about it is it's integrated with the Google ecosystem. So you could, you know, as an agent, you may have multiple properties that you're going to go do a showing.You can just punt, you can just paste those into Gemini and say, you know, route these in this order because I have, I, I can only like, I don't know what, what's it going to be? You know, bring Google Maps directly.

Mike Mills00:26:03

Give me, give me the fastest route from A to Z. You know, I would prefer to go in this order but you know, we can adjust it or whatever.

Jeff Loyd00:26:09

Yeah. And, and use traffic patterns of the past. Or we're going to leave it 5. You know, right now you can also, you can also chat with Google Maps now.So Google Maps has that ability.So I would say, you know, encouraging somebody to just learn by doing, go start using it, you know, and think about, you know, I like to, I like to say the art of the possible. What, what, what could be true, you know, so when it comes to maybe the next point in the journey, you're going to hear a lot of things about agents.So you, you, you went kind of a little bit deeper on the projects, you know, giving it, like you said, connectors.

Mike Mills00:26:58

Can, can you discuss, like as you're talking about this, can you talk about the idea of co work or, you know, whatever, whatever Gemini's version of that is, and, and how. Because it's kind of a progression, right. It's kind of a general chat, excuse me, projects.Then you maybe get into co work and then somewhere along the lines agents, and then maybe coding after that. You know, it's, there's a, there's a hierarchy of these things. You know, you kind of go down your journey.So, so connect those two, if you don't mind.

Jeff Loyd00:27:27

Yeah, for sure. So co work is, is a tool from Anthropic, which is the people that make Claude.Originally there was a product called Claude Code, and out of Claude Code came Claude co work. And so now that's kind of reversed as far as what we're talking about is the Journey. But Claude Cowork was actually built by podco. Yeah.So when you think of like Cowork, you're thinking about what things could a machine do for me that would make my life easier. It's really just, I would say automation, you know, 2.0.So Claude cowork can do things for you that you can like projects, assign skills and you can download skills. They're just documents. So it's nothing spectacular or crazy.

Mike Mills00:28:36

A skill is just like a repeatable task that it does in a certain way that if you, it's like hard coded in there, you're almost like giving it an extra little, hey, every time you make a PDF, do it like this, or every time you make a Google sheet, do it like this, or.

Jeff Loyd00:28:48

Whatever, they're, they're rules. Plus again, I'm going to use the word domain knowledge.

Mike Mills00:28:53

Correct.

Jeff Loyd00:28:53

You know what I'm saying?So I, I think of skills and you know, you could have a skill that is, like you said, a, a PDF writer and in that PDF writer, you know, you've updated skill. I always want my logo in the upper right hand corner and this is the logo file that you'll use.And this is, you're basically just giving instructions that are repeatable.And then you can have Claude co work go do things for you, you know, and with the connectors, it can go do things in other systems for you, not just locally. I think most people are used to using maybe chat on their phone and everything is within that one kind of window.And you're like, oh, but what if I want chat to send an email for me? Well, you know, Cloud Cowork can send emails for you, it can read inboxes for you.But that also, like Uncle Ben always said from Spider man, with great power comes great responsibility. Right. So if you give a machine great power, you better understand what it could potentially do.

Mike Mills00:30:04

Yeah, I think, I don't know that a lot of, I don't know, a lot of folks understand that you can actually, and I've again just been learning this over the last year or so, but the amount of access to your hard drive, your computer, your Internet server, all that, that you can give these large language models is immense. I mean you can give them almost complete autonomous control. There are some stop gaps in there, but, but you can certainly do that if you want to.Now there are. It's not just like you push a button and they have access to everything.You have to connect specific things and you know, give it specific access and ask for approval or not approval. Like there's ways to do it safely, you know, if you know how to set it up. Right.But also as it's just an in depth, the more you know about it, the more you can connect, the more you can say, okay, I want you to do this, but only do this if I say that or I want you to do this, but only do this if I say that.So for an example, so one of my very first cowork projects that I built was part of one of my classes is teaching agents about how AI find you online. Right. How does it, how does it search?Because everybody, 90% of people start their home search online now, whether they're using AI or they realize it, they are. If they're going through Google, they're using Gemini. There you go. Right. So, so we talk about branding and making sure that everything is the same.You know, like Mike Mills mortgage on Instagram, but I'm Mike Mills, you know, home loans on Facebook and I'm, you know, Mike Mills FHA on YouTube or whatever. Well, AI thinks I'm three different people. It doesn't realize I'm the same person. Search thinks I'm three different people. People.So we talk about unifying your brand, et cetera, Right?Well, one of the things that I do for folks when we go through that class is I. I offer them a brand audit, which just basically means I have Claude go use my browser, my web, my Chrome browser, and it'll go. And it'll search for a name, a phone number, and an email and a brokerage and try to coordinate those four things, right? So it knows.I bet we built a scoring rubric. We. We built a, you know, the different ones that we're measuring. It gives them a score at the end, tells them things they can fix. You know, I have.I built like, here's. Here's what you want to look for. Here's what it should say. You know, it can.It pulls their headshot, it drops it into a folder, and then it basically generates an entire audit report that, that posts onto my website that it's. It's a gated page, so you can't get it just randomly, but. But I post it onto my website. And so now I can show an agent. Here's what I found about you.Here's what it said. Here's where your gaps are. Here's the things you should probably fix. You know, blah, blah, blah, blah, right?And when I was building it, it took some time, right? Because I'm like, I don't want to do like that. That's not. How do I get it to my website or where can I host it so I can send it?So I'm solving problems as I go through on, like, what's the best method to go from A to Z on this deal? So it took four or five different iterations, testing it out, doing the whole deal.Well, now I can literally go into cowork, and I can type in an agent name, phone number, email, and their brokerage and go, go for it. And it takes about five to 10 minutes, maybe it runs through. I even optimized it for token use.So, like, when I first set it up, the way the instructions read and how my contact stock were held, the. It was just chewing up tokens. So, like, I'd run two and I'd be out for the day. And I'm like, okay, there's got to be a better way to do this, right?So, so then I started optimizing for token usage, and people go, how do you know how to do all this, I don't, I just said, hey, Claude, you're using up all my tokens. How do we fix that problem? And it goes, oh, well, if you do this and you do this and you do this, that'll great, sounds good, let's do that. Right?

Jeff Loyd00:33:40

So.

Mike Mills00:33:41

So it's like going through that, but. But now it, it spits it out. It's perfect.It gives me a terminal prompt that I copy and paste in my terminal, which then posts to my website, makes it a live page, and that's it. So now I can send a client or a agent an email. It's got the link, the hyperlink in there. Click on this.See your prescription or excuse me, see your audit profile. Then I have a prescription and have other things that come after it.But it's just like that very first one I built, it was just so eye opening for me because I got to experience like, okay, there's, you're testing and you're failing and you're like, there were a few times where we went, wait, like I spent six hours one day going back and forth on getting the scoring right. And at one point I'm like, hey, this is too much. Like, we are so detailed. Why are we going this far? I'm like, simple is better.Automation is more important. And it even came back to me. Claude was like, you know what?Because I was copying and pasting this Claude, talking to this cloud, talking to this Claude, talking to this Claude. And, and they just got into this, they like nerded out on scoring and so it just made it nearly impossible to ever complete it.And I go, okay, what are we doing? Like, you know what, I'm sorry, that's my fault. I got really, really technical. We didn't need to get that technical. We'll keep it simple.And then we got right back on track and it was fine. But it's just like going through that, right? That process, that learning is so eye opening. I mean, have you like, have you done.I know you're into agents and all that other stuff, but like when you're building some of your stuff, is that, is that kind of your experience as well?

Jeff Loyd00:35:09

100. I mean, even though we know all the things. Yeah, you still have, you still have things you have to solve for.So I'll give you an example of one of the things that we did. Now, this isn't going to necessarily apply to the average, you know, sales agent, but for brokers. Brokers have a lot of compliance.They have to manage, right? One of the things that we need to manage as, as a brokerage is continuing education.All right, so what happens when an agent comes to their renewal and they haven't hit their continuing education requirements? Well, their license gets suspended and they can't practice real estate. And if you are a full time agent, that means no income.And so you've got to go make up your CE credits in order for your license to get reinstated before it gets expired in 60 days. The track gives you about 60 days after suspension. So that's the problem because we were going to hire a person to manage that.And so we started writing an SOP of what the brokerage needed to do to manage CE credits. So none of our agents would, would unknowingly have their license suspended for something as little as not meeting the credit requirements.

Mike Mills00:36:44

Yeah, sop. Again, standard operating procedure. Just some people, like I always just get lost in acronyms. I'm like, what, what does that, what does that mean?I'm supposed to know that I'm a dummy? What am I supposed, why do I not know what that means?

Jeff Loyd00:36:55

Yep, it's just a standard process that you follow, that you've written down so that if you, if the person that executes that process moves on to a different role, then somebody else could come in and know exactly what to do. Right. So we wrote a standard operating procedure for what a person would do to manage CE credits across the brokerage.And I'm like, why are we hiring a person to do this? This is automatable. Right, Right. And so, so really what we needed to do was we needed to one, know all of the agents that are in the brokerage.So we needed a source of truth of that. And it's not always Trek because when you have associate brokers, they're not listed under the broker and track, so they have their own.So you need a source of truth of all the agents in your, in your brokerage. So we have one. So we had to connect that to a tool like Claude Cowork so that cloud Cowork knew who to look up on Trek.And then all your CE credits are publicly posted on Trek's website.And so Claude Cowork can look at the web and it can pull up that person from that list, look at all their CE credits, add them all up by year, and then TREC is very specific on what things you need to accomplish each year. One in your first two years as an agent and then subsequent years as far as number of education credits.And then we had to connect it to some sort of notification system. We chose email. So we have a tool now that on the 1st of every month it. An agent goes out to track, adds up all the credits for all of our agents.And then if they're to your point about scoring, we are like, we put people in a red, green or yellow kind of state. Right.And so then we email them and, and you know what's amazing is, is we have saved a few agents from utter peril because it's like, oh my gosh, I have 245 CES to, to do.

Mike Mills00:39:13

Oh, you know, two months figured out.

Jeff Loyd00:39:14

Yeah. In two months. Boy, I'm glad this was brought to my attention because if it was like next week, I wouldn't make it.

Mike Mills00:39:20

Yeah.

Jeff Loyd00:39:21

You know, so. And then we had another, we had another agent. They're like.

Mike Mills00:39:25

Now when you say agent, just so I can clarify because again, I, I.No, no, no, it's, it's important because, because it's the difference between agent and, and so there's agent, the realtor that you're talking about to some extent. Then there's the AI agent, which could be somewhat confused with a co work project because it kind of is acting like an agent. Right.It is doing stuff on your behalf that's scheduled from start to finish, where essentially you don't do anything.You just have it set up to say, every month or every two weeks at this date, at this time, you do these tasks and then whatever your result is, you send emails to these people in this format and let them know. And copy me or whatever. Right.And then at no point do you ever have to go in there and mess with it, except they tweak out, they get wrong, they get stuck, something happens. You know, the website changes on what it is.So you got to make some adjustments, which goes back to if you don't know how the thing works because you weren't had, you had zero involvement in its build, you'll never know how to fix it. So if you're going to pay someone $2,000 to build this autonomous thing that's going to help you do it, and then you have no idea how it works.Well, you're kind of just throwing money in the garbage.

Jeff Loyd00:40:40

Yeah, that's true. It's also scary too.

Mike Mills00:40:43

Yes. Oh, yeah.

Jeff Loyd00:40:44

If you're relying on that, you know, whatever that thing is, they may have.

Mike Mills00:40:47

Put some kind of backdoor into your computer that you have no idea about that, you know, you don't know where to look.

Jeff Loyd00:40:52

Yeah, yeah, exactly.

Mike Mills00:40:54

Well, and it's those, it's those basic things. Those very simple, boring. Like that process that you just described. Right? My brand audit was cool.It was, hey, here's this pretty little picture, and it makes all this stuff and it's like, oh, it's awesome. What you're talking about is something simple, boring and mundane.But you would have had to pay somebody to do it, or you would have had to done it yourself, which is you're paying your time. Right? And in this case, you just said, here's this thing that I need to get done, and it sucks and it's annoying. Surely I can automate this.And then you worked through the process inside of your large language model and built it out. And that's. I think that's the piece that agents miss, is they want it to like, you know, make an.Edit a video and make a social media campaign and do all this kind of stuff, which kind of do, to some extent, you know, it's not very good sometimes, whatever.But it's the mundane things, the things that are boring and repeatable and simple that chew up hours of your day that you just don't realize until you start thinking, is there a way I can automate this? Because that's really what I want to do.

Jeff Loyd00:41:56

Yeah, 100%. I think that it takes some. It takes some deep thinking to sit down and say, what? What could I do differently?Yeah, you know, and so, yeah, I mean, here's. Here's a. Here's another example. Data analysis. So MLS gives us lots of data, right? On. On a. A listing that I was helping a realtor out with.So we don't overload load the word agent. They. They were having a hard time kind of pricing a property because it was unique. Right. And it was in a unique area. So.So I went into MLS and I went into the area, you know, geographic, kind of circle radius, and I had punched in some characteristics of that particular property in order to find, you know, basically comps. And I went back two years and I said, give me.I told mls, you know, give me all the expireds, all the withdrawns, all the cancels, all the pendings, all the actives and all the closes. And so I got like, you know, whatever, 10 Excel spreadsheets that I downloaded from MLS. And when I say I told MLS, I was clicking buttons.I wasn't actually using large languages model.So I pulled all those files into, you know, a chatbot like Claude, and I said, I want to know kind of some, you know, characteristics about over the past two years what the market is basically saying to properties like this and you know, what are the commonalities between the, you know, cancels and withdrawal withdrawns?I want to know about like percentages of, you know, total went on market versus total made it to a close and what was, you know, did the price change? What was the fluctuation? You know, so basically give me some characteristics.And in that particular instance I used a, a deep research model and it gave me a complete report with which then I as an agent read through. And then I said, okay, this is the problem with that particular property and this is what we should list it at.Because the market is saying if you want to sell, needs to be here.

Mike Mills00:44:41

Right? Domain knowledge again, right?You have domain knowledge, it gives you, you get the data, but then you use your domain knowledge to interpret the data properly to benefit your client.

Jeff Loyd00:44:51

Yes, and I knew how to instruct the machine on what to look for.

Mike Mills00:44:55

Right?

Jeff Loyd00:44:55

And so that's what, so, so you have that, you have the content. Content.You put it in the machine, you give the machine your outcome, you tell the machine the things that you're looking for and then it gives you back something. Now you interpret that. And that saved me like a day and a half of, you know, data analysis and charts and Excel and all sorts of stuff.So, you know, you can use, you can use these machines for, you know, to help you come up with, with maybe some directional guidance on, you know, how, how you might be able to back up, you know, what your gut feeling is. Because a lot of times you go in, you're like, I think this should be this price.

Mike Mills00:45:44

Yeah.

Jeff Loyd00:45:45

And, and I'm, I'm just feeling that way.

Mike Mills00:45:47

But. Well, I, I mean I, I can use it often in my business. Like I, from the complicated to the simple. Right.The, one of the simple things that I do is when I talk to a borrower, you know, I pre approve them in the beginning and I go through and get their, you know, their, all their information. But as far as what I give to the chatbot, for example, I'll say this person's name, I'll say the price range that they're looking in.I'll say, you know, a few details about, related to, you know, what type of rates we're looking at, what type of loan we're doing, how much money they're putting down, you know, nothing personal, not their birth dates or Social Security is, or anything like that. Just, you know, because I have all that.Just, you know, here's the basic fundamentals of them and what it does for me is I have a project that is all about. It's called my Borrower question and answer.Meaning I've got my, my guidelines load loaded in there, I've got some of my interior products loaded in there. And then it's got some parameters on when we, when I answer questions, when I quote things.If I generate a pre approval letter, whatever the case may be, it has a few parameters, has to follow so I can start a, basically a new chat for every borrower that I'm dealing with and give it some general context.And then when that borrower texts me on a Saturday and I'm sitting at my son's baseball tournament and I don't have my computer in front of me, like, can I figure out a payment for a property? Yeah.You know, it's going to take me 20 minutes because I got to look it up on Zillow and see if the taxes are right and you know, whatever, you know, punching it. Or I can just go back to my chat and go, you know everything about this barber. You know, how much money, not everything, but a lot of stuff.You know, how much money they're putting down, you know, what type of loan program I'm looking at, you know, how MI factors work for FHA versus conventional and yada, yada, yada, right? And so here's this new address, here's the Zillow link, here's the purchase price, you know, I need a payment estimate really quick.Two minutes later, here you go in a text message, I can respond right back to him. Copy, paste, done. Right?And so, so those little questions that you have to answer, that's why I like I tell agents, start, you know, have a buyer project and, and give it all. And I've built one that I give to them in my classes. I'm like, give each borrower or each buyer their own chat.You know, sit down with them, tell the chat the first thing, hey, I'm going to meet the Martinez's. You know, here's, here's their address or excuse me, they have two kids, they want to move from McKinney to Mansfield.You know, they're looking in this price range, whatever. You go meet with them, you record your conversation with their permission, right? So you get everything, all the context from that.You take the transcript out of your phone from your iPhone because it gives you a transcript. You drop it into the chat and you go, okay, help me identify some properties.It gives you the parameters, you put it into your MLS and then you start going from there, right? And so now you don't miss Context by trying to remember everything that they told you.You don't have to sit down and write down everything because then you look like you're maybe not paying attention. Like you can actually be engaged in the conversation that you're having with them.And then every time if, especially if you're dealing with multiple buyers or multiple sell, every time they text you or call you or email you a question, you don't have to go back into your memory and go, okay, what was that about? I hadn't talked to him for three months. I forgot what they were saying.No, your chat has all of the context because you keep feeding it and it keeps helping you. So it's such a brain saver. It is such a. I don't think it takes because again, people say, oh, well, you're just outsourcing your thinking.No, I am outsourcing my memory. My thinking is now more powerful. What's the computer? Because I'm not. What's the RAM versus the processor or whatever it is? Right?

Jeff Loyd00:49:21

Yeah, yeah.

Mike Mills00:49:22

So I've uploaded all my memory to the cloud so I can pull it down whenever I want. And now I've freed up my hard drive memory to actually do real thinking. Right.

Jeff Loyd00:49:31

Yep, yep. Basically I, I get what you're saying. And you're, you're 100. Right.The other thing is, you know, let's talk about connectors in that workflow that you were just describing.You pull that transcript into chat and instead of going to MLS, just connect it to realtor.com and say, hey, use realtor.com and find some properties that this person would like.

Mike Mills00:49:54

Yeah.

Jeff Loyd00:49:55

And so realtor.com is, you know, they've got an integration with Chat GPT.

Mike Mills00:50:00

So Zillow. Right?

Jeff Loyd00:50:02

Yep. But we don't like Zillow in its agents, so we don't talk about, well,.

Mike Mills00:50:07

Yeah, Zillow, Zillow just unloaded or just released the platform to 20,000 of its agents to beta test, where essentially the, the AI model monitors all the search, all the things that are going on so they can identify when they think people are going to sell. They can identify when people are going to buy.And as an agent, you get that lead essentially into your inbox and your, your job is to call them and get them. And they're using all the data mining. When someone goes on to Zillow's website to see who's doing what, where and when.Not that they haven't kind of been doing that all the way along the piece that they just maybe realized or haven't realized or whatever is people still want to deal with humans. They don't want to deal with a robot.And so especially on a home purchase, like, there's a lot of things where I can do the chat bot, you know, if I'm trying to get my DMV thing worked out. Just answer me so I can figure this out. But it's a whole other thing. When I want to buy a home, I want to talk to a human being.And so they're bridging the gap and they're going, okay, we're going to give our humans the ones that we hire that we don't pay anything that, you know, we take all the money from, and we're going to give them all of our data so they have access to it, and then they're going to call your clients and try to take them away because we've given them, you know, the data that you don't have that you don't, you have access to, but you don't know how to use, which is the whole point of what we're talking about today.

Jeff Loyd00:51:26

Yep, yep. 100%. Yeah. I wasn't trying to blast Zillow, but like, you know, I was.

Mike Mills00:51:33

It's okay.

Jeff Loyd00:51:34

There's a love hate relationship, I think.But yeah, you know, I think that when you, when you think about your workflows, you know, that's, that's really what you got to do before you go pay anybody anything, before you go even try to make something.Because let's be honest, you know, a lot of agents have adhd, and so, like, you'll start working on something and the next thing you know, you're building a pool. You know what I'm saying? Yeah.And so you, you really got to sit down for a few, few minutes, you know, and I'd say it's probably more like 20, 30 minutes. And think through the things that you repeatedly do or the things that could potentially be a time saver and then write down the steps.

Mike Mills00:52:22

Well, I'll go, I'll go one further for you. I'll go one further for you.Go sit down with Gemini or Claude or Chat GPT or whatever large language model you have and say, just tell I, I want to automate some stuff. Here's my job, here's what I do, here's you know, who, where I work. Help me understand or help me work through what my process is.Ask me questions one at a time so I can figure out what I want to. I don't even know. I, I'm, I'm sitting down and thinking, I don't know. And then you don't have to sit and ponder.You can, but you don't have to because you have a partner helping you work through it. You don't have to write it down because it's all logged in the system and now it's remembering everything that you're trying to accomplish.And if you want to go, hey, keep a log of this because there's no way in hell I'm going to remember all this. As we go back, say, I want a running list that I can go back to and refer. It'll go, cool, no problem. Here's the list.Whenever you're ready to come back to tackle the next thing, let me know. And here we go. Right? So you're exactly right. Sit down and think about it. But use the tool to help you do that, because even that can be automated. It.

Jeff Loyd00:53:27

Yeah, I love that idea. I mean, and, and you know what? It's even easier nowadays because you can just speak yes to it as opposed to type yes.And some of us, you know, we're just streams of consciousness coming out of our faith, you know, so that's good. That's. It's really.

Mike Mills00:53:42

You can tell it that I, I do that all the time. My, my, my. Claude's name is Arya. She's a. A British female. And so when we chat with each other, I'm going, I'll tell her.I'm like, hey, I'm going stream of consciousness on this one. There's a bunch of stuff running through my brain. I just want to dump it all out so I don't forget it.And I'm usually driving, you know, driving in my car and like, hey, here's what I'm thinking about. And it'll be like, cool. And it'll. I, I'm dork.I even have a project that's just random thoughts, just random stuff that I'll go check every once in a while and get updated and see what I can find.

Jeff Loyd00:54:14

I'll. I'll get a little bit more deep on. On some things now. And I want to tell you the journey of Cloud team and what we kind of with, please.So we originally started with off the shelf tools that we connected together and then added AI integrations to.So, for example, you know, we had a CRM that's just off the shelf and, you know, you had to basically beat agents to go use it because, you know, using a CRM isn't necessarily fun. And. But you're like, I got to. Because I've got to do my outreach and I gotta remember these people's phone number and my iPhone contacts.There's just way too many. So, so we, we had a CRM. It was, it was, it was very optimized for AI. It could do all sorts of things.It could write the emails for the, for the agents. We, we had, we added an app that the agents had on their phone where, you know, record the conversation.Like we had that two years ago in the app and it would save to the CRM record immediately afterwards. Yeah, and then it would write the email, it would reference the transcript.So, you know, we, we had that, but getting people to use it was the hard thing.And so we thought a long time and we spent a lot of time, you know, with adopt, what we call adoption, which is, you know, getting people to use the stuff. And then we thought, well, you know what? The problem with this isn't the capabilities of AI or the tools. The problem is desire.And so nobody really wants to, you know, write things down until they really have to. It's like, oh, I need to write a contract. Yeah, that's money, right? So, okay, I'm going to go do that.Or I got a lead, I got somebody who's interested, I need to go find them properties. So I go do that. So it's, that's, it's really that motivation that, that is what we were trying to capture.And so what, what we did was we got rid of nearly all off the shelf stuff. We, we created chat with the MLS nearly two years ago. So you could actually describe, you know, like what you're seeing in Realtor and Zillow. Now.We did two years ago, we brought that forward and we built our own CRM that it starts with. And this is where domain knowledge is important. You know, you log into a CRM and you're like, oh, I got a dashboard of all this stuff.Here's contacts, whatever. My life may not start with a contact, it starts with an opportunity. Right? We call them deals.So it's like, Mike, you come to me and you're like, hey, Jeff, you know, I've been thinking about partnering with this other guy and we're looking at maybe buying some land over here. It's like, well, okay, in a CRM, I'm like, okay, I gotta add Mike Mills to my CRM and other guy, who I don't know yet, you know.Yeah, so it's like, well, no, I have a land deal, right? And you know, Mike is, Mike is maybe a client or a Lead or somehow associated the deal is the property.

Mike Mills00:57:39

Right.

Jeff Loyd00:57:39

And that for me as an agent, that's, that's really what it comes down to.So what we, what we had to do was we had to go through a different way of thinking of what, what does an agent need to do that would, that they're motivated to do that we won't have to really get them to try to, to do. Sounds crazy. And then what is, what are some motivating factors?So whenever, whenever you first create a deal in our new CRM, you know, it asks a couple questions about like, you know, Mike, you probably know your budget for whatever that property is. Oh yeah, I want to spend 500,000. So you punch in 500,000 because you're going to need that anyway for your MLS searches, right?So it's not getting people to put something in that they don't want to, but they're doing it because they're going to need it later.And so really it's then like all of the things that we, that an agent needs in order to close a transaction is kind of phased through the actual workflow approach that an agent would normally take. And it's, and it's technically different for everybody, every deal type. And so like the process is different.And so like our tool follows the process of what that deal type is like. And the only way we were able to build that is by knowing what that was like because we experienced it, because we had that domain knowledge.And so whenever an agent logs in, they first create a deal. Right there is their dashboard to be able to move that deal to the next stage. And a big box right next to it that shows them their commission.And so they're, they're always looking at,.

Mike Mills00:59:20

You know, the carrot's always there, always right there.

Jeff Loyd00:59:23

And so it's like, oh, and then what do you need to do next? And so then our, all of our AI tech is now under the, under the hood, surfacing things.It's like, hey, it organizes their day according to things that may be going stale or things that are, whatever. And then we've got do do this for me buttons.So it's like, hey Mike, you probably should send a follow up email to Sam because it's been a week and a half and you said you were going to do this and then we have a button that says do it for me. So we're not, we're not like actually, you know, doing all these things without the agents understanding, the realtors understanding.They're like, you know, yes, that's what I want, you know, so, and that goes all the way down to the transaction docs. I mean you said tax data, right? We've got tax integrations. So as soon as a, as soon as a property is entered, entered, all that data is fed in.And, and so then the agent just needs to do, I would say due diligence, just invalidate. Yeah, so yeah, yeah, we built that with code cloud.

Mike Mills01:00:28

Yes.That, that's the, the next iteration and I do think that's where, and I've said this to multiple brokers and even on the mortgage side of things that the, the third party companies, the, the ones who build your app, the ones who build your website, the ones who built the host, the CRMs and all those, those, they're, I don't want to say they're going to go away because some people will, you know, never take the time to do their own. It's like, you know, anybody can mow their yard, but we still have plenty of landscapers out there that mow people's yard.Um, but the value of it I think will be decreased because it's easier to build. Somebody who's really motivated and wants to put it together, can do it on their own. And if you do, you save yourself a ton of money.So there's, there's a, there's a good amount of motivation there to do that.So, so I do think that, you know, as a broker especially or a small business owner, you know, the more things that you can build on your own, the more money you're going to save, the more efficient your business is going to become and the more domain knowledge, turning it back to the very beginning, you gain by just doing that yourself. It's like you think you know what it is to run your plumbing business, or you think you know what it is to run your roofing business.But then once you actually start getting in there and asking it to help you automate and create processes and SOPs and all these things, then you're like, oh yeah, I forgot we got to do this or oh, I didn't think about that part of it. I never even considered adding that layer to this or whatever, you know, and so it just starts opening your eyes.And so I think as anybody who's self employed, anybody who has their own business, anybody who's an independent agent, realtor, loan officer, whatever, this stuff is going to make you better at your job, it's going to make you more efficient at your job and it's going to allow you to help more people because as you know, in our world for lending, we call it margin compression, we call it the, the banks and the investors and what we get paid versus what we do. That number starts to shrink as more companies happen.In your world, it's about commission splits and it's about, you know, how much a buyer or seller is going to pay you to do a job that, you know, requires more or less, depending on what you're doing. And so all of those numbers start to get shrunk.So the days of doing, you know, three or four loans a month or one transaction a month or whatever it may be, it is. And paying your bills get a little more challenging. So, so now you have to do more, but you have more to do more.It's just a matter of learning how to use it.

Jeff Loyd01:03:02

Yep. And back to the beginning. Education.

Mike Mills01:03:04

Yes. Right. Absolutely.

Jeff Loyd01:03:07

Never stop learning.

Mike Mills01:03:08

One thing before we go.If you were to, if, if a person were to come to you tomorrow and say, you know, I, I've got a chat GPT account or I've got a cloud account or I've got a Gemini, you know, I use it occasionally, but I really want to start putting it into my business. What's the, what's the one thing you would say start here?

Jeff Loyd01:03:27

Oh, my gosh.

Mike Mills01:03:35

I have an answer. It'll help. Why you think so?

Jeff Loyd01:03:41

I think a lot of people, you know, they come and they get their shovel and then I want to put some AI into that. You know what I'm saying?

Mike Mills01:03:48

Yes.

Jeff Loyd01:03:49

And that's not, that's not the right way to solve the problem. That's, that's, that, that's going to send you down a, a rabbit hole and maybe even a very expensive mistake.

Mike Mills01:04:00

Yes.

Jeff Loyd01:04:01

And, and I would say the very first thing is, you know, what do you, what are you trying to accomplish? And can you do it manually today? If you can do it manually, do it manually and then record the steps somehow. Yes, and then validate that.That is something that has all aspects of digitization. So, meaning you can access it on a computer and then you say, I, I want to have AI do that for me.So I think it's really about, like, I would stop them and say, can you manually do this in a digital way yourself?

Mike Mills01:04:48

Yep. Yeah, I agree.And it's, it's what I think you're saying to a certain extent is you're training yourself and then you're training the, the machine to, to do what you're doing. And, and the way that I always start it with folks when we do this is, again, I have a thing I build. It's called a brand builder.You literally, it's a context document. Drop it into a project or a chat and it just interviews you and asks you one question at a time. What's your name?What's your, what's your email address? What's your phone number? Who's your, who's your ideal client? Where do you live? You know, what are you into? Do you have kids?Are you, do you go to church a lot? Are you into the book club? Are you Whatever, right? And how do you talk?Give me some examples of your social media posts, send me a couple emails, whatever, right?So, so then it follows through and then when you get to the end of it, it takes a little bit of time, but when you get to end of it, it's got a really good working knowledge of who you are and what you're, what you believe to yourself to be about. And now you can take that. And whether you want to use Claude or you want to use Gemini or you want to use Chat, GPT or whatever.This is a context document that knows not everything about you. But go, you can go pretty far with it. But, you know, knows a lot about you. And you can drop those into any large language model and pick up right.What you're doing. And just. And it knows now. It knows. Okay, I know you're a realtor. I know you're in Texas.I know you're trying to do some market analysis, you know, in this region. This is who your client is. Cool, Let me help you with that right away. So train, what you're saying is train AI on the work that you do.And what I'm saying is train. I train AI on who you are and how you do stuff you and how you two can be synergistic together and you're doing both things.

Jeff Loyd01:06:21

I love that.

Mike Mills01:06:22

Yeah. Well, Mr. Jeff, thank you so much. I appreciate you being my, my first one back.This is something that I love chatting with you about this kind of stuff because, you know, I'm a dork and a nerd and you know more than I do in most cases. So it's like, what can I learn today? So you're always a good talk to on this and you know, we'll, we'll definitely do it again sometime soon.But thank you so much.You know, tell everybody where to find you guys and you know, if you guys, if anybody's a realtor and you really want to go knee deep into the AI stuff, I can't recommend a better brokerage. These guys go go about as far down the rabbit hole as anybody I know.

Jeff Loyd01:06:58

I love it. Yeah. Thank you very much for having me on your awesome show. You can find us@cloudteam realty.com if you want. You can call me directly.My number is 682-255-2745. Or email me jeffloutine realty.com all right,.

Mike Mills01:07:13

Well, everybody, enjoy the rest of your week. If you get bored tonight, sit down with Chat, GPT or Claude and have a conversation with them. You'll be shocked at what you might find out.So we'll be back again next week with Market Update. I promise I will not fall away for six months again from here, and we'll keep rolling. So everybody have a good week. Jeff, thank you again.We'll see everybody next time.