# Forget Big Apps. These Tiny Tools Are Making Millions

https://www.youtube.com/watch?v=50iWtrj9Mkw

[00:00] This is the asymmetric bet of the
[00:01] century. Cost to start is almost zero.
[00:03] The tools are either free or cheap. You
[00:05] could do this in an afternoon,
[00:06] literally. These can be $10,000
[00:09] businesses, million-dollar businesses,
[00:10] and even billion-dollar businesses.
[00:11] [music] Nico didn't have rich parents,
[00:13] didn't have an MBA, didn't have any
[00:14] experience, and he just did it. He's
[00:16] proved that this is doable for a regular
[00:18] human being. He used to just make pasta.
[00:20] His whole life was hospitality. [music]
[00:21] You don't have to build software. You
[00:23] don't have to invent anything new. Just
[00:25] put your own rims on it. Put [music]
[00:26] your own face on it. If he can do it,
[00:27] anyone can do it. When something works,
[00:29] the window is short before everyone
[00:31] copies it. And the better it works,
[00:32] >> [music]
[00:32] >> the shorter the window is. Cold email
[00:34] back in 2010 was insane. You could send
[00:37] 100 emails and get 15 replies. Today,
[00:39] you've got to send 10,000 emails.
[00:40] Facebook ads in 2014, [music] Google ads
[00:43] in 2005. You could buy a customer for
[00:44] two bucks. Today, it's 40. You got to
[00:47] get it while the getting's good, and
[00:48] that's usually today or yesterday. The
[00:50] phrase nothing is too niche is never
[00:51] been more true. Build lifestyle
[00:53] businesses. They are awesome. Let's talk
[00:55] 12 specific business ideas that you can
[00:57] build [music] using the same playbook.
[01:02] Okay, I just got off a call with a guy
[01:04] who, not that long ago, was making pasta
[01:07] in his family's restaurant in New
[01:08] Orleans. He'd never written a line of
[01:10] code in his life, and today, he's
[01:12] running a software business doing 3,500
[01:14] bucks a month in recurring revenue. No
[01:17] technical knowledge whatsoever. He
[01:18] wasn't even using AI. And the kicker is
[01:20] that anyone listening to this episode
[01:22] can do the exact same thing, literally.
[01:24] And it's all made possible by what I
[01:26] call the great unbundling, which is a
[01:28] concept, framework, whatever, that I'm
[01:30] obsessed with. So, by the end of this
[01:32] episode, you're going to have ideas, a
[01:34] plan, and probably an urge to open up
[01:36] your favorite vibe coding app tonight
[01:38] and build something. So, we're going to
[01:39] talk about what unbundling is, what
[01:41] platforms are ripe to be unbundled, and
[01:44] then at the end, we'll talk very
[01:45] specifically on what business ideas you
[01:47] should build in and around this
[01:49] unbundling concept that I'm about to
[01:51] describe to you. The guy I spoke to, his
[01:52] name is Nico. I'm going to come back to
[01:54] him in a few minutes, cuz his story is
[01:56] kind of the whole point of this episode,
[01:57] but before I tell you exactly how Nico
[02:00] did it, I'm going to show you a picture.
[02:01] It's a screenshot of the Craigslist
[02:03] homepage and on this screenshot are a
[02:05] bunch of arrows pointing to different
[02:07] services, the personal section, the
[02:10] community section, discussion forums,
[02:12] for sale, housing, jobs, etc. Then there
[02:14] are logos pointing to each of these
[02:16] sections, logos of companies you
[02:17] probably have heard of, Zillow, Airbnb,
[02:20] ZipRecruiter, Indeed, Upwork, and many,
[02:22] many more. And yes, this website looks
[02:24] like it was designed
[02:25] because it was. Tiny blue links, we've
[02:28] all seen it before. Bunch of columns of
[02:30] categories all crammed into one page,
[02:31] but what's wild is that almost every
[02:33] category on that page got turned into a
[02:35] billion-dollar company or
[02:37] hundred-billion-dollar company. Billion
[02:39] with a B. If you look at the housing
[02:40] section, apartments, rooms, sublets,
[02:42] rent, sale, that's the whole little
[02:44] corner of Craigslist that got eaten by
[02:46] Zillow and Airbnb, Redfin, realtor.com,
[02:50] etc. Zillow alone is worth what, 12
[02:53] billion dollars? Airbnb, 80 billion? All
[02:56] from one little column on Craigslist
[02:58] that got unbundled. If you look at the
[03:00] personals, strictly platonic, women
[03:02] seeking men, men seeking women, all
[03:03] that, boom, Tinder, 10-billion-dollar
[03:05] company pulled straight out of that
[03:07] column. Jobs, accounting, finance,
[03:08] sales, marketing, Indeed, ZipRecruiter,
[03:11] Upwork. Indeed alone does over 3 billion
[03:15] dollars a year in revenue. You look at
[03:16] the for sale section, antiques,
[03:18] appliances, bikes, electronics, OfferUp,
[03:20] Facebook Marketplace, and many, many
[03:22] others. Multi-billion-dollar businesses
[03:24] or the discussion forums, Reddit. Reddit
[03:26] was worth 7 billion at IPO or the
[03:28] services and gigs section, Thumbtack,
[03:30] Shiftgig, TaskRabbit, multiple
[03:32] hundred-million-dollar businesses. The
[03:34] Nextdoor, of course, like we know all
[03:35] these, okay? First of all, the
[03:37] unbundling of Craigslist is not done,
[03:39] it's still a work in progress for you or
[03:41] anyone else to capitalize on. Second of
[03:43] all, there's so many other things we can
[03:45] unbundle that isn't named Craigslist.
[03:47] Craigslist was unique because it had
[03:49] every feature in the world on one page,
[03:52] jobs, dating, housing, used stuff,
[03:53] whatever. And companies came along,
[03:55] picked one feature, slapped a clean
[03:57] design on it, made it mobile-friendly,
[03:59] and built a billion-dollar business. Of
[04:01] course, I'm oversimplifying, but we only
[04:02] have so much time today. I'm not saying
[04:04] it was easy for them to do this, but
[04:06] they stole the users of Craigslist. They
[04:07] were more savvy. They cared more about
[04:10] one little feature, one little blue
[04:11] link, than Craigslist did, because
[04:13] Craigslist was trying to focus on
[04:15] hundreds of little blue links. Also,
[04:17] side note, unrelated to this episode, I
[04:19] kind of hate how Craigslist is this
[04:21] ugly. Like, people think it's punk rock
[04:22] and it's cool, but they could have been
[04:24] something amazing. They could have
[04:25] provided thousands of jobs. It gets
[04:28] barely, I think, a couple million hits a
[04:30] month today, which is down like 95 to
[04:32] 99% from its peak. They could have
[04:34] innovated. They could have added a lot
[04:36] more value to the world, but now the
[04:37] rich got richer and you know, Facebook
[04:39] Marketplace is is doing that job pretty
[04:41] well. But, that's fine. That's free
[04:42] market capitalism, whatever. So, the
[04:44] exact same thing is happening right now
[04:46] on dozens or hundreds of other websites,
[04:48] this unbundling. And over these last few
[04:50] years, it's happening with AI and APIs.
[04:53] Look at OpenAI, aka ChatGPT, as the new
[04:55] Craigslist, or Anthropic, aka Claude, as
[04:58] the new Craigslist. There's a company
[05:00] called Appify. Think of them as the new
[05:02] Craigslist. Clay, clay.com, Zapier, an
[05:05] API Marketplace, the new Craigslist.
[05:07] Each one of these platforms bundles a
[05:09] thousand different features into a
[05:11] single tool. And the whole opportunity
[05:13] for the next decade is to unbundle them.
[05:15] Pick one feature, wrap it in a nice
[05:17] little app, charge a markup, own one
[05:19] niche. And up until three, two years
[05:21] ago, this sexy unbundling thing was only
[05:24] available to people that knew how to
[05:26] code, software engineers. But, now it's
[05:27] available to all of us, to any of us,
[05:29] anyone that knows how to speak any given
[05:31] language. Any of us can own a niche.
[05:33] Nico, this guy I talked to, he calls it
[05:35] API arbitrage, and my knees got weak as
[05:38] soon as I heard those two words put
[05:40] together, because that sounds awesome.
[05:41] It sounds like opportunity. And it might
[05:43] sound technical, but I'm I'm explain it
[05:45] in Chad GPT language right now. So, it's
[05:49] going to make sense to anyone listening.
[05:50] Okay. So, you go to website like Apify,
[05:53] which is like an app store for scrapers
[05:55] and AI tools. There's 30 plus thousand
[05:57] different Apify actors or apps or
[05:59] scrapers on this website. 30,000. You
[06:01] pick one of those off the shelf. Let's
[06:04] say it's a scraper that can pull every
[06:06] Instagram follower from any account.
[06:08] That scraper's going to cost you like
[06:10] pennies per thousand followers scraped.
[06:13] It's almost nothing. And someone with a
[06:14] lot more technical knowledge than you
[06:16] built the scraper. You don't need to
[06:17] know how to build it. You just need to
[06:19] connect the API key. You build a dead
[06:21] simple website, vibe code it with your
[06:23] favorite vibe coding tool, put a search
[06:25] bar on it, charge customers 1 cent per
[06:28] follower pulled, 5 cents per follower
[06:30] pulled. It's a great deal for them and
[06:32] you have like 95% profit margin on that.
[06:34] The customer doesn't know that Apify
[06:36] exists. If they do know it exists, it
[06:37] might be a little too technical for them
[06:39] or complicated. Or they might not like
[06:41] that it has a monthly fee. Maybe the
[06:43] customer knows Apify exists. Maybe they
[06:45] don't. Regardless, it doesn't matter
[06:47] because they like your website better
[06:48] even if they have to pay more for it.
[06:50] And your website also conveys the
[06:52] assumption that you're better at
[06:54] focusing on only one thing. Whether
[06:56] that's true or not is irrelevant. If
[06:57] you're only scraping one thing, then the
[07:00] customer assumes you're better at it
[07:01] than a competitor or than Apify even,
[07:04] which is the whole thing that you're
[07:06] white labeling, you're using to do all
[07:07] the hard work. That's API arbitrage.
[07:10] You're competing on price, customer
[07:12] support, and how easy your website is to
[07:14] use. You don't have to build software.
[07:16] You're just putting a simple storefront
[07:18] on top of someone else's software. So,
[07:19] back to Nico. He's proof that this is
[07:21] doable for a regular human being. He
[07:24] used to just make pasta in his family's
[07:25] restaurant in New Orleans. His whole
[07:27] life was hospitality. He had no tech
[07:29] experience, no sales experience,
[07:31] nothing. He didn't even know what an API
[07:32] was. He never heard those three words
[07:34] put together. Then COVID hit, he went to
[07:36] Mexico for a 4-day vacation, beach,
[07:39] tacos, you know the the and he ended up
[07:41] staying for 6 months. He was in Tulum
[07:43] and Playa del Carmen, and he started
[07:44] meeting laptop people, you know, people
[07:46] working from laptops in coffee shops,
[07:49] etc. And he's like, "You guys are
[07:50] working? What are you doing?" Like,
[07:52] "Yeah, you know, we're software people.
[07:53] We work remote." This is, you know,
[07:54] before ChatGPT. "We make six figures. We
[07:56] live on the beach." And his brain just
[07:58] kind of exploded. All he knew was the
[08:00] restaurant industry. So, he went back to
[08:02] the states 6 months later, and he's
[08:03] like, "Okay, I'm done with restaurants.
[08:05] I like this laptop life." And he got a
[08:07] job as a sales guy, as an SDR, sales
[08:09] development representative, basically
[08:11] setting up appointments for sales
[08:13] people, for closers. He's cold calling
[08:15] for software companies. It's kind of
[08:16] like the lowest rung on the tech sales
[08:18] ladder. Brutal job, tons of rejection,
[08:21] LinkedIn DMs, cold calling, cold emails,
[08:23] etc. He's learning the game. And while
[08:25] at this job, he learns a tool called
[08:27] clay.com. And he had been getting these
[08:29] spreadsheets with leads and like using
[08:31] the same leads that all of his coworkers
[08:33] were using to get sales, and Clay helped
[08:35] him find anyone's email, their LinkedIn,
[08:38] their company. It helped him enrich the
[08:40] data. So, he had, A, more ways to
[08:42] contact these people, and B, more ways
[08:44] to find new people that his coworkers
[08:46] didn't have access to. Also, this is not
[08:48] a sponsored episode. It's not sponsored
[08:50] for Clay, Appify, or anyone, okay? It's
[08:52] just me talking. And so, he started
[08:53] using this tool, and then his brain
[08:55] broke even more. He's like, "Wow, this
[08:57] is what tech can do." So, early 2025,
[09:00] late 2024, he quit his job, tech sales
[09:02] job, and started building. He just
[09:04] started tinkering with these vibe coding
[09:06] tools, not knowing anything about how to
[09:08] code. He used Claude, he used Codex by
[09:10] OpenAI, and he found a buddy who's is
[09:13] technical and could do the heavy lifting
[09:15] when it got out of reach for him. So,
[09:17] he's the product guy, the idea guy. He
[09:19] tells AI what to build. His buddy fixes
[09:21] whatever AI might break, and then they
[09:23] ship it. And then as they grow, as they
[09:25] scale, his technical friend helps make
[09:27] that happen. Guys, I'm talking about
[09:28] starting businesses every single week on
[09:30] this channel, and one of the first
[09:32] things every new business needs is a
[09:33] website, which means you need a domain
[09:35] name. So, what do you do? You go to buy
[09:37] one and every good .com is already gone.
[09:39] Your business name is taken. Your
[09:41] business name plus your city, taken. So,
[09:43] you end up with something dumb like
[09:45] try-bizname-123.com.
[09:48] And that doesn't look like a business.
[09:49] It doesn't look legitimate. It looks
[09:50] like a fishing link your grandma would
[09:52] accidentally click on. So, stop fighting
[09:54] over .coms. If your .com isn't
[09:55] available, get a .online domain name
[09:57] instead. And not as a backup, but as the
[09:59] better choice. Because a clean URL like
[10:02] yourbusinessname.online
[10:04] immediately tells customers that this is
[10:06] a real business with a real website. No
[10:08] dashes, no random numbers or confusion.
[10:10] Service providers, consultants,
[10:12] e-commerce sellers, millions of
[10:13] businesses already run .online domains.
[10:16] And then you remember that the word
[10:18] online appears in over half a billion
[10:20] search queries every month. So, having
[10:22] that word built into your actual domain
[10:24] name can give you a visibility boost on
[10:26] search engines that a clunky .com with
[10:28] dashes never will. Now, you can buy a
[10:30] .online domain name through GoDaddy,
[10:32] Lovable, Namecheap, and other major
[10:34] providers. But if you want the best
[10:35] deal, go to the link on screen or in my
[10:37] description below and get yours for just
[10:39] 99 cents for the first year. That's 99
[10:41] cents for a domain that actually looks
[10:43] like a real business. Go grab it. So, he
[10:45] started with a Google Maps lead scraper.
[10:47] So, you go to his website, you type in
[10:49] gyms in Texas, and then you have a
[10:50] spreadsheet of every gym in Texas,
[10:52] right? Pretty cool. A lot of demand for
[10:53] stuff like that. There's also a lot of
[10:55] other websites doing it, but he didn't
[10:57] need any technical knowledge. He used, I
[10:59] think an Appify actor, an API on the
[11:01] back end to do all the scraping. And he
[11:03] just vibe coded it into a nice clean
[11:06] little website with a Stripe
[11:07] integration. Charges 35 bucks a month.
[11:09] But, he allowed customers to convert
[11:12] that 35 bucks into credits. So, if they
[11:14] cancel their account, they could still
[11:15] come back 9 months later and use the
[11:17] rest of their credits to scrape more.
[11:19] So, people like that because a lot of us
[11:21] have subscription fatigue. And his
[11:24] business model actually respected how
[11:26] most of his customers used the tool,
[11:28] which is a great learning in and of
[11:29] itself. So, he's been doing this for
[11:31] like 7 months and he's doing 3,500 bucks
[11:34] in monthly recurring revenue. His costs
[11:36] are almost zero. His customer support is
[11:38] basically zero. It's a self-serve
[11:39] product. And yeah, this is the guy who
[11:42] had no tech experience and he's doing
[11:44] well and he's growing double digits
[11:46] every month. He's never done paid ads.
[11:48] He just kind of told some friends about
[11:49] it in the sales space. They started
[11:51] using it. They started talking about it.
[11:53] And he's really found his success
[11:55] through word of mouth. He didn't have to
[11:57] invent scraping. You don't have to
[11:59] invent anything new. Just put your own
[12:00] rims on it. Put your own face on it. If
[12:02] he can do it, anyone can do it. Okay.
[12:04] So, now let's talk 12 specific business
[12:07] ideas that you can build using the same
[12:09] playbook. Then I'm going to do a bonus
[12:11] round of Appify like white label ideas
[12:13] that I'm just going to riff on cuz
[12:14] there's there's 30,000 actors. So, a lot
[12:17] of stuff to talk about there, but we'll
[12:18] focus on the only the best of the best.
[12:20] By the end of this, you should have at
[12:21] least one idea that makes you want to
[12:23] close this video, close Spotify or
[12:25] Apple, and go open your favorite vibe
[12:27] coding tool. That's the goal here. So,
[12:29] idea number one,
[12:30] Appify actor wrappers. Same thing that
[12:33] Nico did. You can do it with Google Maps
[12:34] like Nico, you can do it with something
[12:36] else. There's over 30,000 actors and an
[12:38] actor is just a pre-built scraper or an
[12:40] AI tool that someone already built. You
[12:42] just put a wrapper on top of it. And if
[12:44] that sounds complex, you're just getting
[12:46] a key, a code from the actor and you're
[12:50] pasting it into Replit, Lovable, Codex,
[12:53] whatever vibe coding tool you use. And
[12:54] you're saying, "Hey, connect this thing
[12:56] with this website." That's it. That's a
[12:59] wrapper. Instagram scraper, LinkedIn
[13:01] scraper, Facebook group scraper, Reddit.
[13:03] Buy a $10 domain name, vibe code it,
[13:05] host it on the domain name. Good to go.
[13:07] You could do this in an afternoon,
[13:09] literally. Okay. Idea number two, Google
[13:11] Maps scraper. Like you can do it just
[13:13] all of Google Maps or you could do a
[13:15] dentist scraper, roofer scraper, home
[13:17] service businesses scraper, Texas small
[13:19] businesses scraper, lawyer scraper,
[13:22] anything. Nothing is too niche. The more
[13:24] niche you are, the the you can target
[13:25] your ads. Who's a buyer for this stuff?
[13:27] People like me, B2B sales teams, cold
[13:30] callers, SDRs, marketing agencies,
[13:32] anyone who needs to build a list of
[13:34] local businesses. Charge 35 bucks a
[13:36] month, sell credit packs that never
[13:38] expire, and just run it. Post about it
[13:40] on LinkedIn to your 452 followers. You
[13:42] don't need to be an influencer. Idea
[13:43] number three, Amazon review scraper for
[13:46] direct-to-consumer brands, so they can
[13:48] kind of do intel or reverse engineering
[13:50] of their competitors. Every Shopify
[13:51] brand or Amazon seller, there are
[13:53] millions of these direct-to-consumer
[13:55] founder, they all want to know what
[13:56] their competitors are doing. What are
[13:57] they complaining about? What do they
[13:58] love? What features are missing? What
[14:00] words are reviewers using? Right now,
[14:02] the only way to figure that out for most
[14:05] people is to manually read hundreds of
[14:07] reviews or to manually screenshot and
[14:09] copy and paste and drag it over to
[14:11] ChatGPT and have that read the reviews.
[14:13] Most people aren't even doing that. It's
[14:15] soul-crushing work, but if you build a
[14:17] tool where someone pastes in a
[14:18] competitor's Amazon URL, your tool can
[14:21] scrape the last 500 reviews, run them
[14:24] through ChatGPT, and spit out a one-page
[14:25] report. Get the top compliments, top
[14:28] complaints, common phrases, suggested
[14:30] product improvements. The back end is
[14:32] all handled by an API key on Appify.
[14:34] You're also using an OpenAI API key.
[14:37] It'll cost you pennies per report. You
[14:39] could charge 50 bucks a month, 100 bucks
[14:41] a month, $19 for a one-off report. You I
[14:43] mean, you sell to the same people that
[14:45] are scraping this, right? There's I
[14:47] don't know, three or four million
[14:48] Shopify brands. There's a few million
[14:50] Amazon independent sellers. Your
[14:52] potential customers are the same people
[14:54] wanting intel on each other. The next
[14:56] idea can just be done with Zapier.
[14:58] Zapier is an API marketplace, and you
[15:01] just build a very simple automation. And
[15:03] if you've never used Zapier, you can use
[15:04] AI within Zapier to build this
[15:06] automation for you, or you can just
[15:07] screenshot stuff over to ChatGPT and
[15:09] have that teach you how to do it.
[15:11] But build a customer service bot, which
[15:13] is just a Zapier flow. You pay Zapier 29
[15:16] or 59 bucks a month for thousands of
[15:18] credits or hundreds or whatever it gives
[15:20] you. And if a business gets a customer
[15:23] service email into their inbox, Zapier
[15:25] will respond to the email through Open
[15:27] AI, and its response will be based on
[15:29] knowledge that it gleans from previous
[15:32] sent emails from the same inbox or from
[15:34] the website's FAQ page, the return
[15:36] policy, shipping policy, etc. If you
[15:39] think about customer service inboxes,
[15:40] we're just answering the same questions
[15:42] over and over. So, your cost is going to
[15:44] be like cents or low dollars per month,
[15:46] and you could charge 100 bucks a month
[15:48] per Shopify store. You've got 95%
[15:51] margins. They're going to save hours per
[15:53] week or per month. It's a great win-win.
[15:55] Every Shopify store owner hates customer
[15:57] service. The tech is already built.
[16:00] You're just putting some tools together.
[16:02] Believe it or not, just because you're
[16:03] watching this doesn't mean you're
[16:04] subscribed to my channel. YouTube's
[16:06] going to show you stuff even if you're
[16:07] not subscribed to it. Over half of
[16:09] people that watch my videos are not
[16:11] subscribed. It would mean the world to
[16:12] me if you just hit subscribe. Thank you
[16:14] so much. Next idea could be a
[16:16] micro-influencer marketplace. If you
[16:18] think of big agencies, big companies,
[16:20] they want to work with mega influencers
[16:22] like Mr. Beast, etc. But smaller brands
[16:25] know that micro-influencers usually have
[16:27] a more loyal audience. They convert
[16:29] better. People with 5,000 to 100,000
[16:32] followers. There's not really a good way
[16:34] to find them. Amplify has dozens or
[16:36] hundreds of actors that do exactly that.
[16:38] They can scrape people from Instagram,
[16:40] TikTok, YouTube, etc. that have a set
[16:42] amount of followers. You just build a
[16:44] wrapper on top of that. You could do it
[16:46] for a specific niche only, micro-fitness
[16:48] influencers, micro-food influencers, mom
[16:50] influencers, etc. or you could just do,
[16:53] you know, we scrape micro-influencers
[16:55] that have less than 50,000 followers.
[16:57] You could go that direction. You could
[16:59] turn this into a newsletter. You could
[17:00] automatically scrape everyone and just
[17:02] email it out to everyone that subscribes
[17:04] for 10 bucks a month or whatever. I
[17:06] tried this idea, fun fact, back in 2019,
[17:10] but I gave up because scrapers were
[17:11] terrible back then. Next idea, there's a
[17:14] website called Import Yeti where you can
[17:15] see where companies are importing all of
[17:18] their stuff from, like specifically what
[17:20] supplier, oftentimes what they're paying
[17:21] for it. But their subscription's kind of
[17:23] expensive and there's an Apify actor
[17:26] that will scrape all that Import Yeti
[17:28] stuff for you. So you build a wrapper on
[17:30] top of that to act similarly to Import
[17:32] Yeti, but just to point it to a
[17:33] different audience. You go there and
[17:35] type in Lululemon and see where all of
[17:38] those are coming from and in what
[17:39] containers they're in and how long it
[17:41] takes to get here. It's
[17:42] It's pretty cool. I didn't even plan it
[17:43] this way, but it seems like a lot of
[17:45] these ideas could be sold to Shopify
[17:46] store owners. And you can go to
[17:48] upwork.com and you can hire a virtual
[17:50] assistant that probably already has a
[17:52] spreadsheet of every Shopify store.
[17:55] Because a zillion people like me have
[17:58] asked virtual assistants to scrape
[18:00] Shopify stores. Builtwith.com, you can
[18:02] also buy every single of the 3 million
[18:05] or 4 million Shopify stores out there.
[18:06] So you get one scrape, one list of
[18:09] Shopify stores, you could sell 10, 20
[18:11] different things to them and just see
[18:12] which one resonates the most. Speaking
[18:14] of Upwork, why not an Upwork scraper?
[18:16] Their search tool is fine, but it's not
[18:19] specific enough for power users. So
[18:21] imagine if someone could just use
[18:22] natural language and type in something
[18:24] like find me Ukrainian mobile developers
[18:26] with 5 plus years experience and at
[18:28] least a 4.8 star rating and 50,000 in
[18:31] earnings. Boom. Upwork has all that
[18:33] data, export it to a spreadsheet and
[18:35] charge 100 bucks a month. Agencies,
[18:37] recruiters, startup founders, anyone who
[18:39] hires freelancers. This scraper already
[18:41] exists on Apify. You just need to build
[18:43] on top of it. I mean, what's stopping
[18:44] you guys from building one of these? Why
[18:46] not unbundle clay.com itself? Clay is a
[18:49] service where you upload a bunch of
[18:50] leads and it corrects wrong data or
[18:52] gives you more data about that lead,
[18:54] about that person, a LinkedIn URL, etc.
[18:57] Why not build a clay that does only one
[18:59] of those things really well, really
[19:01] specifically? Because I would bet you
[19:03] the vast majority of Clay's users are
[19:05] only using it for one of like 400
[19:07] different things that they offer, like
[19:09] validating emails or attaching a first
[19:11] name to just an email address so they
[19:14] can send more personalized emails. So
[19:16] your service could literally be we give
[19:18] you LinkedIn URLs or you give us emails,
[19:22] we give you LinkedIn URLs for that email
[19:24] address. And if you can spend some time
[19:26] learning paid ads, then you can test all
[19:29] of these ideas. Why not build five of
[19:30] these ideas? Test all five with 100
[19:33] bucks in ad spend each and then just run
[19:35] with the one that has the lowest cost
[19:37] per acquisition. Here's an idea I really
[19:39] love. The niche Zillow. The more niche
[19:42] Zillow. This is like so obvious it
[19:44] hurts. So Zillow exists for residential,
[19:47] right? But there's no Zillow equivalent
[19:49] for any other category of real estate.
[19:51] If you want to go entirely commercial,
[19:53] you've got LoopNet which is super
[19:55] expensive, only shows you a small
[19:57] percentage of search results, and it's
[19:59] kind of garbage to use. Then you have
[20:00] Crexi which is better to use but also
[20:02] expensive and their sales people will
[20:03] spam you to no end. So yes, we have a
[20:06] couple tools for commercial which is
[20:08] very broad, too broad. And then many
[20:10] tools for residential, Redfin, Realtor,
[20:12] Zillow, etc. But why not narrow that
[20:14] down? RV parks, mobile home parks,
[20:16] self-storage. The Zillow for
[20:18] self-storage, right? As user-friendly as
[20:20] Zillow. That's very important here.
[20:22] Don't undersell how important design is.
[20:25] Companies like Apple, Tesla, Airbnb
[20:27] would not be anywhere close to where
[20:29] they are today without their beautiful
[20:31] design. So yes, there are some websites
[20:32] where I can go shop for RV parks and for
[20:34] mobile home parks, but their design is
[20:36] garbage and they rely on individuals to
[20:39] find them and to list their RV parks.
[20:41] They're not pulling RV parks from Crexi
[20:44] and LoopNet through the MLS system. So
[20:47] they're marketplaces. Marketplaces are
[20:48] hard to launch and most don't succeed.
[20:51] So just scrape and tap in to other
[20:54] categories on Zillow and Crexi, unbundle
[20:56] them, and pull them over to another
[20:58] website. Zillow for laundromats, Zillow
[21:00] for car washes, ATM routes, the
[21:02] BizBuySell for vending routes, the
[21:04] BizBuySell for billboards or convenience
[21:06] stores. Most of these have no good
[21:08] marketplaces. What's stopping you from
[21:10] scraping BizBuySell for every single
[21:13] yogurt shop and putting them on a Zillow
[21:15] for yogurt shops? 3 years ago, it would
[21:17] have been probably pretty stupid to
[21:19] waste your time on something so random
[21:20] that may or may not work, but today,
[21:22] when you just got to prompt it a few
[21:24] times, why not throw a flyer at it? Or,
[21:26] go to BizBuySell, type in one state, you
[21:28] can do this, like Texas, scrape
[21:31] everything, okay, everything, and then a
[21:33] week later to the minute, scrape
[21:35] everything in Texas on BizBuySell again,
[21:37] and then subtract what sold, okay? Then
[21:40] do it again in a week, and do it again
[21:41] in a week, and within a month, you're
[21:43] going to know very quickly which
[21:45] categories are the hottest, which
[21:47] categories of businesses are selling the
[21:49] fastest, because BizBuySell doesn't
[21:51] offer this to you. You've got to kind of
[21:53] back end into it, if you will. Once you
[21:55] learn that, then unbundle BizBuySell
[21:58] with only the hottest categories,
[22:00] because if, say, restaurants don't sell
[22:02] well at all on BizBuySell, then you're
[22:04] probably going to waste your time
[22:05] building an unbundled version of
[22:07] BizBuySell for restaurants. Zillow for
[22:09] duplexes, Zillow for fourplexes, etc.
[22:12] Okay, bonus round. These are, uh, you
[22:14] know, Appify, Actor, unbundled {slash}
[22:17] white label {slash} wrapper ideas. It's
[22:18] all about the same thing. Shopify store
[22:20] scraper. I bet you a ton of the people
[22:22] that go to builtwith.com are just going
[22:24] there to scrape Shopify stores. And you
[22:25] can see in their drop down, it's one of
[22:27] the top listed things. So, they're
[22:28] making it easy for people to find. Build
[22:30] your builtwith.com, but for only Shopify
[22:32] stores. That needs to be a thing.
[22:34] Instead of 450 bucks a month, yes,
[22:36] that's what BuiltWith charges, you're
[22:37] $4.50 a month. What does it matter? You
[22:40] don't have any cost. Or $45 a month, or
[22:42] whatever. Or $5 per scrape in credits,
[22:44] and they can use it whenever. How about
[22:46] a TikTok hashtag trend scraper? Pick a
[22:48] niche, beauty, fitness, finance,
[22:50] whatever. Run a TikTok hashtag scraper
[22:52] every Monday morning, compile the top
[22:54] trending sounds, hashtags, creators,
[22:57] email it out as a weekly trend report,
[22:59] charge marketing agencies 50 bucks a
[23:01] month, 100 bucks a month, or whatever.
[23:02] You do a little work on Monday morning,
[23:04] and that's it. Yelp scraper, every
[23:06] roofer in Phoenix, every chiropractor in
[23:08] Atlanta, phone, email, owner name,
[23:10] review count, average rating, whatever.
[23:11] Sell it to the local service businesses
[23:13] that want to acquire competitors or
[23:14] suppliers who want to pitch them, yada
[23:16] yada yada. How about an App Store
[23:18] scraper or Google Play Store scraper,
[23:20] Trustpilot scraper, an Indeed scraper?
[23:23] Unbundle Indeed. Make an Indeed just for
[23:26] nurse practitioners, Crunchbase scraper
[23:28] for funding alerts, or a Reddit niche
[23:30] sentiment scraper. Brands want to know
[23:32] what people are saying about them on
[23:33] Reddit or on Twitter. Twitter just
[23:35] dropped their API cost, by the way. It
[23:36] used to be kind of unreachable for most
[23:38] people. Now, it's a lot cheaper. If
[23:40] you've been listening to this pod for a
[23:41] while, I'm sure that you've heard me say
[23:43] this, but I'm going to say it again. The
[23:45] efficacy of a growth hack is inversely
[23:47] correlated with its lifespan, okay?
[23:49] Translation is when something works, the
[23:51] window is short before everyone copies
[23:53] it. And the better it works, the shorter
[23:55] the time frame is, the shorter the
[23:56] window is. Cold email back in 2010 was
[23:59] insane. You could send 100 emails and
[24:01] get 15 replies. Today, you've got to
[24:03] send 10,000 emails to get 15 replies on
[24:05] a lot of different offers. Facebook ads
[24:08] in 2014, Google ads in 2005, you could
[24:10] buy a customer for two bucks. Today,
[24:12] it's 40. TikTok in 2020, like, you got
[24:15] to get it while the getting's good. And
[24:16] that's usually today or yesterday. I
[24:18] feel like we're in year two or three of
[24:20] like a 7-10 year time span. And 7-5
[24:24] years from now, there's going to be
[24:25] something else really hot, another
[24:27] growth hack, but it's not going to be
[24:28] this. Think of it this way. entire
[24:30] decade of the 2010s, every venture
[24:32] capitalist on the planet was saying the
[24:34] same thing. They would say it on every
[24:35] podcast, they'd write blogs about it,
[24:37] whatever. They would tell all the
[24:38] founders pitching them the same line.
[24:40] Don't just build a company that could
[24:42] just be a feature of another company.
[24:44] Like, that's what Snapchat did, right?
[24:45] Disappearing text messages and photos.
[24:47] Other companies copied them. Guess what?
[24:49] Snapchat's still around. They're
[24:50] publicly traded. They're doing great.
[24:52] And they were founded while everyone was
[24:53] saying this. But today, it's all been
[24:55] flipped on its head. We used to think
[24:57] that data was everything that these AI
[24:59] models were everything and then deep
[25:01] seek and other Chinese and non-Chinese
[25:03] companies came out and open-sourced AI
[25:06] models that were about 80% as good as
[25:08] OpenAI or Anthropic's. Now OpenAI is
[25:11] scrambling. They're trying to build
[25:12] hardware because if the model's not the
[25:14] moat, then maybe hardware will be the
[25:15] moat. It's all been flipped on its head.
[25:17] The phrase nothing is too niche has
[25:18] never been more true. A feature can be a
[25:20] company. So that advice is officially
[25:23] dead in 2026. Build lifestyle
[25:25] businesses. They are awesome. All these
[25:27] companies, Clay, Anthropic, OpenAI,
[25:30] Apify, they're trying to be the
[25:32] everything platform for everyone, which
[25:34] means that every single feature inside
[25:35] of their platform is just sitting there
[25:37] waiting to be unbundled, begging for
[25:39] someone to come along, pull it out, give
[25:41] it some love, wrap it in a clean user
[25:43] interface, and sell it to a very
[25:44] specific tribe of people. These can be
[25:47] $10,000 businesses, million-dollar
[25:48] businesses, and even billion-dollar
[25:50] businesses. If it could happen in the
[25:51] 2010s, they can happen today. Don't
[25:53] think you're competing with OpenAI if
[25:55] you build OpenAI for podiatrists, which
[25:58] would just be a ChatGPT wrapper for
[26:00] podiatrists. You're not competing with
[26:01] them. It's different. This is like the
[26:03] asymmetric bet of the century. Cost to
[26:06] start is almost zero. The tools are
[26:07] either free or cheap. Venture
[26:09] capitalists are funding these tech
[26:11] companies so they can offer us free
[26:12] trials. Use them and these big platforms
[26:14] can't and don't even want to move fast
[26:16] enough to fight you because who cares
[26:17] about us, right? We care about us. We're
[26:19] doing this. Everyone's winning. So
[26:21] here's my CTA. Here's my call to action
[26:23] for you. I want you to do three things
[26:25] this week. Number one, go find something
[26:27] to unbundle. I don't care where you get
[26:29] it from. Apify, Clay, Indeed, whatever.
[26:31] Don't overthink it. Just pick one. Then
[26:34] go to Replit or Claude or Codex,
[26:36] whatever, and build a simple website
[26:38] with a search bar that has an API call
[26:40] to an Apify actor or to a scraper or to
[26:43] anything. You don't even need to launch
[26:44] it. Just build it. See how far you can
[26:46] get in an afternoon or a weekend and I
[26:48] promise you're going to get further than
[26:49] you think. And if nothing ever comes
[26:51] from it, at least you've learned some
[26:54] awesome skills. And if you or someone
[26:55] you know has a really cool story about
[26:57] building something, either software,
[26:59] hardware, service business, product
[27:01] business, I don't care, that made decent
[27:03] money in a short amount of time, hit us
[27:04] up. Email Molly@cofounders.com
[27:07] or kevin@cofounders.com.
[27:09] Tell them your story and we'll have you
[27:10] on the pod. And if you want to sell or
[27:12] implement AI into small to medium-sized
[27:14] businesses, check out got over 200
[27:15] people in there doing exactly that.
[27:20] Having a lot of success. Nico didn't
[27:22] have rich parents, didn't have an MBA,
[27:23] didn't have any experience, and he just
[27:25] did it. His $3,500 MRR today eventually
[27:28] will be 5,000 and 10,000, maybe 20 or
[27:30] 30,000. Maybe he pivots and does
[27:32] something else. I don't know. But it's
[27:33] growing double digits a month and yours
[27:34] can, too. We'd love it if you hit
[27:36] subscribe, share it with a friend, and
[27:37] we'll see you next time on the Corner
[27:38] Office.