Dominate the local market with the help of AI

Joseph says: “Don’t just build a personal relationship with the local market but use AI to help you build that relationship, so you can dominate like nobody has been able to dominate before.”

What changes in local search algorithms are you seeing at the moment?

“Practically, AI can help you as somebody who may not know a specific local market. Consider people looking for roofing. I’ve never needed a roof for my house, so I’ve never been in that space, but guess what? ChatGPT (or whatever AI tool you want) has the knowledge of the entire internet. That includes anybody who’s ever talked about it, chatted about it, written about it, etc.

There’s information in the database that’s not in my head, or your head. You can go and say, ‘Here’s who I’m looking for. What are they worried about? What are they typing in? Are there any concerns they may have?’, etc.

You can use AI to help you build the best picture and learn more about your prospect than you’ve ever known before. That leads to better keywords, better content, getting in better resources, and even connecting socially. There’s a whole lot that will go along with that.”

Would you use AI to determine which local searches to focus on initially?

“Definitely. Just before this call, I prompted ChatGPT with, ‘Acting as somebody looking for junk removal, what are all the ‘near me’ searches they could possibly type in? Give me 50 of them.’ It gave me 50.

Of course, right off the bat, it’s going to give you junk. The secret is you have to build a relationship with AI. You have to then say, ‘Give me queries for when they’re worried about neighbours thinking that they’ve got a bunch of junk out.’ 2 or 3 different prompts into it, it gave me a whole list.

Then, you take that whole list and go to Mangools, Ahrefs, SEMrush, or any other bulk-checking tool and see if it got any of that right. Meaning, see if there’s low-hanging fruit where people are looking for it, but nobody ever thought of it before. Why? Because it’s AI doing it. AI can pull all of that data together and think like the consumer. It’s like having the consumer sitting there in front of your desk. That’s how I treat the AI tool.

It will help me create the content and know where to focus in ways that I couldn’t see before – even last year. It’s a whole new frontier.”

Are you then using AI as an assistant, a colleague, and a pretend customer, to get different perspectives?

“Yes. It’s funny, there are a few of us who laugh because we actually compliment AI. We treat it like it’s a real person now. You’re chatting to it, and you’re respecting it a little bit more, and you’re coaching it a little bit more. The more we treat it like a colleague or an assistant, the better it does, believe it or not.

If you say, ‘Oh, that’s great! Can you do that again but for X market or Y niche?’, and then you save that prompt, it’ll know what you want and keep doing it. It’s almost like it knows what makes you happy and keeps trying to make you happy, which is weird if you think about it.

We’ve laughed about it. When someone sees a screen grab where I said, ‘Thank you ChatGPT, that was great!’ they’ll go, ‘You compliment ChatGPT? I do too!’ It’s funny because it performs better if you compliment it.”

Does AI know what ‘near me’ means in a localisation context?

“It does. The cool thing about that ‘near me’ AI search is that there are many different ways that somebody can be using ‘near me’. I told ChatGPT to tell me all the different ‘near me’ variations that a person could possibly use when looking for a dentist, because it’s a little bit different in each niche.

It came up with a bunch of stuff that I would have never thought of. ‘A dentist I can walk to near me’ had low domain authority but tons of traffic. There, ‘near me’ meant ‘I can walk to it’. If you targeted that properly, you could get tons of traffic that nobody would have ever looked for.

I asked it to give me 50, and some of them were repetitive and hallucinating, but it gave me about 10-15 ‘near me’ long-tails that were very low keyword difficulty and very high traffic that I never would have seen before.

It’s like putting X-ray vision on the market. You can ask for ‘near me’ searches for dentists, barbecue restaurants, pizza parlours, or whatever. It will give you what you should be targeting that nobody else is even looking at. It’s insane.”

How do you then optimize those pages to ensure they will come to the top?

“You can do what we call ‘EEAT peppering’. EEAT is a seasoning, that’s why it’s called ‘eat’. Expertise, authoritativeness, and trust are seasonings to add to your content.

You can use AI to do that. If intent or any reviews come in, you can get AI to show anything that’s talking about X keyword or Y happiness on your page. There are plugins that allow you to use AI to show certain reviews and things like that on your homepage, which then help connect you to the Map Pack, the rankings, etc.

You can feature other people’s content on your pages using plugins that can filter based on what they’re typing in or the keywords they’re using. Schema markup data is key to demonstrating that you’re the most authoritative business of its kind in the area.

The other thing is automated responses and automated chats. A lot of people may look down on it, but you can have an automated AI respond to a negative or positive review on your Google Business Profile with a fill-in-the-blank pre-prepared statement. We’ve seen that automated review responses saying, ‘Sorry that you had a bad time. We’re going to give you a call.’ will get bumped up.

Then there are the automated chats on your homepage that also filter, whether they are keyword and database-driven or whatever. Those are also a cool usage of AI on your page.”

Is it a bigger opportunity to optimize for third-party platforms, social media, and Google Business Profile, or is your website still the bread and butter for local SEO success?

“If you do a search for your name, your social media profile page will probably come up – your LinkedIn or Twitter. Those will be strong pages that come up. If you type in, ‘Bob’s Junk Removal’, the likelihood is that they would come up on a Google Local Pack. Generally, though, their Facebook, LinkedIn and those social pages aren’t going to show up.

What may show up is a Reddit page. Reddit is unique in that it will actually show up in the top 10 on SERPs, but I’ve never seen a LinkedIn business page. A Yelp page will. It depends on your type of business. If it’s on Angi (formerly Angie’s List) that will show up.

However, if you’re searching for ‘junk removal San Diego’, most of what’s showing up is web pages like Junk MD and Crisan Junk Removal. They are real web pages offering junk removal, and they’re all competing for the Local Pack, but they’re getting tons of traffic and business on the actual SERP itself.

Those businesses have to be strong on the webpage itself, meaning the copy on the site and the SEO. It depends on the market. Dentists also probably have to have a strong website instead of just a Yelp page.

Restaurants, however, might be able to get along with just a Yelp page. They may not need a real page – unless the consumer expects a better experience, like a menu, photos, and a place to find extra services like weddings. A local taco shop, on the other hand, may not need a web page at all. It could rank in the top Google Local Pack.”

Can you optimize your content to appear in the AI overviews within the Search Generative Experience?

“There’s been a lot of discussion about who got a slap on the wrist for using AI-generated content. This just happened over the last few months. What didn’t get touched was the local market. None of the local pages were touched.

That’s because there’s going to be a lot of duplication there. ‘Junk removal in San Diego’ is going to be very similar to ‘junk removal’. There are going to be 50 junk removal services in San Diego, and they are going to copy from each other. Google can’t really slap anybody in that market because it’s going to affect a real local marketplace.

These aren’t people competing over blogs and ad revenue. They don’t want to mess with the carpenter or the plumber, so none of the local content was affected. Therefore, it’s free range to use AI.

We use AI to take the top 10 content that’s out there, analyse it, and rework a brand-new page – using that successful content to create a better experience than what’s out there. We’re not competing over ‘How do I hammer a nail?’ for a carpenter, because that’s an information-based search. You don’t want to be in the questions because those are all information-based, not local.

For any local searches, like ‘How do I find the best plumber?’, it’s going to be the best plumber webpage that ranks. The page that’s got the citations and authority for the local space is going to rank in that area.

You can go harder with using AI for local SEO and you’re not going to get slapped as much, but by ‘go harder’ I mean make it better. We’re optimizers. We’re supposed to make things better, not worse. Use AI to make a page better. Get it to show you what’s missing.

For 2025, I’m calling it ‘gap publishing’. That’s what’s going to get you to rank at the top. Look and see who’s providing the content and where the gaps are. Then, provide a piece of content that’s got those gaps covered. That’s going to get you ranking.

Google wants to see a complete piece of content. The complete, better piece of content wins. Analyse the market, see where the gaps are, and create a piece of content that has no gaps. If you do that, you’re going to rank.”

If an SEO is struggling for time, what should they stop doing right now so they can spend more time doing what you suggest in 2025?

“Stop using an endless number of prompts and focus on engineering 2-3 prompts to get the work done.

We started out with 50 prompts to do everything. ‘Find this’, ‘Find that’, ‘Give me this’, ‘Give me that’, etc. It would take so much time. You sit down and try to figure out what you’re going to do. You have 25 prompts you have to go through, then build more prompts off those, then take those answers and use them. You’re daisy-chaining different prompts to get the results.

The best thing that you can do is start by determining what it is that you’re trying to accomplish. What’s the objective? Is it a list of 50-100 keywords or a piece of content that you’re going to use with a specific intent, area, and target? Have 1 prompt for that.

Have 1 for the content and 1 for the competitive analysis. Work until you get 1 prompt that gives you what you want. Once you have that set, then you can scale. You can go fast. If you just have 2 or 3 key prompts, you can crank those out and get number 1 listings for just about any local market.”

Joseph Kahn is President and CTO at Hum JAM, and you can find him over at HumJAM.com.