YT YouTube2Text

YouTube2Text → Transcripts → The 30-Page Local SEO Hack Google’s AI Is Already Rewarding

The 30-Page Local SEO Hack Google’s AI Is Already Rewarding — Transcript

by Edward Sturm · 15,242 words · 2,180 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:00Caleb Ulku, you are very well known in
0:02local SEO and AI SEO and on YouTube. And
0:06people have been asking for this podcast
0:07for a while. So, Caleb, thank you. Thank
0:10you. Thank you. Thank you for joining
0:11the show. I'm excited to have you,
0:14>> man. I I appreciate you reaching out. I
0:16uh I watch your stuff all the time. It's
0:18funny. I've u my strategist for YouTube
0:21sometimes he'll send me videos like,
0:22"Hey, this one seems like a really good
0:24one to talk about what they're saying in
0:27this video and react to it." And about a
0:30week ago, he sent me one of your videos
0:31and I was just like, "Well, I'm I'm
0:33actually going to be on Edward's uh
0:35podcast."
0:36>> Yeah. I mean, yeah, dude. I'm I'm I'm
0:38hyped to have you. Uh could you share
0:41your background for people who aren't
0:43familiar with you really quick?
0:46>> Yeah, sure. So, I have um uh I went to
0:51undergraduate school in a very small
0:53school in New York City called the
0:54Cooper Union, which no one's
0:56>> I'm from New York. I know it.
0:58>> Oh, all right. Oh, yeah. You said you
1:00you were uh born in Brooklyn. So, yeah,
1:02that makes sense. It's a very usually a
1:04regional school.
1:05>> Yeah. Yeah.
1:07>> Oh, born in Manhattan, raised in
1:08Brooklyn. Well, it's okay that the a lot
1:11of people born in Manhattan would make
1:12that correction, so I won't hold it
1:13against you.
1:16>> But yeah, so um I graduated from that
1:19school uh with chemical engineering. Uh
1:21hired on to Exxon Mobile, moved to
1:23Houston like everyone who hires on to
1:26the oil industry. Uh I worked for them
1:29for 10 years. And
1:32my wife and I, we had a 2-year-old son,
1:35and we really just never saw him. We
1:37both had jobs that required crazy
1:40commutes, uh, long hours. So, we
1:42actually hired a driving nanny because
1:44we couldn't make it to daycare, but or
1:47daycare clothes. So, we hired a driving
1:49nanny. She'd go pick our son up, bring
1:51him home, put him to bed. So, we
1:53basically hung out with him on the
1:54weekends. And we did this for way too
1:58long. It just felt like normal life,
2:00right? And then, blessing in disguise, I
2:04broke my hip. Um, so when I broke my
2:08hip, I had to I was like
2:10nonweightbearing during the recovery and
2:12Exon Mobile at the time was building
2:14this enormous new campus north of
2:16Houston, but like the elevators weren't
2:18done, the escalators weren't done. So
2:20they're like, "You can't come to this
2:21campus without being able to walk." Uh,
2:25so they said, "You have to work from
2:26home." And I worked from home for about
2:30a month and a half. And I think the day
2:32I went back to work, uh, it took less
2:36than a week before I put my two weeks
2:38notice in, uh, because working from
2:40home, you know, I was learning to cook,
2:43I was hanging out with my son, I was
2:45with my family, and it was just like,
2:47what were we doing with the crazy
2:49commute and everything? So, put in my
2:51two weeks and we moved to uh, a town
2:55just outside New Orleans, which is where
2:57my wife's family is from. Everyone ends
2:59up where their wife's family is from.
3:01And yeah, we raised our my my son is now
3:0414. So that was a minute ago.
3:06>> Wow. And so and so when did you decide
3:08to get into SEO?
3:11>> So when we first moved there, um plan A
3:14was, hey, let me go find another job. So
3:16I uh started to look for other jobs.
3:18It's a very serviceoriented economy. So
3:22I wasn't really interested in that type
3:23of a job. And finally just decided, you
3:26know what, let me go ahead and start my
3:27own business. And I didn't know what SEO
3:29stood for at the time. So that wasn't
3:31my, you know, plan A. But eventually,
3:35you know, I was going to chamber of
3:36commerce meetings and things like that
3:37and trying to explain what my business
3:39did. And I would get business owners not
3:42understanding and say, "Uh, but can you
3:44help people find me online?" And
3:46eventually I just said, "Yeah, sure. I
3:47can do that." And started to learn. And
3:51you know, I have this one of my very
3:53first clients um that I got in SEO.
3:56Again, I had no idea what I was doing.
3:58This was a very long time ago. So, I did
4:01all these SEO things that I'd read
4:03about, that I'd heard about, and ended
4:06up getting their entire domain
4:07de-indexed, which, you know, pretty bad.
4:10[laughter]
4:11>> Yeah. Just a little bit.
4:12>> So, they fired me, which
4:15>> Yeah. Yeah. Pretty bad. Yeah. That's not
4:17what you're looking for when you hire an
4:19>> What do you think did it?
4:22>> Oh, I was doing very shady things with
4:24link building. Like I was doing all the
4:26shady things. Yeah. Um and and I I guess
4:29Google didn't like all the shady things.
4:32>> Um so they fired me, which again, right
4:35call. That was the right thing for them
4:37to do. Um I did get their domain
4:40reindexed, which was awesome, but they
4:42didn't reindex, so that's fine.
4:46>> Uh well, I started uh basically
4:48disavowing the shady things I'd been
4:50doing. Uh and then uh the domain did end
4:54up getting reindexed after I did that.
4:56So that was awesome. And at that point I
4:59started buying SEO courses trying to cuz
5:02like hey the stuff I'm learning on, you
5:04know, reading on blogs and watching on
5:06YouTube, you know, obviously I want
5:09something a little bit more structured,
5:10something that I know works. So yeah, if
5:12you were selling an SEO course in like
5:152015, I probably bought it. Like I
5:17bought every course I could find.
5:19>> Which one was the best?
5:20um and and started going through
5:23>> which which one was the best in 2015
5:26that you got?
5:28>> Honestly, I think uh Matt Diggities was
5:30really good. Uh I think it was called
5:32the affiliate lab. It's now defunct, but
5:34that was a solid one. And then
5:38to try to fix the shadiness, I bought a
5:40course from Matthew Woodward about how
5:42to build PBNs, which you know, we don't
5:46really use that as much. try to fix the
5:47shadyness. I started learning about how
5:50to make PBNs.
5:53>> Well, how to do it the right way, but I
5:55mean the right way from 10 years ago,
5:57which isn't what we would do today.
5:59Yeah.
6:02>> Yeah.
6:02>> So, you got So, you got the courses and
6:03then what?
6:06>> So, got the courses. Uh, and honestly,
6:08like I just started applying to jobs on
6:10Upwork. Um, you know, I figured out that
6:14I was not good at a sales call where I
6:18had to take talk to a business owner and
6:20convince them that SEO would be a good
6:22way to get more leads for them. But if I
6:25were on the phone with somebody who
6:26already wanted SEO, I was really good at
6:29convincing them that I knew how to do
6:31SEO and I could make it work for them.
6:33So on Upwork, everyone was posting jobs
6:36on Upwork already deciding they wanted
6:37SEO. They're just looking for the right
6:39provider. You know, I I um I'm sure
6:42you've seen the like Wolf of Wall Street
6:44scene where Leo DiCaprio is like, "Sell
6:47me this pen."
6:48>> Well, I'm the person who's just like,
6:49"Oh, look at how well it writes and look
6:51at how nice the clicking is." I mean,
6:52that's how I talk. Uh so that's how I
6:55was able to sell SEO because, you know,
6:57they'd already said they want the pen,
6:59so I just had to sell them on the
7:00features of it. Um we got a few huge
7:03brands, uh which is great. We got uh
7:06like Skillshare, Adobe, Stanley Black &
7:09Decker from uh Upwork. This method of
7:12marketing is so effective, I had to make
7:14sure it wasn't against Google's rules
7:16before I kept doing it. It's a form of
7:18SEO I call compact keywords. Whereas
7:20most SEO focuses on putting up articles
7:23to answer questions, how, what, when,
7:25compact keywords focuses on putting up
7:27dozens of pages that sell to searchers
7:30who are actually looking to buy. These
7:32pages rank on Google and convert so much
7:35better than normal that when I
7:36discovered this years ago, I couldn't
7:38believe this was allowed. It's less
7:40work, too. The average compact keywords
7:43page is only 415 words. Compact Keywords
7:47is a 13-hour deep course on getting
7:49sales with SEO. A customer recently
7:52said, "Each lesson is dense with
7:54information. You're giving years worth
7:56of experience boiled down into 15 to 30
7:59minute lessons with no filler or fluff.
8:02I feel like I'm gaining a new
8:03superpower. Compact Keywords is about
8:06setting up an SEO funnel that brings you
8:08sales for years and years and years. It
8:11works with AI. It's less work than
8:13traditional SEO and it makes way more
8:16money. You can get it now at
8:18compactkeywords.com.
8:20back to the podcast
8:22>> and the budget that we got from those
8:24larger brands allowed us to
8:26>> You got that from Upwork?
8:29>> All of them from Upwork. Yeah. Yeah.
8:30From Upwork.
8:33>> Yeah. Now, the budget that we had from
8:36all of those allowed us to start doing
8:37split testing, right? Like, you know,
8:39hey, you person ABC says to do this and
8:42then this other person says to do this.
8:44Well, let's do each of them 10 times and
8:47see which one actually works and which
8:49one drives rank. So, we did that for a
8:52couple of years and started to fine-tune
8:55uh our process for how we did SEO, how
8:58we drove results, all non-local uh
9:00because everything was from Upwork. So,
9:03you know, not a lot of like local
9:04plumbers are posting on Upwork. Um but
9:07yeah, we got it took a couple of years
9:09on Upwork to get to seven figures. Um,
9:13since then I don't do much on Upwork
9:14anymore. It's gotten really difficult.
9:17Uh, even with an established profile,
9:20but yeah, you know, uh, I still know a
9:22lot of people who have success on the
9:24platform. It's just harder than it was,
9:26you know, 5 10 years ago.
9:28>> Something that's working in SEO right
9:30now that a lot of SEOs still aren't
9:32doing.
9:35>> Yeah. So it's really and a lot of this.
9:39So the way that we did local or
9:40non-local SEO, right, is we would have
9:43the money page, the page we were trying
9:45to rank, and then we just built all the
9:47supporting content for that money page.
9:50Well, in early 2023,
9:53uh we pivoted completely away from
9:55non-local to local SEO. Uh we did that I
10:00I'm sure Edward, you you know the
10:02timing. Chat GPT came out in November of
10:0422. So, you know, I don't know. I I've
10:07talked to a couple of other SEOs and
10:09they had a similar realization because,
10:11you know, by the time 2022 had come
10:12around, uh, we'd watched Google keep
10:16people on the SERs for years at this
10:18point. One of my favorite examples, uh,
10:20is the search term mortgage calculator,
10:23right? Where you type mortgage
10:25calculator and then there'd be a bunch
10:26of results. You'd click on them and they
10:28had mortgage calculators. Well, Google
10:30decided that they would put a mortgage
10:32calculator on the SER. fabulous for
10:34users, horrible for the businesses that
10:36had spent all the money to rank for
10:38mortgage calculator because suddenly the
10:39traffic went away. So in the first
10:42couple of conversations with chat GPT3
10:44and a half, it's just like man, Google
10:48is going to be able to take this
10:49information, this technology, and keep
10:52people on the SER forever. So we pivoted
10:55very hard away from non-localformational
10:58type searches and said we're going to do
11:00local SEO, right? However advanced AI
11:03is, it's still a couple of years from
11:04being able to replace people's water
11:06heaters. So when we started doing local
11:09SEO, we did it in the same way that we
11:11did non-local. So uh the money page for
11:14local SEO we started treating to be your
11:17Google business profile. So then we
11:19thought, okay, everything on the website
11:20is supporting content to establish more
11:23trust, more authority for the Google
11:25business profile. So we started testing
11:27this concept where whatever your
11:30categories and services on your GBP, we
11:33would build the website to exactly
11:35mirror the GBP categories and services.
11:38So now we call that the core 30 cuz it
11:41for most local businesses it ends up
11:43being around 30 pages. And we have found
11:46that works so well because when your
11:48website matches the GBP exactly, uh it
11:51builds so much trust not just with
11:53Google but with the AI systems as well.
11:55So I think that's the simplest way for
11:57uh and a local business owner can do a
11:59lot of that themselves uh you know by
12:01just prompting AI to write the content
12:05uh and the local business owner knows
12:07obviously enough about that business to
12:09be able to get the AI to write something
12:11unique and interesting about each of
12:13those categories and services and
12:15structure the website to match the GBP.
12:18I mean for 80% of markets that alone
12:22will significantly increase rank and the
12:25number of calls you're getting even if
12:26you don't do anything else. So I think
12:28that's like the lowhanging fruit and
12:30when somebody comes to me and says hey
12:32will you do SEO for us? I mean that's
12:34like step one build out the core 30 and
12:38you know month one results almost always
12:41just look fabulous because it's such a
12:44simple thing and you know the machines
12:46don't know how to interpret messy
12:49signals the way that humans do. So we're
12:51just trying to create the same signal
12:53that makes sense match your website to
12:55your GBP. That was a really long answer
12:57to that question.
12:58>> That's awesome. Could you give uh could
12:59you give like an example of what that
13:01would look like?
13:04Right. So, uh I I do Yeah. So, your
13:07homepage is going to be based on the GBP
13:10primary category, right? So, so say
13:13plumbers. When we started running ads
13:15for local businesses, we ran ads for
13:17plumbers. So, we've had a lot of
13:18plumbing clients. So, my example always
13:21goes to plumbers. So, say say plumber.
13:23So, for a plumber, your primary category
13:24is clearly going to be plumber. And then
13:27your GBP landing page, which is usually
13:30your homepage. You're going to base that
13:32around like plumber city name, right?
13:35Like plumber Houston. And then on your
13:38homepage, you're going to talk about the
13:40secondary categories. And then a couple
13:42of what we call like core services,
13:44services that you really, really care
13:46about. So for a plumber, your secondary
13:48categories might be drainage service,
13:50bathroom remodeling, and then the core
13:53services that you really care about.
13:54Most every plumber I've ever talked to,
13:56the services that they want more calls
13:58for are water heater replacement and
14:01main drain line replacement. So, we'll
14:03do an H2 for each of the secondary
14:06categories and then an H2 for the core
14:08services and then a paragraph, you know,
14:11maybe 50 70 words underneath that H2.
14:14And then we link to a page where that
14:16page is about bathroom remodeling or
14:19that page is about water heater
14:21replacement. So that's going to be the
14:22H1, bathroom remodeling Houston. And
14:25under there, we're going to talk about
14:27all the services that live under that
14:30secondary category on your GBP. Right?
14:32You're when you set up your GBP
14:34services, you set up your categories and
14:36then you set up services under each
14:37category. So the services under bathroom
14:40remodeler we talk about on that page and
14:42then each one of those links to a page
14:44about that specific service. So maybe
14:46you know you have faucet replacement. So
14:48that's a service under bathroom
14:50remodeler. You'll create a service page
14:52for faucet replacement Houston and then
14:55the bathroom remodeler page links down
14:57to that service page. So that creates
14:59this exact same hierarchy structure that
15:02your GBP has.
15:03>> What do you do with businesses that have
15:05multiple locations or multiple cities?
15:09>> Yeah. So for that uh there are that is
15:12one of the two reasons why your GBP
15:14landing page might not be your homepage.
15:17So, uh, essentially every location
15:20should have its own GBP landing page.
15:23Uh, but if you're taking on a new client
15:25or you have an existing business and you
15:28set it up so that all of your GBPs point
15:30to your homepage, we don't move them to
15:32an internal page until we run the rank
15:34map, right? And I'm sure you know this,
15:36Edward, but if you if your GBP landing
15:39page is your homepage and then you move
15:41it to an internal page, your rank is
15:43going to plummet. So, we never do that
15:46if they're ranking, right? So, we always
15:48look at the rank map and if their rank
15:50is horrible, uh, or they're relatively
15:52new, then every single GBP gets its own
15:56GBP landing page. It's basically just a
15:58copy of the homepage. Um, and we build
16:01up the core 30 for every single
16:02location. So, if you have five
16:04locations, that becomes the core 150.
16:07But it's important to keep in mind that
16:10I love multilocation businesses because
16:13as you're building all of this out, the
16:15domain is getting so much topical
16:18relevance and authority that you'll hit
16:20a point when you're adding, you know,
16:22location number four, location number
16:24five, depends on the competitiveness,
16:26etc., etc. But eventually you'll hit a
16:28point where you'll create a GBP landing
16:30page. You'll point the GBP to it and
16:33then you'll be ranking reasonably well
16:34just because the domain has so much
16:37authority and relevance for plumbing.
16:39You don't need the core 30 anymore for
16:41this new location. And then we just go
16:43straight into producing geographical
16:45relevance content.
16:46>> Okay. What's the biggest when it comes
16:48to Google business profile? What what
16:49what's the biggest mistakes that you
16:51[clears throat] see businesses making?
16:55>> Yeah. So, uh, categories and services,
16:58right? Um, the most common businesses we
17:01look at, they'll have one category,
17:03maybe two. Google allows up to 10.
17:06There's this myth floating around that
17:09if you add multiple secondary
17:10categories, you're diluting the ranking
17:12power for your primary category. We've
17:15tested this over and over and over
17:16again. We've never seen we've only seen
17:19positive results from adding more
17:21secondary categories. Uh and then the
17:24other mistake is your services, right?
17:26Uh it and the mistakes we see there are
17:28everything from not enough services to
17:31keyword stuffing, right? So we know that
17:34Google works based on entities. It
17:36doesn't use keywords anymore. So the
17:38days where if you had a personal injury
17:40lawyer, you needed to do car accident
17:42lawyer, car accident attorney, uh car
17:45wreck attorney, car wreck lawyer, right?
17:47So now we have four services that
17:49describe the exact same thing. Google
17:51knows those are the same thing and
17:53keyword stuffing your services. I don't
17:55know that it ever worked, but it
17:56certainly doesn't work today. So, that's
17:58where we spend most of our effort for a
18:00new client is we're going to come in
18:02there, take a look at their categories
18:03and services and make sure those are
18:06correct and optimized before we start
18:08building out the core 30. And it's not
18:10hard, right? Like, but it's just
18:12something that most businesses don't do.
18:15Um, and then the other aspect is just
18:17making sure every single box is filled
18:19out. The most common one that we don't
18:21see filled out is uh holiday hours
18:25because Google doesn't let you put in
18:26holiday hours forever. You can only put
18:28in the holiday hours for like the next
18:30two or three holidays. So, you need to
18:32log into GBP every few months and update
18:34your holiday hours for whatever holidays
18:36are coming up. So, I mean, those are the
18:38the biggest mistakes that we see, but by
18:41far it's get your categories and
18:43services correct. Yeah.
18:44>> How are you structuring the URLs when
18:47you're when you're Yeah. when you have
18:48these like individual pages, when you
18:50have multiple locations, what does a
18:52website look like in terms of URL
18:53structure and and and internal linking?
18:56Yeah, internal linking and hub pages and
18:58all of this stuff.
19:00Yeah. So, the URL structure like I care
19:03very little about it. So, whatever my
19:05developer finds easiest to work with to
19:09keep track of all the pages and
19:10everything, that's fine, right? Cuz
19:12Google doesn't build its image of your
19:15website based on your URL structure. It
19:17builds what your website looks like
19:19based on your internal linking
19:21structure. So I care a lot about
19:23internal linking and very little at all
19:25about URL structure.
19:26>> Do you think that that Google looks at
19:28URL slugs or parent or parent folders?
19:33>> I don't think Google pays attention to
19:35parent folders. Uh we will almost always
19:38make sure the slug is the service entity
19:41that we're writing for. And that's just
19:43because it's pretty common that a lot of
19:45your links coming in will be the naked
19:47URL. So having the keyword in the slug
19:50means you can put your keyword in the
19:52anchor text without being shady or
19:54spammy.
19:54>> So So okay, so you would do So you would
19:57do like forward slash um Oh yeah, give
20:01me give me a service. Give me a service.
20:05>> Sure. So it would be like, you know,
20:06domain.com
20:08heaterreplacement.
20:10>> Okay. So you would do you would do
20:11domain.com/h houston/waterreplacement
20:15Houston.
20:18>> No. Well, if if if if you have
20:20subfolders, then we would just do water
20:22heater replacement in that
20:25slug.
20:26>> Okay.
20:26>> Yeah. But normally we don't do
20:28subfolders
20:30>> unless the developer or client really
20:33>> So, so if you have multiple locations,
20:35then you're not you're not separating
20:37the URLs by location. So, does that mean
20:40that you don't usually do do you have do
20:42you still have a hub page for the
20:44location? Do you have forward/ houston?
20:51>> We might have like the slashorage
20:54serving as the GBP landing page for the
20:56Anchorage location, but in terms of like
20:58all those subfolders and everything, I
21:00mean like I'm not going to go create
21:02that. If it already exists, I'm not
21:04going to undo it either. I don't really
21:06care very much either way. And that's
21:09because the most important you don't
21:12care because so so what are the key the
21:13key elements to putting the keyword in
21:15for you?
21:18>> So the keyword ideally it should be in
21:20the URL slug. Uh but that's the least
21:24important of the really important places
21:26to put it. The really important places
21:29to put it are the title tag, the H1, and
21:31then the first paragraph. We also want
21:34to mention it very early on in the first
21:36paragraph. Uh, a very common mistake I
21:39see among so many local business
21:41websites is their first paragraph of
21:43text is like a history lesson about
21:46their business, right? And I I I make
21:49the joke, right? No one is typing
21:51plumber Houston into Google because
21:52they're interested in reading history
21:54lessons about the local plumbers, right?
21:56So that first paragraph, such a missed
21:58opportunity if you're not talking to the
22:01person who just typed plumber Houston
22:03into Google and you know why you're the
22:05plumber that they should call, right?
22:07Like nobody cares about what year you
22:09were founded and you know why your
22:11grandfather went into plumbing and why
22:13he immigrated from the old world. Like
22:15nobody cares about that but that's where
22:17people start talking about local
22:19business websites. But yeah, so title
22:21tag most important H1 and then in the
22:24first paragraph we'll start talking
22:26about it as we talk to the searcher who
22:28landed on the page also important right
22:31because as you know there's an interview
22:34with Sundar Pachai the CEO of Alphabet
22:36that came out I guess probably about a
22:38month or so ago now and one of the
22:40things he talked about is that he sees
22:42Google evolving to become a home for
22:45agentic search. So I think what that
22:48means you go to google.com and instead
22:51of a search box there's more like a
22:53prompt box and you type in what you're
22:55looking there and then Google sends a
22:56bunch of agents out to find the
22:58information and bring it back to you. So
23:00we're that is reaching a point where
23:02we're no longer writing websites for
23:04people. We're writing websites for the
23:06AI agents to summarize those websites
23:09for actual people. So again, very
23:12important when you're writing website
23:13content to make sure you're writing it
23:16for someone who is searching for what
23:18you're looking for, whether that be the
23:19agents or actual people.
23:21>> Has that changed the way that you do
23:22SEO?
23:25>> Not significantly. Um because the, you
23:28know, the right way to do SEO has always
23:31been to structure the website in such a
23:33way that Google can understand it. And
23:35we always wrote for the algorithm. And
23:39then, okay, let's also make sure we're
23:41writing for conversion for the people
23:43because, you know, if you have the best
23:45website ever that's directed to people
23:47and you're ignoring the algorithm, it
23:49doesn't matter because no one will ever
23:50see it, right? Uh assuming you're not
23:52running ads or something like that. Um
23:55so that like make sure it's structured
23:57properly. Make sure it's written for the
23:59algorithm that stays. Um, and you know
24:03whether a person is reading it or an AI
24:05is reading it, it doesn't really matter
24:07as long as you have the information on
24:08there. So the biggest difference, right,
24:10is like a messy website or messy
24:13structure, people have the patience to
24:16figure that out, whereas an AI doesn't.
24:19And on the flip side of it, the AI can
24:22read your entire website and pick out
24:25the most important parts of it and give
24:27it back to a user. Whereas most users
24:30will spend, you know, 20 seconds on your
24:32site before they decide whether to call
24:33you or not. So that's where we get into
24:36like attribute match matching uh which
24:39is something that we weren't super
24:40focused on before AIS, but now we know,
24:44right? So if somebody searches for like
24:46lead main drain line replacement, the
24:49AI, whether it's chatt or claude or
24:51Google's own Gemini, the AI is going to
24:54want to see that you're talking about
24:55lead main drain line replacements. Well,
24:58if you don't talk about that, if you
25:00just talk about main drain line
25:01replacements without using the word
25:02lead, you might not get recommended. Um,
25:05so that's uh that attribute matching is
25:08something that's relatively new. Uh,
25:11when we're looking at content, building
25:13out a website for for local businesses.
25:16>> Yeah, I always like I like to look at
25:17other pages that are that are ranking
25:20for keywords that I'm targeting and I
25:21see the language that they use. Like
25:23basically like for me I'm a such a good
25:26SEO and I think you're you I'm sure
25:28you're the same where you can go into
25:30any niche and within
25:33like a week or two understand the niche
25:35well enough that you can actually write
25:38pretty expert content in it by just by
25:41taking notes of the the types of
25:43language that appears throughout pages
25:45that are ranking for your target
25:47keywords. And then you make a list of
25:48these. You you have your target keyword.
25:50You know where the target keyword goes.
25:52you have a hierarchy of where the
25:54keywords are going, what's most
25:55important, what's least important, and
25:56then you can build your pages like that.
25:59What do you think about um about AI
26:01images both on uh the website and in a
26:05Google business profile?
26:08>> Yeah. So, the what you just talked
26:10about, we call that uh we've called that
26:12the open book test, right? So, if you
26:14want to rank for whatever service entity
26:16you want to rank for, the best way to
26:18get started is to type that into Google
26:21or type that into chat GPT and it will
26:23tell you, hey, here are my favorite
26:25pages about that topic in the world. Uh,
26:28and then look at what they did and just
26:30do a little bit better, right? Like add
26:32a little bit more information that they
26:34don't have. Uh, and there are variety of
26:36ways that we uh focus on to do that, but
26:38yeah, exactly. the the point is to write
26:40local SEO content as the local SEO
26:44business or not as the local business
26:46owner expert.
26:48>> Actually, another great question is is
26:50are you making uh do do you try to make
26:53all the pages unique to the locations?
26:55Are you giving extra information based
26:57on location or like how do you think
26:58about how do you think about it uh about
27:00doing that versus something like doorway
27:02pages?
27:04>> Yeah. So every page we write for local
27:07SEO is hyperfocused on that location,
27:10right? So even, you know, I talked about
27:12lead main drain line versus main drain
27:14line. So if we had a plumber client in
27:17the suburbs of Boston, talking about
27:19lead main drain line replacements would
27:21be very important because that's a
27:22common issue in the suburbs of Boston.
27:24But for a client in Southern California,
27:27we would never talk about that because
27:28they don't have lead main drain lines in
27:30California. the hoses are a lot newer
27:32and that you know they use plastic main
27:34drain lines. Um so those you know local
27:37differences are so important for local
27:40SEO. Um and we we actually saw an issue
27:44I'm sure you remember is is like in
27:47early May where Google started mass
27:49de-indexing AI slop which you know
27:52fabulous news. So when we build local
27:56content uh we we build it for there's
28:00four different types of local content
28:01that we write. Uh so we write category
28:04content, we write service page content,
28:06we write topical relevance content and
28:09then we write geographical relevance. So
28:12of those four types when Google started
28:13mass de-indexing three of them no
28:16problem. But the geo content we started
28:18to see some de-indexing. So when I say
28:21geo content, what we'll do is we'll run
28:23a local rank map for the target service
28:26entity. And everywhere that you're not
28:29ranked in the top three, uh especially
28:31if you're ranked like four, five or six,
28:33we're going to write hyper local content
28:36about landmarks in that specific area,
28:39right? So you know, again, using plumber
28:42Houston as an example, we'll write
28:44content, you know, plumber in River
28:46Oaks, Houston, plumber near Montro,
28:49Houston. uh plumber near uh 610 West and
28:53I 10 Houston. So, we're finding these
28:56hyper local landmarks like lakes, zoos,
28:59city halls, colleges, whatever. Uh and
29:02we'll write content about a plumber in
29:04that region. Well, for a very
29:07competitive space like Houston, we might
29:09end up with a couple dozen of those. And
29:12it's challenging to make plumber Montros
29:17be a significantly different article
29:19than plumber River Oaks cuz they're so
29:21close to each other. But if you know
29:23Houston, uh, Montros, uh, is a very
29:27different neighborhood than River Oaks,
29:29right? River Oaks is very fancy estate
29:32sized homes. So you could write about
29:35the differences for being a plumber in
29:37River Oaks because they are different
29:38than Montros. But how do you do this at
29:40scale, right? So what we started doing
29:43is just an insane amount of research. So
29:47uh we would pull like US Census Bureau
29:49data, we'd pull Reddit data, we'd pull
29:52all the data and information that we
29:53could find. We started hooking into the
29:56CRM so that we could pull conversations
29:59and leverage all of this data to write
30:02content about, hey, why is being a
30:04plumber in River Oaks different than
30:06being a plumber in Montro? Uh, and we
30:08did that, went back through the
30:10de-indexed content and added all of this
30:12additional research and we got a lot of
30:14it to be reindexed because now it
30:16actually isformational
30:18additive, right? We're not we're not
30:20writing the average of the content
30:22that's already exists with redundant
30:23information. Now we're adding new
30:25information, right? River Oaks uh
30:27plumber is, you know, you have to worry
30:30about walking 200 feet back to your
30:32truck because the front lawn is so big.
30:35Um, and yeah, Montros is, you know, you
30:38might have a fiveplex or something
30:40that's two stories tall in Montros. Uh,
30:42so yeah.
30:43>> Yeah, that's awesome. That is awesome.
30:44And and so how about AI images? Cuz you
30:47see this, you see this from all all
30:48sorts of businesses trying to put in AI
30:51images. Um, but you also see in local.
30:53So what do you think about Yeah. both on
30:55a Google business profile and on the
30:57website?
31:00So obviously we always prefer real
31:02images from the client, but clients are
31:06stunningly bad at providing real images.
31:08It's just like just take a bunch of
31:10photos with your phone and text them to
31:12me. That's all you need to do. Uh but so
31:15bad at it. So
31:17>> AI images are better than no images, but
31:21I would much rather have real images
31:22from the the client
31:24>> and and do you go on?
31:27>> Go ahead. No, I was gonna say if we are
31:28using AI images, we're going to make
31:30sure to add meta data to those AI images
31:34so that they look feel less like AI
31:38images, if that makes sense.
31:39>> Could you talk about that?
31:42>> Yeah. So, I mean, we'll put the geocart.
31:44So, if you take a a photo with your
31:46phone,
31:47>> there's a bunch of metadata is going to
31:49be added to that
31:51>> XF data.
31:53>> Yeah, X. I made I made an episode about
31:55that. Do you know Jay Huntley?
31:59>> I don't know Jay Huntley.
32:00>> Oh yeah, he has he is uh he's uh very
32:03deep in local as well. He did this crazy
32:05study on XF data. Uh but yeah, the study
32:10anyway study said that XF data doesn't
32:12actually make a difference. Um, but uh
32:16it's I think it's a I think it's very
32:18interesting that you that to try to I
32:20never considered trying to put it on AI
32:22images to to make the AI images look
32:25more convincing.
32:28>> Yeah. Well, I mean, you want it to feel
32:31like a real image.
32:33>> Yeah. Right. Do you do you ever get um
32:36do you ever try to get clients to make
32:38videos to record videos?
32:41Yeah, videos are amazing, right? Uh,
32:44especially for
32:44>> I bet your YouTube channel is ranking
32:46for a lot.
32:48>> I said, I bet your YouTube channel is
32:50ranking for a lot.
32:53>> One of my uh largest sources of traffic
32:56on YouTube is Google search. Yeah. And
33:00what's really Okay, so and you know this
33:02Edward, back in March, Ask Maps was
33:05rolled out where, you know, you go to
33:06maps.google Google and you talk straight
33:08to Gemini and obviously no one is doing
33:10that right now like like five people are
33:13but uh I take it to be a signal for
33:17where Google wants to take the map pack
33:19algorithm in the next who knows 6 to 18
33:21months cuz right now the mapack
33:23algorithm hasn't meaningfully changed in
33:25over 10 years. Uh but Google adding
33:28Gemini to the map pack algorithm is
33:31obviously a massive shift and Gemini is
33:34the only frontier model that was
33:36natively trained on video content or
33:39because obviously Google owns YouTube.
33:41Um so Gemini loves video content. So if
33:45you can feed Gemini video content,
33:47really powerful for a local business
33:49trying to get Gemini to recommend them
33:51because so few local businesses do video
33:53content. So 100% that's something that
33:56we we try when we're publishing local
33:58content uh we try to do a YouTube video
34:02for between a third to a half of the
34:05articles that we publish.
34:06>> Yeah. I like that a lot.
34:07>> Yeah.
34:07>> And if if the if the local business
34:10owner can't or won't or doesn't want to
34:12do it, no problem. Generally, we'll just
34:14use AI to generate those videos and then
34:17put those on YouTube. I mean, no one
34:18watches them. They're not for people to
34:20watch. just like, you know, my first
34:23couple dozen YouTube videos, no one
34:24watched them except my video editor. Um,
34:28but it's for Gemini. It's not for people
34:31to actually see.
34:32>> So, then will you put since they're not
34:34for people to see them, will you put
34:35those on the website or No.
34:38>> Yeah, absolutely. We'll uh upload the
34:40video to YouTube uh and then we'll embed
34:43that video link onto the actual article
34:45all the time.
34:46>> Okay. And what are you using um to for
34:49the AI? Are you doing AI avatars?
34:55>> No. Uh we're just using uh it's a
34:57service called Pictory and you know we
34:59can feed Pictory like a two or three
35:01minute script and it does a pretty good
35:03job. Like I mean they're not you know
35:05they're not going to win any awards but
35:07they they're not embarrassing if
35:09somebody watches them.
35:10>> That's cool. Something else. Um, I don't
35:12know if you've tried, this is probably
35:13still hard, but I don't know if you've
35:14tried asking like if uh if a business
35:17owner doesn't want to take video, maybe
35:19you could get them to just record
35:21something. You give them a transcript
35:23and you're like, "Look, just read this
35:24as a voice message. We'll do the rest."
35:30>> Yeah. And then use like an AI avatar.
35:31>> Yeah. Then you could do you could do you
35:33could do that or you could just have
35:34B-roll.
35:36>> Yeah.
35:37>> Yeah. It's a good idea. We haven't tried
35:39something like that. I like it.
35:40>> Uh, are there are there other are there
35:44local SEO tactics that still work
35:46despite [clears throat] everybody saying
35:48that they're dead that that you hear a
35:50lot?
35:55I mean, the way we do local SEO is
35:58incredibly simple, right? We're
36:01primarily just focused on writing
36:04quality content that isformationally
36:07additive. Uh we source quality links
36:10like you know we tell all of our clients
36:12to join the local chambers of commerce.
36:14How many they join is based on how
36:16competitive their space is. Uh to rank a
36:19personal injury lawyer in New Orleans.
36:21We once joined like 17 chambers of
36:24commerce. We joined every chamber of
36:25commerce within 80 miles. Um which was
36:29awesome. Uh cuz it worked. uh will look
36:32for sponsorship opportunities, you know,
36:35like youth sports leagues. For a few
36:38hundred bucks, you get a link from their
36:39website. Usually very powerful local
36:41websites like local charities. Again,
36:44the personal injury lawyer in New
36:45Orleans, we sponsored the uh the Bayou
36:49Bash, which is this annual festival.
36:51They were looking for sponsors is a
36:52couple hundred bucks. Awesome. Um so
36:55yeah, like that's the type of stuff that
36:57we're trying to do in local. How do you
36:59find these these backlink opportunities?
37:01How do you find them?
37:02>> How do you find
37:03>> Oh, how do we find them?
37:05>> Uh we have a prompt that we give to
37:06Gemini. Um Gemini is so much better at
37:09finding them than Claude or Chat GPT. I
37:12don't know why. Uh but if you give the
37:14prompt to claude or chat GPT, it's kind
37:16of middling, but Gemini does a
37:18phenomenal job of looking for these
37:20local organizations that are looking for
37:23sponsor sponsors. And you know, so we
37:26ran the prompt. One of our clients, they
37:28do a house cleaning in Austin and we ran
37:31the prompt and UT Austin was doing a
37:35TEDex talk and they were looking for
37:37sponsors and the minimum sponsor was
37:40like $250 bucks and for $250
37:43we got a link from UT Austin's.edu
37:48domain to our house cleaner. Uh like I I
37:51would have paid thousands for that. Like
37:52their rank went up like three or four
37:54numbers overnight just from that one.
37:57Yeah. And for
37:58>> Yeah. For $250 plus the business owner
38:01got to actually go to the TEDex talk
38:03because they were one of the
38:04>> That's applicable to any business.
38:06That's that's crazy.
38:09>> 100%. Yeah. 100%. And I couldn't put
38:11like I was I was, you know, Gem and I
38:13found it and I was just like, "Oh, a
38:14TEDx sponsor. They're going to want
38:15thousands." But it's like $250.
38:18>> So, okay. So, you know what? Like walk
38:21let's let's let's switch it up. Let's
38:22say you have a SAS business owner and
38:24they want to build backlinks. Walk tell
38:27them what to do to find backlinks.
38:31>> So SAS is different, right? SAS is
38:32non-local and we haven't done non-local
38:35in
38:35>> Yeah, but but you're finding these
38:37opportunities that are still applicable.
38:41>> Yeah, you're right. Like SAS getting the
38:43backlink from UT Austin, I I I'd happily
38:46do that for
38:47>> any business for e-commerce too.
38:50>> Absolutely. Yeah. So, I mean, we'd
38:53probably run some So, the the prompt is
38:55designed very specifically to look for
38:57local sponsorship stuff. I'd probably
38:59just take the same one and delete a
39:02bunch of the local references and just
39:04go find me sponsorship opportunities.
39:07Um, I haven't tried something like a
39:10Chamber of Commerce membership for
39:11non-local. Uh, it's still a powerful
39:14link, but for local, it makes sense why
39:16it's so powerful. Um, I don't know if I
39:19would like it might be interesting.
39:21Certainly worth testing if you know go
39:24join 50 chambers of commerce for your
39:26SAS and and see if your rank gets any
39:28better. I mean that like 50 of them you
39:30know most chambers are between $100 and
39:33$300 to join depending on the size of
39:35the city. So, I mean, you're talking
39:37it's not that much money, five grand,
39:39and yeah, that's a lot, but for a SAS
39:41who's like trying to make a go of
39:43ranking nationally, that's a relatively
39:46split testable uh budget to be thinking
39:49about. I I would I would try it.
39:52>> Where do most agencies waste time in
39:54local SEO campaigns?
39:59>> Uh, writing
40:02writing blog posts that don't provide
40:04topical relevance. Yeah. uh
40:06>> talk talk about talk about yeah I I also
40:08wanted to ask you your thoughts on just
40:09like blog SEO and informational keywords
40:12and top top of funnel searches but yeah
40:14so okay they most SEO agencies local SEO
40:18agencies are wasting their time on doing
40:20blog SEO that doesn't provide topical
40:24relevance so like what what does that
40:25look like?
40:28>> Yeah. So a great example again plumbing
40:31is if you write a blog post top five
40:34ways to winterize your water heater
40:36right that is not providing topical
40:39relevance for a plumber who wants to
40:41provide plumbing services. So, we had uh
40:44a client came to us. Well, a prospective
40:46client came to us and they wanted local
40:49SEO uh a lazy guy surgeon in downtown
40:52Chicago
40:54and they were getting 80,000 hits a
40:57month from Google search, right? They
41:00were paying for SEO and the agency they
41:03were paying was, you know, objectively
41:05doing great. 80,000 is a lot of traffic
41:08to come from Google search. But when you
41:10went and looked uh in the search console
41:12to see where they're getting their
41:13traffic, it was stuff like, you know,
41:15top five exercises to keep your eyes
41:18healthy or best foods to eat to keep
41:21your eyes healthy. It was stuff like
41:22that. And their rank map for like Lasic
41:26Eye Surgery Chicago looked pretty ugly
41:29because they were focused on
41:30theseformational non-local searches. So,
41:34when we took over, uh, their homepage
41:37had used the word Chicago once in the
41:40footer for their ad.
41:42>> That's crazy,
41:43>> right? [laughter]
41:44>> It's insane. Uh, so we obviously rewrote
41:48a lot of their homepage to focus on
41:50Chicago. We started focusing hard on
41:52building geographical relevance. So,
41:54we'd write all this content about these
41:56different neighborhoods. You know, I
41:57talked about the Montro uh River Oaks
42:01example, very similar except, you know,
42:02Wrigleyville and near the Willis Tower,
42:05etc., etc. Um, and their monthly traffic
42:08has gone way, way down. They're now at
42:10like 30 to 40,000 a month, but it
42:13doesn't matter because their rank map is
42:15green across all of Chicago. So, they're
42:17actually getting calls because they're
42:19showing up for local search. So, that's
42:22the challenge, right? If you're a local
42:24business, you don't care how much
42:26traffic you're getting. You care how
42:28many how many local people are finding
42:30you.
42:31>> Could talk more about that that case
42:32study on on what what was done.
42:36>> I that I like so yeah we redid the
42:40homepage like I said to focus on local
42:42and then we just started building
42:43geographical relevance. So when when
42:46we're looking at a rank map there um
42:48couple of things that we're looking at.
42:49The first one is and and it's percent
42:52top three is the metric that we care
42:54most about, right? So, uh because if
42:56you're in position four, that's the
42:57first loser. You're completely
42:58invisible. No one ever clicks more
43:00businesses. So, what percent of this
43:03rank map are we in the top three? So,
43:05the first question is, you know, is this
43:07rank map the right size? So, we look at
43:10the top competitors. Uh usually it's not
43:13the client because, you know, this is
43:15the first rank map we've ever run. Uh so
43:17we'll look at the top competitors to see
43:18what their percent top three is. And our
43:21goal is to see something like 60 70 to
43:2480 90%. Right? If it's much below 60%
43:28then we know that rank map is too big
43:30too big and we need to rerun a a smaller
43:33rank map. Right? Uh our rule of thumb is
43:36we can do at least as well as an
43:38existing competitor but we're never
43:40going to tell a prospective client that
43:42we can do better than has ever been
43:44demonstrated in their area. Right? So,
43:47we're not going to run this giant rank
43:48map where the top player is 30% top
43:51three and show the competitor and say we
43:54can dominate this space. Uh, and
43:56depending on the competitiveness, the
43:57how large the city is, the rank map
43:59might be huge and it might be really
44:01small. Uh, like we'll run rank maps
44:03everything from, you know, 5 600 square
44:06miles all the way down to like 30 square
44:09miles depending on how competitive the
44:11space is. always looking for the top
44:13players to be in that 60 70 to 90% top
44:17three. And obviously if the top players
44:19are 100% then the rank map is too small.
44:22We need to rerun it larger because the
44:23ranking outside the the rank map. So uh
44:26we need to zoom out and and run it
44:27again.
44:28>> Okay. So that's the first step to make
44:30sure that the rank map is the right
44:31size.
44:31>> What software are you using?
44:32>> Then
44:34uh usually we use not usually we use
44:36this software called lead snap uh to run
44:38the rank maps. Uh there's a bunch of
44:40other rankmap tools, but we started
44:42using lead snap because they have the a
44:45really great API connection with
44:47citations. A lot of other tools do now
44:50also, but at the time they were the only
44:52one that did. So that's why we went with
44:54them.
44:54>> Okay. So yeah, so next part.
44:58>> So next part. So uh then we're looking
45:01to try to cross we call it the topical
45:03relevance threshold. So the topical
45:06relevance is convincing Google that your
45:09business entity actually does that
45:11service entity. We're not trying to
45:13establish geographical relevance just
45:14yet. We're just trying to say yes
45:17Google, we are a plumber in Houston.
45:19That's it. So we'll um we we we say that
45:24when our clients top 3% is roughly half
45:29of the market leaders, we've established
45:31that trust with Google, right? So you
45:33know top players in the 80% range. So
45:37once our client is 40% top three cool
45:40topical relevance threshold met and we
45:43build topical relevance. The first step
45:45for topical relevance is of course the
45:46core 30. And if we're in the competitive
45:49enough space where the core 30 is not
45:51enough. We've built out the full core 30
45:53and we still haven't hit that topical
45:54relevance. Then we start building
45:56supporting content. But we build
45:58supporting content that supports their
46:01service entity. Right? So, well, it it
46:03would be something more like what what
46:06to think about when you're hiring a
46:07plumber in Houston, not top five ways to
46:11prepare your water heater for summer,
46:13right? So, it's always focused on that
46:17service entity and providing relevant
46:19content for the service entity. And keep
46:22in mind, right, we might have four or
46:24five different rank maps depending on
46:26the target services. So, for a plumber,
46:29we're always going to have three.
46:30plumber, uh, water heater replacement,
46:32and main drain line replacement. So,
46:35we're doing this process three times for
46:37each of those three main services we
46:38care about. We're not going to run rank
46:41maps for like unclogging a toilet
46:43because most plumbers don't want that
46:45call. So, so we won't track it. And most
46:48of the things like that, if you rank for
46:50plumber, you're going to rank for a lot
46:51of the stuff underneath it anyway. But
46:54once we hit that topical relevance
46:56threshold, then we start building
46:58geographical relevance. And that's
46:59where, you know, let me find the four,
47:01fives, and sixes, uh, where they're
47:03close to the top three, and then I'm
47:05going to build hyper hyper local content
47:08about a local landmark around that dot.
47:11Uh, because and obviously it's important
47:14to use Google Maps for that because you
47:15want to make sure Google recognizes that
47:17landmark. And uh, now we do it
47:20automatically with like a Google Places
47:22API call. Uh, for the same theory, we
47:25want to make sure Google recognizes the
47:27landmark. And if we're calling it with
47:29the Google Places API, then we know
47:31Google recognizes it because Google gave
47:33us that landmark name. Um, and then
47:36we'll just write about yeah, replacing
47:38main drain lines around that lake in
47:41Medford, Massachusetts.
47:43>> How are you doing internal linking with
47:44these posts?
47:47So, uh, it depends on the
47:50competitiveness. For larger, like more
47:53competitive spaces where we might have
47:55dozens of supporting content, we're
47:58going to create like a supporting
47:59content hub. Uh, and then we'll link to
48:02that hub either from the GBP landing
48:04page, maybe the footer, and then that
48:06hub will link out to all of the
48:08supporting content pages. and the
48:10supporting content links back to the
48:11service page we're trying to support,
48:13whether it be the water heater
48:15replacement service page or the plumber
48:17primary category page. Uh the reason for
48:20that is, you know, on the GBP landing
48:21page, if if we have a couple dozen
48:24supporting content pages, we don't want,
48:26you know, 25 links from the GBP landing
48:28page. So that's why we put that hub in
48:30between.
48:30>> Talk about AI. How has AI changed your
48:33SEO strategy over the last 12 months?
48:38So, so far it hasn't done a lot to
48:40actually change our strategy. Uh, it has
48:43significantly reduced costs. Uh, like
48:47our costs to produce an article and when
48:50I say produce an article, like it's
48:51obviously to write all the content, it's
48:53to place it on the website, do the
48:54internal linking, source an external
48:56link, it's do all of that stuff. Uh, our
48:59cost to do that is like roughly 80%
49:01lower than it was in mid 2022.
49:05Um, and AI has allowed us to write
49:08higher quality content, right? 2022,
49:13we couldn't legitimately spend a couple
49:15of hours doing research for every single
49:18article we wrote. But now, we can do the
49:21equivalent of hours of research with the
49:23right data calls and just have AI
49:25assemble everything and pull all the
49:27research and assemble an outline and
49:29then start writing. And we get a level
49:31of quality that would have taken hours.
49:34uh in 2022. Um so so far uh that's been
49:40the biggest impact is that we can write
49:42higher quality content at scale than
49:44what we could in 2022 at a lower cost.
49:47But we are preparing for this world
49:50where AI agents are primarily the ones
49:53reading everything. So I'm in the
49:56process of working through adding a
49:58fifth content type for local SEO. Uh I
50:01talked, you know, right now we have four
50:03and we've had four for 10 years. Um
50:06we're looking at uh building out this
50:09fifth type which we're calling trust.
50:12And the reason I feel like it's
50:13different now is for like a plumber,
50:17the average user would spend 30 seconds
50:19on your page. So you didn't really need
50:21a bunch of content about why that user
50:23should trust you because they weren't
50:25reading it. They weren't looking at it.
50:26But now we can make the assumption that
50:29an AI agent is going to read everything
50:31on your page. So having that trust
50:33content is suddenly quite valuable. So
50:36uh that's what we're working on. So it's
50:37going to be like pricing information, uh
50:40specific stories about things that have
50:42gone wrong in the past. We find that
50:45stories about things that have gone
50:46wrong are really really good at building
50:48trust. A lot of businesses try to hide
50:49that. Um a lot of information, you know,
50:52why service might not be
50:54>> talk about that a little bit. Oh yeah,
50:56that too. Yeah, these things. People
50:58wouldn't expect that. People wouldn't
50:59expect that. So So why this might not be
51:01for you, things that have gone wrong. So
51:04So you're saying include that?
51:07Yeah, 100%. Because right the So we know
51:11of course that when somebody opens up
51:13Google and searches for something, it's
51:14a personal search. Google knows your
51:16search history and it's going to show
51:16you things that it thinks you will like.
51:18Well, chat GPT is that like times a
51:22bajillion, right? It knows who you are.
51:24you've had all these conversations with
51:26it and you're not going to open up chat
51:28GPT and type in plumber Houston, right?
51:31You're going to open up chat GPT and
51:33say, "Hey, my garbage disposal stopped
51:35working. Uh, can do you know somebody
51:37who can try to fix it tonight?" And then
51:40chat GPT is going to go read all these
51:42different websites to find someone who
51:43can fix it tonight. So, if you have
51:46content on your website about, you know,
51:49how quickly you respond, how you can fix
51:51garbage disposals, and maybe you have a
51:53review, even better, if you have a
51:54review on your GBP, uh, that somebody
51:57said, "Hey, I called them about a broken
51:58garbage disposal, and they were there in
52:003 hours." Uh, like that those that
52:02attribute matching is incredibly
52:04important. So telling chat GPT who your
52:08service is not for uh is incredibly
52:11helpful at getting chat GPT to
52:13recognize, hey, the per my user I know
52:16that my user is none of these things. So
52:19this seems like would be a good fit for
52:21my user.
52:23>> And so you're finding that the SEO that
52:26you're doing is getting recommended is
52:28getting your clients recommended a lot
52:30within LLMs.
52:34>> Yeah. So um so we're triing a bunch of
52:37that trust content before we roll it out
52:39as our normal process. Of course we
52:41trial it uh what's the right system to
52:44put in place to write the content in
52:46such a way that it gets picked up. Uh
52:48but honestly on top of that there are a
52:50couple of other things that we started
52:52doing regularly. Uh one of them for all
52:54of our clients we do a monthly Google
52:56news press release. Um just a press
52:59release to make sure it gets picked up
53:01by Google News. Um, we knew the most was
53:04valuable. What are you using for press
53:05releases?
53:07>> We're using a service. It's called PR
53:09Underground. It's the least expensive
53:11one we found that regularly lands on
53:13Google News and that's all we care
53:15about. Like if it lands on Google News,
53:17I want the cheapest service. That's what
53:18I'm trying to get. Um, so yeah, we knew
53:22that helped with Google rank. Uh, but
53:24you know, it's not super inexpensive.
53:26It's, you know, $80 for each one. So for
53:30most local businesses, we didn't need to
53:33do that. Uh but we find that if they're
53:36regularly mentioned on Google news
53:38sources, chat GPT is much more likely to
53:40pick it up. Um so we'll do that every
53:43month for every client now. We add it
53:45like again helpful on Google, but uh we
53:48started doing it for chat GPT. And then
53:51for clients in larger areas uh like
53:55larger metropolitan areas, we use
53:57another service. Um, boy, I think it's P
54:00R A SAP, like press release ASAP, and
54:04they have the ability to get you into a
54:06lot of local publications like you can
54:08publish on the, you know, whatever the
54:10local Kansas City new newspaper is. You
54:13can pay like a couple thousand dollars
54:15and get a feature article in the local
54:17Kansas City newspaper. And again, it's
54:20helpful for Google rank. It's even more
54:22helpful to get recommended by the AIS.
54:25Um, so that's why we started adding
54:27stuff like that. But I mean, it's all
54:29local SEO. Like there's nothing that we
54:32do for AI that doesn't also help Google
54:35rank.
54:36>> Huh. Right.
54:37>> Are there any are there any uh local
54:40marketing tactics outside of SEO that
54:42you just really love?
54:47>> No. Okay. No, cuz I like sometimes I see
54:50I see just local marketing in general
54:52that I think is so cool. Like you have
54:54like branding for example and you have
54:55like consistent branding and you have
54:57these big bright red Vans for example
54:59and the way that the the way that your
55:01technicians dress and uh and the way
55:04that that people behave or talk on sales
55:06call like all of that stuff you know.
55:10>> Yeah you're right. uh a lot of the
55:11branding is very cool and you know
55:13meeting people's needs when they call
55:15you on the phone like being nice to them
55:17and uh because Google is tracking your
55:19calls right half of the US smartphone
55:21market is Android it's not like Google
55:24created Android out of the goodness of
55:25their heart like that data feeds the
55:27Google search engine to make local
55:29search better so we've seen this over
55:32and over again where if you know say one
55:34of our plumbers goes on vacation and
55:36doesn't answer the phone for 2 weeks
55:38their rank map looks
55:40noticeably worse after two weeks of not
55:43answering the phone. uh because Google
55:45is you know we call it goal completion
55:47and it's something that Google has been
55:49doing for over 10 years especially in
55:51the non-local space but even in the
55:53local space if you know uh the uh for
55:56non-local the easy thing to understand
55:58is ping ponging right so you you type
56:00something into Google you click on the
56:02third result you hit back you click on
56:04the second result then you close your
56:06browser right that's a really bad sign
56:08for the third result because you didn't
56:10meet the searcher's needs the same thing
56:12is true for local business, somebody
56:14calls you and you know you were really
56:17mean or rude to them or whatever and
56:18then they hang up and then they call the
56:21second competitor like one of the other
56:22GBP results, that's a bad sign for you
56:25because Google knows, hey, this user had
56:27to call someone else after they called
56:29you. So 100% like when when we bring on
56:32a local client, one of their contractual
56:34agreements that we put is that they have
56:37to have a minimum uh pickup rate where
56:39they can't have too many missed calls
56:41because we can't rank someone who's not
56:42going to answer the phone. Um and not
56:45just because, you know, the ROI is
56:47really hard if they don't answer the
56:49phone. That's obviously true also, but
56:52it's more that like you just can't
56:54consistently sit in the top three if you
56:56don't answer the phone. and answer the
56:58phone like in a real customer service
57:01ccentric sort of way. Yeah. So, I I
57:04would consider, you know, being nice to
57:07someone on the phone, making sure they
57:08don't call one of your competitors after
57:10they get off the phone with you. That is
57:12100% local SEO. Um, yeah. So, and and I
57:16mean, there are cool things that a lot
57:18of people have done in the local space.
57:20Uh, you know, the the Google uh uh LSA
57:24ads, like those are very cool. uh we
57:26don't do them and I don't know how to do
57:27them. Um whenever a client asks, I
57:30usually refer them to someone else. Who
57:31do you refer them to? But yeah, I mean
57:33there are a lot of uh you know there a
57:35lot of really cool Facebook ads that
57:36I've seen uh local businesses run that
57:39have really helped to get uh business,
57:41but again we don't
57:42>> who do you refer who do you refer them
57:44to for LSAs?
57:47>> Uh we have a I have a couple of uh other
57:49agency owners who focus on on those
57:51aspects who don't do SEO as much.
57:53they'll focus more on paid traffic.
57:57>> How do you
57:57>> because it's also very different like
57:59one of the challenges
58:01>> um so a lot of local business owners if
58:03they haven't done a lot of marketing
58:04before they're used to referral traffic
58:07uh and when somebody calls you as a
58:09referral it's a very different
58:10conversation than when someone calls you
58:12as hey they found your GBP online and
58:15then it's another step change in how
58:17challenging the conversation is if they
58:19found you through an ad. So one of the
58:22challenges with running ads is to have
58:24success with running ads. What we have
58:26found and it's not true for everyone of
58:28course but for a lot of local businesses
58:31they need support and help in terms of
58:33you know how to handle a sales call, how
58:36to handle follow-up, which is important
58:39for organic but less so and not
58:42important at all for referral-based but
58:45it's really hard to have a positive ROI
58:47with ads unless you know how to have a
58:49sales call. And most of the time, you
58:52know, those meta ads are just running to
58:54a receptionist or to the business owner
58:57and then they complain that, oh, your
58:59leads are terrible. Well, maybe, but
59:02also, you're horrible selling your
59:04service on the phone. And it's not their
59:05fault. They've never done it before.
59:07They've never had the training and
59:09experience to talk to someone who found
59:11them from an ad. Um, so it's usually
59:15like a much higher price point and you
59:17need to be able to provide the coaching
59:18and listen to the sales calls and we
59:21just don't have the um infrastructure
59:24set up to provide that. But yeah, a lot
59:26of like one of the the guys that I know
59:27that I refer people to, he uh literally
59:31has a staff that will like he'll run
59:34paid ads and his staff will answer the
59:37phone and and sell the the person
59:42calling and then once they're sold
59:44passes them on to the business owner.
59:46>> Um yeah, very cool business model. He's
59:49obviously much more expensive than
59:51hiring somebody to run your ads, but he
59:53actually gives you sold leads instead
59:56of, you know, a random call from from an
59:59ad.
59:59>> What would you say to somebody who
1:00:00wanted to have an AI answer the phone?
1:00:05>> So, what we would say like so right now
1:00:09having an AI answer the phone is better
1:00:11than not answering the phone. Um, and
1:00:14some of the AIs that can answer the
1:00:16phone are are quite good. Uh but you got
1:00:20to watch the data, right? If if people
1:00:23are calling and your AI is causing
1:00:26people to hang up, uh then you need to
1:00:29get a human to answer the phone or get a
1:00:31better AI because again, yes, obviously
1:00:35that's costing you the customer who
1:00:36called, but that's also impacting your
1:00:39rank position. If somebody called you
1:00:40from an organic
1:00:42listing and then hung up on you and
1:00:45called someone else, that's pretty bad.
1:00:48So, uh, you know, if you don't want to
1:00:51pay for like a human to answer the
1:00:53phone, I get it. Humans can be
1:00:55expensive, have an AI do it, but watch
1:00:58the data, listen to the recordings, get
1:01:00the make sure the AI is getting better
1:01:02and better until you don't see people,
1:01:04you know, hanging up on your AI.
1:01:06>> Is there any software that you would
1:01:08recommend?
1:01:11>> Um,
1:01:12we've used a handful of them. Some are
1:01:15are are good. uh some are less good.
1:01:18Usually it comes down to the prompting
1:01:20uh and the information that you're
1:01:22giving it. So I hesitate to say, "Hey,
1:01:25we use this one and it's great because
1:01:28like
1:01:30so now that a lot of my leads come from
1:01:32my YouTube channel, uh we service people
1:01:35in a wide variety of service niches and
1:01:39you know if we had you know whatever,
1:01:41pick one HVAC, right? Um, we don't have
1:01:45the data to be able to set up an AI
1:01:47agent to answer HVAC calls. Um, I have
1:01:51the data across thousands of calls that
1:01:54I can set up an excellent agent for
1:01:56plumbers because we've had so many like
1:01:59well over a 100 plumbing clients. Um, so
1:02:03I mean that's sort of it it's really in
1:02:05setting it up right and you just need
1:02:08data to be able to do that. So
1:02:11yeah, uh an HVAC person, I would say,
1:02:13yeah, we can't help you set up an HVAC
1:02:16AI call center, but we we can help a
1:02:18plumber set up an AI agent because we
1:02:22have the data, we have the prompting to
1:02:24be able to do that. Sorry, am I making
1:02:27sense?
1:02:28>> Yeah. And how do you how about uh how
1:02:30about like an agency on the other side
1:02:32of the world or something?
1:02:36>> Oh, um we haven't tried that. If I'm
1:02:39honest, somebody didn't answer the
1:02:41phone.
1:02:43>> Yeah, we haven't tried that. Um, I mean,
1:02:46I can't imagine it wouldn't work as long
1:02:48as one, the people answering the phone
1:02:51were trained well to be able to handle
1:02:53those types of calls and and second, uh,
1:02:56and I don't mean to come across as, you
1:02:58know, second, you know, the they should
1:03:01speak without a super thick accent.
1:03:04>> Yeah.
1:03:05>> A lot. But we've seen it like even with
1:03:08uh plumbers in the United States if they
1:03:11have a very thick foreign accent uh they
1:03:14struggle to close calls compared to uh
1:03:18an equivalent plumber who doesn't have
1:03:20that and and you know we can lament the
1:03:23state of the world but I mean that's
1:03:25what we see in the data.
1:03:26>> How do you how do you approach reviews?
1:03:30>> Yeah good great question. So we used to
1:03:34uh so the another thing that's uh in the
1:03:36in the contract um we rarely set up any
1:03:40sort of review management type software
1:03:42because the conversion rates the action
1:03:44rates on those are so low. What we're
1:03:46always going to try to do is to get the
1:03:49business owner uh or the service
1:03:51provider from whatever phone from
1:03:54whatever phone number they're used to
1:03:56talking to the customer on. uh we'll get
1:03:58them to text the customer from that
1:04:00phone number and ask for a review. Um
1:04:03the take rate on that is like 40 to 45%
1:04:07that's 10 times higher than a generic
1:04:09you know review management service.
1:04:12Um, but we did change after uh when Ask
1:04:16Maps started to come out. Uh, we started
1:04:19to, you know, dive into it and I've seen
1:04:21several other data dives on where Ask
1:04:24Maps is getting its attributes from. And
1:04:27one of those sources is of course GBP
1:04:29reviews, but Ask Maps seems to care a
1:04:32lot about non-GBP reviews. So the two
1:04:35changes that we made is first instead of
1:04:39having our client hey could could you
1:04:42you know if you enjoyed our service
1:04:43could you leave us a review and then the
1:04:45GBP review link that's all we used to do
1:04:48now we try to rotate review platforms uh
1:04:51so a GBP review a Bing for business
1:04:53review chat GPT uses Bing so the Bing
1:04:56for business review is good for chat GPT
1:04:58uh we'll do a Yelp review we'll do an
1:05:01Angie's review and then we'll come back
1:05:02to the Google review and And on top of
1:05:04that, the ask is also different. So now
1:05:07the ask is, hey, could you tell us about
1:05:10what happened and what the outcome was
1:05:12instead of just could you leave a
1:05:14review? Because what we want the user to
1:05:16put in are the attributes, right? We
1:05:18want the user to say they came and fixed
1:05:21my garbage disposal at 700 p.m. Right?
1:05:24That's what we want. Uh so but you can't
1:05:26ask for that. Um Google says you can't
1:05:29ask for that. Uh so we don't. We just
1:05:31say, "Hey, could you talk about what
1:05:33happened and what the outcome was?"
1:05:35Yeah. So, reviews are more important now
1:05:38than they used to be because of the AI
1:05:40agents. The AIS will read all of your
1:05:43reviews and look for attributes.
1:05:44>> What are mistakes that you see people
1:05:46making with review collection?
1:05:51>> So, um, even beyond just fake reviews,
1:05:54which I hate. [laughter] I've never used
1:05:57fake reviews. That's a really good way
1:05:58to
1:06:00Yeah, people are getting like Anyway,
1:06:02enough about that. Uh
1:06:03>> no, no, talk about it. Talk about the
1:06:05talk about the results with the fake
1:06:06reviews. Talk about what you've seen.
1:06:09>> Yeah. Well, I mean, uh, Google will
1:06:11remove a lot of them and then there have
1:06:13been several stories about people whose
1:06:16main focus are fake reviews, uh, getting
1:06:19caught and I can't remember by some
1:06:22government agency getting sued by some
1:06:24government agency uh, cuz fake reviews
1:06:26are, you know, they're so there are a
1:06:29lot of things that SEOs do that are
1:06:31maybe against Google's terms of service
1:06:34and at least as of now, Google's terms
1:06:37of service isn't law.
1:06:39So you can break Google's terms of
1:06:41service and if you get caught and Google
1:06:42gets mad at you then you know whatever.
1:06:45Um you might get your entire client
1:06:48website de-indexed.
1:06:51Um but you know fake reviews are
1:06:55actually against the law. So that's a
1:06:57very different bar. uh beyond which
1:07:00often they just don't work very well
1:07:02because Google wants to see reviews done
1:07:06by PE like somebody who lives 500 miles
1:07:09away and posted a review of a plumber
1:07:11just doesn't make any sense. So I mean
1:07:15it's hard to so anyway uh so fake
1:07:18reviews is the big one. Another one that
1:07:20we see all the time is review gating.
1:07:22I'm sure you see this too. um where you
1:07:25know it's hard to get caught, but we've
1:07:27seen people get caught. Review gating is
1:07:29against Google's terms of service. Uh if
1:07:31you get caught review gating, they'll
1:07:32typically scrub all of your reviews,
1:07:34which is pretty brutal. Uh and again,
1:07:36it's hard to get caught. A lot of people
1:07:38still do this, but you know, we we try
1:07:41not to do something risky like that. So,
1:07:43we don't review gate. And it's almost
1:07:45funny, right? So we've seen that the
1:07:48best conversion rates uh generally
1:07:50happen when your rating is between like
1:07:5247 and 49. Uh a perfect 50 rating
1:07:56generally has fewer calls. So I have
1:08:00given advice to clients before to like
1:08:02be really mean to the next few people
1:08:03who walk in. Like you have a perfect 50
1:08:06rating that's too good. We need to get
1:08:08some onestar reviews cuz you'll get
1:08:10better conversions. So, just be really
1:08:12mean to the next few people who come in
1:08:14to get get some one-star reviews. So,
1:08:17>> if you get bad reviews, that's part of
1:08:19doing business. Just get more good
1:08:21reviews.
1:08:22>> Yeah. Um, and then I mentioned the
1:08:24review management software, which yeah,
1:08:26it's easy, but the action rates on those
1:08:30are generally like 5%. Uh, so you just
1:08:33need, yeah, just have whatever phone
1:08:36number the customer is used to dealing
1:08:39with, have that phone number, text them
1:08:41and and ask for a review. You'll get 10
1:08:44times as many reviews.
1:08:45>> Did you see how uh Google is asking,
1:08:48does this business offer rewards in
1:08:50exchange for reviews?
1:08:53>> I did see that. Yeah, 100. And obviously
1:08:55that's also against their terms of
1:08:56service. So, uh, a lot of, you know,
1:08:58I've been to a restaurant and they'll be
1:09:00like, "Hey, if you leave us a review,
1:09:01we'll give you a free bowl of queso."
1:09:04>> Yeah.
1:09:05>> Yeah. And you got to be careful with
1:09:07that because if if too many people say
1:09:08yes, then uh, you know, Google might
1:09:11come and remove a lot of your reviews.
1:09:13>> So, you're also you you also talk about
1:09:16uh about how to build up an agency,
1:09:18right?
1:09:21>> Yeah. Sometimes. Uh, so if somebody lost
1:09:26all of their clients tomorrow, how would
1:09:28they rebuild, Caleb?
1:09:32>> Yeah. Uh, good question. So, a lot of
1:09:35that depends on how fast they're trying
1:09:38to rebuild, right? So, I know this
1:09:42sounds kind of silly, right? But the
1:09:44best way to get clients in my experience
1:09:47is to have started a YouTube channel
1:09:49like 2 years ago.
1:09:52Uh, and you know, obviously starting
1:09:54today is better than not starting, but
1:09:57it takes time to build up. I'm sure.
1:09:59Yeah, you've uh you've been there,
1:10:00Edward. It takes time. Uh, but when it
1:10:03starts working, it starts working like,
1:10:05you know, nothing else that I' I'd ever
1:10:07tried.
1:10:08>> So, beyond that, uh, you know, going to
1:10:12every single Chamber of Commerce
1:10:14meeting, um, not trying to sell
1:10:16anything.
1:10:17>> That's huge. But the people who attend
1:10:19chamber meetings are typically local
1:10:21business owners and it's typically the
1:10:23same group of people who attend over and
1:10:25over again. So just go listen to them
1:10:28talk about their business. Everyone
1:10:29loves talking about their business. Um
1:10:32maybe run rank maps, email them rank
1:10:34maps. Uh uh email them, hey, I got a
1:10:37couple of ideas. Change your title tag,
1:10:39change your H1. Um so many local
1:10:42business owners. So, we uh I have this
1:10:45series on my channel where I do like SEO
1:10:48in real life. So, I'll literally pull up
1:10:50Google, I'll do a Google search, and
1:10:52then I'll just talk about usually the
1:10:55number one ranked business, the number
1:10:57four ranked business, the first loser,
1:10:59and then somebody who's ranked in the 20
1:11:01plus range. And especially like the
1:11:04people who are ranked on the 20 plus
1:11:05range, the most common title tag I see
1:11:08is the word home because that's the
1:11:10default title tag for so many website
1:11:12builders uh for the homepage. And
1:11:15literally if you run a rank map and then
1:11:17change the title tag to be, you know,
1:11:20plumber Houston instead of home, you're
1:11:22go and then rerun the uh rank map,
1:11:24you'll see improvement with that 30-
1:11:26secondond shift. Um so just make friends
1:11:30with the people who attend. talk to them
1:11:31about how you do SEO, give them free
1:11:34advice, you know, make these changes, do
1:11:36these things. And again, it takes time,
1:11:38but you will be able to grow a local SEO
1:11:41business doing that. Um, and I suppose a
1:11:45third way, when we started in local SEO,
1:11:48we just started running ma ads. Um, but
1:11:50then, you know, you need an ad budget.
1:11:53Uh, we probably had spent 5 10k in ads
1:11:56before we had developed an offer that
1:11:58worked. Um I had a sales staff like
1:12:02people who are answering the phone
1:12:03taking these calls. Uh obviously if you
1:12:06do the calls yourself you can avoid that
1:12:08that cost. Uh but it takes a huge amount
1:12:10of time right like my my sales guy he
1:12:13was doing you know four or five calls a
1:12:16day um every day six days a week for
1:12:20months. Um, so just it's it it gets
1:12:23challenging, it gets difficult, but you
1:12:25know, we landed uh once we came up with
1:12:27an offer that works. Uh, we landed like
1:12:3197 plumbers in the next 6 months. Um, so
1:12:36meta ads can work, but you have to
1:12:38budget for it. You have to do your split
1:12:39testing. You have to figure out the
1:12:40offer. Uh, overall just, you know, being
1:12:44present, you know, building actual
1:12:46relationships, uh, where business owners
1:12:49go. Uh, that's probably the approach I
1:12:51would take if I were starting over
1:12:53because assuming, you know, starting
1:12:55over, you might not have 10K to drop on
1:12:58split testing meta ads to find something
1:13:00that works. And, you know, it's even
1:13:02training on yourself to get better on
1:13:03the phone, right? I talked about how
1:13:05hard it is to close people who call from
1:13:07an ad. And that goes for the agency
1:13:09owner, too. You have to learn how to get
1:13:11better on the phone. You have to listen
1:13:12back to your phone conversations. Um,
1:13:15you know, now you can give transcripts
1:13:16to AI and ask AI for advice. You know,
1:13:20what did I do wrong? What did I do well?
1:13:21What do I need to improve on? Um, all of
1:13:24these types of aspects. Yeah.
1:13:26>> Let me ask you, what what do you think
1:13:28separates a sixf figureure SEO agency
1:13:31from a sevenfigure agency from an 8
1:13:33figureure agency?
1:13:36>> Yeah. So, interesting. So, six figures
1:13:40you can usually get to five or six
1:13:42clients, right? I mean, that's $8,300 a
1:13:45month. Um, so a six-figure SEO agency is
1:13:50very easily can be a oneperson show. You
1:13:54know, maybe they have a developer or
1:13:55something like that or a VA, uh, but,
1:13:58you know, 90% one-person show. And how
1:14:02they got clients, you know, referrals,
1:14:04the chamber of commerce stuff I talked
1:14:06about, you know, just very local serving
1:14:09their local community, uh, doing the SEO
1:14:12things, right? So, um, getting to six
1:14:17figures is really about learning how to
1:14:19be consistent with prospecting and
1:14:23learning how to get better on the on the
1:14:25phone and and closing sales. So, now
1:14:29growing from 6 to 7, obviously, massive
1:14:32difference. Now, you need $83,000 a
1:14:34month. So, if your client is on average
1:14:382K, you need what? 42 clients in order
1:14:42to hit that figure. So going from six to
1:14:45seven uh is learning how to grow a team,
1:14:49right? Who do you hire? How do you hire?
1:14:52How do you train people? Uh like you
1:14:54figured out how to prospect. That's how
1:14:56you got to six figures. Now you need to
1:14:58bring a team on board and prospect even
1:15:01more, but have your team do a lot of the
1:15:03prospecting. Have your team manage
1:15:05clients. have your team uh do a lot of
1:15:08the fulfillment and going to eight
1:15:10figures. I don't know, Edward. I'm I'm
1:15:12not there yet.
1:15:14>> Be there soon.
1:15:18>> Maybe. Um but yeah, so
1:15:22growing above seven figures, a lot of it
1:15:24has just been, you know, how to make
1:15:26sure that you're not the bottleneck,
1:15:28right? And because even at even at eight
1:15:31figures with 40 odd clients, you can
1:15:33still be involved with every single
1:15:35client. You can't talk to all of them.
1:15:37Uh obviously, but you can still be
1:15:39involved. You can you can be at the
1:15:41strategy meetings, you can review what
1:15:43you know the plan is for every single
1:15:45one of them over the next month. But as
1:15:47you start to grow 50, 60, 70 clients,
1:15:50you just you suddenly have dozens of
1:15:53clients that you've never even heard of
1:15:55the client before, right? you you have
1:15:57no idea what's happening and your team
1:16:00needs to be just that much better to be
1:16:02able to manage them without even you
1:16:04knowing that they're a client of yours.
1:16:07>> Last question, Caleb, and thank you
1:16:08again. You are awesome. I hope we do
1:16:11many more of these episodes. This was so
1:16:13much fun. Uh last question. 5 years from
1:16:16now, what will the best SEO agencies be
1:16:20doing that the average agency isn't even
1:16:23thinking about today?
1:16:26So, okay, great question. I have to
1:16:30start by saying, you know, I mentioned
1:16:31the Sundar Pachai interview from a
1:16:33couple of months ago. One of the things
1:16:35he said in that interview, which was
1:16:36mind-blowing to me, is that, you know,
1:16:39obviously Google used to plan 5 10 years
1:16:42out into the future, but they don't do
1:16:45that anymore. Now, Google's long-term
1:16:47planning is 12 months in the future
1:16:50because AI is changing so much that when
1:16:52they plan more than 12 months out, they
1:16:54have to change the plan anyway because
1:16:56AI did X, Y, and Z and it got this much
1:16:59better. So, that changes their, you
1:17:01know, their two, three, four year
1:17:02projections. So, my first answer is,
1:17:05man, Google's worth $3 trillion and they
1:17:08can't plan more than a year out. You
1:17:10want me to plan five?
1:17:13Um but beyond that, like even thinking
1:17:1712 months, 18 months, 24 months, um it
1:17:21really feels like going toward a world
1:17:24where websites aren't visited by people
1:17:27very much anymore. Um that right now we
1:17:30write websites for Google's algorithm,
1:17:33we write websites for the LLMs, we write
1:17:35websites for actual people. And it feels
1:17:39like that's going to collapse to where
1:17:41we're just going to be writing websites
1:17:43for AI agents.
1:17:46And uh the AI agent is going to get
1:17:49whatever relevant information they need
1:17:51and then give that information back to
1:17:53the user. Uh so everything that we'll be
1:17:56doing is going to be optimized around AI
1:17:59agents. uh being able to easily
1:18:03see, understand and regurgitate the
1:18:07information that we're generating which
1:18:09you know uh quite different from writing
1:18:12websites for people and I've heard
1:18:15people talk about you know websites
1:18:17being like infinitely customizable uh
1:18:20we're seeing some of that now where you
1:18:21can build AI generated websites uh that
1:18:25serves as your Facebook landing page and
1:18:27then based on you know you can set you
1:18:30set up Facebook ad campaigns that are
1:18:32running only to a very specific
1:18:34demographic and then you set up your
1:18:37landing page to be focused on that
1:18:39demographic. Um, so that like ratcheting
1:18:42up so that websites are more and more
1:18:44custom, but you know that seems like a
1:18:47stepping stone to people just not
1:18:50visiting websites anymore. Um, so I
1:18:54think that's going to be largely what we
1:18:56look at in the future cuz and I had a
1:18:59conversation with someone else and they
1:19:01said, "Man, do you really think SEO is
1:19:03just going to be content and mentions in
1:19:055 years?" And if you step back, right?
1:19:09So if you're ever really bored,
1:19:12uh, having trouble falling asleep, uh,
1:19:15Larry Pa's original PhD dissertation
1:19:18from 1999 is available online. it's
1:19:21public and you can read it. And he sets
1:19:23out the foundation of Google search
1:19:25algorithm. And we all know this today.
1:19:28It seems obvious, but the incredible
1:19:32uh flash of genius that he had was to
1:19:36use links as a sign of your how quality
1:19:39your website is, right? Uh more links
1:19:42from quality websites means your website
1:19:44must be more quality. Quality websites
1:19:46don't link to spam. Uh so that obviously
1:19:50is the foundation of Google search
1:19:51algorithm. Humorously they tried to sell
1:19:54that algorithm to Excite for like
1:19:58$150,000 and Excite turned them down. Um
1:20:02anyway, so um from that foundation,
1:20:07Google has spent the last 27 years
1:20:11trying to because you they they work
1:20:12hard to get their own users, right?
1:20:14Chat, GPT, Claude, Google, they work
1:20:16hard to get their users. So now you have
1:20:20to build a machine that can recommend a
1:20:23trustworthy business to its user based
1:20:26on nothing but what that machine can
1:20:28find online. So yes,
1:20:32SEO is going to optimizing for those
1:20:35machines, whatever machine it is,
1:20:36whether you know, and I don't subscribe
1:20:38to GEO, AEO, whatever acronym soup you
1:20:42want. I I I it's all search
1:20:44optimization. Whether it's optimizing
1:20:46for the LLMs or for Google or for Bing,
1:20:48it's all optimizing for search. And for
1:20:50all of those tools, they're all trying
1:20:52to figure out how can I decide whether
1:20:55this is a trustworthy business based on
1:20:57nothing except what I can find online
1:20:59about that business cuz that's how they
1:21:02work. That's all that they can work on.
1:21:04So even in 5 years the question is how
1:21:07do we optimize so that your online
1:21:10presence is trustworthy enough that
1:21:12whatever that machine whatever you want
1:21:14to call that model that one is
1:21:16recommending you because of how it sees
1:21:18your online presence. I don't know. I
1:21:20don't know.
1:21:21>> That's a great answer. I I I want to ask
1:21:23aside from the backlink stuff that you
1:21:25talked about, are do you go hard with
1:21:27digital PR and yeah, you talked about
1:21:29press releases, you talked about
1:21:31backlinks, but are you doing any
1:21:32anything else?
1:21:35>> Yeah. So, you know, so the the PR
1:21:38underground for the monthly Google news
1:21:39reports and then the PAP,
1:21:41uh they're a little bit well, not a
1:21:43little bit, they're quite a bit more
1:21:44expensive. Uh, but for the larger
1:21:46metros, yeah, $2,000 and you can be in
1:21:48the Dallas Star or whatever the Dallas
1:21:50newspaper is called. Uh, really, really
1:21:53good for the LLMs for building that
1:21:55level of trust because, right, uh, Joe
1:21:59the Schmo doesn't get an article about
1:22:01him in the Dallas newspaper. Um, and you
1:22:06know, I I I mentioned briefly earlier
1:22:08how much I love multilocation businesses
1:22:10because it's much easier to optimize for
1:22:12Google's algorithm if you have multiple
1:22:14locations. It's even more powerful for
1:22:17the AI systems because a lot of like
1:22:21local businesses, a lot of their ranking
1:22:24is based on proximity, right? That's a
1:22:26major rank factor, just how far away you
1:22:28are from the searcher's location. But
1:22:31Gemini, the AI systems, they don't rely
1:22:34on proximity nearly to the extent. So if
1:22:37you have a 50 location business or a
1:22:41thousand location business like
1:22:42Rotorooer, how much easier is it to
1:22:45build trust for Rotorooer versus Joe the
1:22:49plumber? Right? So that's one of the
1:22:52challenges as you go to LLMs making all
1:22:54the recommendations is they like larger
1:22:57businesses so much more because they can
1:23:00find so much more information about
1:23:02them. So Joe the plumber uh is going to
1:23:06struggle to compete with Rotorooer uh
1:23:09unless you know the sooner they start
1:23:11getting their name out there optimizing
1:23:13their online presence the better it is
1:23:15for Joe the plumber uh when cuz yeah
1:23:18right now local businesses 90 95% of the
1:23:21traffic is still coming from the map
1:23:24pack the mapack algorithm same one that
1:23:27hasn't changed in 10 years um that's
1:23:29going to change
1:23:30>> maybe the answer maybe the answer is
1:23:31like what are Uh what our SEO is not
1:23:34thinking enough about right now is
1:23:37is viral linkable assets. And everybody
1:23:40thinks that a linkable asset is a
1:23:42calculator. It's not just a calculator.
1:23:44It's a stunt. It's it's it's something
1:23:46that's newsworthy.
1:23:48>> It's it could be it could be a research
1:23:50report. It it's but it's something where
1:23:52people are going to write about they're
1:23:56going they're going to write about the
1:23:57linkable asset. Sure, but they're going
1:23:58to write about your brand. And when you
1:24:01have something that's viral, that means
1:24:02lots of people are writing about it and
1:24:04making videos about it and talking about
1:24:06it.
1:24:08>> Yeah, 100%. Like, what could you do in
1:24:11your local community to get your like
1:24:14maybe it's not the Dallas newspaper, but
1:24:16like for example,
1:24:18um my business a couple years ago, we
1:24:21had an article that was about my
1:24:23business in the Huffington Post, and we
1:24:26live in a relatively small city. So, I
1:24:28reached out to the business editor of
1:24:30the local newspaper and said, "Hey, I
1:24:32thought this was pretty cool. I thought
1:24:34you might be interested in this." And
1:24:35then they wrote a feature about my
1:24:38business because it was featured in the
1:24:39Huffington Post. So, if you're a local
1:24:42plumber or a local business owner, you
1:24:45know, what can you do that might get you
1:24:47noticed even by your relatively small
1:24:49local paper? Like all of these mentions
1:24:52are incredibly valuable to building the
1:24:54online trust basis.
1:24:55>> Caleb, you are awesome. Thank you so
1:24:57much for coming on the show again. This
1:24:59was so much fun. You shared so much
1:25:02great stuff and uh yeah, thank you.
1:25:07>> Yeah, it was a blast, Edward. Really
1:25:09honored that you asked. Thank you so
1:25:10much. Um I'll I'll uh I'll have to have
1:25:13you on my channel and we can have uh
1:25:14more conversation back when I'm
1:25:16>> Yeah, that would be fun. And and where
1:25:18should everybody find you?
1:25:21>> So, luckily I have a weird name, Caleb
1:25:23Ulu. Uh u ku. So, you can find my
1:25:26YouTube channel, uh, my website,
1:25:27calebalcu.com. It's it's it's hard to
1:25:30get me confused with someone else.
1:25:32>> Once again, thank you. This is episode
1:25:341,85
1:25:36of the podcast. 1,085
1:25:39days in a row doing this show. Let's go.
1:25:43If you watch this on YouTube, thank you
1:25:45so much for watching. If you listened on
1:25:47Spotify or Apple Podcasts, thank you so
1:25:50much for listening. And I will talk to
1:25:52you again tomorrow. Bye now.

About this transcript

This page contains the full transcript of The 30-Page Local SEO Hack Google’s AI Is Already Rewarding by Edward Sturm, generated from the public captions YouTube serves with the video. The transcript has 15,242 words across 2,180 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.