Jesse Cunningham’s Microsite Blueprint for Local Markets

0:00In this video we're going to build and revise a microsite. Okay, we need a new logo off the bat. What are we doing? We're going to have this. We have a Claude code instance that we're going to use that basically has a direct connection to our website. When you have a direct connection between Claude or Codex to your website, it makes your life easy. All you have to do is talk to the machine. What we did initially is get this right here in the classroom in the Rank Expander Academy. This file right here, I'll zoom in as a zip file,

0:26which is the brains of the operation, which tells the LLM how to build this thing very, very specifically, right? We're going to go over this, we're going to go over that. All right, let's keep it rolling. Claude is working right now, we'll come back to it. We have a lot to talk about, but I I want to build with you. So, the anatomy of a microsite is a machine. It's not a pile of pages. Often times I see people make these websites, right? A local SEO website, and it's just like a pile of pages with a blog and this and that and another. We don't

0:52have blog posts on our websites for a reason, right? So, we have one skeleton, infinite markets. So, through the blueprint, we have a fixed structure. It's reproducible and it's fast to clone. That's very good because this concept only really works if you scale it. I don't want one website, I don't want 10, I want hundreds of websites. We'll talk about it. So, every microsite is three layers. We can probably come back to this. Let's go to Claude and see what it's saying. All right, it's downloaded final selections at high resolution. I asked it to use its vision. So, I asked

1:25it to do a new logo, change the website, and then go back and use its vision because Claude can actually take screenshots and see if it makes sense, like a sanity check. Now, you can see here, I also asked it to go out to free stock imagery, and the key here is use your vision. So, it's going through here and it's saying, "Ooh, this logo worked, or rather image worked, this one didn't." You always want to have it have, if you can, two agents in parallel. One does the work, one checks the other agent, right? Very easy to do nowadays. But, let's keep it rolling.

1:54Let's come over here. All right, so every microsite is three different layers. So you have eight about eight core pages, right? You have home, about, contact, estimate, all those things. Most local businesses have those core. And then this is where people screw up, right? The whole reason our websites rank so well is what I'm describing to you right here. And our websites need to rank well in order to rank in Google, in order to get leads, in order to make money, right? It all starts here. So we're going to have one to six service pages. This is about what we do. There's

2:25a lot of questions on this, right? You can make a microsite that's focused onto one particular service, but you can also do one that's like a little higher, broader niche, and then get sub niches under it, and you can rank for more words. There's a balance here. And then we're going to have 20 to 30 location pages, so like where you do it. So one per neighborhood or city or suburb, because we always want to give Google two signals, what we do and where we do it. I can see that Claude just made the changes instead of having SVGs we have

2:55these stock images. Let's come down. There should be another one. I This is the first time I'm looking at it. Okay, and it also did, I guess, alt text under it. That can work for me. And then here's another one. It doesn't look bad. It looks much better than it did. And then we have the logo up here. It actually It actually edited the logo, which is crazy. You can give it a PNG and it can chop the PNG, change it. Very interesting. But I want to show you our services. This is what I was describing with services. So this website has four.

3:19It has sewer line replacement, trenchless replacement, sewer line repair, da da da, right? And then we have the service areas, which kind of sprawls out into different areas. So we're always going to affix two things, the service and the location. Funny thing is, I refreshed the page and it's actually still working, so the Vision Agent is not done. I'm curious if it's going to change anything else. But let's keep it rolling. Let's come back over here. All right, so the core pages make you a real business. It's all about trust signals. It's all about conversion pages and support pages, right? So

3:48that's layer number one. Layer number two is the money. So one page per service and a hard cap. I don't think it's wise to build one of these microsites with say 20 different services. The the more distilled you can make a website, the more potent it is, the more Google knows exactly what the website is all about because sometimes Google gets confused. You'll see this in your ranking volatility when you first build one. And then location pages blanket the map. So, if we have one to six services times 20 to 30 locations, it's a huge local coverage. Okay, so I think Claude

4:21is wrapping up giving us what it did and all these things and it does quite a lot of work. I'm using Opus 5 right now with bypass permissions on. Very cool. It says worth flagging those diagrams explain pipe burst and blah blah blah were more precise than the stock photos. And it's crazy that it knows all these things because it has the vision. Very very strong. Opus is strong, Fables even stronger, but I like Opus for these type of things cuz it's more affordable. Now, why did it take? It took like 18 minutes to do this change. Now, the reason for

4:49that is because I have like a breadcrumb concept. Like the fable of the breadcrumbs in the woods, that type of thing so you can see your way out. If you don't have version control on Claude, what happens is the contextual window, the tokens, it gets too big. And then the the AI forgets what it was doing. It's like, I don't remember the rules for this. So, for every change, everything we do in Claude, we need to have a system that allows it to to cross-reference and kind of log away every change that happens. Now, this is old school. If you've been around,

5:18you've seen this before. The hub and spoke, this is classic, right? And it works for local SEO, too. So, the homepage is the hub. The homepage is the strongest thing you have going on. Services and locations cross-link and you can pause this page, right? If you want to check it out. Internal links are the top factor. The cool part is, you know, that might back in the day it was a pain in the butt. It was terribly complicated to do that well, but now because we have just a zip file, more or less, it's like a it's like a stamp.

5:43Every website gets it appropriate. That's how we're able to scale. Now, this is interesting because back in the day, we rank and rent was typically not typically, but a lot of guys were running and girls were running single page rank and rent websites. I understand why because it was a pain to build websites back in the day. Now, right, a whole site beats a single page.

6:05That's obvious, right? Very, very obvious. The reason we can do it is with AI. So, AI can write the frame, a human can add the truth and spice it up and the structure. And we're going to get into frequent ask questions here in a second. The structure is the ranking factors. So, service pages hit dedicated page per service. Location pages hit geographical relevance. So, those two right there is like top dog, right? That is so important to rank on Google or Bing or AI. The hub and spoke hits internal linking. So, if you start to push backlinks, which I mean, I'll show

6:36you if I come over here, I come back over here, and I come to the classroom. We have a module, it's new, where I gave my whole Rolodex of backlinks right here. I'll click on it. So, there's DR 30, 40, 50, 70 backlinks. It's 800 plus vetted sites that I have used personally. But maybe that's too in-depth for this video. But hub and spoke is internally linked correctly, which means that, you know, Google understands relationships of pages, and if you ever buy backlinks, it's going to send the juice throughout the whole website. And then the type scope hits the topical focus. So, we're

7:09not going to do a 200 page website. There's no point in that. Let's keep it somewhat small for topical focus. And then 40 real pages hits content depth without tipping it into junk. Google right now is is like on this mission to to go after scaled content abuse, which is not this. These are very small websites in comparison to what many people are doing in different concepts outside of local SEO. The next part has a lot of value here, I think, but I want to change this page. This is the service areas page. So, talking about hub and spoke, this is the service areas page

7:43and then the specific areas are underneath of it, right? This is an important page because it it acts as like the connector for everything and I think the page kind of is boring right now. How would I describe this to Claude? So, I'm going to take this HTML, right? I'm going to take the HTML, I'm going to come over to Claude and say, "This page is boring." Right? And then let's just see what it does with it.

8:03This is the crazy part. If we give Claude something very abstract like the idea of being boring. Like this page is boring. What does that mean? That's a very subjective comment. I'm curious how Claude is going to handle it. Here's what it said. It said, "Fair. Three near identical gray link lists in a row." Let me look at what's actually on the page.

8:24So, it's going to audit the service areas page structure first. Okay. We have the hero. So, it already it already has the structure. Remember we're using the blueprint, the zip file. It already knows what the structure should be. And here it is. And coming further down, the problem is it's a bare directory. 26 identical tiles, 484 words, no information. I already have a pretty a real per city data sitting in the generator. It's just not surfaced. Let me put it to work. So, all of a sudden, we gave it like a weird abstract, this this page sucks or this page is boring

8:55and Claude goes, it investigates and it's going to change it. Okay, so if you want to make money with this though, we have common questions. A lot of people come into the community. By the way, the Rank Expander Academy, it's a 7-day free trial. So, come check it out. I spend a lot of time in there. All right, so question number one. There's so there's no search volume or a map pack or too much competition. So, low search volume.

9:18Like so, what we teach is if you want to do this, you find a niche first. You have to focus on the right niche. If you're going to go after like general electricians services, it's not going to work. Go after general roofing services, it's not going to work. It's too much competition. We're not playing that game. But so, what we're doing is finding low competition things. So, low search volume is fine. So, even even if the tools that we use have zero search volume, we call them zero search volume terms, that's okay if you use your common sense and you're like, "No,

9:45there's definitely search volume here." Because remember, everyone out there is using probably the same tools you are using to see, "Oh, is there search volume here?" Right? The people who make the most money are the creative ones. Right? A map pack is not a wall. It's okay if there's a map pack. I just get concerned when there's like LSAs and Google Ads and map pack and an ad in the map pack and then organic and and even worse off you throw in AI overviews, that's a problem. And then competition, you have to remember this, means there's money. But we just need we need the

10:14right mix of like enough competition because it's volume based but not too much competition. Claude is still working, so we should probably just keep going with this. So, the next question, a common question, is one site per area or one big site with lots of location pages. So, should you have one site per area or like a huge site? Let's just go after many services. We're going to go after local sites that are tight. We're not doing one regional giant website.

10:40What we are going to start doing though is directories, which is a whole different conversation. So, we want tight websites, right? Do I need schema? Do I need photo location tags? Like in in all the technical stuff? This is what I love. 80% is is structure. 80% of this is the structure of a website. You can go read studies on this stuff. Look at I think it was What was it? WhiteSpark recently had a a study on what experts think matter in organic search. The structure of the website and what the copy says on the website matters more than most things,

11:14right? Internal links, all of it. The 20% is the nice to have. Schema, photo location tags, meta tweaks, but right? How hard is it to have Claude just do those things too? Not hard at all, right? And what do I charge? So, if you have a website, one of these micro sites, that is ranking, pulling in leads, what do I charge? So, you can charge for the value and not for the cost. Like, if you own one of these, you own it. It's your property. You can charge whatever you want for it, and value determines the cost. I can tell

11:42you the three basic ways and then one extra. You can charge per lead flat fee. You can charge like a bundle of microsites. Say, I'll build 10 of them for you for 1,500 bucks a month. You can charge per um per job at a revenue basis. And then the fourth one, the bonus is if you can qualify the lead, and I've seen this once or twice, and it's harder to do, but if you get leads into your website and you qualify them yourself and you run the leads, you can charge a lot of All just sent this to me. This is what the revised service

12:10areas page looks like. I think there's pros and cons, and when you do an abstract thing, remember I just said like this page is boring, do something about it. It kind of went off in left field a little bit. It didn't adhere to what I would like in some regards. Can you spot what I don't like? Right, I'll just show you. Let's just scroll down here. Dallas neighborhoods looks nice.

12:28We have proper internal links. Look, it says sewer replacement in Dallas. If I click here, it will go to that page. That's what I want. That's good. I want the hyperlink. I want the the anchor text to be very specific, which it is. That's good. I like all of this. It looks much better than what it was, but I don't like the one thing that sticks out to me is the H1. Like that should be like not there. 26 areas and what is buried under each one. Probably not the best, but everything else it crushed on.

12:55So, whenever you give Claude an instruction that's abstract, you should expect that it's going to go out into left field and do something. Sometimes it's good and sometimes it's not so good, but I'm fairly happy with this. And the whole point of me showing this to you and how I'm going back and forth between Claude is I have two computers that I work off of, right?

13:13And maybe you do, too. And I work on one and I build on the other, and I go back and forth, and both have Claude instances up because the tact time that it takes for me to actually work with Claude and build websites is is like a hundredth of the time that Claude is processing. So, as I'm working, I'm doing other things like talking to you on YouTube or whatever, right? But if you want to learn how to do this, how to make money with AI, how to build microsites specifically, come into the classroom. There's a 7-day free trial right now. You'll have access to the

13:42microsite playbook right off the bat, right? This is where you'll get your feet wet. You'll build your first microsite. And then, eventually, you'll have access to the microsite empire. You come in here, you click here, and you come down to this zip file.