How to Make Money with Claude Code (Full Breakdown)

We're going to go ahead and bring on our next speaker, Johan. And uh yeah, how do you want to do it? It's up to you. >> Uh I can start on the laptop and then I'll just walk around. >> Yeah, sure. Here you go. >> Yes. Uh thank you everyone for coming and big thanks to Ian Kernis as well as Anthropic for hosting this event. I made some slides and I use cloud code for the entire thing. I just told it make me some slides, but I wanted to have something to show. But I'll be talking

about it. Use the laser pointer so I can uh look professional. >> Ah, yes. >> I feel my IQ just went up.

about it. Use the laser pointer so I can uh look professional. >> Ah, yes. >> I feel my IQ just went up.

Okay. Well, a little bit about me. My name is Johan Si Nathan. I've made over six figures of revenue with Cloud Code. I built and sold a SAS company in 12 months. And now I'm using the same playbook to grow three different SAS companies. The main thing I focus on is how can I make revenue with cloud code because building cool products is amazing. But I wasted a lot of money. I've built over like 19 20 products and maybe only four of them made money. So now my main focus is ship a product as

fast as possible and then use some of these playbooks and stuff that I'm going to share with you on how to actually make money with cloud code. So once you make a product, a big thing is how do you get traffic to the product? Because if you have an amazing product, but nobody knows you exist, you're never going to make money. People have to know you exist. There's so many products that are just like living in a graveyard because they are amazing products, but no one ever discovers them. So, they

make zero dollars. But there's some horrible products that have amazing distribution that people see them, they're doing all the marketing, and they are making killing crazy money. So, how do you get traffic to your website? That is the first goal. You made an amazing product. What do you do next? How do you get traffic to your website? And there's a lot of stuff that used to take a long time, many, many months. The three things I'm going to focus on are SEO. You can pay SEO agencies like tens

of thousands of dollars, get scammed, or you could do it yourself. You can hire freelancers, but I like using cloud code, and that's what we want to talk about. So, the first one is free tools. The general idea about this is let's say you have like a email copywriting tool or an email outreach tool. I'll give an example of one of the tools that I have, which is a Twitter cold DM tool. It's one of the tools I'm going right now. It is made for Twitter. So, some of the

free tools that we're making are like Twitter bio generators, tweet generators, thread generators. And the idea is that your target audience is searching for those tools. They can discover you via SEO. And once they find you, you can have tons of pop-ups, lead magnet to get their emails, or even just having your branding. I think the general rule is like it takes seven different touch points with your customer until they trust you. So them seeing your free tool, discovering you

from SEO, seeing your brand, that just builds more trust and helps you eventually convert with them. So free tool the workflow for this I I abuse cloud code. I use it extremely like I don't want to do any work if I don't have to. You want to work smarter not harder. So free tools you give in your product to cloud code say this is my product. This is what it does. If you are in your directory, you can literally say like un read my codebase, understand all my features, understand my ICP. You

can also go to Claude and like try to get a better understanding of your full ideal customer profile and have it brainstorm 10 15 different free tools that you can make and then from there you just want to focus on three free tools and have it generate the free tools. These things claude can generate in like 10 minutes. So I hope that some people go home today if you have a product generate some free tools because you can make maybe three free tools this afternoon and then after that you can go

into ah refs which has amazing like metatag description updaters like it improves your metatags. So you make your free tool go to ah refs have it optimize your meta tag your metad description that way you can help rank faster. So that's a little bit about free tools programmatic SEO. So this is the big thing this laser pointer is so addicting. Um, the big thing with programmatic SEO is how can you make a large amount of pages? Here it says a thousand thousands of pages, but it can

also be hundreds of pages, tens tens of pages, whatever it is. And my favorite thing for programmatic SEO is competitors. So the workflow for this is you want to steal your competitor's traffic. Chances are if you're starting out, there is a company that has been around longer than you. They're getting tons of organic traffic from Google and they are making money from it. So steal their traffic. They've done all the hard work. You just want to take money from

them. And the idea about this is let's say for example chatbt and claude let's say claude just came out yesterday chatbt has been dominating the market for a long time chatbt is getting a lot of traffic from chatbt but there are people who search chatbt and then they'll search up chatbt alternatives you want to rank for that because that means that someone is actively looking for an alternative so how you can do this with claude is you have to do some manual work and find your competitors

them. And the idea about this is let's say for example chatbt and claude let's say claude just came out yesterday chatbt has been dominating the market for a long time chatbt is getting a lot of traffic from chatbt but there are people who search chatbt and then they'll search up chatbt alternatives you want to rank for that because that means that someone is actively looking for an alternative so how you can do this with claude is you have to do some manual work and find your competitors

find maybe four or five competitors the general idea with programmatic SEO is that you only have two files. One file that has all your variables and all the data. So you want Claude to essentially say, "Here are my five competitors. Make a variables file that has all five of these competitors, their websites, their links, their pros and their cons." And then obviously if you're doing a competitor page, you're not going to make yourself look bad. You want to make

yourself look good, right? So you want to do competitor alternatives and then competitor versus you and have that variable file that really just makes your product look better. That way if people search competitors to your product, they come, they see your page, and you can hopefully convert convert them there. Have a lead magnet pop up so you can get their email or even if it's just a page lally explaining why you're better than the competitor. Chances are they're going to end up signing up

hopefully, right? And then targeted blog posts, high intent keyword coverage. So this is something that you can use claude or claude code for, but the general idea is there's different parts of the customer journey, but when people are exactly ready to buy, you want to target keywords for that. A good example of this is someone searching like best AI chat tools, right? That means they are ready. They are like they're not just searching about the problem. They

hopefully, right? And then targeted blog posts, high intent keyword coverage. So this is something that you can use claude or claude code for, but the general idea is there's different parts of the customer journey, but when people are exactly ready to buy, you want to target keywords for that. A good example of this is someone searching like best AI chat tools, right? That means they are ready. They are like they're not just searching about the problem. They

know exactly what they need. A better example would be like cold DM tools. So again, I have that Twitter cold DM tool. If someone is searching up how to automate Twitter cold DMs, they are not like thinking about, oh, how do I get more sales? They actively know, I want to do Twitter cold DMs, how do I do it? They are ready to buy. So, you want to make targeted blog post for this. And the best way to do this is get maybe 20 different high intent keywords, feed it

into cloud code, and you can make a skill for this to really have like SEO optimized content as opposed to just AI slop and have like 10 different very high intent blogs that just target customers who are ready to buy. So this is like the first step. How do you get traffic to your website? You have an amazing product. These things can help you get traffic. And SEO usually takes months, but one cool little trick with the free tools, again, this might take some weeks or months to really get

traffic with SEO, you can buy an age domain. There's a SEO expert in this room who told me you can buy an age domain and help you get traffic a little faster that way. But with free tools, another thing is just make a demo video, post it on Twitter, post on Instagram, post in different directories like Product Hunt to help get traffic today. Like if you make a free tool today, you post in these directories, you post on Twitter, you will get some traffic. It might not be crazy, but you'll get

traffic, which is the goal with this. So, the second thing, traffic into signups. Imagine you get a million or a million website visitor website visitors to your product, but your design is horrible, your copy is horrible. No one's going to convert. So, you just wasted all that time getting traffic and now you left tons of money on the table, which is going to absolutely suck. So, the next thing is how do you get traffic? Yeah, traffic is useless if the page doesn't convert. Cloud code giving

me amazing copy. I didn't even write that. Um, but how do you turn visitors into registered users? One thing that Cloud Code released like two weeks ago was the front-end design skill before. If you design stuff with AI, it looked horrible. Like people I I made the mistake of like designing one of my products with AI. I posted on Twitter and I got absolutely roasted. They were like, "This is generated by AI. Take this down right now." Tons of design agencies were like, "We will redesign

this for you." But with the front-end design skill, you can design very, very high converting landing pages that actually look good. They don't look like AI slop. So check out Cloud Code's front end design skill. You make way way nicer design. That way once people come to your website, you give off that like luxurious branding, premium branding. They see it and they'll convert. Another thing is cloud code generating variants. And a big focus that I want to talk

about is Post Hog MCP. So Post Hog is an amazing like monitoring analytics tool. It does a lot of crazy stuff. And they have an MCP that can connect directly to cloud code. And once you set that up, abuse it. There is so much that you can really optimize and that's what I want to talk about. So AB tests for websites. If you're not running AB tests, you're leaving money on the table. Once someone gets to your website, you're rolling the dice if they actually like the copy, if

they actually understand it. So what I typically do is like very very opposite AB tests. One will be explaining the problem one way, one will be explaining it the exact opposite way, just to see what people like more, what converts better. And I used to use chatbt come up with variants but now you can do it all with claudin post hog mcp. So you can literally say create ab test for this landing page and it'll do the entire thing for you. Obviously you have to

connect postto hog mcp which is pretty simple. Should be pretty straightforward and cloud code will write the copy for you. You can explain how you want to run those AB tests. So if you want them to be like very small changes or very drastic changes to really test what converts better for your customers. And the whole goal is that you design very fast with the front end design front end design plugin front end design skill test as much as possible right so run

tons of AB tests with the copy and then just keep repeating like this whole process I don't know 2 years ago 10 years ago when I was a kid would have taken very long but now you can do it like you can run tests every single week and keep optimizing your conversion rate that way if you get a million visitors the first week you get 10,000 signups the second week you get 15,000 you keep optimizing and optimizing that way you're the whole the whole point of this is growth, right?

Step three, there's way more steps than that, but this is just like the rundown. From signups to revenue. So again, another problem. Imagine you get a million people signing up to your product, but no one pays you. You just waste the time again. So another thing is how do you guide the customer from signing up to the point where they're pulling out their credit card and they're throwing it at you? You got to be ready to catch it. So post hog mcp again you can use cloud code and tell it

to create a funnel with post hogmcp and what this will do is create post or cloud code will understand your product and know the entire customer journey. So from user signup what are the steps they need to take to end up pulling out their credit card. Again I'm going to give the example of my Twitter coldm tool that I have for us. People have to connect their account they have to scrape leads and then they have to start a cold outreach campaign. If people do not

connect their account, they're not going to do the rest of the stuff. And if they don't do the rest of the stuff, they're not going to pull out their credit card and pay. I could manually make a funnel with Post Hog, which would take maybe 10, 20 minutes. Or I could tell Claude code, look at my codebase, understand it, and map out the customer journey and create the Post Hog funnel. I thought it was those four steps. When I looked at the funnel that Post Hog made, it was

like 10 other steps. It gives you a lot more in-depth funnel, and you can understand the drop off and every single part of the funnel. So once you have the funnel, obviously run it for like 10 days, 15 days, collect as much data as you can. And then you can go back to cloud code and have it analyze the funnel and see where are the biggest drop off points of the funnel. So for us, the big thing was accounts not being connected. So that's the first step. After people sign up, they weren't

connecting their accounts. If they don't connect their accounts again, they don't do the rest of the stuff. So once we had that data, seeing that they were not connecting their accounts, you can have cloud code optimize that. How do you convince people or like push them across that first step to connect your accounts? That way you can get them further along the journey. And that's where Claude's going to help you improve as well as onboarding emails. So you can

have Cloud Code give you suggestions saying, "Okay, look at this data right here on the first step. People are not connecting their accounts. What do we do to help improve it?" And Cloud Code's going to brainstorm a lot of stuff. You don't want to like raw dog it and execute first. Like have it planned for you and brainstorm different improvements that can have. And a big thing that I left money on the table when I sold my first thing was onboarding emails. Proper onboarding

emails and like life cycle emails to help just guide the customer. If they sign up, keep emailing them. Um maybe not the best example, but until they buy or die or unsubscribe, right? Like just keep emailing them. That way you can get them to subscribe, reactivate, and cloud code can help with that whole thing, right? So analyze the funnel, see where people are dropping off, and then improve each step. And if there is something where it can't really improve that much, add an email. So like

for us, people who aren't connecting their accounts, there's only so much we can really do. We can optimize the UI. We can optimize if people don't really understand it, we send them an email after I think it's like 24 hours with a video saying, "Okay, here's how you connect your account." That way we can guide them to the next step, to the next step, to the next step. And this is an ongoing thing. Again, just like on the last slide where it said design, iterate, test, whatever it was, it's an

ongoing thing where you just want to keep every once a week, bi-weekly, have cloud code analyze that funnel, see any drop off points or any things that can be improved. Have it brainstorm improvement stuff, whether it's fixes that it can make itself or you might have to implement if it's onboarding emails. less friction, higher activation, more revenue because again if you have a million people signing up, you want a million people ideally you want a million people going through the

entire funnel and pulling out their card. There is going to be drop off points. So you want to understand where those drop off points are, how you can optimize it and cloud code with post hog MCP makes it a lot easier. Build once automate exit ready. So right now I'm going three different SAS products and amazing thing is that cloud code automates the business not the code. Let's say five years ago for me to run three different SAS products. Realistically, I would not have the time

to do it. I would have to hire developers. I I do work a lot with offshore teams, so it would be a lot cheaper. But you're still looking at a pretty decent monthly expenses just to have these three SAS products running. But with cloud code, you can focus on automating the whole business, not just the code. You can have cloud code understand your documentation and just see which parts to optimize. So like auto autogenerated docs and SOPs. So some things that aren't even necessarily

related to the software you can have it fix. For example, customer support that's not even related to the software, but if you do have something like intercom, I use bento now for all my stuff. You can have cloud code write really good SOPs and add it directly into the codebase because the more organized your business is, the higher multiple you're going to get when you sell. You do not want to be like the business does not want to be relying on you. You want to be able to go to a

buyer and say, "Look, I have this business that can run entirely without me. As soon as you step in, like you can generate money. You don't even have to run it." That's what I did with the last one where I was not even in it for like the last two months when I sold it. And it was very, very incentivizing for the buyer because they were able to buy it and just hold a cash flowing asset where they're just making money doing whatever they want, living on the beach, I don't

know. But the end result, calm, profitable, sellable anytime. So, you can build something with cloud code, get traffic to it, optimize it, and then really just spend time spend time optimizing everything where you don't have to put as much time into it. It's entirely automated so you can keep building calm, profitable, sellable any time. That's that's one of my models. Calm, profitable portfolio of businesses. And that's what I've been trying to do with George Lobo, which is

know. But the end result, calm, profitable, sellable anytime. So, you can build something with cloud code, get traffic to it, optimize it, and then really just spend time spend time optimizing everything where you don't have to put as much time into it. It's entirely automated so you can keep building calm, profitable, sellable any time. That's that's one of my models. Calm, profitable portfolio of businesses. And that's what I've been trying to do with George Lobo, which is

right here. And yeah, that was everything I have to share. My name is Johan Nathan. You can check out some stuff at yan.fyi. If you click these, nothing's gonna happen because you can't click them. But you can go to the website and get my socials. I've been making vlogs as well with Ian. So on YouTube, I post some different vlogs, which is just sharing the journey of building a portfolio of startups in Chiang Mai. Hopefully going to sell another one within the next 12

right here. And yeah, that was everything I have to share. My name is Johan Nathan. You can check out some stuff at yan.fyi. If you click these, nothing's gonna happen because you can't click them. But you can go to the website and get my socials. I've been making vlogs as well with Ian. So on YouTube, I post some different vlogs, which is just sharing the journey of building a portfolio of startups in Chiang Mai. Hopefully going to sell another one within the next 12

to 15 months. Thank you guys for coming. I appreciate your time. Hopefully it was good. See you. [applause]

to 15 months. Thank you guys for coming. I appreciate your time. Hopefully it was good. See you. [applause]