Copy the formatted transcript to paste into ChatGPT or Claude for analysis
Today I'm going to show you how to rank
in Google's AI mode and chat GBT. And
yes, classic SEO still matters, but
ranking a brand in AI generated
responses requires you to master new
variables that didn't matter before. And
for this training, I picked a random
personal injury lawyer to show you how
AI search optimization actually works.
Also, check the pin comment below to see
how you can get access to my new free AI
SEO and geo checklist. Let's get right
into it. First thing I want to look at
is in the local pack very specifically.
And you can see there's already some
problems here. So, first when we look at
Steinberg, you can see they're
absolutely crushing it. They have 487
reviews, which is double, triple the
next best competitor. Okay. And you can
see the top three. There's a couple
things going on here. We've got the
proper category, which is personal
injury attorney. We can see the website
that we're working on, the law firm that
we're helping here. They do not have the
proper category. So just a a quick
category switch could probably give them
a nice lift. We're talking five maybe 10
spots in the local pack just by
switching category. It is that
important. Aside from that though,
they've actually got a really good
address. They're they're actually in
Goose Creek. So they're really solid on
that. And their phone number aligns with
that location. Okay. So you can see
every single one of these top three is
actually located in Goose Creek and they
have that 843 number. So, when you're
comparing performance here, the only
significant gap we're seeing, at least
in the local pack, is is this review
gap. Okay? So, 87 reviews doesn't even
match the third best competitor, let
alone the top one. Okay? So, when we see
a competitor that has nearly 500
reviews, we've got some work to do.
Okay? And that's the only way that
you're going to be able to catch them is
you have to narrow that review gap. It's
so so critical on the local level. And
then we want to talk about the the
traditional SERs. And as you can tell
once again, Steinberg is crushing it in
the local SERs, okay? They're ranking
really, really well. And this is
something that's a very different than
before. Like your standard SEO used to
be all about ranking here. It was a good
thing. Um, but there there's kind of a
different reason for ranking in the SERs
now. It used to be for traffic. Okay,
that used to be the good thing. You'd
rank and you'd get clicks and you'd get
organic traffic and you would make
money. Not really the case anymore. uh
people are clicking a lot less and
Google has deprioritized these SERs like
they're getting harder to find when you
go to Google because Google has now
frontloaded the AI mode and that's what
it wants people to use right and even in
the case of local you know you're still
going those results are going to get
pushed down by the local pack anyway and
and a bunch of other features but what's
really important is understanding this
highly technical term that I promise is
not that technical okay the reason why
we need to still show up here even
though we may not be getting a lot of
clicks or traffic is because of this
retrieval augmented generation or rag
for short. Okay. And essentially what
this is is when we go into chat GBT or
any of these LLMs and you run a search
and you'll see that sometimes it will
there it will not go to the web to get
information. Sometimes it will just give
you an answer directly because there's
two things that happen when you when you
use chat GBT. When you run that query,
it's either going to go to the static
training information. So, whatever it
was trained on at that cutoff date, and
if it knows the answer, it's going to go
there, and if it if it already knows,
it's going to give you an answer. Okay?
If it doesn't know the answer, it's
going to go to the web. It's going to
retrieve some sources. It's going to
come back and then it's going to give
you this augmented response with all
this this context from live results.
Okay? That's why it's so important
because typically when it goes to the
internet, it's going to search first,
right? So, it's going to go to Google
and Bing and Duck.Go or whatever it may
be and it's going to look at the top
ranking results a lot of the time and
it's going to it's going to use those
results to generate that response to the
user. Okay, so this is this is the main
reason why classic SEO is still really
critical because you still need to rank
really well so that then you can
ultimately influence the answers that
are generated inside these AI platforms.
Okay, that's the main reason.
Now you'll see here this is something
that I highly recommend doing. Okay. So,
you can tell when we look at this
particular keyword here, which is
personal injury lawyer in Goose Creek,
that is a very old school type of search
query. This is not the way people search
based on my experience, based on what I
know so far. This is not the way people
search. Even anecdotally, I don't search
like this inside of Chat GPT. Okay? It's
much longer, multiple sentences,
sometimes a whole paragraph. Sometimes
I'm doing the voice and it's like a
dissertation. very different behavior
going on. And that's why when we start
to analyze the other AI platforms, we
can't just throw one of these this
traditional query that we're used to
into there and think that that's the way
people search. It's not likely. So, what
you want to do is you want to convert
that specific seed into something that's
much more realistic that someone might
use in conversational search. So, this
is an example from Claude. I actually
really like using Claude to create
these. These are really what you would
call a synthetic a synthetic query.
Okay, so basically you go and you
convert it to a realistic query that
someone might use. So you can see, you
know, personal injury lawyer in Goose
Creek gets converted to I need help
finding a personal injury lawyer in
Goose Creek, South Carolina. Okay,
that's still a little robotic. One that
seems a little more natural though would
be one kind of like this. I was in an
accident, need a personal injury lawyer
in Goose Creek area. Okay, that's much
more realistic and you can see even gets
a little more contextual even in some
cases. But the point is you want to grab
one of these and if you're going to go
and benchmark your performance in chat
GBT or claude or perplexity, you need to
use one of these more realistic types of
queries that are much more
conversational. Okay, so here's an
example of me running this through
Google's AI mode. Okay, as I mentioned,
you know, I showed you the web earlier.
This is, you know, web results. Now,
check this out in the actual navigation
here. Google doesn't even show web
results anymore. It's actually buried
here in the more section. So, to even
get to the web results, you have to
click the dropdown, then go to the web
results, and you've got your blue links,
right? And and the truth is, no one no
one's going there. Okay, let's just be
honest here. No one's clicking there. No
one's going to use that. Um, and so what
we need to focus on is appearing in this
generated answer. So you'll see some
people saying, "Oh, well, we should do,
you know, AEO, which is answer engine
optimization, so we can get this
citation here." Citations are just one
piece of the equation. And truthfully,
most people are not going to click
through on these citations because
they're already getting the answer that
they need. So the citations here are
that's not a good goal. That should be a
major last secondary goal. The primary
goal at this point for for optimizing
for AI search is to get the brand
mentioned in this actual response. So
once again, we can see Steinberg Law
Firm here and I'll tell you some of the
things they're doing to be here, but
this is once again proving that um when
we the type of query that you want to
use when you run these searches. Okay,
so in AI mode, you can see I used a much
longer query to ultimately see what
those results would end up being.
Truthfully, you probably could have run
the shorter one and the results probably
would have been pretty similar, but just
to get a more realistic picture of, you
know, how people actually search to get,
you know, just a better idea. Okay. And
then the same thing here. I gave a much
longer one to see the AI overview. Once
again, same thing. We've got the
Steyberg Law Firm, and this one's even
more extreme because we see Google's
actually referencing its own products.
So, the Google business profiles are
being mentioned here. once again proving
that if you want to do well on local in
the AI products within Google, which is
simply Google's AI mode and in the AI
overviews, you just need to have a
really good Google business profile.
That's it, right? So, you know, all the
stuff that I described earlier, that's
going to be the that's probably like 90%
of the the performance that you'll see
in these in the generator response um e
either here in AI mode or in the AI
overview. Okay, so your Google business
profile on local has become even more
important. It's always been important,
but it's even more important now. Okay,
so I want to talk about how Steinberg is
doing so well and specifically with the
Google business profile. Okay, so check
this out. So first things first is they
have a perfect address. Okay, so not
only are they on Goose Creek Boulevard,
but they're actually located in the
Goose Creek City as well. You could not
draw up a better address for Google
Business Profile. It's absolutely
perfect. They've got the 843 area code,
which is also perfect, and a personal
injury attorney category. This is this
is you can't do better than this. This
is as good as it gets. Okay. Then
they've got 335
more reviews than the next competitor.
Okay. That they're crushing them. So,
you know, if you want to compete with
them, you need to come with that same
level of velocity with review
generation. Okay. And the other one
that's really important is in the
description. Okay. I really like a lot
of the things they're doing here. This
is it's pretty aggressive with the kind
of uh entity injection. In other words,
keyword injection, but there's a lot of
things that here are actually really
really smart. So, I want to kind of
highlight this. So one thing that's
really important with the AI platforms
is understanding entities. Okay? And an
entity is a person, place or thing and
or even just a topic like personal
injury lawyer which is I guess is a
thing. But you can see here that I've
highlighted the different entities and
these associations. You want to start to
associate your entity with these
relevant ideas. So on the local level
there's it's very obvious we want
Steinberg Law Firm to be associated with
personal injury lawyers that thing but
also specifically in the area Goose
Creek South Carolina. Okay. So you can
see these different entities working
together which creates a picture of of
of relevance. Okay. And so you have to
deliberately do this. This is really
really important. And you can't just do
it you know in your blog posts. You got
to do it in all your descriptions across
all your directory listings, in your
press releases. There's all kinds of
opportunities to kind of shape this
entity association. Okay? And then you
add in the secondary topics. Car
accident lawyer, truck accident lawyer,
you know, XYZ. Okay? And you can see how
this all starts to connect and build
really really good relevance which you
know to no surprise they do really
really well in the AI platforms as well
because they've done a really good job
of associating their entity with these
topics along with the other variables
we're talking about. And then the micro
variables on the local level they're not
linking directly to their homepage
they're linking to that Goose Creek
page. Okay. So the Goose Creek you know
Google business profile is linking to
the Goose Creek landing page. They
respond to every review. Um, as I
mentioned, they inject relevant
entities. Be careful, okay? You can be
too aggressive here. So, just be easy.
There's nothing wrong with what they're
doing, but just make sure you're not
going too crazy. Like, they do mention
Goose Creek, uh, you know, three times
in the description.
The the AI NLP, you know, extraction,
you know, the the NLP is what the AI
platforms use to understand this text.
um they probably don't need that much,
you know, information to understand,
right? You don't need to keep slamming
the same thing multiple times. It's
probably good enough to understand just
after one instance. Okay? Uh and 95
years in business, they've got 52 images
and videos. They post two to four posts
per month on the Google business
profile. They, you know, they labeled
themselves open 24 hours and they've got
they're linking out to their four
primary social media profiles. And so on
the homepage or on the on the website,
you use that same as schema so that then
you can once again get that entity
association, right? Steinberg law firm
is associated with the with the Facebook
page and the Instagram and you know
whatever it is. Okay, so you can see
this this what they're doing is really
really tight as far as entities and as
far as entity association. So this is
this is like perfect. Okay, they're
doing a really, really good job. And
this is what you want to model if you
want to do a good job as well. Okay, and
just to kind of re-emphasize this point
a little bit more, um, you can just see
the domination that's going on here. We
can see this is chat GBT. We can see
this is Grock once again. And you can
see I'm using that long query and
they're just crushing it across the
board. And there's more reasons than
just the Google business profile and the
reviews, which I'll share here in a
second. Once again, here we go. This is
perplexity and then this is clawed.
Okay, so they're right here, highest
rated. You can see that specifically.
And then also here in in Perplexity,
they're number one. Okay, so very very
important. And this is this is the
byproduct of all of that work that
they've been doing with review
generation and really trying to fortify
their reputation and sentiment. Okay.
And finally, you know, the next thing
you want to be thinking about is uh how
do you reverse engineer this? Okay. So
when you see that someone's crushing it,
how are they doing it? Well, the good
news, the AI platforms kind of tell you
what they're doing. Um, and you can go
and just look at the citations. Now,
this is the part that's really
important. I I mentioned earlier that
there's some people saying like, oh, the
goal of doing all this like, you know,
GEO or AEO is, you know, to get in the
citations. I actually don't really agree
with that. I I don't think people are
going to click on citations. And in
fact, this what you're seeing over here
on the right hand side with Claude, I
had to open these up, right? They were
hidden. You couldn't even see the
citations. So, I had to expand them just
to be able to see them. But the the real
utility of the citations is so that you
can reverse Well, that's that's not why
they're doing it. But I'm saying for my
purposes and and as an SEO, we can see
the citations that are being used and we
can then go and look at those and see,
are we being mentioned in these
citations? Right? That's the idea. So,
we're getting kind of the intel that we
need and we can see, okay, they they're
on uh they went to Justia, they went to
Super Lawyers, they went to Fine Law,
and they also went to this KM Law SC or
whatever it is. Oh, and then there's
Yelp. Okay, so like 80% of these
citations that are used, we can also get
listed there. So, there's no reason that
we shouldn't just look at this as
opportunity to show up in the same spots
that our top competitors are. It's it is
actually that simple. Okay. All right.
Now, this is very important because I
showed uh why Steyberg is doing so well
for a couple reasons. Obviously, great
reviews, great reputation, but there's
something that's really fascinating. So,
I went and looked at their whole link
profile and I categorized every single
backlink that they have by the type of
backlink that it is. Okay? And as you
can see, there's a lot of very important
things here. Number one, editorial
links. These are basically any
nationalbased link, which would be like
getting a link on Huffington Post or
Forbes or something like that. They
don't have a lot of those, okay? There's
not a ton of those in their link
profile, but they have a lot of
localized links, about 13%. And those
localized links, they range. It's uh
sponsorships, uh scholarships, and then
some other stuff sprinkled in there.
Okay, so those are important. We should
clearly be pursuing those localized
links, but of course, it's a smaller
pool of opportunities. But you should
absolutely be trying to get as many
localized links as as you possibly can.
Okay. And then we obviously have the
directories. Directories as I showed
here are very important. We can see all
the directories that are being
referenced uh to ultimately generate
this response. Okay. So we need to be in
those directories and we also want to be
trying to get reviews on those
directories to get that review
diversity. This is very important for
the AI platforms. They do not just rely
on Google reviews. They're looking
across the board. So, they're going to
pull every directory and they're going
to basically bring those all into one
place and say, "Okay, Steinberg has
reviews on Google and Yelp and Justia."
And they bring that all together and
that's how that's able to understand
that, you know, Steinberg is is
something that's worth recommending in
their generated response. Okay. So very
simply, we should be on every directory
that our competitors on and we should be
trying to drive reviews on other
directories um and other platforms other
than just Google. Google's number one,
like hands down, not even close. But we
also need that diversity as well if you
really want to influence these
platforms. And then the thing that's
kind of crazy is the press releases.
Okay, they do a ton of press releases,
this particular brand, and they're all
uh very they have a lot of positive
sentiment about their brand. Okay, now
these press releases are published on
third-party platforms. This is not what
they're saying about themselves on their
own website. This is what uh what
they're putting out into the ether onto
the internet to ultimately market their
law firm, right? And of course, when you
control the press release, you can
control the narrative and you can create
positive sentiment around your brand.
Okay? So, the reason why this works in
my opinion is not because of the link
that you score from a press release. In
fact, the link probably has very very
little value. What matters is that it
creates this very interesting
environment where you're getting
positive sentiment. Okay? So like for
example, these are all the templates I
pulled from various you know headlines
that they use. But like Steinberg Law
Firm recognized as best personal injury
law firm. Okay. So someone gave them an
award, they put out a press release and
distributed that idea out there as much
as possible, right? Very smart. Okay.
Next one here is about just news about
the brand. Now this is very important.
You know, a Steyberg law firm opens up
new office in Goose Creek, South
Carolina. Okay. This is very important
because it it it trains the AI platforms
when they're actually using training
data and they're coming out with their
new uh version. It's it's creating
information that helps the AI understand
more about this entity. And I'll show
you how you can audit your entity
information inside of chat GBT here in a
second. But but this is very important.
You should be you should be feeding
these platforms and feeding Google and
all the other uh platforms trying to
influence as much as possible about your
brand. Okay, you can control the
narrative. Okay, press releases are
actually really useful for that
apparently. Uh, another one here, just
another kind of update about their their
firm. And then this one here, results.
Okay, results, results, results. So,
this is once again just kind of proving
that this controlling the sentiment and
the um and the freshness of that's
that's kind of swirling around the brand
is really really important. and
ultimately trying to dict and and trying
to uh create accuracy around the brand.
Okay, so so important. I cannot stress
this enough. All right, so it's not
about the link. It's likely about what's
being distributed that's influencing the
platforms. Okay, so let me quickly show
you something you can do. This is a a
little tool that I built inside of
ChatGBT. Okay, and it's very simple.
It's a free GBT. You just go go into
ChatGBT and just search uh geo brand
audit, okay? And you'll find it. And
this is really important because what
it's going to do is uh I've I've uh set
it up so that only pulls from chat GBT's
training data. So you this is not going
to uh it will not go and search the
internet and bring in live information
to enhance the response. It's purely
based on its the information that was at
the cutoff date. Okay. Uh and
specifically 40 which is their most
popular model. Um, it's also their
oldest model, I believe, right now. So,
the reason you want to do this is
because you want to see ultimately what
does the AI know about us. And you can
do this across all platforms, but chatb
is the most popular. So, you should
probably start there. So, it's very
simple. We're just going to go in here
and we can just run actually the same
exact one we've been talking about,
which is Steinberg.
Uh, make sure I get the right brand in
here. So, Steinberg Law Firm is the
exact name. And typically, if it's
local, I might add just like a little
local modifier. So, Steinberg Law Firm
in uh we'll just say Charleston,
SC. Okay. So, this just kind of makes it
a little more specific. I think it's
pretty important, especially when you're
working with local businesses where they
don't have a lot of brand recognition.
Okay. So, you run that through and it's
going to measure the overall brand
knowledge of this particular entity.
Okay. So you can see what we're trying
to look for is consistency and
ultimately we're trying to make sure
that this is all relevant stuff to what
this brand actually does. Now obviously
a little executive summary here but the
really what we're looking at is like
core brand knowledge. So you know what
is the core focus of what they do. Um
and then the one that I add in here
that's really important is also the
entity association. Okay. So this is
basically that the platforms know what
this brand actually does. Okay? And I'll
show you an example of, you know, a a
gap that's not good. Uh which is like
our software. Okay? But you can see you
can use this completely free. And I
think I'm using there's like 25
uh questions that are run through this
GPT that helps it generate these
results. Okay? So, keep in mind there's
also uh some checks in here to prevent
hallucinations as well. So, this should
be pretty accurate, okay? Should be
pretty accurate. And plus, it's free.
So, if you see any bugs with it, let me
know. Um, but for the most part, I've
we've run a lot of tests and it works
really really well. And it just gives
you a nice baseline for you can you can
use this and say, "Okay, we've got some
issues here with inaccuracy. We've got
some issues with how our uh brand is
being associated with various topics."
You get the point. Okay. But if we run
like let's say Rankability, which
Rankability was uh was very new when the
last cut off date for 40 was. So there's
almost no training information about our
brand because we hadn't launched really
yet at all. Um and if we run this,
you'll see it's going to be wrong. It's
saying, you know, we're a digital
marketing service uh that, you know,
does SEO auditing and it's just not
right. Okay, this whole thing is
completely wrong essentially. Okay, so
this is an example of like we've been
doing a lot of work since that point,
but it's still very very off. And so the
goal here is like if it's off right now,
you have to start working really really
hard so that then you know you can you
can only be in this training data and
that the AI is going to know more about
your brand. Okay. All right. So now
let's move on to the the geeky uh
technical stuff here about the website
that we're working on. So, I went and
ran a crawl on uh Screaming Frog. And
basically, right away, I noticed a few
challenges. Okay. So, and I'll show you
kind of what this looks like. So, we're
going to sort by color. And
unfortunately, I don't know if this
website went through a migration or
something, but they've got a a problem.
Okay. Okay. And the first problem, it's
actually a really simple problem to
solve, but it is a big problem, which is
uh you can see that they're not using
the subdomain, which is the ww version
of their uh domain. And then here they
are. Okay. Now, the problem with this is
you only need to have like you need to
have one preferred version of your
website. And typically, whatever that
preferred version is, all other variants
get 31 redirected to the preferred
version. So in this case, we've actually
got both versions that are being indexed
in Google. So we'll go to the detailed
Chrome extension. You can see there is
no canonical set up. There is no
redirect setup obviously. And
redirection is the correct move here by
the way. Um and you can see they're
they're both they're both available both
versions. Okay? And you can see that. So
this is a a serious technical issue that
needs to be resolved and it will cause a
lot of problems. And so first thing is
figuring out what is the preferred
version of the domain. Then you go and
you 31 redirect the non-preferred to the
preferred and that fixes the problem.
And if you do that essentially you'll
eliminate 26 pages that are creating
basically duplicate content across the
site when it's completely unnecessary.
Okay, so that's the first thing is just
like clean that up and that's going to
give some quick wins right away. All
right, next thing I want to look at is
on the topic authority level. So, we
know that they're going after Goose
Creek. And so, first thing we can do is
look at the competitor. Just run this
quick site search and then put Goose
Creek in quotes. And this is what I
typically do just to get a kind of
highlevel overview of like how many
pages do they have targeting this this
location essentially. Okay. And you can
also, by the way, if you want to get
tricky, you can actually run your
competitor through Screaming Frog and
then export uh their whole site map and
then you get a you can get a really good
idea of like, okay, they've got, you
know, 40 pages in about Goose Creek and
personal injury essentially. That can
give you a baseline for what you should
do. Okay, that's a little bit more
advanced. But if you want to get a high
level review, you can just do this. You
can say, "Okay, they've got at least 10
pages." All right, and then you just
compare that to yours, right? And we
look at this. It's it's uh there's some
problems going on here because like I
said, I don't know if they just went
through some sort of migration, but the
way they had this set up was actually
correct. Uh they had, you know, they had
the the core uh service offering and the
core location and a dedicated page
specifically for it. But for whatever
reason, when we go to this, it redirects
to the homepage now. And this can be
very problematic because now we're
sending irrelevant pages and signals
back to the homepage when we want these
signals to be hyperfocused and and these
pages be hyperfocused on one topic,
right? And so I I would personally I
would try to roll this back out so that
these pages still exist because this is
going to be way more effective for
ranking. So, you know, obviously this is
not my client by the way. So, if I don't
even know if they're going to see this,
this is more just for demonstration
purposes, but this is not the right
move. Um, and it's going to cause some
some lack of performance. Okay, it
actually was set up correctly. So,
that's a part on the topical authority.
Um, now let's talk about the
architecture. So, this is uh what I'm
calling an architecture upgrade. So, you
can see here when we look at this, look
at the title and the metad description.
We can see they've there's a lot of
duplication. Uh, it's not actually very
relevant. like they don't even mention
Goose Creek in a lot of places. Um, and
it's got this kind of old school, you
know, 2012 type of setup. So, this is
not you don't need to do this and it
just doesn't really work. Um, and then
on top of it, the metad description,
there's tons of duplication. So, the
good news is that uh, you know, AI can
make this a lot faster. And so,
basically, I'll show you what I did. All
right. So we can see here with the
architecture upgrade what you want to do
is first get the practice areas in
place. This could be if you're an HVAC
these would be your you know your
service offerings. Okay essentially. So
you put those in place. Those are your
broad categories and then your core
location that you're trying to go after.
That's it. That's all you have to do.
Now we're going to look at the old URLs
and we're going to see they've got that
you know.php.
We want to get rid of that. Um and then
just have nice clean URLs. So, funny
enough, um this isn't necessarily the
way that I would personally structure
URLs, but I let the AI do it and it's
it's fine. Like, you could you could run
this. The key to having a good URL
structure is like only one objective.
You want to have the core topic in it
and you want to have the location. This
is if you're doing local, you got to
have both. And you'll see some brands
that like they'll just have like, you
know, slash wrongful death. That's fine.
But the the thing is it imagine you're a
bot and you took away any context and
all you saw was wrongful death. Well,
that's not exactly super helpful, right?
So, you want to try to make it as clean
and as natural as possible. And based on
what I'm seeing here, like if you remove
the root domain, and I saw SC, Goose
Creek, and wrongful death, those three
entities together, it would be easy to
kind of conclude this is likely a
lawyer. Okay? you could certainly come
to that conclusion very quickly as
opposed to just having like slash goose
creek, right? That's not enough context.
So the URL creates context, creates
relevance. And so this is actually a
fine structure. I typically like to go
right off the root. It it doesn't
matter. All that matters is you get the
core topics in the URL. And you can
slice and dice it however you want.
Okay. Um and then we look at the old
title. Not exactly a great setup. I'll
go and zoom in here a little bit.
Not exactly a great setup, but then this
is what the AI produced for us was Goose
Creek SC. This is a I don't really like
this format personally, but once again,
I wanted to show you like raw AI stuff.
Um, I would probably make it more
natural, which would probably be like
construction accident lawyer in Goose
Creek SC. Okay? Like keep it kind of a
natural language a little bit. Um, but
once again, this can work. Like this
absolutely can work because we still
have the core location and the core
topic. So metad description, same thing.
Old meta, now we've got unique meta
descriptions. Okay, completely unique.
Good to go. All right. And so how did I
do this? Very easily. Chat GBT go in
here and you just say something very
simple like this. I need you to revise
the URL structure, title, metad
description to maximize SEO performance,
and please put it in a table. So then I
just slap it in there. Okay? And you can
pick whatever model. For this one, I did
04 mini. Um, but you'll see within a few
seconds
already optimized it pretty well. Okay,
I'd say this is like a solid 8 out of
10. Okay, not perfect, but can get the
job done, right? This can get the job
done and you don't need to overthink
this. Okay, so once again, very simple
way to kind of clean this up. And
honestly, at this point, there's just no
good reason to have duplicate
metadescriptions because chat GBT or AI
can just easily make them slightly
unique. Okay. And then the next thing I
want to do is now that we've got the
architecture all squared away, we want
to go and start looking page by page and
looking at the relevance and the quality
of those pages. Now, of course, I'm very
biased here, but I use Rankability to do
this. Okay, so very simply, I ran their
uh their most important page through
Rankability specifically for personal
injury alert in Goose Creek. and I
wanted to see how well it was optimized
and it was not. Okay, so they're they're
missing out on a lot of core topics that
their competitors are talking about.
Okay, so for example, when we look at
this like they're not talking about dog
bite. All right, and we go over here
examples and we can see that almost all
of their competitors are talking about
dog bites. Go over here, almost all of
their competitors are talking about
truck accidents. Okay. Um and so on and
so forth. So these are not just for SEO
purposes. This is also for business
gaps, okay? And more than likely they do
this like they actually do truck
accidents and slip and fall and all
those things. But the problem is they're
not really talking about it in the way
that they should. And so these are gaps
in relevance that they need to narrow.
So what I would do, let's just say this
was my client. I wouldn't try to like
look at this and like take one top and
try to like inject it in here and like
go back and forth. I think that's just
not a good idea. Sometimes it's better
just to kind of build from scratch. So,
let me walk you through how I would
actually go about this. So, the first
thing I would do is I would go into
Claude or any really any AI platform
that you prefer. And here's what I will
do. I'll use this very simple prompt.
So, essentially create an AI content
prompt for rankabuilders AI writer for
the keyword and then I'll put in the the
seed that matches the intent. In this
case, it would be commercial intent and
is positioned to sell this business type
products or services. So, this would be
personal injury services. Okay. It
should be no more than 6,000 characters.
That's because the prompt window inside
of of Rankability is uh 6,000 character
max and just some other details that it
tends to do uh when it comes to Claude.
Okay. And then the most important part
is right here. Okay. So, some context.
So, you want to go to that law firm,
which is what I did. I went to the law
firm. I looked at their about page and I
grabbed a bunch of information about
their brand. Okay? And this is really
important when you're generating AI.
When you're using AI to generate
commercialbased content, you want it to
be highly relevant to the brand because
you're trying to actually sell some
stuff. So, it can't just be generic, you
know, AI junk that gets spit out. You
want to make sure it's highly relevant.
Okay? So, that's what you do. Then,
you'll see that it will generate this
nice little, you know, AI content brief.
You copy this AI content brief and then
you just go into rankability.
And the way this works in rankability is
you actually enter the keyword in the
content optimizer. And when you click
run optimizer, it just gives you the
choice. You can do it manually or you
can use the AI to create the first draft
for you. Okay. So this is in the case of
running AI and you can see I just picked
service page and I put all of that claw
generated context inside of this context
window. Okay. And then it's going to go
and produce this content and it will
consider this context here in the
context window. Uh and it will also uh
what happens is we actually go and crawl
the top ranking results and extract the
relevant topics. And so this draft is
created based on what your competitors
are actually talking about. Okay? And so
you can see here this is the first draft
that it produced. And you can see right
away the score if we look at the
original one which was 14. Very little
coverage on the core topics. This one
here is extreme coverage across all the
core topics. Okay. And you can see that
this particular you know asset is not
this is one thing that makes our AI
generation unique is that it doesn't
just spit outformational content. It
understands the intent. So the intent
here is is commercial and so the intent
is to sell. Okay. So you can see that is
how this is actually structured. It's
it's built to sell and this is the
advantage to using you know you know in
particular like I said I'm very biased
but I'm very proud of what we built
because it keeps that relevance tight
and if and it matches the intent. That's
always been my dream is to have an AI
you know content tool that actually
matches intent. And so that's what this
does. And on top of it, it automatically
optimizes the content, puts in all the
appropriate entities and relevance. And
so, it's pretty cool. And now, keep in
mind, this is a first draft. Treat it
like a first draft. If you're going to
uh I would not just take this and slap
it on your website, okay? I would have a
human come in here and run some human
edits through it, make it relevant, do
what you need to do, and then push it
live to the site. Typically for us, most
AI content is going to go through three
to four rounds of edits before it goes
live on the site. Okay? But this, as you
can imagine, if you're if you're
building out, you know, if we go back
and look at the um the architecture that
we were building, you can see in the
architecture, we actually have to build
15 pages. Okay? So, imagine having to
write 15 unique assets. Um, and
truthfully for these commercial pages,
I've created an ungodly amount of AI
content um, even before chat GBT came
out. And I can tell you now, the
difference for commercial pages like for
a human writer versus AI is pretty
negligible at this point. Okay, most if
you use a good prompt, use a good tool
like I just showed you, it's going to do
the job. Um most people in fact most
people using rankability a lot of them
on the local level are pretty much doing
it for commercial pages um because it is
so so effective at doing that. Okay so
that is just a very highlevel overview
of what you can do to perform well not
just in the local pack but also perform
well in Google's traditional results but
also more importantly in the AI
platforms. There is um so much more that
you can do that I am just scratching the
surface today and I don't want this
video to be you know 7 hours long but
there's some good news. I actually
created a checklist. So this checklist
here, this is version 1.0 and the next
version um actually probably you know by
the time you're seeing this video I'll
probably have already upgraded this to
the next version because I'm going to be
adding like tools and various things in
here. But essentially this checklist is
can walk you through all the core things
that are likely to influence the AI
platforms. Um I've already gotten some
feedback here. People want to know like
what matters the most. So there's going
to be uh some new things added to this
which is like the impact of that action
or the likelihood that it's going to
influence AI uh or even you know Google
for example. Um one thing you're going
to notice is like and and you've
probably heard this before that like
some people say that geo is just SEO.
They're not wrong per se because yes,
you need to do SEO. Like SEO, the
traditional way to rank in the
traditional SERs is still critical. You
should keep doing it. Keep doing
everything you're already doing. And
what you're going to find is a lot of
this stuff in here is still classic SEO,
okay? Just good SEO. But there are some
things that are new, okay? And the
things that are new is what I'm calling
like the extra 20% of things that you
need to be doing. And when we look at
this, we'll go down here. I would say
most the firstparty signals are pretty
like I'd say like 99% of the first party
signals um are standard SEO. Okay,
you'll you'll see that a couple new
things in here, but nothing really that
different. What's different is when we
get down here to the third party
signals, okay, and so this is where
things get very very different. And you
know, just to give you some perspective,
I have been able to rank all kinds of
pages that don't have any brand, uh,
don't have reviews, whatever it is. And
I've been able to rank them for the last
decade. Never once needed reviews for
the brands that we're we're trying to
rank. Um, and reviews have never
influenced traditional search results.
Uh, they influenced the local pack, but
traditional it didn't really matter. Now
it does. Okay? and not for traditional
but it matters for rank or for appearing
in the AI generated results. Okay, so
reviews are critical and you can see
I've got a whole blog and there's a
reason why it's the first block because
it is so so important. Um listicles are
huge. Okay, once again getting links on
listicles has always been a common link
building strategy. That's not anything
new but what's new is that you actually
don't even need a link in the listicle
to get the benefit. Okay, so that's
what's different. Obviously, get a link
if you can. It's going to be much more
beneficial. But even just getting a
brand mention in that listicle on a
website that's highly trusted also can
influence these platforms quite well.
Okay? And then so on and so forth.
There's a bunch of other ones,
obviously, UGC. I'm really just giving
you a bunch of opportunities that you
can think about, but the point is
there's there's some new variables that
we need to be thinking about. Um, and
the good news is this checklist is free.
So, you'll just have to go to uh the
there'll be a link below the video in
the pinned description. You can go enter
your email, get access to this free
checklist. Oh, and make sure you save
this checklist because all the updates
that I'm going to do are going to be on
this checklist here. So, I'm just going
to like add like every version will just
be this one checklist. Okay? So, every
new update, I'll I'll keep you posted on
when I do a big update, but but this is
like an ongoing project as I discover
more things that tend to influence. Like
I said, it's probably 80% classic SEO
right now that does the influencing. The
other 20% is the stuff that's a little
bit newer, the unlin brand mentions, the
sentiment, and all the other stuff I
showed you in this video. So, hope that
helps. Once again, link will be below
the video. And thank you so much for
watching. Appreciate it.

Neil Patel