Efficiency in Land Investing: Using AI to Maximize Deal Flow

I once dined on steak with a candle soft glow but the bill left Mercedes despairingly low now wisdom has taught me so simple, so brand Eat baloney Buy land Eat bologna Buy land Trade the fork for the deed in your hand Skip the filet let the sandwiches stay and your fortune will grow when you buy your own land yeah, yeah. Oh, yeah.

Well, that was a rousing intro. Welcome to Land Geek Night. Cap, we're here. We're gonna have a great show tonight. We have some awesome guests joining us, right? But tonight, just, just remember, if you're watching the replay, we go live every Tuesday night, 9pm Eastern on YouTube and Facebook. Tonight's topic, we're getting a little geeky. We're getting, getting a little techy. So if, if, if folks out there are spending, you know, thousands on mailers for 2 to 3% response rate, what if AI could cut those costs for you by 7 to 12 times and boost your call volume five fold? That's what we're talking about tonight with Fred from Reworked AI. That's a mouthful, Fred. We worked AI Fred's a former land investor who built an AI system that predicts which owners are actually ready to sell. So this might be a big leap in land investing efficiency for all of us and we're ready to dive in here. So, Fred, welcome to the Land Geek Night Guy.

Salute. Thank you. So, just a couple corrections. It's not geeky, right? It's really easy. I'm not a technology person and I understand it. And I'm a current land investor, right? I got into this as, you know, as a land investor and, you know, I've got a funny meet cute because four and a half years ago, I was sending 12 to $15,000 worth of mailers a month. You know, listen to the talking head saying, mail, mail, mail. This gentleman called in and said, you know, and I'm giving him my spiel. I can buy her land cash easy. You know, it's a, you know, and he says, look, I'm not selling, but I'm just curious. I keep getting these letters, what are you guys doing? And so I was like, well, we buy land for below retail and we may do something to it, but then we're more than likely going to sell it. He said, but why did you mail it to me? I said, well, you own a piece of property in a state in a county that I like, that's the right size, it's got the characteristics that I like, the right valuation. And so I mail everybody. He said, why me? Because I'M not going to sell this land. So it turns out he's a data scientist and a computer engineer and my current business partner. Right. That's how we met because I sent him a dang mailer and we spent the next six months he was in Maryland and I was in Miami just talking about machine learning. Right. And this was about the time that all the LLMs ChatGPT were coming out and hitting the market. I didn't know the first thing about AI, and I really still don't. But what he did is he looked at all my data. And what we quickly determined was with a little bit more enrichment for demographics, we can make a pretty good guess of who's going to respond on who's going to want to do a deal. And it's. It's machine learning, which has been around for 70 years, and that's how we started.

Well, that's amazing. Well, we're so glad you're here with us tonight, Fred, to give us kind of a new perspective on. On mailing and, and response rates and getting the deal right. It's all about securing the deal. I would be remiss. I need to mention we have a very special guest with us also tonight accompanying you, Fred, Mark Podolski's back.

Mark, I'm back.

You're not a guest, but you're a guest because you're not a guest, man.

I know, I know. I took the summer off and it's great to escape the Scottsdale heat and come back and it's under a hundred, so that is a huge win. And I got to experience a lot of Europe and. And of course, my favorite place is Bali, but there's no place like home. No place like home. And I get to hang out with you guys.

There's nobody lonely.

Sandwiches and buying land.

Did that.

Did we, like, bring that up?

I think we did.

While you were gone. You're like, what is it like? Apparently this is just about bologna sandwiches now.

Yeah. But it's like, to Fred's point, like, who's eating bologna?

I didn't know they made it anymore.

Everybody who doesn't want spam?

Yeah.

But it is fun to say.

Very good. Well, we do have some folks chiming in tonight as well, so thanks everybody for tuning in. We have our. Some of our regulars here. Daniel, Brian, Jim. Jim Lockton's here. Good to see you, Jim. Shannon Morris is here. Don Thrasher. Good to see the regulars and good to see some new folks out there. If you're new, introduce yourselves. Tell us where you're from. But let's, let's get back to Fred and Fred, you know, take us back. You know, when you started land investing, how was the business for you? I mean, how was the business for you? And, and when you implemented this new system with your data scientist mailing friend?

Yeah.

How did it, how did it change things?

Yeah, so I got into land because I wanted to get away from the corporate grind. And my wife was looking at various ways and she said, let's flip some houses. And I said, no, I can probably change a ceiling fan, but that's the extent of my capability. And then she found some information about flipping land. And I saw some videos online and being a natural cynic, I said, these guys are full of crap. This is, this doesn't work like this, right? It's just impossible. Who's going to sell their land? $0.30 on the dollar. But I tried it. I made all the classic mistakes, right? I did it all wrong. Spent too much money, spent eight, nine months trying to prove that I was smarter than everybody that was on YouTube doing this. And after about nine months, when I made all the mistakes and I got into a groove, I started to get some, some, some deal flow, right? I started to find a groove. What worked for me because as you guys know better than me, every land investor does it a little different, right? They've got their little niche, they got their, their, their way that they approach it. And I just found a, a, a methodology that worked. Very simple, low key. My third year, I did 72 properties in and out and they were all small, right? I never had the courage to go for the home runs. I was getting singles. 5 in, 12 out, 7 in, 18, 20 out. Focused on Texas and Florida. Looked at counties surrounding the major metros so I could get it and then sell it. Because anybody can buy a piece of land, but as we all know, you have to sell it, right? You've got to figure out what to do with it once you buy it. You can't be like, oh, I got this great piece of land. What are you going to do with it? Right? So, you know, all these things that I learned and you know, I enjoyed it and I was thinking just before I was making dinner with, for my kids and I was, I don't do much land these days. In the last three months, I've done very little because we've been pretty busy. But you know what I miss about it? I miss that intense satisfaction of getting a piece of property under contract and then getting it out there in the market. And selling it and getting that wire, getting that check in, getting it from the closing agent, that, you know, satisfaction of, you know, I did something right. It worked. I saw it from identification, mailer, call, whatever you do on the phone, close, market close, get the money. That's a great feeling, right? There's nothing better. When you get that check in, you get that wire, you're like, I just made 10, 12, 13, whatever that number is. It's just, you know, it's really satisfying. And I think in what we're doing now with rework, I don't get quite that. That rush of like, wow, I just did something. It's. It's more esoteric, and it's a little bit further away. And I was just thinking that since we've just launched a rebrand and we've done some things with roofers, that I would get back into doing some more land just because it's kind of like fishing for me. It's fun. I enjoy it. It's good. It's a good thing to do. And if you're going to make a little money on it, then all the better.

It is like fishing. It's a lot more relaxing than homes.

It's certainly better than golfing. You get some money out of it versus giving it all up. But, you know, the. The. The adage that I. That I like to share with with potential clients and clients is, who gets rich at a gold rush? The people selling the shovels. Right? And one of the things that we really wanted to stay true when we put this together was, you know, and you said, it's got the seven to 12 times. We want a customer of reworked who's using Betty, who's using the machine learning to be able to say, I just saved 7 to 12 times. I spent $149 with rework to take my 3500 mailer, turn it into 1100, 1200 mailer. I'm putting 8, 9, maybe 1100 dollars back in my pocket, and I'm still getting my deals, right? I'm still gonna get the deals that are in there. And the other thing that this removes is the inconsistency, right? I'm sure you've all got a mailer ready to go. It's out the door, it drops, it hits the usps. And you're thinking, oh, I know I got some good, and it's a dud, right? You get very little call back, and you just. It's crickets. And you're thinking, you know, what did I do? Am I a Failure. This is not going to work. And then the next one, maybe you get some good responses, but this removes this, right? So we've had dozens, hundreds of customers do this, and then they'll reach out, like, wait a minute, you said it's about half, but I just put 8000 in. And Betty, our software saying, hey, only mail 1800. And the answer is, you're welcome because you only had 1800 people out there worth mailing to. The rest was trash and vice versa. It could be, I put 8,000 in. Betty says mail 6,200 because she doesn't grade the properties and the owners on a curve. She's not saying, oh, you know, Mark just gave me 8,000. I've got to get him 4,000amail. She's looking at the individual attributes of each owner or owners or entity. It by itself, right? And that's where the scoring and the propensity score comes in to say, you know, here's your 8,000 that you put in. Here's 8,000 back. It's the same exact file. It's reworked based on your propensity score. And then you can make decisions about how you want to attack this file. Do you want to spend a lot of time on the, on the top 15%, put some offers in there? Do you know, get a better mailer. You want to do the other ones, just a regular mailer. You want to text the bottom half, do whatever you want. But now you can approach that with some thought, some tactical approach versus just a spray and pray.

So if we can just go ahead. Mark.

Yeah, Scott, I was wondering though, like, like, you know, like I'm looking at Zano, John, right. Who are super geeky, I love chat, GPT, Rock, Gemini, Claude. But sometimes they hallucinate. Like, why should we trust Betty?

That's great. That's a great question. So the reason that LLMs hallucinate is because they're uncontrolled. Right. They're not supervised. Right. In machine learning, both ends of the equation are supervised. What the machine learning algorithms are going to see, they know the algorithms know what the input is. Right. So we have defined what the input is to the algorithms and define the output. So there's no opportunity to hallucinate because it's all controlled. It's a very defined environment. What we're looking at. When a list comes up at a minimum, we need owner's name and owner's mailing address. That's the triangulation process. That's the first step to say, John Doe 123 Main Street, Peoria, Illinois, whatever the zip code is. We check that. Does that John Doe still live there? Does the USPS have a moving notice for that John Doe? Is that address deliverable? Right, because we've all gone online to buy something and we're putting our address in for it. And we put 123 Main street in and that pop up, comes up and says, did you mean this? And you look at it, you say, well that's what I put in there. But the USPS for some reason sees it a little differently. You always pick the recommendation because you're not an idiot. So you picked it. That's what we're doing to check the mailing address. Because nothing worse than getting those yellow stickers on your, on your mailers coming back. So that triangulation allows us then to go to our data providers and we have five. We're going to our data providers to look for age range, marital status, credit score, income range, did the credit score move in the last three months? Or properties they own cars, they own membership subscriptions. Are they a member of a country club? Do they own a boat? Do they? All of these things that are readily available in the great United States for a price. We're not scraping the web, we're not getting this for free. We're paying for this information. We hit it via APIs that gets populated on your file. Then the machine learning runs, then the score is created and then the file comes back. None of that information that was used to create the score can come back because then our price goes from 0.02 cents a record to A$20 cents a record. Mark, did I answer your question about the hallucination of why the this doesn't see hallucination? Because the supervision in and out.

No, no, absolutely. And I'm glad you did because you're going to be at our dirt rich summit in Austin and so is Zano. And I was afraid if you didn't answer correctly, he was going to do jujitsu and like bring you down. But now it's like I don't have to worry about him being violent. Like that was a great answer.

I mean that that was the hardest part of putting this business together. It wasn't creating the algorithms, it wasn't me creating the algorithms, but it was getting the access to the data. And I like to tell the story. The first company we called and we said, look, we're looking for this slew of data, this demographic data in the United States. And we said, do you have that information? Yes, we have that information. We have 98% accuracy, updated every month. And they said, what part of the country are you looking for? And we said, well, the entire country, Everything. Okay, that's no problem. When your wire, when your ACH clears for 1.2.

It was gonna be a.

Big number, you know, Sorry, Fred, you.

Frozen the big number.

1.2 something.

Yeah, it was. It was $1.2 million to get access to this information. I mean, if I was sitting on $1.2 million, you know, I would be a really successful land flipper and I wouldn't have done anything else. But we had to go out and find companies that would sell us this data on a transactional basis, right? We use it, we calculate, and then we dump it. And they audit our code to see that we dump it. If. If Mark puts a list for 8,000 and Joe puts a list for 8,000 and there's 4,000 overlap and it's done within an hour of each other, we can't. We have to ask for that information twice. We have to pay for it twice. We can't use it as an overlap. But it's the data. The data is the key. Anybody that's a geeky nerd in anything will tell you the key to successful machine learning is good data. In.

Can you talk a little bit more about the demographics and the data that make for a good response? What are we tossing out on this? If I give you a list of a thousand names, what am I tossing out and who am I mailing?

The first step when we were creating these algorithms was I had my three and a half, four years worth of data that I shared with Sham, who's my business partner. I said, here's all the deals I've done, here's all the mailers I've done, and let me tell you what an ideal customer is. I had in my mind, look, I'm looking for someone that's. That's plus 50, doesn't live in the same county as the property, has owned the property for 15 plus years, has a credit score that, you know, it's not great, right? They're not well off, possibly not college educated. You know, just. That's what I had come across. That's what I had gleaned from my experience. And so we took my information, we populated the demographics with it, and it turns out I wasn't entirely correct, right? In the. The biggest indicator in our data is the credit score, right? And it's specifically, did the credit score move up or down? Something happened, right? Got a Job, lost a job, had a baby, death in the family, came into some money, something's happened. Whereas four or five months ago, if you had sent them a mailer, they were in a good financial shape, they got their W2, it's a hold, right? No, I'm not selling. No, forget it. Something happens, they lose a job, they get behind in their mortgage payment, that maybe they don't pay that credit card bill. Credit card credit score pops down a little bit. Now they get the mailer. You know, I really wanted to hold on to this property, but I think I better do something with it, right? I think, you know, maybe it's time to give these people a call and have a conversation. Credit score goes up, hey, got into a job, I don't need to worry about that land anymore. I'm going to get one less hassle, right? But the, the very long answer to your question is I don't know how Betty scores land anymore, right? I don't. Because over the last three and a half years, the learning part of machine learning is our institutional investors. As part of their arrangement with us, they have to every month download to us what they've mailed, what their inquiries were and what their deals were. And that's an ongoing flow and on a regular cadence, they have to mail the bottom half of the list to make sure that Betty's not a self fulfilling prophecy. So that goes in. She learns from what she's seeing, she adjusts. And now she has her own, her own loop going on that I genuinely don't know what she's doing. Right. And I'm okay with that because if you guys say, well, hey, you know, we have customers all the time say, well, I'm a, I'm kind of a doubting Thomas. I'm going to test it, I'm going to, I'm going to run a list through that I mailed two, three months ago. I got some deals on it. We just had someone was it last week, you know, for some reason he had gotten a thousand credits for free because we had done a podcast with some people. And he said, I'm just going to put a list in there. And he came back that afternoon, he says, dang, yeah, it got all of them. And you know, okay, now, now I'm ready to go. So we love when people try to break it. We love when people try to test it and you know, go for your life. But I can't tell you what she's using for her. Her, how she's weighing everything out, you know, is it 10 points for this, minus five for that. I genuinely don't know.

Fred, are you saying that the assumptions you had were wrong about the ideal client?

Wrong is a strong word. Right. What we, what we were able to discern is the, is classic bell shape for that credit score. Right. The people in the, in the top of it would, had a fairly normal range on their credit score. They weren't the high end, they weren't the high eights and they weren't the twos. Right. Those people in the center, you kind of average Joe and Jane, those were the ones that were more likely to engage in a productive conversation, to sell. Because remember, we're not predicting the inquiry, it's just too wide of a variable. We're predicting who wants to have a genuine conversation about an off market below retail property. The different delineation between inquiry and doing a deal. Because we all get inquiries. Sure. I'll take full. Full market. You know, it's worth a hundred thousand dollars, right. You know, no, who's, who's ready to talk turkey and, and get a deal done.

Yeah.

Like John Burnett.

Right?

Six kids. So each kid, you know, Betty's gonna be like, oh, male John, but he's wealthy.

I'm at the top of every list.

But, you know, because like there's, you know, the, there's constant change in his life. But would Betty know? Like, okay, he's got a good credit score, though. He's stable financially.

Yeah.

But he's also stressed. Betty's also going to look at the size of his house. Right. Is he an owner? Does he own a couple cars? Right. All these things that get recorded and transcribed by some entity that is being bought by another entity to create a picture. If John or somebody else has six kids and everything else in their life is a dumpster fire, then it might be a different score. But you could have six children and be just fine. Right. I couldn't. I got two. And that's, you know, I'm full. But yeah, there's, you know, one of the things that Shyam, who is the data scientist, he always tells people. There's never one single piece of data that will determine the, the outcome when using machine learning, it's the entire picture. And this is. If I were to give you gentlemen four properties by four owners with all of this demographic information, and you had 20 minutes, you could look at it, you could deduce through your experience and everything you've seen and make some assumptions and probably assign a Betty score, could you do it for 8,000? Right. And could you do it for 20,000? We have clients that are upload 50 to 100,000amonth. And what it's saving them is time and money, right? I, I remember watching this video four or five years ago. Jared, he used to be Seth Williams sidekick, right? Great guy who was, who would sit around and talk about how he and his brother in law would spend three to four days scrubbing a list. So I was like, all right, I'm gonna scrub, I'm scrub these lists, right? I got my Excel sheet and I'm gonna, I'm gonna scrub this two days in. I'm like, what am I doing? I have no idea what I'm, what am I scrubbing? I'm probably just making it worse. This really sucks. I don't like staring at an Excel spreadsheet for two days. I didn't get into this to stare at an Excel spreadsheet, much less a computer for two days. So I just listened to the other person on YouTube who a different philosophy is mail it all. Mail it. All right? One deal, you get it, it pays for it all. Which is true. One great deal out of a twelve thousand dollar mailer, you, you're gonna pay for it. But what if you didn't have a $12,000 mailer? What if you had a $3,000 mailer or $4,000 mailer and you still got the deal? It's all about efficiency. It's being a little smarter.

I think it's interesting. I think it points to a bigger, it points to a bigger. We make assumptions in our business all the time, right? And if we don't look at the data, it's like we just assume this is the way it is. So what you're pointing out here, I think it has bigger ramifications, right? I mean, obviously we're talking about very specific part of the business, the mailing, but this could look at our whole business. Like we just make these generalized assumptions and maybe if we dug a little bit into the data, there's some things we would unravel that we just took for granted. So I think it's really a great.

Point look at, I mean one could probably trace a lot of the things that go wrong in our society to human bias, right? We all have it, right? We all have glasses on and we're seeing the world through our historical perspective and our, and our bias, right? And it's generally crap. It's not. Right? So anytime in your business or your personal life you can validate something with. And whether it's a Friend, Right. A, a spouse, co worker, say, hey, what do you think about this? And as long as that person's able to give you, you know, maybe their bias is off as well. But between the two of you can get to something. Right? Right. Just to kind of lower that noise in our own brain. Right. I mean, how many times have we all gotten. I know I have gotten worked up about something. This person said this and it totally wrong. It's nothing to do with it. I agree with you. The human bias, the approach to a problem. We see it, we, we think we know the answer. We're not even close.

An old friend of Scott and I used to say, data, not trauma. Right, Scott? And this is what we're talking about here.

Yeah, I, I always say, you know, the only thing I'm certain of is my uncertainty and Scott Boston's genius.

Right.

And good lucks.

That's it.

That's so nice.

Everything else uncertain.

Right, Right. Go ahead, John. And then we do have a couple of questions from the audience that I'd like to throw up on the screen as well.

Just two more comments before we get into the questions. One is, and this will probably come up in the questions is it took us about a year and a half to be able to crack the non Persona owners, the LLCs, the INCs, the trusts. Right. The schools, universities. And that came about because of that feedback loop. We had investors say, look, I mailed this. Betty scored it low, it was owned by a school. Betty scored it low because it was owned by a school. And what we learned was not every school is the University of Michigan, not every school is Notre Dame. Some schools are little, small, own a piece of property, may want to get rid of it. And so that feedback loop and being able to get about 11 data sets, maximum on entities, LLCs, trust, incs, it was tougher, it took a little bit more time and it's not as sharp as a person as you can imagine. But don't scrub your list and remove all those things. Just remove duplicates and put everything in there. Try to do as little to your list as possible because all you're doing is restricting the flow into the pump. And the other one, the other. I always, always tell people this. If you use us and you don't get the results that I'm talking about, let me know, we're going to give you your money back. We're not here to get rich off of $149 from this person or 289 from this and keep it and tell you to pound sand. Nope. Here's the money. Give it back. If you could let us know what you did, what you were dissatisfied with, that's good feedback from us. We take it, we learn from it, we move on, we walk away. No hard feelings.

I think Mike's being served a drink over there. Thank you. That's awesome.

New notebooks.

New notebooks.

Oh, new notebooks.

Good.

More note taking.

Yeah, I noticed there was a lack of note taking tonight.

Now, this is going to be resolved right now.

Pull the pen out. Yeah, yeah. Very good. John, I think you had a question for Fred.

Yeah, I had a couple that came to mind. So one thing is that you talked about is looking for recent changes. So a recent change in credit score and stuff like that. So I guess with that my question is with as we are submitting our list to Betty and retrieving them back from her. So are. Is it the kind of thing where I would upload a list one month and then a month later I would upload that same list and I would upload a list one month and then a month later I would upload that same list and I would say, oh, here's 20 people that are on the list now that weren't on the list last week or last time around. And so maybe they're.

They're more.

Right.

Some people do that. Right. I wouldn't recommend doing that one month to the next month. Right. If, say, you, you know, say you had, you know, we've got the three tiers, 3500 list in for $149. So you're going to run 3500, and then the next month you say, I'm going to see how this list changed and you're going to pay 149. Again. I would advise run the 140, you know, run the 30,500, action it. Get another 3,500, action it. Because if you're going to run that same list again, you're kind of scraping at the bottom of the barrel. Just get a whole bunch of new dirt, see if there's any diamonds or gold in there, mail it out. Because that is in. In our whole land flipping world, data is like teas in golf. It's the cheapest component of what we do. Everything else costs money. We've got golf clubs that are thousands of dollars, balls that are 20 bucks. The tea is the cheapest part. The data in our world is the tea. Get the tea. Get lots of teas. Don't run out of teas. Run them through the machine. That's where your eye, you know, that's where your opportunities are. I just came up with that. I. I kind of like that.

Oh, yeah, let's trademark it. Write a book.

Get the teas that's coming. I just actually got the domain, Fred, just now. Get the tease dot com.

No. You could buy it for Mark.

No, no.

No way.

After millions. Watch this, this podcast. Are you kidding me?

Right? Right. Yeah.

No way.

That's great.

For sure. For sure. So, Fred, we have, we have a few questions coming in from the audience here. I think one of the questions is, is like, this is all tied to property owner, but have you explored, you know, like looking into property characteristics? So can you enter parcel numbers from a certain county and filter properties that way to determine, you know, which properties are worth mailing to?

Yeah, so that's a great question. Right. And something that was on my wish list in the beginning, right? It's the. To kind of the holy grail of is this property worth a dang. Right. Is it landlocked? Is it on a slope? Right. Is it. Is it 10ft wide and 500ft long? Is it swampy? You know, and the answer is we attempted in the beginning to do such things as a box coefficient. Right. Is the box coefficient 1? Is it 100, 100ft by 100ft? Right. But at the end of the day, what we determined, and I was again biased, I reversed myself on this because we would have people say, look, I want you to filter out all landlocked properties. And I would look at them and say, why? Well, I don't buy landlocked properties. But why? Well, I just don't do it. Why not? As long as you know how to sell it and you get it for the right price, I don't care what's going on in the property right now. If it's a bad property and it's on a slope and you don't think you can sell it, don't buy it. But don't artificially remove something from your list of opportunities. I bought swampland in South Miami before for $5,000, sold it for $55,000. Right. So it's just don't think that you're smarter than the market now. Don't sign the deal.

You.

And don't send the check if you don't know how to get out of it. Right. But looking at the property characteristics, that's up to the individual investor of what they're willing to do, how much they're willing to spend. I mean, there's a Lot of good people out there trying to develop a way to value that property, and they've made progress. But I may want to get into a property for $5,000, and you may think, I'm not touching that for, you know, for anything more than two. Right. It's all about our own individual profit level, what our overhead is and what we want to do. There's just too much individuality in that. You got to figure out what you want to do, get. Get the types of properties you want. And that's totally in. In the land investors playbook.

I love that because I always say, don't be a land snob, which John Burnett might be.

I mean, that's a. That's a real elegant way to say it. Don't turn your nose at something. You just don't know what's there.

Yeah.

Figure out what you can do with it.

Be it.

Be a. Be a solutions person. Right. Just figure out how I can get somebody else, how I can sell this. Because I mean, to me, that that's the. That's the art in land flipping, not buying any orangutan can buy. How do you sell. How do you convince somebody that this property has value for them and they should pay you $30,000 for a piece that you just bought for five?

So, Fred, we basically would have our list. We send it over to you or put it through your software, and then you send back that list and it's graded as to, you know, like you said earlier, you might have a certain portion. You do one thing, text, whatever, whatever the technique of the individual investor might be, but you get it graded in terms of potential for response, and then you make a decision based on that. Is it just. You get a list broken up into segments that comes right back to you?

Yeah. So pretty much your 8,000. You go on to rework AI, you create your account, you pick your plan, you put your payment in. You go membership or you go subscription or not, no monthly commitment. You upload your, you know, you go to the. Click, browse, you click the button, you go to your computer, grab it, throw it in. Betty will tell you, I got it. I recognize the file now. Now, she does use a large language model to read the headings. It's proprietary. All of our algorithms are proprietary. Sorry, it's this crazy water I'm drinking. So she'll read it and then she'll say, I've got it. And then 15 to two hours later, depending on the size of the file and her activity with the APIs to get the Demographics. She sends you an email that says your file's done. You go back to the website and it's in your dashboard and it tells you the potential savings. It's the same file. You download it, CSV or Excel, and it'll be scored and ordered based on the propensity score. And on the far right hand side will be the updated address. If you had an address that was wrong or the person moved, the updated address will be there. So if there's an address there, use that address because that's the one that USPS says that person is either at or they can deliver to.

Do you know, Fred, about what percentage of addresses generally get updated?

That's a great question. I don't know, but I should know that, right? We've got enough data now to see. And that only came about because one of our clients asked, hey, can we get. If there's a. Because we told. We've always been checking the address to get the right person. And this person said, hey, but you could you tell me if there's a different address? And I said, well, why? He says, well, that's who I want to mail to because no one else's letters are getting there. I was like, oh, yeah, that's great, but we should just tell everybody. He says, no, don't tell everybody, just tell me.

So, you know, we had a quick.

Conversation, Shyama and I. I'm like, hey, if we've got this information and you know, should we charge for it? Nope, this is just a. This is a feature, right? Throw it in. Right. It's already there. It's just another value add because I made a mistake early on. Did a mailer, had a bunch of returns. They all went to my virtual mailbox. I saw them there. 3, 400 returns with a little yellow sticker on the bottom. And then to add insult to injury, I didn't dump them from my virtual mailbox and I got charged storage on the damn thing. It was just insult. Remember I told you I did all the mistakes? Stupid things like that. It comes into your virtual mailbox. If you know what it is, dump it, shred it, recycle it, whatever, but don't pay storage on it or don't have it opened. That's even worse, right?

That's awesome. Well, thanks, Fred, for all this information. I just went to reworked AI. That's correct Website. Correct, yes, Reworked AI. Interesting. Though the first thing I see is that there's reworked AI Betty for roofers. Yeah, interesting. Give us a 30 second rundown on Betty for roofers.

Yeah. So three years ago we were saying, okay, how do we grow? How do we get bigger? How do we keep growing the company? Do we go vertical? Do we go reselling data? Do we build a CRM? Reselling data is a very low margin business. A lot of land flippers are very tight with their data source. Didn't want to try to get into that. There's a lot of great CRMs out there. Didn't want to compete in a space very difficult to build, very intensive. So we said, what other industries are using direct mail and are experiencing these problems? Right. Solar roofing. And we got into roofing over a year ago, but we had a problem. We were using images that were free, they were not timely, they were not high resolution. So we could predict what was going on in the mind of the owner. But what we're selling is we predict the line, the mind of the owner and we tell you the owner that's got a bad view. But we couldn't nail that. We built our own computer vision program, but we couldn't nail it. We ended up creating a collaboration with Eagle View, which is one of the premier aerial imagery companies in the United States. So we buy their data and their computer vision output. So no, now we can send, we, our customers can send a mailer or call or knock or whatever we want to do for a home owner and say, hey, Mark, did you know that your roof is failing and we can help you out with that. And how many people know exactly the condition of the roof? Right now we're land flippers. We're not going like, hey. But not many people are aware of it. Right? So there's 1% of the people that know they need a roof and there's 99% that say they don't know until the mailer shows up or the phone call or the doorknock says, ma', am, your roof's about to go and it's $22,000 to fix it today. And if it fails, it's $45,000. So yeah, we're really excited about that. Was just two weeks ago that we signed that. And so, you know, we're in insurance, we're in solar, all real estate. And we're just going to keep bringing our machine learning core competency to the essentially the direct mail outreach because there's 80 billion pieces that go out a year and the vast majority of it is quite dumb.

That's awesome. I just sent that to my friend who owns a giant roofing company.

So there you go. Talk to Me, I think Mark Podolski should get into roofing. Yeah, well, I am never too late. I'm part of his friend. Yeah.

It kind of reminds me of the days, like, remember we, we would, we just didn't know anything about anything. And then all of a sudden, Google came along and like, oh, look at the searches. Like, oh, people are searching this. And it became a thing to know your avatar better. And now, you know, Facebook, like, these algorithms know know us whether we know ourselves. And now Fred's taking this to another level where we can know, like, oh, this person's more likely to sell us their land. This person's more likely to replace their roof. This person's more likely to get insurance. I mean, it's, it's the best time ever to be alive.

It's.

You know, Mark, I, I, I echo your enthusiasm and your excitement, right? And when we first were getting into this, it's just, you know, like the vast majority of people in the United States, I wasn't paying attention to what direct mail came to my house. Now I'm a direct mail nerd, right? I want to, I want to look at it. I call companies because I'm like, look, you sent me a direct mailer, and early on, I got a direct mailer, and I was living in a condo in Miami. I got a direct mailer for pool maintenance. I'm in a condo. My man don't send it to me. And then I got one for a cemetery plot. Now, I mean, I'm 50. At the time, I was 52. Right? But I'm not thinking. It's. Can you just think about where you're going to send it to? Right? You sent. Somebody sent me an H Vac mailer, and I just had the air conditioning replaced eight months ago. Pull the permits. Don't send it to me. Send it to Mrs. Johnson next door, who's had the same H VAC unit for 14 years, and say, Mrs. Johnson, your H vac is costing you this much money. I can come in, replace it. I'll finance it. I'll save you $32 a month, and you're better off. Give me a call. Boom. Call to action. Do something with it. Don't just send a mailer out and says, I'm Fred and I'm a roofer. Yeah, so what, what, what are you doing with that? Kind of, you have to be just. Anyways, I get fired up, Fred, you.

Think that, I mean, look at. Nobody would be bigger than credit card companies. Are they doing this to us right now?

Yeah, but it's, they're doing it on a very, we've had conversations with a couple of them, right. But they're super proud. And you know, I get the same bloody mailer from Citibank every week. Every week, right? And I have never done a lick of business with Citibank and they hit me every week. And this is, this is, you know, a standard mailer with couple letter, you know, a couple papers in there, these offers in, in. So they may be doing what John was saying earlier. They're taking that list and they're just hitting it, hitting it, hitting it, hitting it, hitting it versus, you know, they're not doing it every door direct. So okay, great, you're better than Joe the Reefer. But are they really doing anything different? And their budgets are massive, Right? Yeah.

Now Fred, I've interesting question and I don't know if I'm going to get the story right, but is there any risk, like privacy risk? So for example, a woman was shopping at Target and Target, based on her shopping habits, knew she was pregnant before she did and sent her coupons to buy diapers or get ready to get, you know, whatever, you know, like whatever pregnant women need when they're pregnant. And it was like a thing, right? Oh yeah. Here's Dana Ryan, Synchronicity. So is, is this, you know, so.

There, there are some guardrails with information, right? Anything to do with health, right? So we had a three month engagement with a company that was adjacent to a medical company. And their lawyers, the business people wanted to work with us, but their lawyers said no, can't, they cannot share anything related to health. But if you're in Target and you're buying composition book or whatever you're buying, you have no reasonable expectation of privacy, right? You're transacting in public. If you're choosing to be there, the record is being created. You know, there's a record being created, right. Unless you're paying in cash, you know, but if you've got a Target card and all these things, of course they're saving it. What do you, how do you think Amazon say hey, suggestions for your shopping pleasure. How do you think Netflix predicts that I might like this funny movie, right? It's that old adage, right? If you're not paying for something online, you're the data, right? And so there are guardrails. We are sensitive about it. We, you know, how many times have, have we all gotten inquiries from people? You sent a let. I'm going to call the Secretary of State for my state, and you sent me a mailer. And I am, I am. I'm going to sue you. You can't, man, you know, take off the tinfoil hat and go get a hobby. Right. It's just, I paid the USPS to put it in your mailbox. It. It's totally legit and I'm sorry. Right. Yeah. So as long as you're not. It's fine.

What about texting?

Would that be.

Yeah, I was just going to ask. Can you. Can you. If we give you a list, can you give us a phone number and an email?

Now? We, we looked at skip tracing phone numbers two years ago. We said, hey, that'd be a great, you know, one stop shop. Get it, you know, get it, get it. Scored with Betty. Get a skip trace with Joe. Right? But it's deep water. It's murky, it's sloppy, it's ugly. And there's some good companies out there that do it, but the risk that they take on has only gotten bigger. And so I was like, I don't, I don't. Again, don't want to play in that area where it's. So if. Plus my bias again, I would never answer a text, an unsolicited text. I get them all the time because I own property. Right. I get text all the time coming in, hey, you want to sell this property? But yeah, if you. I tell people, if you're going to use skip tracing for phone numbers, take your data, run it through Betty, take your eight, turn it into four. The four, and skip tracing. You'll save money on your skip tracing. Don't skip trace it before you run it through Betty because then you get a pay to skip trace. Eight, run it through betting you're only going to action for.

Awesome. We did have a question earlier from. I think it was Don. So what exactly is. Is the average response rate you're getting on a mailing campaign after using Betty, Fred? Is it 7 to 12%?

Yeah. Yeah. Easy. I mean, it. Obviously there's. There's variance in it, Right. And I always encourage people to look at things over time. Right. But yeah, it's 7 to 12%. But again, don't get too hung up on the response rate. We were trying to predict response rate in the beginning, and our engineers were looking at it. We're only predicting it between 75 and 83% of the time. We're only getting it right. That much. But one of our engineers said, okay, we're over here. We're a B student for inquiries, but we're killing all the deals. We'd only miss like one deal in 75 deals that were done. We predicted them all. It was a real aha moment. Don't try to predict whether somebody's going to call or inquiry. Predict what they're going to do with the money. Are they going to transact? Because at the end of the day, that's what we all want. Eddie says if your list, if Your score is 100 or above, action it, mail it, call it, text, do whatever you want. Me for my list, I only send 240 and above because I'm not being a male, a land snob, but I would love to bring my inquiry to deal ratio to one to one. Never going to happen. But it's just all about conservation, being efficient with time. If I could just get the people that really. That Betty predicts will want to talk about doing a deal to inquire and do the deal. That's where I want to be. That's my happy spot. If I miss a deal, it's okay. You know, we all run across land investors like, oh, but I got to mail everything because it could be that $75,000 property in there. Yeah. Or there could be 3,000 nothings. More than likely it's 3,000 nothings, but go for your life. Yeah. Could you.

Is there a way to do any of this on the other side, selling property so we only target the people that actually want to buy it?

Yeah, that's a. You're not the only one that's asked that question. Right. I'm assuming there is.

Right.

I'm assuming it's. It's. It's. But where's the. I mean, where's the data coming from? Where's that raw data? How do you. How do you. I don't think it's an unsolvable problem. I think the, the trying to price out a piece of property is pretty tough, but trying to predict who would buy it. I mean, Facebook is cracking that code. Right. I mean, to put your stuff on Facebook and, you know, use some of their algorithms of who are the knuckleheads. Want to buy an acre an hour and a half out of Houston to get their man cave and get away from their, you know, spouse. Yeah, it's probably solvable. I. I just don't know how.

Well, I'm gonna try it and I'll let you know in November how it works, because it sounds great.

I'm gonna.

I'll see you there. I'm gonna give this a whirlwind.

I mean, it's, it's irresistible.

Why.

Why wouldn't you use it?

Right, Exactly. This seems crazy to me.

Yeah.

We're gonna save a lot more than 150 bucks.

Yeah.

Or 289 bucks, so. Yeah. Mark. You know, Mark, think of the millions you've wasted since the year 2000.

I mean, well, also think of the trees. Think of the diesel, the. At the fuel. Getting this stuff out there.

It's.

It's, you know, where. Wherever you land on your thoughts about ecology and the climate and everything, it's. It's. There's a lot. Right. And if we can each do a little piece to do a little bit less, it's all the better.

I learned a lot tonight. Like, I am no longer shopping at Target because I don't want them, like, get it? I didn't know they could do this. Like, I'm like, cash only.

Like, no way.

I'll tell you what, though. Like, I'm. I think, like, we, we shouldn't be making any decisions without Fred. Like, where am I going for dinner tonight?

Like, right.

You know, how do I know this is even like an A level restaurant? How do I know the people that haven't gotten sick? Like, every decision should actually go through rework?

Well, in the beginning, you, you joke, Mark, but I mean, it's, it's multiple data sources, right? We go to Yelp and we see the, the reviews. Oh, this is a 4.3. It must be okay, but if we had the time, if we had the ability, we'd want to look at seven or eight reviews. We want them amalgamated. We want to get in the synopsis. We want to get that. You know. You know, maybe everybody that's going there likes to get drunk and party, and I just want to go have a nice quiet steak with a glass of wine. Right. Okay, that's not said.

So.

Data, data, data. Right? And I mean, we're, we're seeing it. I mean, Gemini now. You. Gemini Search or anything is, you know, you. You were talking about Google searches. Mark, my children who are 10 and 12 will never use Google. They're just not. I mean, no, I wouldn't. That's antiquated. That's their phone book for them. That's the encyclopedia that we grew up with or National Geographic. It's, It's. I mean, it's still huge. It's still 90 of what happens out there. But why would you type in and look. No, you're just gonna ask a large language model. Hopefully you're smart enough to Realize that there might be some hallucination in there or some, you know, afterburn. But, yeah, it's so much better.

Well, this has been awesome. We're coming up on close to an hour. I have one more question for you, Fred, and this goes along with what Don is asking here. So have you thought about getting into market analysis, real estate market analysis, with this tool? As far as, you know, pricing analysis in certain areas, counties, subdivisions, that type of thing? Because that would make our lives even much more simple.

Yeah. We've been asked many times, can you tell us where things are transacting, especially in the land market. Right. Because the vast majority of our real estate customers are in the land. I mean, we have single family, multifamily, commercial. But can you tell us. And we've looked at this.

Right.

In the beginning, we didn't have enough data, but now we have enough data that we could identify where things are transacting fast, where there's a large spread, what's going on. But here's the rub, guys. That's not our data. It's yours. Daniels. It's Shannon. It's Don's. It's not ours. We're processing it. We're caretakers of your data. We're looking at it. We're giving it back to you. It's not ours. So to use the amalgamation of your output and what. Where Daniel's, you know, honeyhole is. Sorry, it's just. It's not worth the juice, not worth the squeeze, because every, you know, early on, people would ask question, are you looking at my list that's coming up? I'm not looking at. Your list is coming. They don't even let me into the. The database. Right. Human eyes don't see it. It goes in, it gets processed, and it comes back down. So we've got enough data. It's not ours. So go for your life. Put it in there.

Betty's not going to say 90 of the people are mailing here, here, and here, but 10 are only over here. So there's a.

And they're killing it over here. Yeah, they're getting a 14 response rate over here. Stop mailing outside of Dallas, Texas, because it's. It's saturated. Right. Go over here. We have that information, but I. I don't know where it is. I still might. I still do. My tried and true. The counties around the major metropolitan in. In Dallas because I got to sell the property.

Yeah. And just for all the people listening, can you clarify? You're not. Look, sounds to me like you're not buying under 5,000. Selling an owner financing. You're following a slightly different model. Can you clarify?

Yeah, so I, I try to get in for 5 to 7 and out for 12 to 14. Okay. Yeah.

Okay.

I mean, I've done some properties with owner financing and I've, you know, I've sold the financing. I'm, I'm. Because I've gotten really busy and reworked. Right. I'm trying to be efficient with my time and I still, like I said, I haven't done anything in the last three months. This, this license agreement with Eagle View is sucked up all my time. But, you know, I'd like to do 12 properties a year, just vacation money and just put it aside. It's. And, and it's fun. Like I said, it's. I feel good when I buy and then when I sell and I don't want to over complicate it. And there's, But, I mean, there's so many good land investors out there that are just wicked smart. And you guys are that, right?

Because they're wicked smart.

Well, I was born in Connecticut, so yeah. And I'm, I'm really looking forward to the summit. Right. I'm really. The timing is lovely. Austin, a great city to be in. Great time of the year. You know, I've heard many good things about it. We're really pleased to be a sponsor and it's fun to meet new people and you know, and see what's going on and, and learn some new things. So. Really excited about it, Mark. Thank you.

Two, two things I wanted to say here at the end is, number one, it's really nice to see that, you know, Mark, that land investing can fit in anybody's. It could be vacation money, it can replace your income. It really has the flexibility to what you want it to become. But also someone had mentioned to me, and I think this is, I should have said this at the very beginning. Like, if they just tuned into like any one of these random night gaps, it'll be nice at the beginning for us to give a little synopsis.

Maybe we should do the clock 30.

Seconds because people are like, what are they talking? It's the first time you've ever listened to anything. You'd be like, what the heck are they talking? Like, I think at the beginning it wouldn't be bad if we gave like a 30 second rundown. Like, this is what we do. You know, we used to do that little clock. Mark, how quickly can you tell them? Because if I was first tuning in here, I might be. It might be a great thing to listen to, but I might be missing, like, scratching my head saying, what do they do?

I thought that was what the banjo was for. I think, Mike, it's just the baloney baloney.

And they buy land. I don't care.

It's a little bit confusing to people. It's kind of a big.

And we could have had the inside.

Joke that, you know, we should.

Yeah.

Just to create. Just saying it could be a good idea. I'm all.

I'm all for it.

I'd love to have you back, Mark.

Thanks for having me back. Thanks for having me back.

Yeah. This was amazing, Fred. Let's see what we're talking about next week, shall we?

Thank you.

Next week. September 16th already. Middle of September.

Next week we're talking what type of pumpkin beer.

Oh, okay, good. I'm glad to hear that. Next. Next week we're talking land investing. Just. Just simple land investing versus traditional real estate. Oh, man. We're gonna. We're gonna go off next. Next week. That's gonna be fun.

Yeah. If you.

If you.

If you're. If you're right now. Traditional real estate and you're.

Yeah.

You're tired and stressed out.

Come to us.

Come.

Come to this next nightcap.

We'll.

We'll have you. We'll have you. Right.

Yeah.

Right.

Is rainbow twist you. Twist you on. Thanks, everybody, for coming again. Check out reworked AI. Come to the Dirt Rich Summit first weekend in November and keep doing land deals. Land is hot, right? We're buying and selling, man. Let's keep. Keep it up.

Yeah.

All right. Thanks, everybody. Have a good night. Cheers. The outro. Here we go.

An outro.

It.