How Jesse Cunningham Uses Claude Code for Google SEO

0:00In this video, I'm going to teach you how to build an extremely powerful local business website using Claude Code. It's not hard to do. About two videos ago, we built this thing together. If we scroll down here, you see it's pretty solid and you come here and it has good call to actions. It looks professional and it took less than one hour to build this.

0:15I mean, we quite literally timed how long it would be. And it was based on this website, which is a real company, a public facing website, and we just gave it a nice facelift. And I can almost guarantee it would rank better with this structure and format. Now, if we come over here, someone commented on that video. They said with Claude Code, you can do much, much more.

0:32You can have competition analysis. You can pull their data also and use these keywords also to populate the website. So that's going to be the challenge. And so starting out, I asked Claude Code. I have a nice skill that I asked that can draw up some, you know, this kind of visual stuff. So it said local business site structure and where to make it stronger.

0:50So this is the current structure of the website that we are building. We have four hubs. We have the services hub, which is roof replacement repair. So this would be any service you offer. So any local business, whatever they offer, very important to have a services hub. And then we have service areas hub, which I think is hyper important. And then we have trust and company, the typical about YS reviews, and then the conversions, the call to action.

1:13So I asked Claude, basically, what improvements can we do? Now, before we go on, I want to explain to you what a hub is. So roof it in Austin, Texas, that is a service area hub. So basically with every local business, you want to show Google, you want to tell Google more or less, where do you do business? So if we have four different hubs, we're talking about this one right here, service areas hub.

1:34On this website, there's seven, Houston, Dallas, et cetera. And this is what it looks like. So we're going to click here roofing in San Antonio, Texas. Now there's something here a lot of people get confused on and they kind of freak out on it. And it's if I go to San Marcos, Texas, you're going to see the same templated format. Now, for a lot of websites, that's a problem.

1:53But for a local business, a templated format is welcome. You are welcome to do this because Google has been seeing this for decades, right? So this is how we do it. We put a little bit of spice in here, right? San Marcos sits at the heart of the Texas Hill County corridor, blah, blah, blah. We talk about the specificity of this area, but then as we move down, it's very, very particular on its H2s and P-text.

2:16The whole point of these four different hubs, right, which we're going to improve on. And we're going to do some automations based upon, remember this comment here, but the four hubs are just ways for us to understand what is the website made out of, right? We have services hub, what we do. We have area hub, where we do it. And then we have trust and conversions.

2:33Now, before we go much further, this is data for SEO. We're going to use this API so we can pull in real world data, right? So based upon the comment earlier, they wanted to say pull their data, also use these keywords to populate your website. So instead of kind of drawing in the dark, instead of knowing exactly what we should target, this will allow us to target real keywords that our competition is doing.

2:54So we're going to talk about it, but before we do the improvements that Qualled Code had upon this website, it said this is good, but it said, okay, fix the orphan pages, right? Give trust pages a real hub, mega menu navigation, resolve duplicate CTAs, local NAP on every Metro page and guard the money matrix. The money matrix is the service city pages need original local copy, internal links, or Google filters them as a thin doorway.

3:19We talked about that. I would propose, I would kind of push back because we do have some unique nature here. And quite frankly, I have websites that are very, what is the word, kind of copy pasta that rank really well in Google. So I'm not too concerned, especially with us having this right now. So what we're going to focus on, well, we're going to go to Qualled and we're going to tell it, let's come over here.

3:39We're going to say, do all of, well, actually, you know what? I can just talk to it. Do all of your recommendations and then give me a breakdown of what you did. And so because I have, if I come here, bypass permissions on, it's just going to do it because we're not going to be asked a million times. Can I do this?

3:58Can I do that? Qualled code will just do it because we have bypass permissions. And I have a cool little thing here. I can press a couple of buttons and talk to the darn thing instead of typing all the time nowadays. Now here's a cool little hack. I'm going to scroll in. Do you see where I say BTW? How do I give you API access to data for SEO?

4:13You can see above it, it's still working on the request we gave it earlier. You can see now replace the header, blah, blah, blah. So basically what you can do with the slash BTW is you gave it something to do, right? And if you have a small little question, just do slash BTW and ask it. And it can do both things at once, right?

4:29You can do a new instance of Qualled code, but this is kind of a faster way to do really small side projects. So I'm pretty happy with Qualled code, what it just did. It has a company hub that ties everything together. I don't often see this and it's kind of a light bulb moment for me. It just links to our about us, why hold it, duh, duh, duh, duh, duh, duh, duh, right?

4:46And then service areas has an all service areas page right here in the dropdown as does services. Very, very nice. And this was all based upon improvements, fixing the orphan pages, et cetera. Now what we're doing, let's come back here. I gave it the API credentials and it says, okay, we have the account balance. It's 49 bucks. We are ready to go.

5:06Now this is where it can get fun and this is based upon the comments. So with Claude code, you can do much, much more. You have competition analysis, you can pull data and also use keywords to populate the website. Well, how we do this is right here, right? Very, very important. We have AI optimizations, which is kind of expensive. We have SERP, which is the search results, keyword data, domain analytics, and so on and so forth.

5:28So the question, it's not really a question anymore of, you know, can you do it? It's a question of having, you know, the right idea, like what should we do? Because we have at our fingertips all of this data, we could feed it the Claude code and we could kick back anything. We could do anything we want, basically. So it's an imagination issue.

5:47So what I'm going to do, I'm going to ask Claude code its opinion. What do you think the lowest hanging fruit will be? So let's come over here. I'm going to say this right here. Let's control tab. Let's come over here. I'm going to ask it, based upon this YouTube comment, what do you think the best thing to use for data for SEO API would be?

6:04And then it's going to print that out. And then we're going to paste it here and let's see what it says. So Claude came back with a lot of stuff and too much stuff to read right now. And I don't like reading things if I don't have to. I'd like to visualize them because that's how I learned. So I have a Claude skill.

6:20We'll talk about that in a second. But what I'm going to say is, okay, put all that into a nice diagram so I can understand it. And what this skill is going to do, right, I'm going to hit enter. We'll have to exit here. What this skill is going to do, it's going to, it's going to work up a diagram of what it proposes automatically.

6:34Okay. There we go. Pretty cool, right? I mean, I didn't have to do anything except just request this. I'll get into how we do that in a second. So best uses for the data for SEO API. For a local Legion site, use the data to decide what to build, then write original local content itself. Makes a lot of sense. We always want to use data if we can, but it's usually very expensive to do so, right?

6:54This hopefully will give us a cheap way to do it outside of like SEMrush and Ahrefs. All right. Which API is ranked for local? It has labs, API, the workhorse, cheap. I like cheap. Keyword ideas. The ranked keywords and the keyword gap. Pennies per call. Pennies per call. And if we have what? 50 bucks in here already. Pennies per call is perfect.

7:14Business data. Then we have SERP API live. So use surgically. Who actually ranks page one in each city on page API, one time tech audit, and then backlinks API later. Yeah, that should be later. All right. Let's scroll out a bit. Let's see what else we have. So the research and pass build a list, a build list rather, pool volumes. So six services and seven metros plus top suburbs.

7:37Real Google search demand for every combo. So that's going to allow us to know how many people search for these specific keywords. This is very good for identifying niches and how you're going to build your website out. Drop the dead combos. No demand equal. Don't build it. This is what guards the money matrix. So if you're out in the sticks in America somewhere and your niche is something odd and the town is very small, maybe you shouldn't build a page around it.

8:01Maybe that would just be fat. That would be better left off of the website, right? Keyword gap versus competitors. I like it. Local pack snapshot and output, a prioritized build list. I mean, it says a $50 balance equals a thorough research pass with room to spare. Labs is cheap. Ration live serve. I'm curious how much money it's going to cost. I highly doubt it's going to be much at all.

8:22The core move, 42 service slash city combos, great filter by real search demand, and then build only the winners. I like it. I like it a lot. There's a lot more we can do, but this is like the first pass to build a website. Basically what it's saying is where is the demand? Where do people actually type for your product or your service and then it's build around that.

8:43Now I can't help. There's a little bit more to have here, right? That's very, very basic in my opinion. I think we can do a bit more. So what we want is an automated approach where every week or every month or every day, depending upon the process, we look at the SERPs, we decide what's working, what's not working. We hook into Google search console and then we revise the website automatically.

9:02I never want to have a human in the loop. I just want it all automatic. All right. Let's see what that's going to do. Boom. I love just being able to type and rather just talk into the machine. It's very, very fast. So the idea is if we're going to build an automated website, like an actually automated website, we need automations. So we're going to have to probably get a VPS, right?

9:23We're probably going to have to have a virtual machine running this and it knows the time of day and knows when to loop through these things, but I'm very curious what Claude is going to say. Okay. Claude says it's totally buildable. It says we already have most of it. We have the GitHub repo. We have a stage in sight. We have this, that, and another.

9:40Okay, cool. What is the loop? The loop is observed, diagnosed, act, verify, learn. I like it. Google search console API, your actual performance. We diagnose. It seems very, very doable. No human in the loop is a fine goal, but the safe version doesn't mean nothing to watch as it changes. It means you replace the human reviewer with automatic guardrails and auto rollback.

10:01It depends. It depends on your appetite for risk. I guess you could say, it depends if this is a microsite, right? Just a rank and rent website. Depends if this is a client's website. So the guardrails are the autonomy concretely, stage and first, validation, a change budget, every change committed to Git, da, da, da, to design the right version for you. Two decisions actually shape the whole build.

10:24So it says, how should I automate changes? The live site, stage and first, then auto promote, direct to production or direct to production, no rollback. So we have a lot of options here. I'm curious your thoughts. If you had a website, if you had a local business or a microsite, how would you like the automation to truly play out? But as I'm thinking about this, right, we've already done all the improvements that it reckoned we should do.

10:45And it said, look, we can grab data. We can grab data with these APIs and I can tell you what pages we should build based upon data, based upon volume. So I do think it's worth the effort, right? The second iteration, the one that takes a bit more time, let's talk about it in the comments. Tell me what you want to see.

11:01But what I would really love on this website is to have a custom portal, right? A custom portal where it draws in the data from data for SEO and I can log in and see, all right, well, here's the volume for this. Here's the volume for that. Let's build this, that, another. I want a graphical user face. I want something to look at, something that's pretty.

11:18I don't just want data. So I'm going to ask it to build it right here. Now, as it's building, because I already requested it, come over here to the RankExpand Academy. You'll see the Microsite Playbook. I think it's worth the effort to build out more service location pages like right here because on our microsites, we have upwards of 30 plus different service location pages on a local business.

11:41In other words, if I have this roofing company website and I only have seven, right, seven right here, in reality, we often run many more than that and it's very cool to tie the idea of grabbing data for which location pages we should run. Check it out. This is what it gave us. I'm pretty stoked on this. So what does the demand say?

11:59So for how much was this, 7.5 cents? Do you see this? This is how much it costs. This is why it's ridiculous, right? Absolutely amazing. So instead of, let's just pause here for a second. Instead of paying for SEMrush and Ahrefs, I hate to say it. Why not just get your API yourself and do these micro blasts of keyword data that you need, right?

12:19So we have 378 keywords, build a list, one API call and then it figured out 57,730 monthly, I think volume, right, searches across 42 Texas cities. So here we go. It had different phrasins. That's why we had that many keywords. So 42 cities times nine phrasins equals that, Google search volume and what was it, right? Build Q tier A, Houston roof repair, San Antonio roof repair, nice graphical interface that I can see.

12:50It's not on the website itself, but I'm going to settle for this on this video. Tier breakdown A, build now B and then C, skip these combos. So all of a sudden, we know exactly what we should build. Let's see what this says. The headline finding, roof repair is the demand king. It's bigger than roof replacement in every single market. Now we're going to come down here and ask it, build it out, do it as best as you can based upon the data, right?

13:12I'm going to hit enter. We're going to let that cook. Now, I know we talk about a lot of different things in this video when I was moving fast. If you want to go deeper or if you have questions about it, just join the Rank Expand Academy. I'll show you. Let's just control if I can find it, right? Over here, you'll see that we have tons of different modules start here past 2000.

13:30There's two types of people that join this. There's the builders who just enjoy building and they want to build a valuable asset. And those who want to grind it out and make money off of Upwork and other things through their AI SEO skills. Regardless, there's a seven-day free trial. Come check it out, right? What do you have to lose?