Nate Joens - Peer 2 Peer Real Estate's Podcast

8:08PM May 27, 2022

Speakers:

Nate Joens

Keywords:

ai

structurally

machine learning

conversation

zillow

leads

seller

people

real estate

data

predict

company

software

computers

conversational

question

property

type

buy

built

Hello, everyone, welcome to peer to peer real estate Show.

I'm your host Willie Morales.

On today's show, I got Nate Jones. Nate is the co founder and head of innovation of structurally a conversational artificial intelligence that responds to qualified and nurtures your online leads, under his leadership, structurally experienced a 10x growth in conversations process in real estate industry, and built a proprietary conversational AI application that 99.9% of consumers believed was human. I'm pretty sure I would have been one of those two. Nate, thank you so much for being on peer to peer real estate show. How are you?

Yeah. Thanks for having me on. William. Looking forward to it. I'm doing good.

Yeah, no, listen, I when this opportunity came, I definitely want to learn more about algorithms, AI and all that. So I definitely you're the guy. So Nate, early on, did you know you wanted to be an entrepreneur? Was this something that you want it to do? Or just fell into your lap? as you got older?

Yeah. So kind of the founding story quick and structurally, was myself and my co founder, Andrew, in college. So I've known Andrew, since third grade. So we've kind of been friends ever since we knew we wanted to start something we didn't know what, in college, we decided we wanted to try our hand at real estate investing, which makes no sense because we had no money. So that quickly failed. But we ended up building a lot of relationships with real estate agents, lenders, people from the space, kind of the general property space, I guess, if you will, right, and kind of pick their brain here and there and constantly kept hearing, you know, hey, if you're going to do something, I really don't like following up with my leads, you could solve that. So we did with artificial intelligence, conversational AI, our kind of idea there was, you know, call centers have been around for the last 2025 plus years. That's a tiring job. It's kind of a mundane, exhausting, high turnover, high burnout job. And we said, can we solve this with AI? We thought at the time, aI had advanced enough to do so. We're gonna we found some really smart technologists from our respective schools who are PhDs in math, stats, computer science, and we set out to build a conversational AI to qualify leads, basically.

That's amazing. So early on, once you knew you wanted to do this, did you get the support that you needed? Or any pushback saying, Hey, you should just get a nine to five? How was that? How was your support early on, when you decided that you wanted to do this?

Yeah, I think there's an old, kind of saying that your parents will never understand what you do as an entrepreneur. So there was a little bit of pushback, especially right after I graduated school, after going for four years and said, Hey, I'm purposefully not going to look for a job. That was a weird conversation to have. But the way I always like to say is, is structurally going from college, I was already eating ramen noodles to starting a business, you're just gonna go eat ramen noodles again. So it was an easy transition. I thought it was a decent time to try my hand at a company. I never family, and no one to depend on me. It was just me and Andrew. So little pushback, not a lot of pushback. But then we got a really great amount of support from some local entrepreneurs in Iowa, which is where we're from. And they've been with us kind of ever since and still are,

you know, and wanted to ask you like, how did you get into this space? What was it about the AI field machine learning? What was it about that technology? If you could call it that? Tell me if I'm wrong, but what was it about that space that drew you in? Instead of let's say, being real estate investors and going to meet ups and all that stuff? What was it? What drew you in?

Yeah, that's a really interesting question. That had never been asked before. But I think about a lot. I think, kind of, in hindsight, one thing that I really liked about software is it's infinitely scalable. I like real estate, obviously, a lot. But to manage a portfolio of real estate or buy and sell real estate, or even be a realtor or lender or anything related is very involved. I guess I'll say you have to be emotionally involved with your client who need a lot of emotional support. They have a ton of issues. You know, you're managing property. Door breaks, you gotta go fix the door. It just like it's a very involved thing like that is your job to be in real estate. Full time is a full time plus job. Not to say that's not the case for software, but it's like, you know, when when something breaks in software, you still have to fix it. But you can fix it from the convenience of your home by typing on the right kids type of thing. So I think that, that seems a lot more enjoyable, scalable. And I think partially like, why we got into it, why I was interested is kind of twofold is, like I mentioned earlier, like, call centers have been there, done that been around forever. Yeah. And, like, it's a mundane job. I think the world of AI, although it sounds scary, and people always say, Oh, it's here to replace that says you to replace us, it's not your gig, it's here to take the mundane tasks off your plate that you don't like doing, like calling 100 leads and getting told no 99 times, and a lot more explicit explicit words than that. And the AI handles that. And I think that was a huge, like, goal of mine. And then the second is, like software itself, like applications, you know, like CRMs, dashboards, whatever, email, it's kind of been, like, developed in a really structured way, like, you know, you have a button to click to send a message or login and logout. Right, but what hasn't been developed yet is like a framework, a common framework for building an application around a conversation, like, how do you structure a conversation in a, in a software sense? And I think that was a really fun and interesting challenge to say, like, how can we map out a potential conversation? How can we think through everything that someone might say, at any point in the conversation, I think that was a really fun challenge to think there to still is. So

before we move on to the next topic, so when you talk about conversation, so you talked about between a buyer and a seller, because as an investor, you know, I gotta read a script and all that, and sometimes I might miss something, or, you know, you get excited that you might have a deal, and then you blow it because you missed the line. Um, can you explain that more about conversations, let's say, between an investor and a seller or buyer and a seller?

Can you? Yeah, open that? Yeah, absolutely. So in our context, it's between a consumer and our AI, which we generally call, it's like an AI assistant. A lot of our customers give it a name, give it a persona, and really treat it like a person on their team who is like an inside sales agent, or just an assistant generally. And it's having a conversation with buyers, sellers, investors, people who don't even know they want to sell or buy yet renters. People looking for a loan, literally everything and anything in between, we're having conversations are Converse conversations. While we have like a big library of pre built scripts, kind of like you mentioned, that you can use out of the box, you can also tweak them to your liking, or write them entirely from scratch. So if you wanted to ask some really specific questions and a really specific way, maybe you use the word like y'all, for example, where you can customize that level of granularity in our scripts, in ask and answer very specific questions and very certain ways through a product like ours, or you can use our default pre built ones too.

Right. So you know, like one of the main things when I talked to the seller, and we talked about creative financing, where we buying, you know, for sale by owner, and we talked in terms, most cases, if not all, the sellers not going to know what that is. So we have to dumb it down and say maybe, hey, if we send you the pros and cons of selling to us through monthly payments, is that something you guys can work with, with investors that say, Hey, I'm a creative guy, I don't use banks, I'd rather deal with the owner. Sets slash seller.

Yeah. So this kind of comes back to my original point of things that I like, love to think about. And that is think about, like, FAQ, frequently asked questions. Yes, yeah, this is super common thing on like, every website everywhere. But same same type of thing goes to what you were just saying there are frequently asked questions that your perspective for sale by owner leads will ask you as a person on the phone or whoever your assistant like, what is creative financing mean? What does that actually mean? Right, an FAQ. In our product, we develop something that we call custom answers, but you can actually think of them like conversational FAQs, right? Where, where we have over about 200 of these common questions that you can actually customize the answers to. So one of our frequently asked questions would be like, you know, what does financing mean? What is I think we have something like that. So you could write in a reply that are AI would use in that case, to say, you know, here's what'll happen if we want to buy your house that would be unique to you and your business and to your use case, maybe you have a bunch of different types of leads that might want that question answered different ways that can be set up in a product like ours. So it really becomes like tailored conversational FAQs.

Wow, that's amazing. So, um, if you can, so is that part of the machine learning? When I read your description of some of the things we wanted to talk about? Is that what they call real estate machine learning?

So I would say yes, and no, there are parts of it that are definitely machine learning. And in our world, that's essentially categorizing the responses that the lead is saying into certain data points. So if it says, Hey, I'm looking for a two bed, two bath home in Des Moines, Iowa, our product would cat classify that as buying home beds equals two baths equals two location equals 2.4. But if a lead said something like, Hey, you know, my approximate credit score is 700. What can I do for financing, we've categorized that as tell credit score, credit score equals 750. And ask about financing, for example. So that's the machine learning part where we're tagging, we have historically tagged upwards of 10 million 10 million individual messages by humans that are labeling basically messages from a lead to structured data, some some common classification of things that leads are saying, and then our product has kind of pre built responses that it will say, based on how that message was categorized. I was just gonna

ask that it. So it seems like it just can I say it's pre programmed to anticipate what a buyer or seller might ask,

not necessarily ask. But it's pre programmed to account for any number of things that they might ask at any time. Yeah, with certain responses that can be customized. So there's a there's a whole world in them. There's a whole kind of vein, in the world of machine learning. Yeah, that is called generative AI, which is something that we actually don't believe in at this time. Okay. And that is, that is commonly used for AI, or conversational AI or chat bots, that you want to basically think of, it's where you want to make it create its own reply on the fly, based on a bunch of data, you never know what it'll actually say next, right. And that's scary. There's been a lot of really horrible examples of that. Gone Wrong. There is one bot out there that Twitter made called Tay, this was like two or three years ago, right, was built like that. And it became a racist, terrible person on Twitter. And he viewed terrible things back and they had to shut it down. So there's a lot of bad history behind generative AI. Yeah, it's not fully understood or researched well enough yet. In theory, that's kind of how a lot of people think about machine learning mystically, as it's this box, it's just going to think of replies on its own. That's not, that's not the case at all.

Right? So is, is the future of investing for real estate is to algorithms? And if so, can you explain a little bit more about that? Because, for us, we're looking for, like I said, For Sale By Owner or distressed properties. You know, we look online, we look through whatever, you know, data is given to us. If you could explain more about algorithms, and real estate, if my question made any sense at all?

Yeah, I think there's all kinds of answer in terms of like, there's a different there's a bunch of different areas of what I'll call artificial intelligence, okay, and the way that our machine learning engineer, his name is Pete and he is a co PhD in math and stats who study computational theory. So he knows this stuff in this world. And he defines AI as pretty simply something that a human used to do that a computer now does. So that could be a whole bunch of different things. One example is like predictive analytics, Zillow. Zestimate is a great, great with password, because I know some people got a few asterisks. It's a solid example of what Predictive analytics can do. It takes a bunch of data points and says, Based on all these things, it's going to be worth this very common example of predictive analytics. There's natural language processing, which is what we use for our conversational AI. There's your vision, which is machine learning to do You know, look at pictures and identify what's in pictures or drive self, self driving cars, things like that. Right? Those are a few of the examples. So I would say that, like in terms of using different algorithms to help with investing, it would depend exactly on what you want to do. If your idea is you want to just simply like, give someone an estimate on the property that you might buy it for. That would be an example of predictive analytics, not too dissimilar to Zillow, Zestimate, for example, and that's kind of an done, that's kind of a bunch of times by a bunch of companies. Yeah, one thing that's really interesting that I've seen in the real estate, machine learning world, and I always forget the name of this company, and I'll think about it or think of it at some point, but they basically kind of map a lot of a city, or I think it's called city builder, city builder, I think. And they basically, I think they layer in a bunch of public record data, MLS data, a bunch of other data and say, what is the best and highest potential use of this property, this particular parcel, and they predict what you know, what, what type of ROI, what type of outcome could actually come from a change in zoning or a new development or something in this area? So they're trying to, like predict city building, in a way? And how it might, you know, turn out to investors, for example, who want to buy some of those lots or properties? I think that's a really creative example of that is, yeah, have one. I think that I think the company's name is city builder. Don't quote me on that. But that's the that's, that's coming to mind.

So are we seeing using the type of data that you talking about AI machine learning algorithms? Is that the wave of the future for real estate now? Or is I mean, is it more relevant now than it was 510 years ago? What you're seeing, or maybe conversations that you're having with other people about this?

Yeah, yeah, I would say. Yeah, I would say that a lot. I would say that AI has come a long way. In the in the, in the recent years, I think, when we launched structurally, we were kind of at the top of the hype curve, curve of AI. And then if you follow, kind of, I forget what chart this is called. But then it went into the trough of digital disillusionment where everyone was like, AI isn't so great. It's not doing exactly what I want. And then it kind of comes up again, to like, level out. I think we're there. We're there. Again, we're AI. We've been through the hype cycle of AI. And we're at a point where people believe it, they understand it, they trust it. I think we have to still be careful as an industry to not overhype it, for sure, we've kind of been there done that. So as long as we're really just really succinct, really direct and really honest about what AI can and can't do. One of the things I like to say on every podcast that I like to dispel as a theme is AI is not self learning, that is not a phrase that makes any sense whatsoever. Hear it all the time, even if the AI appears as though it is learning on its own. It is it has historically been trained by a bunch of humans who labeled a bunch of data. It's just you didn't see that part. And typically with AI applications, you never see that part. So it appears as though it's learning on its own, but it never really is. There's always a human in the loop, constantly training data, data is the oil that makes machine learning go,

No, I love that. I am glad you explained it that way. Because I think the biggest thing for like some of my friends that are in business, you know, real estate, and they want to invest outside of New York, where I'm from, and I don't know if you've been here, Nate, but most of the house prices look like phone numbers. And so I this, would this type of software that you guys have will come into play for investors that are looking to invest outside to get the data that they need, like, hey, is this a good place to invest? What are the price points that you know we can afford? They might question make any sense at all? You know, I mean, go ahead.

Yeah, I don't think that necessarily our product is the best for understanding that type of stuff. what it could do is kind of automate how you reach out to prospective buyers. It can help farming in a way if you will, yes. So if you want to farm a neighborhood and reach out to see who's interested in possibly selling and appending that with other data, like stuff from city builder for example. To say like you know, here's what could happen in your your market, here's what your property could be valued at. Is that something you'd be interested in ever or what what have you that If that is something that you can run through a product like ours, no, that

would be perfect, I think, because you want to get a feel for a neighborhood and whether you know if the pricing is right, or if the sellers willing to sell and you have that type of data available, I think that's that'll work just as fine. I guess the next step for us will be, obviously to get boots on the ground in that area. So they can come see the place and say, hey, you know what, this is a good place. I definitely go for it. Um, can we? Or can AI or machine learning? Can it make predictions? I think and you touch base on this, can you make predictions in the market? Going forward? You know, whether it's, whether it's, oh, we're gonna see another, you know, 5%? House value go up? Or 20%? Yeah, is that possible, making that type of prediction?

Yeah, not was a product like ours, but absolutely was products again, I keep kind of bringing them up. But like city builder, they have the ability to understand like, where the best development opportunities are for different cities. And then they're, you know, again, I hate to bring up the Z word Zillow, but ability to historically map and I think even today predict, like the future potential price based on typical appreciation in the area of, of your property. So it's not terribly challenging. I say that. And then I think companies like Zillow ran into a lot of issues where they couldn't predict it very well. They were trying to buy houses at scale. And, you know, flip them, but the problem they had was they could never really estimate what it would sell for in a very accurate way. And they would Oh, yeah,

they took a bath. And that part of the colder what I buying Yeah, I remember I did a show on that, because I read some articles. Yeah, please continue.

So I think it's Yeah, I think it's tricky. I think it's tricky for even the biggest companies in the world like Zillow to try and predict the value of any asset, especially real estate that is influenced by so many different micro and macro economic impacts, you know, they picked a funny time to launch a new, huge initiative in a pandemic, where things were changing drastically, right? So we've never seen that type of thing before. So their models probably were just going haywire.

I remember I saw an article was two, two years ago, maybe longer, where some, I think it was a seller. I don't know if it was a seller or buyer. And hopefully, I'm not misquoting him, I think it might have been the seller. Price on Zillow might have been, let's say, 350, I'm just throwing a number at you. But nobody was offering more than 323 10 Whatever it was, and they sued Zillow again, they got it wrong, I apologize, you know, but if it was through the Zestimate, that the seller or, or, again, buyer went with, and you know, their will completely wrong. And, again, they sued Zillow, and never found out what happened with that case. And again, hopefully, you know, I'm not spitting in the wind here. But I can see where what you're saying, you're making a prediction that how high the house is going to go in value, you know, whether it's six months a year, you know, I mean, I'm hearing anywhere from 20 to 20%, gain and house value in the next two to three years, then I'm seeing all by the end of this year, early next year, there's gonna be a price correction. So you don't know what to predict. Right. So, um, personal computers, and personal computers. Again, if my question makes sense, ready for machine learning? If so, what's needed to move forward? What, what will we need if I wanted to use AI? You know, machine learning, you know, again, this is all new to me. So my questions are, up and down. Please, please assemble my apologies.

Yeah, no, I think it's fair question. We use we have historically used our own pre our own custom built machine for machine learning. It basically had the same graphics cards as a lot of gaming computers does. That's kind of a funny fact. Funny, known, unknown. Kind of known fact, in the machine learning spaces. A lot of the super gaming computers are actually great for machine learning, because they use the same things. But we ended up pretty quickly. It was great for proof of concept type stuff, right? But we went to the cloud, and the cloud is someone else's computer. And they have bigger Amazon, Google, and other companies have way bigger, powerful, more powerful computers than ours. So it made sense to eventually do all of our modeling, machine learning, training of our models in the cloud on basically Google's or Amazon's computers.

Wow. You know, and it's funny because we always hear about Amazon, Google and how big they are and all that. But I think, given a chance with, structurally, it could work as well, just for people like us people like myself, because we don't know if we want to always go to Google and Amazon, if that makes sense. You know, we want to try something that might be suited for us. If, again, if that makes sense at all.

Yeah, absolutely. We've trained a lot of our models specifically for the property, space, property, leasing, investing, mortgage, whatever, what have you. And you know, a lot of there's a lot of different chatbot builders out there by like Google and Facebook and others. And they're, they're trained on the world of data. They're trained on very broad stroke, tons of data, but not very specific data. And I think that's one thing that is really interesting about narrow kind of niche AI companies is they take a really laser focused approach on we're gonna have really strong conversations in this space, and be really good at it and having a lot of training data on it that Google and Facebook has a hard time getting their hands on, because it's so good. It's so niche.

Yeah, I, you know, this, to me, this is exciting. This is some different. Where, you know, you have these conversations, and when it tells me a little bit more about how to 99.9% of people that that will talk to a human, if you can talk more about that, because that was great. When I read that.

Yeah, absolutely. So we have, through our product, we have the what's really interesting is we have the ability to granularly, see what people are saying in a conversation. So historically, if you if like, so we've had about 6 million conversations with leads, historically, that was basically all the data that was in those conversations was locked in those conversations, you could never write software to say, like, tell me of all those 6 million conversations, how many people are about to have a kid or mentioned they're having a new child? They're not approved for financing, and they are ready to sell in the next three months. Right? How would you ever do that? But with AI, because we're labeling every single message, basically, we can actually write searches to tell us, okay, show me all the leads that match that criteria. And the same is true for we'll call it like bad conversations. Yeah, a lead says something which happens common, you know, not too commonly, point 1% of the time in our product, where people say like, Hey, are you a robot? Are you a human? What are you it's kind of a weird conversation. You know, sometimes it happens, we're not perfect. But we actually have a label for that. That's called Ask if human and we have a response that says, Oh, I'm the digital assistant here for XYZ, company, whatever. And usually people are like, Oh, I'm sorry. Sorry for asking us back to the conversation. But then on the back end, we can go and write searches to say, show me all the people who asked if we were human. And that's where that number comes from. Right? Obviously, we're trying to get that number as small as possible. And we're really happy where it's where it's at now with 99.9%. Yeah, roll is 99.9999%.

Yeah, gotcha. When did you guys, when did you form the company? When did you guys what was what was the founding of it?

was about five years ago. Okay, when we graduated, me and Andrew graduated from school.

That's amazing. So in the five years that you've been in business, I mean, my God, the technology, the AI you I mean, you must start from I'm not saying the beginning. But pretty much it's, it's just growing. My God, why on a hourly basis?

Yeah, it's crazy. I think there's some really fun Tech Tech laws if you're interested, like Moore's Moore's law, which says something like, every year, two years, the number of semiconductors or something in a chip in a computer chip doubles. And that's basically saying at some point in our lifetime, you know, computer chips are going to be microscopic, but equally as powerful as one of the most powerful computers today. And it's not a theory, it's actually happening. Like Moore's law is tracked pretty religiously, and we're on track with it actually happening. So I think that's been interesting to see just like the power of computers thing in our time. But then also, AI has advanced quite a bit. There's really amazing bleeding edge research companies in AI. One of my favorite is open AI, which is a former Elon Musk company. Okay, and they just recently launched a new, the world's largest natural language processing model called GPT. Three, if you're really interested in reading what that is, it's basically a model trained on the entire, as much of the internet as they could scrape all of the words. And its goal is to basically predict. It can write books for you, if you say, hey, write me a children's book, and I'll write a children's book for you. It's just like crazy. The amount of things that it

can do anybody can be an author down the road.

Yep, exactly. And there are a lot of companies that are taking advantage of that type of tool to help with like writing blogs, writing emails, yeah, writing legal contracts, things like

that. Oh, that would be great. Yeah. Oh, that'd be amazing.

So I think so even in my lifetime, I think in starting structurally, they went from GPT, one GPT, two to GPT. Three, I don't even know what the next ones are going to look like, they're going to be insane.

No, I definitely believe that. So what's next for you and the company? What are you guys looking to do in the next three to six months? I mean, I know it's hard to predict. But if you had any goals going forward?

Yeah, I think from a tech perspective, one of my one why I wake up in the morning for structurally is kind of related to one stat. So there's a there's a survey out there research out there by salesforce.com biggest CRM in the world, they know salespeople better than anyone knows salespeople. And they have done research to say, salespeople spend 1/3 of their time actually selling so they spend two thirds of the time not selling. They're doing things like chasing leads, scheduling appointments, updating their CRM, basically non commissioned generating activities. So my goal and our goal it structurally is to flip that.

Oh, you're gonna say can you read my mind? And say, I guess you want to flip it? Yeah, yep,

exactly. So whatever we can do to help make salespeople, whatever industry more effective, more efficient, more, give them more opportunity to close deals. Since salespeople work on commission. That's what we want them doing. It's closing. We want to do that. So we're obviously trying to take care as much of as we can qualifying leads for you teeing them up on a silver platter. But then there's also other things we can do. Like we can maybe leverage things like GPT, three, to write your emails, to write your scripts to write your sales copy to write a bunch of stuff for you. So you don't have to do that. It's basically our goal to just flip that.

Yeah, I mean, it's basically outsourcing that to you guys. Which is great, because, you know, I'm pretty sure you know, this. It's it's, you work on your core, I guess, core genius, whether it's, you know, being out there talking to people that are writing a blog and sending emails you have structurally to do that. First of all, Nate, thank you so much for being on p2p Real Estate show. This was amazing. any books you would like to recommend? If you have one in you that you wrote, please say so? Or if not, you gotta write one.

Yeah. So oddly enough, even though these are all books by me, I actually don't really like to read. These are all for display only. I'm not a huge reader. I like to read blogs in short things like tweets and stuff. Because once I kind of read a book, I just, I like the few bits of it that are important, but a lot of it is kind of a waste of time, in my opinion.

But I agree, I would say about 80% of the year. It's just babble. I'm like, Yeah, okay. Yeah.

gave up on it after a while. So no real good book, book recommendations. Some of the things that I like to read in terms of like software are, there's, there's, I don't know if that's really interesting, but there's like a Joel on software. He's a former I think he's a former tech exec who talks a lot about management is on software. Joel Joel, software, thank you. And then there's another much more specific one related to software as a service businesses, which is what we are SAS businesses, but really great management techniques. Again, the the blog, or entity is called a Sastre s a s t r, and it's run by a guy named Jason Lemkin sold his company to Adobe. So he definitely knows his stuff. He has great management techniques and styles as well. So it's not just related to software companies. It's really more broad if you want to, if you want to look at it that way. Yeah, no, thanks.

I'll definitely put that in the show notes. And if somebody want to get in contact with you, what's the best way? Yeah, so I

always like to put my money where my mouth is. I always want people to try the AI themselves. Okay, have a conversation with it. So go to test your ai assistant.com. Test your AI assistant.com Fill out the form there. That's a great way to get in touch with us. First of all, since we'll get your contact info that way but You'll be responded to by Rei. And I always like to say this with one piece of caution. Try and break it have a tough conversation, but within a relative amount of reason. Sometimes when I give this link people will say things like, you know, hey, what, when are we going to find aliens and what language they speak? Okay, yeah, I got it. We'll have some response for that and we'll have a conversation. It's just like trying to be a little bit like a lead. Definitely have a hard conversation, but one within reason.

Okay, so let's test your AI. Assistant assistant.com. Okay, test your ai assistant.com. That's great. I'll definitely put that in the show notes. And Nate again, thank you so much for being a peer to peer real estate show.

I really appreciate it.

Absolutely. Thanks for having me on, William.

Yeah, no, it's my pleasure. Thanks.

Well, everyone that was Nate Jones, and you could find a match structurally.com That's structurally.com. Nate, thank you so much for being on peer to peer real estate show. Really appreciate it. You can find me at PTP realestate.com That's period number two period state.com. Check out our past shows and check out our blog. Also, when you get a chance please go to Apple podcasts please subscribe leave Review tell us that we can make this show better. And before I go, guys, there's a couple of more things. Do not give up on your dreams fight for it. God protected Don't let anyone talk you out of it. And I really believe keep the momentum going good things that happen on behalf of peer to peer real estate and winning mirallas Until next time, thanks, everybody. Have a great day and please stay safe but