Claude Code built me a $273/Day online directory

I sit down with Frey Chu to go deep on how to use Claude Code to build AI-coded directories, specifically how to tackle the hardest part: getting valuable data. Frey walks us through three real-world directory examples (a funeral home directory, a senior living directory, and GasBuddy), we play a game guessing their traffic and monetization, and then he does a full live walkthrough of the seven-step process he used to build a luxury restroom trailer directory in four days for under $250. I also ask him about the future of directories in a world where LLMs are changing how people search. Timestamps 00:00 – Intro 02:15 – What you’ll learn 03:00 – Directory Game:Parting(Funeral Home Directory) 05:42 – Directory Game: A Place for Mom (Senior Living Directory) 08:00 – Directory Game: GasBuddy (Crowdsourced Gas Price Directory) 12:32 – The Data Moat Thesis 14:02 – Luxury Restroom Trailers: The Niche Directory Demo 15:52 – Before & After: WordPress Directory vs. Claude Code Directory 19:04 – Cost Breakdown: Built in 4 Days for Under $250 21:23 – Step 1: Scraping Raw Data with Outscraper 22:25 – Step 2: Cleaning Data with Claude Code 23:27 – Step 3: Using Crawl4AI for Automated Website Verification 28:01 – Step 4: Enriching Trailer Inventory Data 31:33 – Step 5: Scraping & Verifying Images with Claude Vision 36:33 – Step 6: Amenities, Features & Filter Data 38:31 – Step 7: Service Areas 39:15 – Niche Directory Ideas: Dementia Care, ADA Bathrooms, Tap Water Quality 43:38 – For Naysayers: Is Building a Directory Worth It in 2026? 47:51 – LLMs, AI Search & the Future of Directories Links Mentioned: Outscraper: https://startup-ideas-pod.link/outscraper Crawl4AI: https://startup-ideas-pod.link/crawl4AI Key Points Data is the moat for any successful directory — and with Claude Code plus Crawl4AI, the hardest part (data cleaning and enrichment) is now dramatically faster and cheaper. Every successful directory helps people save time, save money, or make money — and price transparency is a massive, underserved opportunity across boring niches. Frey built a fully enriched luxury restroom trailer directory in four days for under $250, a process that would have taken 2,000+ hours of manual work. Monetization depends on the niche: lead generation, vertical SaaS, agency services, ads, debit cards, affiliate, and marketplace models all work. Directories remain strong in an AI search world because users browsing a directory are in the decision-making phase, especially in high-stakes niches like health, legal, finance, and senior living. Building a directory is one of the best playgrounds to learn Claude Code, SEO, and lead generation — even if the first one is just a learning exercise. The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/ FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/ FIND FREY ON SOCIAL X/Twitter: https://x.com/freychu YouTube: https://www.youtube.com/@FreyChu/featured ShipYourDirectory: https://www.shipyourdirectory.com/

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Greg Isenberg?

  So you want to start a startup that does anywhere between $2,000 and $10,000 a month passive revenue, something that you only have to work on 10, 15, 20 minutes a week. But you don't have a lot of money to start. Maybe you have $200, $500, $1,000. Well, what kind of business would you start? I would recommend an online directory. A directory is basically a website with a bunch of different links and data. And in this episode, I talk about how you can use Claude code to actually build an online directory that is going to get traffic on autopilot. And I brought on Frey, and Frey is Mr. Directory.

• 0:39Greg Isenberg?

  And he shows us step by step how to use Claude code to create a very big directory. I really enjoyed having him just take us through the entire process. What are the prompts he uses? How to think about data enrichment? How do you think about cleaning the data? So that you can create an online directory that generates traffic every single day. Now, I know what some people are thinking. They're thinking, "That's really boring. An online directory? Come on, I want to build a mobile app. I want to build a SaaS product." But what they miss is that building an online directory is going to get you thousands of visitors on autopilot. And once you have that, then go and vibe code a SaaS product, then go and vibe code an

• 1:24Greg Isenberg?

  app. In the beginning of this episode, we share multiple examples of people making millions of dollars a year using online directories. And some of them have apps that go with it, too. So this is an incredibly important, yet boring, episode of the podcast, but those who stick to the end, I think, are going to have an unfair advantage. I totally recommend you take this link and share it with a friend, because this is alpha that is not shared anywhere else on the planet. You're in for a treat. Enjoy this episode with Frey.

• 2:04Greg Isenberg?

  It is a treat because we have Frey on the pod. By the end of this episode, what are people going to get out of it?

• 2:12Speaker 01

  Yeah, thanks for having me on, by the way, Greg. Yeah, by the end, I really want people to have a solid understanding of how to use Claude Code to build AI-coded directories with an emphasis on what I think is the hardest part, which is getting the valuable data. And we're going to go deep on that. But first, I think it's only fair, we run it back. The last time I was on the pod, we played a game where I showed you a few cool directories and you guessed the traffic. So if you're down, I'll share three cool directories. And if you want, you can try to guess the way that these monetize as well, just because I think that'd be pretty fun. So I'll go ahead and share my screen and we can dive into it.

• 2:55Greg Isenberg?

  Let's do it. I can't say no to a game, Frey, you know that.

• 3:01Speaker 01

  All right, so the first one is parting.com. And we can see it's a funeral home directory. This is kind of what it looks like. Click around. And how much traffic do you think that this website makes?

• 3:18Greg Isenberg?

  Well, it does say it's the number one. If you go to the, if you go back to the homepage, I think it says the number one funeral home comparison website for price and quality.

• 3:32Greg Isenberg?

  I'm going to guess it gets at least 100,000 visits a month.

• 3:38Speaker 01

  That is a solid guess. Let's take a look.

• 3:43Speaker 01

  All right. All right, 61,000 organic monthly visitors. So pretty close. And what's really cool about this directory, I'll kind of share a little bit about this one is, by the way, any guesses on how this might monetize? I'm just curious before I go into this.

• 4:03Greg Isenberg?

  Well, two things. I did see some ads pop up, but I don't think that's where they're getting the bulk of the revenue. I would say it's a lead gen directory, so they sell leads to funeral homes.

• 4:17Speaker 01

  Yeah, that is partially it. Definitely the ads don't make up the bulk of the revenue, so. This is reported to do over a million dollars a year in revenue. There are some quotes that go up to five million dollars, but they even raised money. They raised 1.5 million dollars in 2021.

• 4:37Speaker 01

  But the bulk of their money actually comes from a software. And if we look at Parting Pro, this is kind of their vertical SaaS that they offer for cremation arrangement software for funeral homes. What's really like interesting about this is like this software just helps them do things like dropshipping urns, like that's, that's a thing. And then it's kind of got an agency backend where they just help you create the website and then do online marketing, basically lead generation. And I think we talked about this last time, but I still believe that building a directory

• 5:18Speaker 01

  and creating a micro SaaS or vertical SaaS is a super solid way to go about it. And with Claude Code, it's never been easier. But the agency directory combo is a potent one and I see it all the time. Some of the biggest directories that I know monetize this way. But yeah, this is what's bringing in the majority of it. So yeah, that's the first one. Solid guess. Let's move on to the second one. Drag it over here. This one's called a place for mom.com and senior living, massive industry. What are your thoughts on this one?

• 5:55Greg Isenberg?

  Is it just for moms or for dads too? You know, what about dad?

• 5:59Speaker 01

  That's what I'm saying. It's messed up.

• 6:02Greg Isenberg?

  It's messed up.

• 6:03Speaker 01

  It's for everyone, hopefully.

• 6:05Greg Isenberg?

  Yeah. I would say 50,000 visits a month or more.

• 6:11Speaker 01

  Okay. And are you thinking lead generation as well?

• 6:16Greg Isenberg?

  Yes, uh, there's a huge business around lead gen for senior living.

• 6:25Greg Isenberg?

  It's, it's a huge business and there's huge publicly traded REITs, real estate investment trusts that, you know, you can see how much they spend on marketing. But yeah, there's a, it's a big business.

• 6:37Speaker 01

  Massive space and you know, care.com, that's the famous one. Let's take a look. So placeformom.com is getting 824,000 organic monthly visitors. The estimates on this are wild, like some, some websites claim that this makes anywhere between like a hundred million dollars a year, 500 million dollars a year. My gut tells me like 50 million dollars a year is probably right where I, I feel like it's more, more accurate and the way they monetize is basically purely from lead generation. So they have 18,000 or over 18,000 senior living homes that are listed on their directory.

• 7:23Speaker 01

  And they basically have all of these kind of sneaky ways to collect information and then pass them off to these senior living homes where they collect a portion of the first month's rent or like a fixed fee for every referral that they give.

• 7:38Speaker 01

  So just a beastly, but very simple and straightforward directory and like super classic, like I think most people think of lead generation when people think of directories. So I love this one. It's just so well made. It's also built on WordPress, I believe. So for the WordPress users out there, this is some inspiration. So just a really cool one and I'll leave the last one over here. The last one's going to be gasbuddy.com. It's a crowdsourced directory to give you gas prices near you. So if you type in like a random zip code, you can just find the gas prices, save some money and it's as simple as that.

• 8:23Speaker 01

  So what are you thinking on this one as far as traffic?

• 8:29Greg Isenberg?

  Uh, this is, this is a big one, dude. Like, I see right here 100 million downloads and counting. I've heard of GasBuddy. I think that this, like, data mote is a really interesting one, just as like a business. I'm a co-founder of an agency called LCA, and we built some really popular AI products. And like, one, one that we're working on right now is called Stretch AI. So it's grocery prices around you. And it's by the co-founder, uh, one of the top guys at Waze, which is also a data play. So anyways, point is, I think that if you're number one, you have the data mote.

• 9:10Greg Isenberg?

  This is getting, I'm gonna, I'm gonna say one, I'm gonna say 3 million. No, I'm gonna say 2 million visits a month.

• 9:19Speaker 01

  Like, that's, that's a good amount. I mean, okay, yeah, but that's, that's actually a really good guess. And I totally agree. I, I think on the surface, this looks simplistic, but everyone needs gas. I mean, pretty much everyone. But this directory is getting 1.1 million organic monthly visitors. And they monetize in two main ways. They make a ton of money on ads. So with this amount of traffic and tier one traffic coming from the US and Canada. Yeah, they're probably, I wouldn't be surprised if they're pulling in like, on the lower end, $30,000 a month, just in ads, uh, which if we clicked around, we'd start seeing. But their big play is actually a debit card.

• 10:02Speaker 01

  And so, um, if we go over to GasBuddy Plus Premium, they have this debit card that they partnered up with Mastercard to create. And it basically allows you to save money on gas. It's $10 a month or $99 a year. And so they have a ton of members here. I love this one. This is my favorite one of the list because... I think crowdsourcing gas prices is so simple, but extremely difficult. So I went back to 2004. They started in 2002, but I wanted to understand how they even incentivize people to get gas prices. Cause that's like, why would anyone want to take time out of their day?

• 10:44Speaker 01

  And back in 2004, all they did was create a public leaderboard, which they still have today. If we go to gas tools here, we can see all these people. Prince 33, leading the pack. And they basically did giveaways. So in 2004, they gave away e-bikes. And the way that it would work is that if you updated or reported any gas prices, you'd get like 150 points. And then, once you reach 1,000 points, that gives you one entry into the giveaway. And they're like, yeah, you could get an e-bike. So I thought that was super interesting. It's like very simple psychology. Everyone loves a public leaderboard.

• 11:25Speaker 01

  We all can't be Usain Bolt, but if you want to be the best gas price reporter in the world, then you know, you can be that. And let me see if I can quickly find the wayback machine tab here. Yeah. So check this out. I thought this was kind of cool. So this is a, yeah, so this is a snapshot from 2006. And we can see this little leaderboard here. Prince 33 was active back 20 years ago. Gary, Indiana, shout out. But now, you know, consistency, persistence has gotten him to number one.

• 12:07Speaker 01

  And what's insane is this guy hasn't missed a day in almost like in over 5,800 days. Like this guy wakes up and he's passionate about gas prices, apparently. And it's just insane to see the super fans. These days, they do give you, the giveaways in e-bikes anymore. They give away a hundred dollar prepaid debit cards for gas. But I just thought this was super interesting and it's actually a rare case where the directory monetization is super aligned with the reason why users are coming to this website because a place for mom.com, partying.com, they have to go and do like outbound

• 12:50Speaker 01

  sales to like reach out to business owners and create an offer that way. But these guys just were really creative with it. So I love this example and I wanted to show you these three because like all three of these directories execute this framework that I think all successful directories do, which is they help people save time, save money, or help people make money. And if you look on any one of these like listing pages here, we can see like more specifically, they've done a really good job with bringing price transparency as a form of data enrichment in areas where it's just really hard to get pricing information.

• 13:33Speaker 01

  And I think this is actually a massive opportunity for directories. It's incredibly unscalable and manual, but there's so many industries and niches where price transparency just doesn't exist. Tons of boring ones. And so for me this year, I think like data can be a moat, it can be a differentiator, and price transparency is a big one because it's so hard to get. So that kind of leads me into what I wanted to show you guys today, which is how to get

• 14:10Speaker 01

  valuable data. And let me just switch tabs over here to what I built. Let's do this.

• 14:19Greg Isenberg?

  Good segue because I'm bought in on the thesis. Like I agree. I think the data is the moat. I also think that you can use AI in clever ways to get the data. So.

• 14:34Greg Isenberg?

  I'm intrigued, Frey, I'm intrigued.

• 14:38Speaker 01

  Yeah, it's definitely the most boring part and the reason, it was actually kind of hard to decide what I wanted to talk about because with Claude Code, you're doing so many things that I wanted to go deep rather than wide. But yeah, I would say for directories, the mode is definitely data and your SEO, like if you have really strong backlinks.

• 15:02Speaker 01

  But yeah, like also just quickly, another reason I wanted to show you those three directories is because I think the number one question that I saw reading the comments from the last pod was how do you monetize these directories? And the answer is really unsexy, but it depends on the niche. I don't think it has to be display ads or Legion. It obviously can be really creative. I've seen tons of different offers created on the backend of directories. But yeah, it's super cool. So anyways, I built this directory in four days and it literally started from a raw, massive, unorganized CSV of data

• 15:44Speaker 01

  of over 70,000 rows that I scraped for the entire country for porta potty suppliers. And I just wanted to start with an example where I show you what a directory might look like with high quality data and one that looks, yeah, and what one would look like with low quality data. So let's just start with this one. This is my Claude Code directory, luxury restroom trailers, boring but awesome niche. People are spending a thousand dollars, $2,000 a day renting these out for their weddings. You know, there's corporate events, film sets, really all sorts of stuff. Just kind of showing you what you can create in record time.

• 16:30Speaker 01

  And yeah, the listing looks great. Super, super clear with the amenities and features. We have some images, the types of stalls that you can go and rent. And then of course, the lead form, some service areas. And I think this looks solid. But let me take you a year and a half back when I didn't know Claude Code. And I created this beauty. This is porta potty match.com. I built this in WordPress and it still has lormipsum on the front page. You know, we got some listing pages that are looking pretty sparse. And all of this is just AI generated, like sentence structures all the same. It's a bad directory. It's ugly. It's bad.

• 17:13Speaker 01

  All these images are the same for every listing. So I built this thinking like no one could reasonably trust this directory and like inbound as a lead, right? But to my surprise, I just built this and left it up. And I got like some leads coming in. And this is one example here, kind of blocked out the email, but I actually got a bunch of leads that just inbounded. The guy named Eric wanted a restroom trailer for, with multiple stalls. He has a film shoot, so that was kind of a cool way to look at the use cases.

• 17:54Speaker 01

  Another film shoot guy coming in, asking for a trailer for a film shoot. Martha looking for a reasonably priced porta potty with a sink or hand wash station. And then this is the big one. I got a lead from the New Mexico State Fair. This is the order that they wanted. And this is like $20,000 plus of porta potty slash luxury restroom trailers that they wanted. So just an insane lead and I was surprised. So despite my really poor data curation, it kind of just showed me that there is a

• 18:38Speaker 01

  massive need in this space and it's not really easy to go and eat, compare the different porta potty types if you're in the market for one. And there's also like a vast amount of like use cases. But I knew that things were kind of falling through the cracks. And so if all these leads inbounded with this crappy looking directory with lormipsum on the front page, what would it look like if I actually built this thing out now that I know Claude code? So I vibe coded this, kind of how I built this directory that outlines this seven step process on how I'm using Claude code and an open source GitHub repo called Crawford AI to.

• 19:19Speaker 01

  Handle any kind of data curation. So this is where I kind of just want to show people how to tackle that data side, because I've seen probably a thousand or 2,000 people build directories in the last year. And 100% this is where people either quit or give up. And I totally empathize. There is a point in my life where I was waking up and being like, "All right, I'm going to sit down in this chair for six hours and click on this website and just manually verify if this is a luxury restroom trailer." And it was hell, like it was awful, terrible, terrible use of time. But as you can see from these stats, I mean, I built this in four days. Like I mentioned, I probably saved over 2,000 hours in what would have been manual data cleaning,

• 20:03Speaker 01

  manual data enrichment, also record low costs. I mean, I would have to hire a developer to like write custom Python scripts before, and it's just not necessary nowadays. So I built this in under 250 bucks. A hundred, yeah, a hundred dollars of that was my Claude code max subscription. Another hundred dollars was for the data. And then 50 dollars for some Claude API credits to do some deep cleaning, which we'll talk about in a moment. So all you really need to get started is your directory niche. Mine's luxury restroom trailers. You need Claude code and you want an understanding of what drives decision making?

• 20:45Speaker 01

  Like, what do people actually need to make a decision around a luxury restroom trailer? And you can just achieve that by looking at Reddit forums. I think TikTok's great, Facebook groups, Instagram, anywhere where the conversation online is being had around these, around the topic that you're building your directory around. So as we can see, like this whole process I'm about to go over is just the second step. For me, I think my structured approach to building directories starts with finding the idea, collecting the data, building the website, SEO optimization, and then figuring out monetization. So let's just start with the first step. The first step is just scraping the data.

• 21:29Speaker 01

  I just went straight to Outscaper. I don't think.

• 21:33Speaker 01

  There's nothing new here. I went over this in the last pod, but Outscaper is still the cheapest option for me. I've used Appify and other kind of alternatives, but just keep it simple. I think Outscaper is a great place to start and there's tons of videos online where people can learn how to scrape data with, from Google Maps using Outscaper. So I went and did this. I got 71,000 rows and potential listings and covered the entire state. I build nationwide directories and the whole game plan is to sell leads in this high ticket niche. So pretty straightforward.

• 22:11Speaker 01

  The problem is 71,000 rows is massive and we definitely have to do a lot of cleaning as far as data goes. So this is when I introduced Claude Code into the mix and I used to have to manually clean data and I would have all of the obvious junk data removed. So remove things like listings with no business name, address, city, state, permanently closed ones and any like obvious ones that don't relate to my niche, like big box retailers. And so this is the prompt that I wrote. And obviously people are going to have to kind of retrofit this to your own niche, but I'm just telling Claude Code,

• 22:53Speaker 01

  here are my five CSVs. Go ahead and just look at every single one and use this criteria to clean the data. And so this simple prompt got me from 71K down to 20,000 listings. And if you are going to use this prompt, I think anyone can benefit from this particular part, which is just getting rid of obvious junk data. And yeah, what we're left with is just still a massive piece of data. So the next round is where it gets a little bit more interesting.

• 23:31Speaker 01

  And this was like where I had the biggest problem a couple of years back, which is, okay, we have 20,000 potential porta potty businesses. I've kind of exhausted all of my ways to kind of superficially clean the data. And so at this point, I need a way to automate the process of manually going to every single website and verifying whether or not these are luxury restroom trailers. And so the way that I did that is I installed Crawler AI. This is an open source, LLM-friendly web crawler and scraper, totally free, which is insane. And this is something that you just install locally to your computer.

• 24:12Speaker 01

  I'm brand new to AI coding, like I'm only six months in. And so if you're just getting started, the way that I literally did it is just took this link, gave it to Claude Code and told it to help me install it. And it took about 15 minutes. So now that we have this installed, Crawler AI is kind of the engine that allows us to, at scale, look at every single website. And then Claude Code is the brain, so we can prompt Claude Code and say, "Look at every single one of these 20,000 websites and just go and identify the luxury restroom trailers." So as a quick little demo here, um, oops.

• 24:56Speaker 01

  I'll just pull up my Claude Code here. And I just wanted to show people the workflow instead of just talking about it. But I'll just do it on a smaller sort of sample size. So here's a CSV of just 10 restroom trailers, or sorry, here's an example of 10 porta potty companies. And basically, I'm trying to figure out which ones are restroom trailers and which ones aren't. So I'm going to go ahead and just run that prompt that I showed you here. And I'll just copy and paste it here. And great, you're not very technical from what I remember. I'm not, I'm not.

• 25:37Speaker 01

  Like, that's the funny part is like, when I learned Claude Code, it was like, even to this day, I don't really think I use Claude Code to what it could be. But yeah, like, I've still been able to accomplish like a ton and...

• 25:59Speaker 01

  Yeah, I probably should explore other other tools, but I just try to keep it simple and do it well. With tools like this. So yeah, I'm totally not technical, which hopefully gives any reason to anyone watching this who's less technical to learn Claude code. Cause if I can do it, I can. You definitely can do it. So I use cache results. So this came back super fast, but basically I gave it 10 businesses. Here we can see it identified three luxury candidates and seven standard porta potties, but it might take longer if you run it for the first time. And essentially I just kinda told it to

• 26:40Speaker 01

  go through this, go through every single website and identify like restroom trailer related keywords. So it's kind of helpful to know like all the synonyms of the keyword you're trying to look for. And a couple of pro tips, if you are using crawler AI, there's extra modules that you can add. Including this one. And this is one of a few that I used, but async web crawler allows you to just crawl multiple websites concurrently. That saves a ton of time. But yeah, this was kind of the point where I was like, oh snap, this is insane. And I can't believe it's free because I set this up in probably 15, 20 minutes.

• 27:25Speaker 01

  I wrote this prompt by just talking to chat GPT. And then it just ran in the background for three hours. And this is really what would have taken me like a thousand hours, just a couple of years back. So the result was 20,000 websites down to 725 luxury restroom trailers. We can see the CSV here. And yeah, it even gives us like verification confidence. It's just nice. It's nice with it. Like I saw this and I was like, this is so cool.

• 28:01Speaker 01

  But that's not it. And that's not really enough to build a directory. So we move on to step four, and this is when we get into trailer inventory. For me, that's trailer inventory, but for you, it's just like, for you, it's the first round of data enrichment. Like I need to figure out. Out of all of these 725 restroom trailers, what do they offer? Like, what's, what, like, if you look online, most people are looking for two-stall and three-stall trailers, because those are good enough to cater to like a 50 to 100% wedding. But some people are looking for more. And so I wrote this prompt, again, it's very casual. You can tell that I just had a conversation with ChatGPT.

• 28:45Speaker 01

  And I just used Crawford AI in the same way, but this time, instead of looking for keywords like restroom trailer, I told it to look for the full like fleet of products that a business offers. And, um, yeah, so I believe I have a prompt as well. We can actually just take this right over here, just to show people. I won't do this for every single one, but just to show people kind of the workflow. I can say, um, let's take those three candidates for luxury restroom trailers.

• 29:27Speaker 01

  And use this prompt, I'll just do that. And this prompt again is just telling it to go and find all the different types of luxury restroom trailers that a business offers. And I do like to add to just like get Claude Code's opinion. Give me your game plan. Tell me if I'm missing anything. And just let me know before we we go for it. Before we burn tokens. Before we burn tokens and we have to upgrade our Claude. Yeah, so

• 30:03Speaker 01

  this is basically,

• 30:08Speaker 01

  yeah, so, uh, cool, yeah. So, yeah, it basically came back with.

• 30:14Speaker 01

  A1 containers, there's a two-stall trailer and a four-stall trailer. Again, this is gonna take more time, but you can set up whatever logic you want. And actually, there's a lot of nuance with this process. So just to kind of go over that, like, there's a ton of edge cases that you wanna check after every single pass. And one of the first mistakes that I did when I started enriching my data with Crawford AI was I would just give it this massive laundry list of things to get and just gave it like a massive CSV. And I was like, get the trailer inventory, get the images, get any amenities, features, pricing, all of that stuff. And it just didn't work, like, it was super low quality.

• 30:58Speaker 01

  So that's why I'm kind of going slowly, one at a time. And this is also helpful, because after this pass, I could examine the results. You always find edge cases, just these one-off situations where you should let Claude Code know that it messed up, it done messed up, so that you can fix it. And you might have to rerun this. I reran this, I think, two or three times until pretty much all of my data was good. And I had all of the stall information for every single listing.

• 31:34Speaker 01

  So yeah, this, this process just continues one by one. For me, the next one was trailer images. And this one was cool, because not only did I use Crawford AI to scrape the images from the business websites, which, by the way, I know is a gray area, but I planned to go and reach out to these businesses and ask for permission and just have them claim the listing, which basically gives them, gives us the, you know, green light to use this. But you can scrape the images. And what I did was I just told Claude Code to look for any alt text, file names, examine the page that it's on, and scrape the highest quality images.

• 32:16Speaker 01

  And then what I did is I told it to create, I told it to take the top three candidates for trailer images. And then I just sent it over to Claude Vision. And this way, it was able to look at these images. And identify the best ones because the first time I did this, I got like logos and crappy images and Fabicons and it was like, am I really about to clean this image data? Like this sounds so bad. And this is the workaround that I came up with. This cost about $30. All I had to do was go to Claude and connect my API key. Super simple. And this is the prompt that I used.

• 32:59Speaker 01

  I let this run overnight, but yeah, this took probably about an hour or so for 700 listings. So if we take a look, this is kind of where we're getting brick by brick. We're just starting to get all of this good information, stalls, and all the green columns here are kind of the ones that are valuable. Let me see if, oh yeah, yeah. So all of the, yeah, all of the green highlighted columns are of the images. And I think it did a really great job. So if we just click on any one of these, they're pretty solid. And this was not the case the first time.

• 33:44Speaker 01

  So we're starting to see how this comes together.

• 33:47Greg Isenberg?

  Just on the scraping image piece, you know, I'm not a lawyer. I don't play to be one on the internet. Frey, not to my knowledge, is a lawyer. Is that correct?

• 33:59Speaker 01

  I'm not.

• 34:00Greg Isenberg?

  He is not. So I think, you know, but I, but I think that you, if you scrape people's images, you are, the only way you're, well, let me take a step back. You're supposed to have the rights for images that you post on your websites, right? So sometimes those rights are available, like.

• 34:27Greg Isenberg?

  You know, if you look at the terms of services of some websites, those are available. And sometimes these are, these are like small local businesses, the chances are they might even have a terms of service. So, so yeah, just so you know, there is some risk associated with scraping images without the rights.

• 34:45Speaker 01

  Totally, yeah. And I think you don't even need images for a directory to rank and get traffic. I just kind of wanted to show this use case and I'm sure there's ways and there's different niches where you can scrape images where it's less risky, you know, public images and such. Or you can just use like stock images. I mean, we saw from my crappy little directory, no pun intended, like this got leads without needing a ton of images on the listings. So 100% not advocating for this and definitely some risks. So just know that going in and I think that was a good shout out.

• 35:25Greg Isenberg?

  Before we get to the next app, I just want to make a quick note about your quote unquote crappy other website.

• 35:33Speaker 01

  Yeah.

• 35:33Greg Isenberg?

  I will say the logo on this website does slap. You've got this mascot and he feels very approachable and interesting. So I am feeling, I don't know if the audience feels the same way, I am feeling a bit of a loss of a little character in the new, you know, fancy Apple style website.

• 35:57Speaker 01

  100%. I worked, I worked hard on this little guy and yeah, I just generated it with chat GPT. I'm sure you can use Nano Banana for this, but yeah, I'm actually planning on bringing this guy over and adding some human touch to this directory because it's definitely unfinished. It's probably like 60% finished, but I appreciate that. That's the, that's a compliment to my creative side, which, which does exist in the world of directories, but it's just maybe less common. But yeah, like just kind of going back to this though, it like the process just kind of continues stacking. Like I did this with the amenities and features.

• 36:42Speaker 01

  Again, like, the process wasn't smooth the first time. It would miss some amenities and features here and there. But this is the very casual prompt that I wrote.

• 36:53Speaker 01

  Basically, just, you know, mentioned that the first time I ran this, it didn't do a good job. There was a bunch of weird words like it and and and the. Like, that's not a feature. That's definitely not a feature. But this pass worked really well. Once I started to kind of understand the edge cases and say, I want you to go and identify all of the best pages. Go on the homepage, look for any page with restroom trailers and go deep.

• 37:22Speaker 01

  And was able to create these cool amenities. And this is what it came up with. Tons of different features that would ultimately become the filter on my directory. So if we check out

• 37:38Speaker 01

  that page that we're checking out here, like, I give people the option to choose by stall. So this is the stall data. But if we also go back and click on running water, for example, it'll just show all of the luxury restroom trailers with that specific amenity. And people can click multiple if they want and then sort from there. So that was kind of the reason I chose this one.

• 38:06Speaker 01

  See if there's any sort of mentions here. Yeah, like,

• 38:14Speaker 01

  this one was pretty straightforward and this didn't really take too much time. But again, massive time saved compared to manually doing this. And then the last one was also pretty simple. It was just the service areas. A lot of people want to know if a place can drive out like 100 miles to a remote location where they're having a outdoor wedding.

• 38:36Speaker 01

  So yeah, this is the prompt that I wrote.

• 38:39Speaker 01

  And just asked it to go and find the service areas.

• 38:45Speaker 01

  The only challenge with this is that some businesses were like in different states. So if a business was based in Florida, sometimes the first run would show me like Florida, Texas, Arizona. And so I had to make some adjustments, but this is. These are the new columns that it created. I broke it down by city, region and radius. So also just good to know for anyone who's shopping around for a luxury restroom trailer. And so that's pretty much the full seven step process. And I think it might be eight steps or four steps, depending on the type of niche that you're building a directory in, or how deep you want to go with the data enrichment.

• 39:29Speaker 01

  But I think if you can just use Crawford AI and Claude code, it'll serve you well. And more importantly, it kind of just opened my mind up to the types of directories that I can build. Like for a long time, I had some ideas where I could go and create like very, very niche directories in competitive spaces. And we saw a place for mom.com killing it.

• 39:55Speaker 01

  But unless you're like an expert SEO, have unlimited budget for backlinks and have a lot of time to wait, you're not going to rank for senior living homes. But you might rank for senior living homes for people with dementia, which if we look on Ahrefs actually gets like over a thousand monthly searches. So it's in high ticket niche. You're kind of taking your little pocket of the industry and doing that really well. And I think this kind of idea works well because for the people looking for this, like if they don't assist with dementia, that's a deal breaker. So what are like the deal breaker features that people make decisions off of?

• 40:38Speaker 01

  And how can we create like very specific directories around that?

• 40:43Speaker 01

  Bathroom contractor, same thing, super competitive. We got Angie's List and Home Guide and all of these other directories. But what about like an ADA accessible bathroom contractor? Like way more specific. And now that we have the ability to go and clean and enrich data by looking at the website content, like this is all possible now. I'm also a massive nerd of like public databases. Like data.gov has an unreal amount of just public data. That's super like horrible to use. The UI and experience is rough. This guy named Andy in our community built a tap water quality directory. No backlinks, started it back in, I want to say November.

• 41:27Speaker 01

  And he's getting over 40,000 monthly visitors, got accepted into Mediavine. Selling like water filters through Amazon affiliate. And yeah, it's super boring, but like air quality directories. Like there, there's a bunch of like mundane, but important directory ideas that I think this unlocks. And yeah, event directories, I think event directories suck. They still suck. I, I don't know if this is like, like too much information, but I went to like a singles event like four months ago. And I went there and I started talking to the people. I talked to some guys and I was like, how did you learn about this event?

• 42:12Speaker 01

  And they're like, oh, we went through Eventbrite and just scrolled and found this event. I'm like, that's crazy that you do that. Like I just, I, I, that couldn't be me. Because I feel like these event directories are horrible. But now you can essentially set up scrapers that scrape from multiple sources and you just curate a better version. And with Cloud Code, you can make anything that you want. So I think there's an opportunity in event directories as well. But um, yeah, just to show people the full process. I mean, once you get the data, you have the heart of your directory. At this point, you know, I created a cleaned up

• 42:53Speaker 01

  version. There's tons of data that you're gonna get. So I stripped away all the columns of data, and then I just gave it to Claude Code and said, hey, use this to create the Superbase database, use the exact columns. And then the fun part started, where, once you have the database, you can just go buck wild and create whatever you want and design it however you like. So that is the full process. That's the most up-to-date way that I've been kind of scraping, cleaning, and enriching my data. And yeah, it's the most important part. It's just kind of the most boring part as well.

• 43:38Greg Isenberg?

  So if you've made it to the end, you're probably bought into this idea around, okay, a directory could be a really good entry point into building something that doesn't require a lot of my time. It could be high value if it's a really good niche. And then I could build a MicroSaaS to go and, you know, really amp up monetization and enterprise value. But I suspect there are some people who are like, not bought in. So for the people that are naysayers about directories and creating a directory right now, specifically

• 44:21Greg Isenberg?

  in this moment in time in 2026, what do you have to say to them?

• 44:25Speaker 01

  Yeah, that's a good point. And I would actually say, if your timeline is to make money in less than six months, I would not build a directory. And I'm just saying that because I've been building them for a few years, and I just want to save people the time. With that said,

• 44:42Speaker 01

  I don't know if there's, I'm obviously biased, but I don't know if there's a better playground and structured kind of approach to a project where you can learn how Claude Code works and learn SEO and learn how to sell and get into the business of lead generation, which will always be around, than directories. And the first directory that I built with Claude Code was just a playground for me to learn. And because it's so simple, like you're able to

• 45:17Speaker 01

  like use this as like a vehicle to learn that high leverage skill, which I think is AI coding and SEO. So at the end of the day, the way that I think about directories is that it is still to this day a playground for me to learn how to get people's eyes on my projects. And I've built other businesses and I've spent like a lot of money on Facebook ads in my first business. I would argue the top three ways to get distribution nowadays are ads, organic social and SEO. Like those are the big three. So choose your distribution. I made the mistake of not focusing on distribution first and directories are a distribution first model.

• 46:00Speaker 01

  They kind of have an unfair advantage when it comes to SEO because if I create this directory or when I publish this directory, I kind of have this topical relevance where I have like a thousand pages all around luxury restroom trailers. So it's really easy to start ranking. So it's really easy to start ranking for luxury restroom trailer Bakersfield, California, and then work my way up to like luxury restroom trailers, Los Angeles. So I think it just gives, it kind of just gives you this really full process of like how to take an idea, build an online asset that you can sell. If it is monetizing, they can monetize in multiple ways.

• 46:43Speaker 01

  Generally, pretty high margin, you know, you can do it from anywhere in the world and it's pretty low cost now. You can build a directory for under a few hundred bucks. So in a way, these are small bets, but they can be massive businesses as we saw in the very beginning. But again, just to save people time, SEO generally takes time, especially if you're just getting started. So if your timeline is under six months, if you got fired from your job or you need money immediately, I would say go to garage sales and flip items or something like that. Cause that's what I did when I really needed money and yeah, that's what I would say.

• 47:22Greg Isenberg?

  A hundred percent. I think in general, no matter what startup idea you're working on, you can't expect it to be overnight in general. No matter what business model, marketplace, social network, mobile app, directory, Microsoft. You know, this is a thing that you have to iterate towards and find and find the channels and scale it and stuff like that. And some things are out of your control too. One thing I wanted to ask before we head out was in a world where L L Ms, like, you know, you mentioned S E O is a huge traffic driver for some of these directories.

• 48:05Greg Isenberg?

  In a world where people are finding products and services through L L Ms like perplexity and chat G P T and Gemini. You know, where, what is the role of directories there, right? Cause you know, is chat G P T just going to scrape luxury restroom trailers dot com and just put it in there? And then how are you going to be able to generate revenue?

• 48:28Speaker 01

  Yeah, totally fair question. I would say I have two points on this. And the first one is that I think if someone's browsing a directory in 2026. They're sort of past the discovery phase. Like what I think ChatGPT perplexity and all these other LLMs do really well is they help you discover things or get more information on the thing. But by the time someone's on a directory, I think they're in the decision making phase. And with any more complex decision where the consequence is too much to risk if you mess it up, they're gonna do their due diligence and look at all the options. And so this can play out in a couple of different ways, like senior living homes.

• 49:12Speaker 01

  Like if you're finding a senior living home for your parents, are you just gonna go on ChatGPT and just like, choose the first one? No, you're probably gonna take your parents' preferences in mind and they might have deal breakers like the dementia example. Lawyer, anything in like health, finance or legal, the consequence and the risk of choosing a bad accountant or financial advisor is just like not worth. Like you're not gonna just choose the first option. And so you're gonna compare all the options near you. And I think that's what pushes people to Google and like really go out of their way to look for what could be your directory. Pricing is another thing.

• 49:56Speaker 01

  Anytime someone can save money, I think that drives a lot of just going on Google and looking at directories. And yeah, like that's, that's a hundred percent a thing. I was listening to the podcast, a podcast around like, you know, these millionaires driving across town for cheaper blueberries at the grocery store, like this is always gonna be a problem. And then the second point is like local SEO or any version of local SEO hasn't really changed that much. And what I mean is LLMs are decent, but if we go to like haircut Los Angeles, for example,

• 50:38Speaker 01

  this is now kind of looking at things from an SEO perspective and why directories might continue working for a long time. This is a local query and there's not really much different. Like you kind of have your local pack here, some sponsored posts, but then you have your organic listing results. But if you say like, I don't know, hair gel, something like that. Google's completely changed the way the serps looks over the years. And this is the reason why when I build directories, I don't choose product-based niches. Because it's way harder. You're just competing with so much other content and like socials and all this other stuff.

• 51:20Speaker 01

  So from an SEO perspective, I think like any kind of local SEO is still pretty low-hanging fruit. And I still think directories are just gonna be around in some capacity. And maybe we'll see a different direction. I don't know if you saw Mark Lewis' Trust MRR, but it's essentially a directory. And he leveraged his personal brand and audience to create it. But that's been going crazy. He turned it into a marketplace where now you can acquire or sell businesses. Kind of playing on this idea that you can monetize. You can monetize, once you have traffic, you can monetize in any way that they want. And people will go out of their way and just email you sometimes and say, hey, can you like add this to your directory?

• 52:05Speaker 01

  And that might inspire a kind of monetization idea.

• 52:10Speaker 01

  So that's what I would say. But it's definitely a real concern. I'm sure it's gonna get way better. And we might get to a point with agentic search where it's just unreal. And yeah, like we'll, we'll, we'll have to figure it out at that point. But I don't think lead generation as a business is ever gonna go away.

• 52:29Greg Isenberg?

  Yeah, you could make the argument that in a world where people are going to LLMs for search or AI search takes over, that in fact, the niche directories are the ones that stay. And the ones that are horizontal directories actually get hurt. Because the question, the prompting is very, very specific. And what's gonna happen is the data.

• 53:00Greg Isenberg?

  That, you know, ChatGPT is gonna have to reference your website no matter what. They can't just like fully steal all your data. So they're gonna, they're gonna give you the link. And the beauty is, instead of getting, you know, a thousand little blue links on Google and an AI search world, you're getting three, you know, you ask for a question, where can I find the best, you know, luxury restroom trailer in my area? You're getting one or two or three responses. So I think that if I'm building a directory this year, I'm trying to build it super niche so I can play into that new AI search era.

• 53:40Speaker 01

  Hundred percent. Can't be ignored. That is for a fact.

• 53:46Greg Isenberg?

  Frey Chu, thank you for, you're an angel. You're, thank you for coming on and being so well prepared and sharing as much sauce as you do. You're basically a spaghetti factory for directories and God bless you for it. We appreciate you. Frey, if people wanna learn more about you and your ecosystem, what's, what's the best place for that?

• 54:09Speaker 01

  Yeah, man. Always appreciate you having me on.

• 54:14Speaker 01

  Seriously appreciate it. But yeah, if you wanna learn more about directories, I do make a video every week on my own channel. And I also created a free directory community. There's over thirty two hundred people in it. So a bunch of smart people in there. If you have questions and you're just getting started or you wanna learn some tips, get some feedback, that is always an option. Otherwise, that's pretty much it. And yeah, happy to share what I'm up to.

• 54:39Greg Isenberg?

  Appreciate you. I'll include links for that in the description and in the show notes. And yeah, if you like, if you like this content, you want more of it, feel free to like, comment, subscribe, and I'll see you hopefully, you know, I think, what year did you come on, Frey? Did you come on last year or the year before?

• 55:01Speaker 01

  It was literally like a year ago, like almost to the, almost on the dot, actually.

• 55:06Greg Isenberg?

  Yeah. So would love, would love, I'd love this to be a yearly tradition.

• 55:11Speaker 01

  That sounds good.

• 55:13Greg Isenberg?

  And, you know, and see how the world of directories is evolving. And I know people love you and, and, and love, love you coming on. And I think, I think it's just because you're, you're really clear on explaining things. And, and it's, and you've, you've amassed a tremendous amount of knowledge in the space. So, so thank you again, and, and I'll see you next time.

• 55:37Speaker 01

  Appreciate it. Super kind, and I'll catch you later.