Josh Pigford, Creator of Maybe.co

Hey, everyone. Welcome to the podcast. Today, I am talking to someone that I have known for a little bit in the bootstrapper world, Josh Pigford. So I first ran into Josh when he started a company called Baremetrics. 2Pole used to be a customer of Baremetrics, and they do SaaS analytics.

Come January, like, last month, I was feeling real nostalgic about all the stuff we had built because I was like, it was I knew we had shut it down earlier than we needed to. Needed in the sense of like, it had not had enough time to be proven if the product could work. We just ran out of money. And so like, I was like, man, I'm so bummed. I was feeling real bummed about it.

And I mean, that's a great

It took off.

That's a great little line.

I that yeah. I mean, that'll that'll get you.

That works.

And it did.

It did. That was the top of hacker news.

I was well, I was surprised. My my biggest takeaway from that, or I guess my biggest surprise from that, was how little people, tried to, like, tear it down of, like, that's stupid. Like, no way does that cost a $1,000,000. No almost no one commented on the price tag of of that. It was just sort of like, oh, okay. Cool. Yeah. Like, this exactly, which is true. I mean, we spent a $1,000,000 on that.

So But you were anticipating that response?

I was I was anticipating a massive amount of blowback from people just, like, saying that there's no way that something like that could cost $1,000,000. And

Yeah. Wow. That's funny. I I also tend to mispredict negative things like that.

Yeah. Well, people get but they'll but then they'll get negative about negative about totally random stuff that like, what are you picking that apart?

Yeah. Yeah. Sometimes sometimes I I actually, I'm probably often taken by surprise at the stuff that people get mad about.

Yeah. Yeah.

And so it's like I'm kind of wasting my time trying to predict it. Yeah. Or, like, changing my behavior because I'm worried that that will be the response.

Sure. Just gotta do it and let it happens.

Yeah. Okay. So what happened next?

So that brings us to almost today. I

Well, talk talk about that that day though. Talk about the next 24 hours.

Yeah. Yeah. So in the in the 24 hours after that, it's like, very quickly starts taking off. Like, you know, I think we had many thousands of, like, GitHub stars, which, you know, fake Internet points. But I mean, like, still it's people being like, oh, this is kinda cool.

Had a

lot of feet had a lot of well, and a lot of that's around like, like SOC 2 compliance, SEC regulatory stuff because we're giving advice. We have to store every conversation with an advisor for, like, 7 years. So that stuff still exists on a like AWS servers.

What's an example of the infrastructure requirement you have to, like, satisfy those things?

Yeah. So a lot of that stuff is around, the conversational side.

Like data retention?

Yes. Like and and and, like, data retention in that week have no ability to modify Postgres database and call it a day. Like, we have to prove that we can't modify the data. Interesting.

Do people offer this as a service, just like backups that are untouchable somehow?

Amazon does.

Yeah. Makes sense.

For Fintech companies. So but let's just you know, but we just have to build all this stuff in, like and, and you have to like, so we're working with third party compliance companies to, like, do all the paperwork to prove that we're doing this correctly. And,

yeah. The SEC doesn't demand that you, like, use microservices or, like, a certain kinda back They

don't know. Not that kind of stuff. No. But, like, even some of the data aggregators, so, like, we use Plaid and then FENICITY. So FENICITY is Mastercard's take on Plaid, essentially.

Okay. But so it blows up in terms of interest. Like, at least definitely passive interest. People read about it on Hacker News. They leave a comment. They start again.

I like the idea of a React based personal finance software that I could download and

Yep.

You know, hack around on. Yeah.

Okay. Is there is there interest from people, like, doing things? Does anybody open a poll request or file issues or

get to work? So so when I when I threw it out there, it was like, no. I'm not I'm not gonna turn into like a like a repo manager here of like trying to field pull requests. If you wanna do something with it, be my guest. But like, I'm a whatever.

Yeah. What was the license that you picked?

Agpl3.

What does that do? What does that mean?

So it's like if you can use it commercially, but if you do anything with it, you've gotta shove it back in or, like, you have to give us access to it essentially. So you can't just, like, fork it, then do whatever you want with it and sell it or anything like that. Like, there's some commercial, stuff that it keeps someone from Like you could launch your own version of maybe with it, but you have to give us modifications that you make to it.

Interesting.

Gonna deploy it.

Give them how? Like, what if you don't, like I give you like, let you see them? Like, if you meant you don't want them to

Send them back to the, like, send like, I don't know. Whatever the method could be, whatever, I guess. But, like, ultimately, we have to have the ability to, like, get your changes.

Sure.

Back. You have to

Do I have to open source mine too or no?

Yes.

Okay.

So I don't know. There's, like, a million sort of one day, I would love for us to, like, list out just have a bulleted list of, like, what ifs. What if I do this?

This seems important for you, for your company. Having this license to do what you want to be is like seems actually really kind of vital.

Right. So it's this very this is different than like, MIT or something where, it's essentially just do whatever the heck you want Right. With it. You know? As long as you leave a copyright copywriter, like, linked to the license, whatever. Like, this has to come back to us. Like, you can't just this isn't just take it and do whatever you want.

Okay.

Be done.

Alright. So

and I based like, a lot of that was sort of ad hoc. Like, just sort of quickly threw that together based off a bunch of other open source projects that I commercial open source projects.

Okay. Okay.

That I'd seen at the time. But so yes. So within, like, 24, 48 hours of doing this, I realized, like, hold okay. There's a ton of interest here. Yeah. People are pumped about this. What if we could make it work? Still was not like, let me turn this back into a business. It was, what if we could just get it working again? That'd be kinda neat.

You're like product managing more or less.

I started yes. I'm just like, okay. Again, if the community would like to do this, I have no ability because this is I haven't touched this code. I don't know how to I can't get this thing working on my computer. So, like, if someone else can, you know, I would post, like, the errors I was getting on the command line trying to get this thing to work. And Okay.

Yep.

So that's sort of what kicked it off. And then the number of people who were, like, interested in it, I decided to throw, like, a a discord server up just to have a spot for people to chat this stuff out and get it working or whatever. Just because it's like, almost like a, from an experiment perspective, it was kind of fascinating to watch.

Yeah. It It sounds exciting. Like, yeah, someone's coming along and fixing up your code base.

They're like, that's pretty neat. We'll make it work.

I'm pretty into that.

So so then that that but that snowball kept growing. Like Uh-huh. A number of people interested just. You know, I've posted this a couple of times on on Twitter of, like, our GitHub star chart thing. Like, it literally just goes straight up to 10,000 in, like, a few days.

That's cool. That's cool. So But that's like yeah. That's like likes, though. You know?

Oh, sure. It's like Instagram. Oh, yes. It's whatever. It's complete vanity metric.

The the people doing stuff, though, that I mean, I mean, pull requests. That's some real

People doing pull requests and yes. And we had, like, over a 1000 people in the Discord server very quickly who were, like, asking questions about how to get it working.

Wow. That's way more than I thought. I was in the wrong order management.

Yeah. Yes. And, well, again, this is like all of these things that sort of, again, on their own are are vanity, but they all sort of point towards

But altogether they they start to coalesce

to Okay. So maybe there is something. Like, let's let's, like, let's twiddle a couple of knobs and see what happens.

Yeah. Yeah.

You know? Okay. It's it's like AB testing with humans in the sense of, like, we've now got something to be maybe statistically significant if we try some stuff.

Yeah. Yeah. Okay. So So what do you start trying?

Well, just okay. Maybe we can get this thing running. Oh, oh, wow. It is running now. Like, everybody's able to sort of relatively consistently get it at least to work locally. Yeah. And, so I at that point, it was like, what if I I started thinking through the, like, what ifs? What if for this thing to work, we are ready. We're gonna have to rip out a ton of, like, all the advisory stuff because

So you've decided advisory's a dead end. You're out.

Because of the regulatory overhead.

Fair enough. There's a

massive amount of compliance

stuff that comes up. So delete all the all that.

Delete literally, I mean,

it was a whole app.

And Yeah. Boop. It was like it was like 20,000 lines of code or something stupid. Did you

could you delete one directory and it all disappeared and there were no other changes?

For the most part, yes. There were some, like, database bits. Like Yeah. Fair.

We had to

do some migrations to remove

some Fine.

But, like Okay.

Still. From

a certainly from a front end perspective with the UI, the entire interface for that. And a little even back end ish stuff because of the express. Either way. Yeah. For the most part, it was just deleted directory and you're done.

I like that. Yeah. So That sounds like good architecture, man. If you get a good red diff when you remove a feature, like, some I think I think you built it right. That feels good.

So what as far as JavaScript based?

No. Just in terms of, like, constructing an app.

Okay. Yes. From an organization perspective.

There's something about that that feels really clean. Like, you didn't it wasn't all you didn't have to go change everything else around it to pull it out. Like, it was they were decoupled in their own separate operating things.

No. That's correct. And that's, you know, to credit for Zach, who ended up he's now back at maybe, like, but he was the original

Dude, this story is nuts.

Full stack engineer. Yeah. So, but either way, so yeah, so we just start deleting stuff left and right.

Okay.

We have to rip out plaid. You got

to rip out plaid? Were you excited?

We had to.

It had to be therapeutic.

It was, but it's also like so much function this is so much functionality was tied up in that. Yeah. And it's impossible to use on an individual level. Like, because you have to have an enterprise contract with them. And, so either way, that was sort of the first, like, people started saying, okay.

And what have you landed on?

So we initially replaced it with a tool called or data aggregator called Teller. Teller is similar to Plaid, though they're only checking savings savings account support, but they have a have an individual use plan. Like, you could use it for free for just your own bank account stuff.

Okay.

So that's what we we integrated that. This is still all under the React Next JS thing. Like, we

have not

hit the point where we decided to

start off here. By, like, saying we're gonna get right to the rewrite. We're we're almost there. It's been half an hour. But this story is good. I think this is legit. I have no regrets.

Yeah. So that either way, so we are heading down this road of, like, let's rebuild it. It's still all in Next. Js. I hire so I mentioned Zach, who's the original Let's

rebuild it. Why like, what yeah. So you decide there's a a bigger rewrite or something needs to happen.

So this plays a role here. So, like, I I'm still in Next. Js React world Okay. Trying to build this thing. I decide, like, okay. If if I'm going to start this thing up again, I need to become functional as a developer. So then I start spending a few days trying to, like, really dig in and teach myself React and, like, modern modern Next JS stuff. Good for you. Yeah. And you would if you, like, watch my tweets around that time, like, it's just me getting angrier every day.

and you won't connect to my bank account and tell me what's in it, but I type it in manually or you'll do use teller?

Initially, yes. So it initially, it's like add manual stuff. We'll eventually re add some of those aggregators.

But the way is, like, maybe you can just get people to type it in and save yourself, like, the biggest technical challenge ever.

Right. So but that's this is where, like, the hosted for pay version of Maybe comes in as, like, the data aggregators where we will automatically do it for you.

Okay. Okay.

So, and then there's some stuff you can do around, like dragging a CSV export of your bank account or, or even upload a PDF and we can pull out the stuff from it. These are some solvable problems, overall. But at the so at this at this point in time, I don't know, week and a half in or something away from since open sourcing this thing, and it's just still straight up on, like, every graph that I am tracking of anything. And I'm trying to, like, wrap my head around what it means to build you know, we've got a sizable React Next JS app thing here. It's just there's a lot.

Talk more about that.

So so a VC, named Joseph Jacks, but run the name of the VC is like OSS Capital. And it's like the specifically invest in open source commercial open source products. So, like, probably big example of that's like cal.com. Big open source project backed by them.

Got it. So the model they like is there's an open source version I could use for free, but we also host it and

They'll have a commercial version. Yes. And, believe that that ultimately allows for, like, highly efficient use of capital. So, like, you don't need massive teams, especially if you can build a community around your open source project. Interesting.

Totally. Yep. Yeah. It seems like the incentives point the same way, which is pretty nice.

Yes.

Like and, like, if I'm a developer, getting landing open source contributions raises my profile, like, raises my value. Right?

It makes it you know, you when when when I'm hiring an engineer, I send me your your GitHub profile. Right? Like, show me what you've built.

Yeah. Totally.

And this is an opportunity to show that you've built things. Right.

So you're getting, yeah, free code contributions. They're getting reputation boost leading to salary.

Exactly. Which I do think, yes, I do think it can have a direct effect on their salary. Like, they're

Oh, yeah. Totally. No no doubt to me. And then also, if they're users of the products, they can

They should've been.

They can add a feature they want.

That's right.

I could fix a bug and be like, this bug is obnoxious. I fixed it for you.

I'll I've I've done that. Yes.

Wow. I'm in. This sounds great.

Yes. So, and I and I it starts

Hold hold hold on, though. Hold on. Hold on. Okay. So it sounds great. But what Yes. Well, how how might this also fail? Like, where might this kinda go bad?

The commercial open source concept?

Yeah.

Yeah. So if you're not doing a decent job, and I'm not saying I'm doing a decent job yet, but I like if you're not doing a decent job of managing pull requests and issues

Totally.

It very you you can just burn down the whole product in, like, bloated stuff. Like You have

to say no a lot.

You have to say no a ton. And this politely. This is sort of the crux of what started to show cracks in the React Next JS.

Okay. Great. Perfect.

For us.

Perfect segue.

So, after we started, like, getting some big stuff done and there was, you know, dozens of pull requests from lots of pea from from people on lots of different issues. It it despite there being a lot of work to be done and, like, those being listed as issues to tackle, people very quickly started focusing on the, like, semantic stuff. Like, you know, how do we format this function? Well, that's not the way to do it these days. Like let's go into let's refactor all this stuff that like changes zero functionality.

know, style wars erupt, basically.

Yes. And and at the same time, again, when we started with Next. Js, it was like they have their pages router and their, like, app router thing, which app router essentially adds a bunch of back end functionality. We were using the original pages thing, but it was because we had so many disjointed different pieces, it would have it made most sense for us to try to convert a bunch of this stuff over to their like new app router thing. And but even if we didn't do that, it was starting to become very clear.

Yeah. Yeah.

We're having to rip out a ton of stuff and rewrite it anyways. And,

and are you change is this most are you changing user facing functionality or is it mostly rewriting how it's done? This is basically a huge refactor.

That's essentially it. And, and so, like, that led to a lengthy conversation with Vercel, who's, like, behind Next. Js. And they were great, like, super smart people there. And, trying to like, just talking through, like, they were familiar with our code base, spent time looking at our code base.

Cool.

Had was, like, actually commenting on issues in the code base that people were having confusion over Next. Js stuff and, like, they were being active in the community. Like, that's just fantastic. Yeah. But talking to them, like, it became clear that if we're trying to build something that's very easy to contribute to, that requires relatively little sort of like wrapping your head around where things are, It's hard to stick with something that is changing very fast because we can't we can't just, like, stake it in the ground of, like, here's how we're doing this because the reality is some of that the newest, like, Next.

of this thing.

Like Are

there not good conventions in JavaScript worlds that, like, let people drop in and contribute because they just know where stuff is expected to be?

That would be great, but no. Not when you're like, to me in the so if especially because of the stuff that Next JS has has introduced in the past, that's probably 12 months, that sort of blew up the entire convention. And so there were lots of arguments. I'm like, well, now this is the way that you do it in Next. Js.

Yeah.

Be functional. We could stick with potentially a framework that's been around for 20 years and does have a lot of strongly set convention. Whether you like the convention or not, like, it still has a lot of this is the, in this case, rails way of doing things.

Yeah.

And,

So you're like, we're making a rails app.

That's right. So I was like, okay. Do you

wanna hear something funny?

Sure.

So I ran a product at Thoughtbot called FormKeep, and it was a form back end as a service. It was like an endpoint you could post to with some form data and we would like forward it on and like send you an email or like trigger Zapier or things like that. Pretty simple app. We built it at a time when, like, everyone was, like, we were using it as, like, a learn JavaScript experiment playground.

Sure.

Yep. It was maybe Backbone app or something. It was a little bit later than Backbone, but I mean,

I remember that.

You know, it was like not that would not way later. And it was like we wrote this really complicated JavaScript front end to show you a bunch of form entries. And it was kind of like a pretty basic crud interface, and it was pretty darn complicated. And when I took that product over, the first thing I did was rip out the JavaScript and replace it with just basic rails views.

Sure.

And it made it so much easier to work on.

That's right.

Some of that's my personal bias because I'm a rails expert, but not a JavaScript expert. So, like Yep. Some of that was, like, I made the same decision that you were making, which is just like, I know I could do this one. There's definitely trade offs here as it's not a pure win. But, like, for making a damn app, it's Rail's a pretty good choice. I still feel great about that choice.

Well, and I and especially, you know, this I think in, like, the rails sort of 6 era, it was rough on the on the front end stuff. And, but with with all the Hotwire and all of the whatever, you know, I I use Hotwire in the, like, umbrella sense of turbo stimulus strata, I think they have now. Like, it's become much more of like a first class citizen on building these like really nice UI bits in a railsy kind of way. And, I think having spent the last 6 months of, like, floundering around figuring out what on earth to do with maybe the entity, and experimenting with all sorts of stuff. I was doing a I mean, there was a time where I was, like, doing rails new, like, weekly.

who knows?

So I want to I well, so I want to be clear there that I'm not saying React developers aren't good developers. Okay. I think because there is so much it's such a popular thing, it it means that there is a lot more entry level availability because there's a ton of people. That's the thing they learn. And they're like, let me go contribute to something. It is popular with beginners. So you end up skewing

Yeah. Sure.

Towards a lot more entry level people.

Right. Especially if anyone can submit a poll request or something.

That's right.

And you're open.

So I think rails, if only because of that single dynamic, tends to skew a little bit more towards people who are, slightly more skilled.

That's probably that's probably true. I feel like that wasn't exactly true in in my day. Like, rails was the thing, but then I think this has just shifted. This is just the new reality.

That's right. So, and so far, that's actually I I would say has proven to be the case.

Cool. You've been happy with the caliber of Rails people you're seeing?

That's correct. Now there's still certainly some people it's like disagreements over how to do this or people sort of overcomplicate things or whatever, but that's programming.

Do conventions for open source things I mean, like or for I mean, for anything. It's just, like, if you just have strong app conventions, I think that's such a huge win. It makes it so much easier to drop in and do something productively because you just know where stuff will be and where your changes gotta go.

Exactly. So that's sort of how we ended up on the rails thing. I think I think we covered the how we ended up.

I'm convinced.

Did we touch the bases on that?

Mega Rails f. You sold me.

Yeah. That's right. That's right. So we'd literally started from scratch.

Oh, you did you okay.

I mean, I I rails nude it.

Yeah. So you're you're not refactoring a thing. You are rewriting from scratch.

We're rewriting from scratch. And the caveat there, though, is, like, that I think makes us different than oh, man. You're just like, you're having to remake all these decisions. We're not having to remake all these decisions because our well, one thing we're very confident in was, like, the data modeling that we had done, in the previous app. And there's a lot of decisions around product functionality that, like, have already been made, like, how we're doing this stuff. And so

do you know that you've made something people want yet though, product decision wise?

I believe that we have. But I mean, like, there's almost it's we can't know that because we never built a successful business. Like Right. Right?

Yeah.

So there's still some there's still a lot of, assumptions being made here. And Yeah.

So you're gonna def you're gonna just, like, accept all the product decisions you've already made. You're not, like, also changing functionality yet.

Essentially. Yes. Same same product decisions. However, we will because we're doing this rewrite, and so that means we're still building a feature, then the next feature, then the next feature. We will put this out in the public far sooner. Right? Like, you can you'll be able to start using this stuff way before it gets to the feature parity to the preview.

Oh, okay. Great. Cool. Yeah. Because it seems maybe maybe weird to reimplement product decisions if you're not confident about them, but also trying to change them as you rewrite it also seems insane to me. So

There's there's a lot of variables there that's like, okay. How can we ultimately prove that people want something? And Yeah. We're essentially doing that by making similar product decisions but testing the waters with them before there's a lot of decisions to get feedback on.

Okay.

So, you know, that's starting off by just, like, being able to add a bank account and the transactions with them, and then let's see what people say about that. So, like, very basic stuff.

Okay. And how you get in the transactions? So you're hosting this and charging people for it. Right?

Eventually. Yes. That's not there yet. Yep.

Gotcha. Okay.

I mean, we should be there by the end of the month, but, yeah. So the initial manual input will be balanced based, and then you can input individual transaction at a time or upload CSVs of transaction data.

Okay.

So that'll be the initial pass at that.

Cool. Okay. And your hope to and you'll hope to sign up some customers who want to pay for whatever

that is needed. Real I mean, even, you know, not not at whatever full price we decide is what you should be paying for. But just to, like, say, if somebody's willing to pay $5 a month, then, like, that gives some level of, validation. Like, if they can if they'll give a credit card, then okay. We've proven a little it'll something.

Mhmm.

So, but I, so to go back to the rebuilding the same features, I say that, The way that we view this is still is is a little bit different this time around. Like, instead of it just being this sort of before, maybe it was much more on the, like, wealth management side of things and, you know, let's grow our investments and, like, figure out where to put money and so that it'll grow. And, like, this is part of having a financial adviser who can help you make decisions about what to put money where. We are changing that to be so, like, our first big feature will not be the forecasting stuff. It'll be budgeting.

I think you have to be fast, though. Right?

No. And that's the thing. Like, again, we're You know what things are? That by the end of the month, we'll have something there to start using. And then, like, the first thing we're building on top of that is budgeting.

Okay. I guess what I'm saying, I think this opportunity of the the ex maintenance customers probably is short. Like

I agree with that. Yeah. Yeah. No. For sure. But it's just like the market is that that was the thing that everybody was coming from before. When we would talk to people 2 years ago about maybe it's like, what are you using now? Mint. Mint. Mint.

What do you what do you wanna use maybe for?

So for us, it's a let's track all of investments, ultimately. Like, that's the that's the component or module that I'm personally most interested in. So

What are you doing for that now?

We nothing great. I mean, the closest thing I've sort of found is Kubera. So it it does a decent job of aggregating lots of data sources, but I don't think it's just sort of like reports on a bunch of numbers. And I don't think it it doesn't really do anything to help, affect those numbers, I guess. And I we sort of view, providing stuff for you to be more hands on with your money or with suggestions or, like, reminders that, hey.

Well, yeah. And you just got another $1,000,000 and you have 2 paid people? Or, like

1,500,000 and, and she has 2 full time from a salary perspective. Yep.

And then a bunch of open source contributions, which

That's right.

Are those are those overall, like, been very helpful? Like, are they making, like, good decisions, big strides, helping you out a lot?

So what I've found is you start seeing, like, a few people who've are heavily invested as an open source contributor. And so they will you don't necessarily see them tackling every issue, though they do tackle some of them, but they're, like, commenting on code of like, hey. We did this and this other commit. Like, here's how we can sort of stick with that convention that this app has. So I would say that they're yes.

Any results from that yet?

Yeah. I mean, we've paid out 3 almost $3,000 in bounties over the past couple of weeks.

Cool.

I'm still trying to figure out, like, what's a reasonable amount that's not, like, offensive. Like, you wanna only pay me a $100 for this? Like, well, it's not so much the I just want to, like, convey sort of a thank you for your time and that this is important to us. But, like, you know, it's I don't know. It's that's a weird line to, to walk on, but, you know, so we're we're experimenting, I guess, with that at least.

What's next in your mind? What's coming down the pipe?

Get the thing working. Like, you know, having a functional personal finance app.

Meaning hosted somewhere.

Hosted. Yeah. Like, you could go sign up at beauty.co.

That's right. Yeah.

Yeah. So we'll get there quickly. It'll just be really basic and but it'll also change really fast. Like, it took us a long time to build that sort of React version for a lot of different reasons, not just because, oh, it's like React's slow to develop with. Like, I'm not saying that.

Cool.

Yeah. It's just get the start making some money, like, not wait 18 months again to try.

Are you gonna charge people right away? Uh-huh. Customers instead of testers?

Yes. I mean, that's the that's the plan. It's like, okay. We'll, you know, have some sort of, like, early early customer discount or something because it's, you know, very basic functionality. But and figuring out what that looks like around, like obviously, we'll wanna test out freemium and and most financial tools start with some sort of free component. So, like, we're competing against that.

I would want to see you charge for it, for everybody.

I mean, that would be great.

Just to make sure that like people value this and you're making something they actually want, like get get the money. You

know? Yeah. And then we can also, you know, play around with the free side Totally. Down the road.

Yeah. Yeah. Cool. Awesome, man. Good luck. I hope it works. Yeah. Yeah. That's the the whole what a story. That was great.

It's bonkers. It's totally bonkers. It doesn't make any sense.

It's awesome. I I hope it I hope it works out because I want the ending to be great, you know. Like, just I just finished the story. Alright? Like, you need to make this a huge success.

I know.

The the framing of, like, what what if it was the thing you used for your whole life is interesting. Like, it's it's kind of audacious. It's like, yeah. Like, our our LTV is gonna be, like, $100,000 because they're gonna be a customer for 4 years.

Like, I love

yeah. I love the idea of it. Yeah. And and, like, I don't know if you'll be able to pull that off, but almost like the orientation of, like, trying to do it.

Trying to.

You know? For sure. Like, aim for the moon sort of thing. Aim for the stars.

It's like, imagine, the Internet 20 years ago. Like, the idea that somebody would have built something 20 years ago and that they're still using that thing today is super rare. And I think I think of, like, I have an eBay. My my eBay account might be 15 years old or something like that, but like, sure.

Yeah. My Reddit account is real old. It's in the teens.

I see. Mine is probably similar. But, but again, like, you're directionally if we're if we're making decisions that support that type of outcome.

Exactly. Right. That seems good. Like, you won't you'll, like, be like, I'm not gonna make the short term trade off ever because what I care about is the

long term. That's right. And and I You're like,

I I'm yeah. Exactly. I'm already rich, and we raise a bunch more money.

I'm so I'm so hesitant to ever be like, well, I've got a lot of money. Like Sure. But it matters. But it it does. It changes it changes the types of problems that, like, I'm I'm willing to tackle because I don't it doesn't matter to me purse if I personally make some money.

that's how I

that's how my brain works. But, like, at the same time, I don't personally be it's not like, well, we need I need a paycheck. So we need to figure out how to do something today in the short for a short term win. You know? So

Cool. I guess just yeah. The alarm bell rings a little bit, though, which is like like, that seems great, but also don't go slow.

Sure. I

You know?

My brain doesn't my my brain also doesn't let me move out. Like, I have to not be bored. So I have to be working towards and, like, seeing progress. It's just I don't like the feeling of, like, well, we need to figure this out, tomorrow. Sure.

Well, I'll

give it a few days. It's okay.

Nice. Alright. You talk about detangle just a little bit because it seems like that kinda is working. And I play with it a little bit and found it, like, nice.

Yeah. Yeah. So, you know, you're familiar with this where, especially in business, you just get contracts for all sorts of stuff even if it's just even if it's like an NDA. But any kind of business business acquisitions or just business agreements or employment agreements, whatever it is, you get a lot of this stuff. That was the impetus for, like, well, it'd be cool if you could, translate this into, I I say, like, human speak.

Yeah.

Here's, you know, all the money stuff that's in the document and here's

That part that part's pretty cool.

Yeah. I I think it is pretty cool overall.

Yeah. I think it's a cool

And it's right now, it's a it's one off charges per document, so there's no subscription component. It's one of those things that

How much revenue has it done?

I don't at this point, I mean, it's not a ton. It's like maybe 4 or $5,000 or something. Like and these one off couple months.

2 months? A

little bit less. Yeah. Okay. It's the kind of thing that's like it feels like there's something there. We're probably not the people to tackle it. I just don't know what to do with it now. Because it's also it's part of maybe finance inc. Like, it's under that umbrella. Oh, geez. It complicates things.

Is anybody, like, asking you like, are people asking for new features or, like, raving about it?

Yeah. People want subscriptions.

Like Do you wanna do a bunch of docs every month?

Right. And, there seems like Sounds

like nothing.

Wanting to be able to ask questions of the document, like, the sort of typical sort of chat, like

Yeah. Yeah. Yeah.

Especially if you've got a long document, it becomes kind of valuable to be able to like, wait, what was that thing where it said this thing about the thing?

Totally. Yeah. I can see that.

We played around with, like, a couple of big organizations. Open Door was one that we had a lot of conversations with where they do the, like, buy and sell houses. Like, they'll buy your house from you, and then you can also buy houses from Open Door. So they they're dealing with a bajillion different documents around homeowners associations and, also buying houses. And so, like, they want to be able to dump in tons of documents and pull out information from them.

Very much there is to me. Right.

And so

I hear that you're not into it. I'm not trying to sell you on it. It just sounds interesting.

No. But it's like I do think there is something there, and I I hesitate to just be like, well, that was the opportunity cost. Like, shut it down. You know, whatever, maybe is a bigger opportunity. That's like, I don't know. Like, in fact, that's what

hate is. Though? I mean, just just I

don't know.

I I mean, maybe I am trying to sell

you a little bit.

Maybe I am trying to sell you a little bit, which is that one of these things already is making money. Sure. And like yeah. It's it's got to

be I think but that's where this goes into the, like, what what do what's like I want to tackle, like, moonshot stuff.

Yeah. Totally.

Right? Yeah. So, but I but it is also the reason why I haven't just, like, well, bye. We'll shut down shut down the servers and call it a day. You know? It's because, like, I I do real I still feel strong that there is something there. I just don't know what our angle is.

Does chat gpt give you, good enough results?

So so there's the that of, like, will chat GPT get to a point where, like, we've we've heavily massaged the the, the prompts on the back end

Sure.

To be able to spit out this stuff in a usable way. But does chat gpt get good enough or, like, do these things like templates where you can just be like, here are my pre written blocks. Let me throw this out a document and get the output of it. Well, then that's essentially what detangle's doing at the moment. Like, we're sort of taking chunks of this and and just asking different questions of it and then summarizing it in these little blocks.

I guess what I guess what I'm asking is, is the stuff that's coming out about the legal documents, like, legally smart enough? Is it, like, enough like what a lawyer would tell you that knew what they were doing?

Well, I mean, the the the line that we sort of walk there is saying, like, this isn't a replacement for a lawyer. If anything, it's to, like, augment your conversations with your lawyer so that you'd know what to ask them about.

Mhmm. That's I thought and and are you and it sounds like you are delivering that. People seem to

I think so.

Feel that way.

Very consistently, it's been like, man, this was amazing. I don't have any more documents to detangle, but I love it.

Yeah. That was me.

Sure. I mean, that's that's I

would buy something. Yeah. And then, yeah, I had one.

That's the crux is like

But but it were I mean, when I was doing sales for Tuple, I had a lot of documents in my life.

Sure. I so I think there is a commercial component where you could sell to companies and just say like and especially companies that were at the point where, like, they didn't have full time counsel or something. Like, Yeah. They don't have it in Yeah. Which is like every Probably

a lot of them.

Small business. Yeah. So Yeah. You know, there's there are so many angles that have been left unexplored with that. And

I But you only get one life, so you gotta work on the things that gets you.

I know. I know.

But but traction can be really exciting. Right?

I don't there's either way, I'm I'm all in on maybe. So but it's just but I'm but I'm not so all in that I am willing to, like, throw out detangle. Not that I'll keep working on it, but, like, I want someone else to do it. I just don't know how to format that.

Yeah. Interesting. Cool, man. This is great. I love chatting.

Yeah. For sure. I'm glad to chab you on, man. You're only what? You're a couple of episodes into this thing, right?

Yeah. I think this is 4. Alright. Or 3. I forget. Anyway, yeah. You're early. You're single. Did you're low you're below 5 for sure.

Look at that, man. I'll I'll I'll I'll make note of that somewhere.

I'm gonna try to make this I'm gonna try to make this cool for you. So that one day you can be like, I was on episode 4.

I was I was in single digit episodes.

Alright. Take it easy.

Alright. Thanks, man. Have a good one. See