April 23, 2025 • 43 mins
In this episode of Over The Bull, we pull back the curtain on the AI-generated content boom—and the hidden dangers behind it. From lifeless language to content “inbreeding,” we break down how agencies are quietly replacing human creativity with algorithms, and why it’s eroding trust, killing brand authority, and flushing your E-E-A-T score down the drain.
Plus: how to spot the red flags, what the “Habsburg Theory” has to do with bad blog posts, and how to use AI the right way—without losing your voice or your credibility.
Over The Bull is brought to you by IntegrisDesign.com. All rights reserved.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
SPEAKER_00 (00:00):
You're listening to Over the Bull, where we cut
through marketing noise.
Here's your host, Ken Carroll.
SPEAKER_01 (00:10):
How artificial intelligence and design agencies
are conspiring to give yourbusiness cancer on this episode
of Over the Bull.
You want to stay for this one.
Now I've said, and I stillcontend, and I'll stick with it,
that design agencies in largepart are out to create big
swaths of content that make youthink that they're doing a lot
(00:32):
more than they are.
Now, that theory, the reason Ihave it, first of all, is
because, frankly, it's true.
I can't tell you how many timesI've talked to search engine
optimization companies or evenlarge companies who are
mass-producing artificialintelligence-based content, and,
frankly, it's going to destroytheir company.
(00:56):
I know that's a littlesensationalistic, but if they
don't put it in reverse and getit figured out, it can cause
irreparable damage.
So let's talk about this alittle bit.
So obviously there is a rise ofAI content across the spectrum
of the internet.
(01:16):
Now we know it's true.
Some of you may be producingcontent or articles using
ChatGPT.
Some of you may be using closedAI systems.
Some of you may be using moresophisticated tools.
But the point is that that toolin itself and that process in
itself or agencies who are doingthat on your behalf are not
(01:38):
doing you any favors.
Now, to understand it, let'sdive into something.
I want to talk to you aboutEEAT.
Like the word eat, but with oneextra letter E.
Now, that stands for expertise,experience, authoritativeness,
and trustworthiness.
Now, this is the gold standardof quality content.
(02:03):
That means that it has to betouched by a human being.
Because remember, Google's goaland all these search engines'
goals are to provide the mostrelevant and meaningful content
when someone searches forsomething.
And so generic content that'sspit out by artificial
(02:23):
intelligence, although valuable,is one of those things that
Google is getting better andbetter at sniffing out and
determining.
So let's talk about agencies fora little bit.
The one thing you need to knowas a business, and this is
absolutely critical.
If you don't hear anything else,you need to hear this and
understand this.
(02:45):
Agencies are out to producecontent and do things with the
minimal amount of overhead.
This means if they can churn outgeneric articles, post them to
your website and not involve youin the process, this is better
for them.
Because it's less work.
It's less communications.
(03:05):
It's pushing a button,generating content, sticking it
on your website, and saying,hey, look what I've produced for
you.
It's a fraction of the time,effort, and energy, and
completely negates possibly themost valuable thing you can have
when you create content.
That is the EEAT factor.
(03:26):
So it's good for the agency.
It's bad for you.
It's almost, you know,meaningless content in some
respects.
So let's talk about this alittle bit more.
Let's move into the risk ofinaccuracies in AI-generated
content, especially when itdoesn't pertain to your specific
(03:47):
business in your specific areas.
If it doesn't relate to thepeople you're trying to serve,
it's useless.
Okay, it doesn't matter what itsays.
If it doesn't relate to thepeople that you're trying to
service in your target area,that content is useless.
So what that means is that youneed to be involved in the
(04:09):
content.
You need to be involved, why?
Because you are the expert ofyour company.
The agency is not an expert ofyour company.
And so, therefore, they couldcreate the best content in the
world using artificialintelligence.
Although technically correct,it's going to miss the EEAT
spectrum.
(04:30):
Okay, so also, just to throwthis out, when you look at AI
content, look, it's good.
We use AI.
We use it in very specific ways,and it saves hours of time.
But it's also often inaccurate,right?
Sometimes it doesn't relate tothe business.
Sometimes it puts in things thatdon't even pertain to the
(04:52):
business, but rather theircompetition.
You see, AI content, althoughpowerful, if it's not carefully
reviewed and edited, it can leadto all kinds of issues, maybe
legal issues, reputational risk.
All kinds of different issuesare associated with just
churning and burning AI contentby giving it a subject, saying
(05:13):
work in a few of these keywords,and then spit the article that I
can put on a website.
You do not want to do this andyou certainly don't want
somebody that doesn't understandyour business doing this.
So what's the magic answer?
The magic answer is using thepower of artificial intelligence
combined with human sentiency.
(05:35):
And that human sentiency usuallyhas to be partly you as the
owner of a business.
And the reason being is thatDesign agencies are not someone
who understands everything aboutall businesses.
Even niche-based design ormarketing agencies don't
understand your business.
(05:57):
They may understand all thetechnicalities.
They may understand all therules, but they don't understand
you.
You see, that's the missingpiece here.
You're the authority element.
You're that person that's goingto bring in that trust factor by
incorporating you.
into the process.
So one thing that we do with ouragency is we'll use artificial
(06:21):
intelligence to create anoutline.
And then what we'll do is we'llcreate the scaffolding of an
article.
And we'll let artificialintelligence rough it in.
But between the outline and theactual development of the
article, We actually injectURLs.
We inject different things aboutthe business.
(06:42):
We inject different processes tomake sure that it's close.
And then we also remove anythingthat's not relevant to that
business to the best of ourunderstanding.
Then once we create thatcontent, we pass that on to the
client.
We ask them to humanize it, toput in their experience, to put
in their personal energy so thatthat article becomes more, well,
(07:06):
human.
And then what happens issomething really magic.
The EEAT factor starts to go up.
Now it becomes more originalcontent.
You didn't have to go all theway over to the idea of creating
an article from scratch, but youwere able to use the content
that was created as a basis foryour humanized content.
(07:31):
And then that's what wepublished.
So a quick side note, there arealso powerful tools that allow
you to go in and actually lookat the best articles that are in
your area that are beingsearched for the most.
So you should never go in tochat GPT or one of these things
and just ask it to write anarticle based on something that
(07:52):
you think is good.
you can actually find out whatpeople are looking for in terms
of content and then buildmeaningful content.
So a little bit of meaningfulcontent is going to far
outweigh, you know, just tonsand tons of generic content
you're spitting out andbasically just guessing on
whether or not that content isgood content.
(08:15):
Now, here's the sad part.
A lot of agencies, they guessand They just look at your
keywords.
They'll try to generate a fewarticles based on those and just
keep feeding the beast.
But that's like feeding cancerwith sugar.
You don't want to do that.
Meaningful is always better.
(08:36):
Now, just like, you know, inhistory, you know, we tend to
look at people in antiquity andthink they weren't as
intelligent as us.
But let's hit the pause buttonon that for a second.
And let's look at Machu Picchu.
Let's look at the pyramids ofEgypt.
(08:56):
Let's look at some of theseancient structures, which would
be extremely difficult, if notimpossible, for our modern
abilities to reproduce.
Now, there's a reason thatthere's these rocks that are
really, you know, so tightlyknit together still after
centuries.
(09:16):
thousands of years where youcan't put a piece of paper
between the two rocks or thegeometric rocks that are not
squares.
You know, the reason we havesquare blocks is because it's
easy to produce.
These asymmetrical elements thatare in some of these structures
are much more difficult toproduce.
(09:37):
So we would do well by learningfrom our ancestors and not
having this idea that justbecause we have different tools,
that somehow we're moreintelligent or more insightful
than people that have lived onthis earth before us.
And this is really important.
So when you look at somethinglike these cautionary tales,
(10:01):
such as Icarus, When you look atthese, I think we should take
those and then carefully thinkabout those within the construct
of our business.
Now, if we look at it from thatperspective, one of the modern
things that have come in issomething called the Habsburg
(10:21):
jaw of content decay is kind ofwhat I like to call this.
And basically, the Habsburgwanted to, there were a regal
group And what they wanted to dowas maintain their purity.
And so what they did was theykept inbreeding with themselves.
(10:42):
And rather than increasing theirroyal lineage, they actually
bred themselves out ofexistence.
And one of the visuals of thatgroup was this kind of like
extended jaw.
And that's where you get thisHabsburg jaw.
So just like the Habsburg keptmarrying their cousins to
(11:05):
maintain purity and ending upundermining their entire
destiny, some businesses andagencies are now inbreeding
their content by relying tooheavily on AI without fresh
human input.
Now, the content that lookspolished but lacks soul is
(11:26):
something that that AI looks at.
And if AI content is going touse AI content and get in this
feedback loop, then the contentbecomes more and more generic
without human input.
So search engines, although youthink they may want to play the
game of artificial intelligence,They do to an extent, but they
(11:50):
also understand that humansentient expertise is much, much
more important and more valuableto the reader than just this
churn-and-burn style content.
So the modern-day Habsburgtheory, which is a little
tongue-in-cheek, a little bit ofphilosophical debate— More just
(12:14):
fun stuff to talk about ispossibly a big warning sign
where, you know, the ancientsirens would call out to sailors
and lure them to the depths ofthe ocean to their death.
Perhaps this is something thatis a warning for small
businesses and businesses of allsizes.
(12:35):
that if you create a model ofartificial intelligence to whip
out content, then maybe you'repushing yourself into the depths
of your own demise for yourbusiness and the people that
trust you to keep your businessafloat.
So that's real strong when inreality, most businesses just
(12:55):
want to have a product, sell aproduct, and then push on.
Now, as far as gosh, I don'tknow, 10, 15 years ago, in our
industry, the marketing anddesign industry, it was already
being pushed that journalismdegrees would be huge.
They would be phenomenal.
And the reason being was thequality of content.
(13:18):
And the idea is that everybusiness kind of becomes their
own outlet for that content,which is true.
So if you're going to play thecontent game, you need to play
it right and not pay someonejust to kind of keep cranking
out content over and over again.
So if we really look at it, theexpertise and experience, it
(13:38):
just can't be fake artificialintelligence.
And it can't be faked byagencies pretending to know your
business.
I mean, I spend probably, Imean, with coaching and
retainer-based clients, I spendtime with them every other week,
once a month, and I truly try tounderstand and document things
(13:59):
about their business.
And I still...
and not the expert of theirbusiness.
And I don't pretend to be.
See, it's really liberating whenyou go in and you don't say, I
am an agency that specializes inelectricians, or I'm an agency
that understands you so muchthat I can create content and it
sounds and acts just like you.
(14:20):
You see, I'm putting myself in aplace where I can't do that
because the minute that businesswalks out, I'm talking to a
bookstore, or I'm talking to aplumber, or I'm talking to, you
get the point, I'm talkingtalking to different businesses
and to go through that deal withhundreds of businesses and then
still think that somehow I canfocus on your niche business in
(14:43):
your area is frankly impossible.
Now, we do our best, but we loveclients who participate.
Clients who participate alwaysdo better because their
resonance is sent out there.
Now, that sounds a little hokey,but it's true.
Because when people read yourword, they read your expertise
(15:03):
that goes into it, it helps.
And so part of what we do is wedon't really care if a client
understands the best grammaticalrules.
As a matter of fact, I workedwith a particular person.
I won't mention her name, butshe is huge in the copywriting
industry.
I mean, absolutely phenomenal inthe copywriting industry.
(15:25):
And I remember one time, I askedher about a sentence structure,
and I can't remember if it was acolon or a semicolon, but I
said, you know, you reallyshould have this here, I think,
in order for it to be accurate.
And I was kind of shockedbecause we're talking about a
person who$3,000 for a half hourof her time and has written
(15:47):
stuff for very largeorganizations.
And she basically, with theconfidence that she had, she
goes, I really don't care aboutthe grammar in some cases.
I care about the pace.
And she said a colon causessomeone to pause.
And I don't want them to pausein that instance.
(16:08):
She inserted an ellipse, youknow, that dot, dot, dot, and
then continued.
Now, the reason she did it wasbecause it kept the pace of the
content going in the right way.
You see, that's expertise.
That's something that's kind ofhard to fake.
Now, granted, I get it.
You're watching all these thingswith, you know, songs being
(16:30):
reproduced and all this otherstuff.
And you're thinking, well,surely they could reproduce her.
Yeah, but can they reproduce herexpertise?
Can they reproduce all thethings that go into that human
sentient experience to createthat content?
So don't let grammar or beingimperfect with grammar get in
your way.
Matter of fact, I tell ourclients, don't worry about it.
(16:51):
That's not your job to begrammatically sound.
Your job is to give me thathumanized factor and then let me
take it and I'll run it over thegoal and I'll polish it up a
little bit.
But you see, that's the process.
It's not a churn and burn.
Okay, so let's talk about thepeople that are working with
you.
Now, I get it.
(17:11):
Businesses, and I'm the sameway, I am jealously protective
of my clients.
And the thing about businesses,most of them, they're really
kind of protective of the peoplethat serve them, that agency,
that freelancer that comes inand sees them and shakes their
hands and smiles and talks aboutthe ballgame.
(17:33):
I get it.
You like your agency.
You like your freelancer.
That's okay.
There's nothing wrong with thatunless it inhibits your business
ability to be successful.
Remember, you're not running acountry club.
You're not running something tobring friends in and joke.
You're trying to grow yourbusiness and protect the people
(17:53):
you work with.
So let's talk about some flagswhen it comes to agencies that
may be using artificialintelligence-based content only.
The first one, every articlesounds the same.
The tone, the structure, and thevocabulary feel eerily similar
(18:14):
across blogs or pages, even fordifferent topics.
You see, the thing is, is thatartificial intelligence
generates content, but it alsokind of brings in that tone that
is similar.
Now, some of the more advancedtools that we use when we're
creating those outlines and thatscaffolding for an article, so
to speak, we can actually put inthe ability to alter the tone
(18:38):
that we give it.
But even at that, it's not goodenough.
And so don't let anyone tell youthat they can produce content,
create the right tone for thecontent, send you the content,
and then that content ismagically great.
No, it's not true.
There's no tool that replicateshuman sentiency.
(18:59):
The next one.
vague language with nospecifics.
You know, those articles thatsay things like it's important
to have a strategy withoutexplaining what that strategy is
or giving real examples.
So, for example, if I were tonot give you our example of how
we build content, then thatwould be more generic.
But I'm giving you a real worldexample of how we actually use
(19:22):
AI content with the client toproduce something meaningful.
There's no first-hand experienceor personalization in AI
content.
No mention of things like casestudies or customer feedback,
employee quotes, or thatbehind-the-scenes process.
Again, podcast is a goodexample.
(19:44):
I'm giving you content that I'mbasically rattling off based
upon what I run into each weekat our design agency.
And this is one of the abusesthat I've noticed this week, and
I'm simply sharing it with you.
But I'm doing it in more of ameaningful way that is bringing
my personal experience, my over30 years experience, into the
(20:07):
conversation and sharing thatwith you in hopes that you
properly harness it for yourbusiness.
Let's talk about emptycall-to-action sections.
You know, the call to actionsare generic.
Contact us today for more infowith no context or
personalization.
You see, the idea, and of courseyou're going to use some of that
(20:27):
stuff sometime, so that's alittle bit ambiguous.
But the idea is that, well, letme run you down this road real
quick.
One of our clients, they havemultiple locations.
And I got to be very, veryabstract about this.
And I apologize, but I just haveto, to protect their
intellectual property.
(20:47):
You see, I won't share with oneclient what another client is
doing because actually I thinkthat's a breach of trust.
And in this case, someone couldlearn some very powerful things
if they knew the company I wastalking about.
And so location number one,location number one was doing
really good out of the gate.
(21:09):
Um, When I say that, we'rerunning Google Ads, and through
Google Ads, they were gettingimmediate conversions.
Conversions equal meaningfulactions.
So clicks, although are goodactions, they're not meaningful
ultimately to businesses becauseit doesn't necessarily show what
they're converting to in termsof real sales.
(21:32):
Location number two was gettinga lot of activity, i.e.
clicks, but no conversions.
running the exact same businessin two locations, exact same
structure almost, minus a fewthings we did to basically
create the localization basedupon feedback and things like
(21:55):
that.
No success.
So what we did was we startedlooking at the demographics in
area number two versus thedemographics in area number one.
And what was interesting wasarea number two actually should
have been more successful thanarea number one, but it wasn't.
(22:16):
And the reason being was theproduct was actually good for a
demographic that they, thecompany didn't think that it was
good for.
So then we started engaging withlocation number two and through
their input, not our magic wandAI thing, but we were able to
(22:36):
work through using theirintelligence, not only
additional content, but thosecall to actions that address the
audience that was moreinterested in their product,
that the business wasn't awarethat that was more valuable to
them.
And so by doing that, theystarted getting conversions.
(22:57):
Now, that took a lot of work allthe way from the keywords we
were choosing to the ads we werebuilding to the landing page it
went to to how the landing pagewas built and rebuilt, the way
we used heat mapping to look atwhat was going on when a client
came to it.
(23:17):
adjusting the call to actionsand adjusting those messages.
So you can imagine that's quitea bit of work to get
conversions.
And someone who just says, well,we're going to run Google ads to
your landing page is one ofthose things where it's almost
comical when you get into thisbecause you realize it's like
saying, I'm going to work on thefirst two links of a chain and
(23:41):
then we're going to use rubberbands and then put a couple more
links in place.
It doesn't work like that.
So when you look at these callto actions and these empty call
to actions from an agency thatwe really try, when I say that,
I mean, we really, really put aneffort in understanding each
client we work with.
We don't call them by someautomated number.
(24:03):
Can I get your account number oranything like that?
We know our clients by name.
So we really put that effortinto doing that.
And when we do that, we can comeup with something that's good.
But even then, Even then, wehave to bring the client in to
help us figure out certainthings.
Now, when it comes to content,if I were to whip out content
(24:26):
using even the history of thatbusiness without bringing in
location number two, I wouldhave completely missed the mark.
If I use the main location, Imay have completely missed the
mark.
I had to use...
that one specific local place inorder to make this thing work
(24:47):
and put all the pieces together.
So you can see how importantthat all this stuff plays in
together.
And so generic call to actionsare just another one of those
things where, although we allhave the call now, the book now,
you know, all those buttons thatare there, But if the lead-up
(25:08):
point, if the crescendo is notbuilding toward the right
demographic at the right momentwith the right message, then
it's going to fall flat on itsface.
Another one, weird keywordstuffing.
All right, so keyword stuffing.
Just think of it as an articlewhere you start reading it and
(25:28):
let's say, for example, you'rean electrician in Hoboken.
And so what you do is you lookat the article and it says
Hoboken electricians,electricians and Hoboka, blah,
blah, blah, blah, blah, blah,blah, all the way through.
And so by the end, it looks likeit's just been like crammed full
of these words that arenonsensical or they don't flow
(25:51):
very well within the contentthat was created.
This is called keyword stuffing.
Now, Google's gotten really goodat looking at things like
keyword stuffing.
And a matter of fact, just abonus tip, the more you use a
keyword in an article, no, thatdoes not make your article more
effective.
And a matter of fact, when youlook at certain tools, It'll
(26:14):
actually look at the amount oftimes you should use it, and a
lot of times it's a lot lesspercentage than you think.
And so if you see weird keywordstuffing, that's an indicator of
a couple things.
One is AI content, but numbertwo, it's showing that whoever
you're using is old school, andthey don't quite understand the
way things have changed over theyears, and they're not keeping
(26:35):
up with Google's algorithms.
So no formatting or visual flow.
You know, I'm talking aboutthose long, long walls of
unbroken, no bullet points, anduse headings that are just at
the top of the page, things likethat.
Layout matters.
Content matters for thosearticles.
Using photos, using imagery,using infographics are really
(26:58):
important when creating content.
Now, it has been shown that youdon't want to be perfect in your
content.
You want to be better than thenext guy.
But one of the things that youcan obviously tell if it's AI
content just being pushedthrough is if you see these
weird generic images or you'reseeing these long flows of copy,
(27:22):
you're seeing a lack of...
you know, proper, just look atthe page.
If it looks like it's just beensplattered on the page, then our
odds are it's just been droppedin using an AI tool or just
copied and pasted with a coupleof generic modifications like
making the headers bigger,things like that.
So obviously that's not good foryour business.
(27:45):
Now, here's the big one.
It doesn't sound like you.
Okay.
It's got to sound like you.
So when you read it out loud,does it sound like something you
would say or your business wouldsay?
Or does it sound like somethingthat an agency would say or an
artificial intelligencemechanism would say without your
input?
(28:06):
Are you calling that person upand saying, well, this really
isn't that accurate.
This isn't that good.
And they're not really involvingyou in the process.
See, for us, it's so much betterto involve the client because
then we can say, Please lookover this.
Add your own input.
Strip out anything that's notgood.
(28:27):
Put your voice into part ofthis.
And it's simple.
It's really simple to do that.
And then when you do it, theproject just comes alive.
And that EEAT score starts toclimb.
And everything starts going inthe right direction.
So even if you're doingsomething technical, you need to
put the human sentiency spin onit.
(28:49):
So here's a fun one.
If you want to find out if youragency is good at it, what you
can do is ask them a question.
And ask follow-up questions.
You don't have to be an expertto ask questions.
You're paying people to dosomething, and they should be
able to give you good answersfor those questions.
And the question you could askis, can you walk me through your
process for ensuring thiscontent reflects our voice,
(29:13):
experience, and our goals as abusiness?
Now, if they can't provide youwith a good answer, all they can
provide is we use AI tools andwe plug this in and, you know,
just kind of churn and burncontent.
You can read through it and youcan tell if they're not using
real processes that aremeaningful to your business.
(29:36):
I mean, for example, you come tome with that.
If you're working with ourcompany, what I'll typically do
is when I start off, the firstthing we ask are, what are your
goals?
What are you trying to achieve?
And so right out of the gate, wewould have a good answer for
that question about how are youmeeting our goals?
They're baked into everythingthat we do.
(29:57):
If a client asks how we givetheir tone, we're constantly
asking for their tone.
Oftentimes with contentmarketing programs, we send them
the content that we outline andask them to help humanize it and
make sure that that's put intoit and make sure we're not
saying things that areinaccurate.
You see, If you bake it into theprocess, those are really easy
questions to ask.
(30:18):
If you're not baking into theprocess and all you're doing is
trying to create the illusion ofwork.
then those questions are goingto be a lot harder to answer.
And dive into it.
You don't have to acceptsomething that's ambiguous in
their answer.
They should be able to give yougood follow-ups.
I can't tell you how many times.
(30:39):
We've recently started workingwith a pretty large group, and
it's common for them to ask aquestion and then in real time
do a follow-up.
I mean, they want to know thatyou know.
The good companies, they don'tcare if they look foolish in
asking a question.
You know, me, when I talk tosomebody else, if it's something
I don't know about, I'll say,talk to me like I was a kid.
(31:01):
Don't worry about it.
Say whatever.
If I say something foolish, justtell me it's foolish, but
explain to me why it's foolish.
I want to learn.
These are not bad things, andyou should never be fearful of
someone who carries an air orpretends or maybe they do know a
ton of stuff about internetmarketing and design.
But if they do, they should beable to carry a conversation
(31:21):
and, frankly– be excited aboutsharing that content with you.
I mean, we get excited abouttalking to customers and
educating them and equippingthem and showing them exactly
what happens and look under thehood.
Because for me, what that doesis it takes the pressure off.
because I no longer have to beMr.
(31:44):
Authority guy on everything.
Now I can be that guy who says,well, this is what we offer, and
this is exactly what it does.
It's not a magic bullet.
And guess what?
There are no magic bullets.
Artificial intelligence is nomagic bullet.
So if we look at it from thatperspective, it's a lot easier
(32:04):
to kind of bring things intofocus.
Now, I will tell you this.
AI is phenomenal in a lot ofways.
I'll give you a couple of themright off the bat I've used it
for in just the last week.
I'll have a client send me a lotof historical data about their
business.
And then what I'll do is I willengage with artificial
(32:29):
intelligence and I'll ask it toparse data in certain ways.
And then it'll usually give me aresult and the result is not
what I'm looking for.
So then what I'll do is I'll askit to adjust the content and
then base that content on otherthings across the web.
And then I use that content toreally kind of go through
(32:53):
thousands of lines of code inorder to come up with this
boiled down, powerful, bit ofdata that can be used in a
meaningful way for a campaign.
And so we use it, for example,for isolating zip codes.
We can use it for isolatingdifferent cities, different
(33:14):
style of content, all kinds ofdifferent things you can use
with AI.
And then normally that wouldtake just a ton of time to build
pivot tables and sort throughdata and then collect the data,
you know, all those things.
And so it's really good for dataaggregation.
It's also good for brainstormingideas like this podcast.
(33:37):
I regularly communicate withartificial intelligence when I'm
driving down the road to go towork.
It's great because I can ask itfor analytical data and I can
ask it to organize and outlinethings and do things like that
and put it together in ameaningful way.
So when I get to the office,I've already got a lot of
(33:58):
content that I can sort throughas I'm driving.
So I save a lot of time in thatgoing from point A to point B.
And actually, ChatGPT is prettygood about that.
I mean, some of these, uh, AItools, you'll find that they,
uh, you know, the creations isonly as good as the creator,
(34:20):
right?
So the idea is if, um, an AImodel, uh, has a, it presents a
certain philosophical view.
It's because a philosophicalview is baked into the AI,
which, which is dangerous.
It's like George Orwell, 1984stuff.
Um, Yeah, I'll give you an ideaof that.
(34:41):
I have a friend of mine, a dearfriend of mine, and her
husband's going through someserious stuff.
And I was asking artificialintelligence to give me some
data based upon what's going on,based upon some recent
(35:02):
situations.
And it did not give me the data.
It gave me...
The data it presented wasactually generated by the places
where I was asking questions.
So let me try to be morespecific.
If I was asking, are cigarettesbad for you?
And all I got was case studiesthat was built by cigarette
(35:24):
manufacturers, then obviously Iwould know that that data is
skewed.
Well, this particular AI model,what it was doing was giving me
data that was funded by thepeople that I was questioning
what was going on with andreally had to do a lot of
dancing in order to get the datathat I wanted to get.
(35:46):
And AI, if I would have justtrusted it, it would have given
me this philosophical preceptthat was inaccurate.
Frankly, it was completely wrongbecause of information that was
actually released.
It showed the opposite of thesoftball pitch that AI was
(36:06):
giving me.
And so, you know, chat GPT, whenyou look at it.
Yeah, it's going to give someinaccurate stuff.
No, you shouldn't use it forjust generating articles.
None of that stuff.
But when you look at it for likedata aggregation and kind of
organizing things and thingslike that, if you get one of the
(36:26):
more advanced models withChatGPT, it can be quite
beneficial actually.
But not just to set it andforget it or just let it do
things on its own.
For example, regularly I willask it to organize this podcast.
As I'm talking, I'll say, well,this is what I want to do.
And then it'll pull in some dataand then I'll do it and we'll
(36:48):
massage the article using AI.
But then a lot of times thesubjects, the content is so
vanilla.
It's just so not us that we haveto go in and say, okay, really
this is what it is.
Really artificial intelligenceis a cancer for small
businesses.
(37:09):
It's not that it's bad practiceand this is the best practice.
You see the difference betweenthe two is one's going to kind
of give you this lukewarmversion of content and then the
other, which is human-basedcontent, is going to be more You
know, if you take the book ofRevelation, when it talks about
(37:34):
the seven churches, theLaodicean church is one of the
things that Christ had said was,I would rather you be, you know,
when he told John what to write,you know, let me try to clear
that up, was that he said, I'drather be hot or cold, but not
lukewarm.
So the meaning is that you don'twant to be lukewarm in your
(37:55):
content.
You want to be meaningful.
You want to provide somethingthat's going to help somebody
out.
I honestly won't when you'redone here because most of you
are never going to engage withour agency.
And the idea is I want you totake this and go back and build
a more powerful business becausein doing this for as long as
I've done it, here's what Iknow.
(38:15):
Businesses put everything on theline when they grow.
For small businesses, it seemslike everything's always against
them.
There's no lobbies, no realisticlobbying that's going on for the
small business.
Typically, they're the low guyon the totem pole.
It's very hard for a smallbusiness to make it.
(38:38):
If you're being attacked fromthe inside, Or if you've got
that cancer growing in yourbusiness and you don't know that
it's growing in your business,then you have no way to combat
it.
You have no one telling you,hey, quit eating sugar.
You know, that could be fuelingyour cancer.
You don't have that person andyou really need that person on
(38:58):
your side.
So frankly, God's been good tous.
And what I see is a way to kindof contribute and help you help
your family and help thosepeople that work with you.
All right, so let's wrap thisthing up.
EEAT is important.
EEAT is the gold standard.
(39:20):
You want authoritative humancontent based on expertise and
trust.
If your content does not haveyour voice, does not have your
tone, and it's generic, I don'tcare what it reads like, it's
not right for you.
And you need to make changes foryour business before you get too
deep in it.
(39:41):
And don't feel alone if you'redoing this.
I talked to this one company,and I'm 52, so I can officially
say this kid.
So you could tell he was young.
He was so excited aboutartificial intelligence.
And he was in charge of a mega,mega big project.
And I remember him talking to meabout how they're going to use
(40:03):
artificial intelligence to tospin up these localized websites
all across the country.
And they're going to rule theworld through AI generating over
100 pages per location oflocalized content.
And they're going to dominatethe world.
And I asked him, I said, well, Ihad questions.
One of them was about somethingcalled private blog networks and
(40:26):
PBNs based upon his interlinkingstrategy.
But the other one was EEAT.
So you see, even companies thatare big are making this mistake.
They're going too far into it,and they're not thinking about
quality over quantity.
They're thinking somehow theycan build the magic bullet.
(40:49):
So it's kind of like this.
If I gave you a million dollars,it would help you in your
business.
Now, if I gave everybody amillion dollars, Everybody.
Then how valuable is a milliondollars to you?
Not valuable.
Okay, so that's a challenge insearch engines' eyes because if
(41:12):
everybody uses AI content, guesswhat?
Everybody starts at that newbaseline.
And now search engines likeGoogle have to go, okay, what
makes the next group of contentmore valuable?
And that always goes back toEEAT.
And that goes back to theconcept of work.
You know, it seems like thepeople work so hard to avoid
(41:34):
work.
And in reality, work is whatmakes the difference.
Putting that effort in, workingthose extra hours when your
business is starting up, puttingin that thought, looking at it,
not being afraid to say, I'm notsure that this is right for our
business.
Putting money into it to giveyour agency time to do the stuff
(41:54):
that's right.
And never, ever, ever And thisis huge.
I use this all the time.
Never do things halfway.
Stay away from thosedo-it-yourself builders.
Stay away from that stuff thatjust sounds awesome and it's
going to save you all kinds ofmoney by not having to bring in
a professional.
That is a dream.
(42:15):
That is a fantasy.
Just like E-E-A-T is a fantasy.
No one ever enters a Pinto inthe Daytona 500 and says, okay,
if I finish in the top 10, I'mgoing to put more money into it.
No one ever does that.
And why is that?
Because they know by enteringthat car, they're never going to
break the top 10 to begin with.
(42:37):
They have to put their besteffort in, the best tools in,
the best people in, in order tohave a chance, not just for the
top 10, but just to make thequalifying field.
That's what the world is today.
And so make sure you work onyour content.
Make sure you check this out.
And don't get buried in thisavalanche of AI-generated
(43:00):
generic content that Google'sgoing to crawl and they're going
to recognize it because, yes,they are recognizing AI content
more and more every day.
So to the best of luck to youand to your success, I really
appreciate you tuning in andlistening.
You can subscribe and thenyou'll get notifications.
(43:21):
Just subscribe to our podcast.
We're on all major platforms.
If you go to integrisdesign.com,that is spelled I-N-T as in Tom,
E-G-R-I-S, design.com, you canalso sign up for there.
Just look for the Over the Bullpodcast.
Thank you again so much fortuning in.
To your success.
SPEAKER_00 (43:42):
Thanks for tuning in to Over the Bowl, brought to you
by Integris Design, afull-service design and
marketing agency out ofAsheville, North Carolina.
Until next time.
You're listening to Over the Bull, where we cut
through marketing noise.
Here's your host, Ken Carroll.
SPEAKER_01 (00:10):
How artificial intelligence and design agencies
are conspiring to give yourbusiness cancer on this episode
of Over the Bull.
You want to stay for this one.
Now I've said, and I stillcontend, and I'll stick with it,
that design agencies in largepart are out to create big
swaths of content that make youthink that they're doing a lot
(00:32):
more than they are.
Now, that theory, the reason Ihave it, first of all, is
because, frankly, it's true.
I can't tell you how many timesI've talked to search engine
optimization companies or evenlarge companies who are
mass-producing artificialintelligence-based content, and,
frankly, it's going to destroytheir company.
(00:56):
I know that's a littlesensationalistic, but if they
don't put it in reverse and getit figured out, it can cause
irreparable damage.
So let's talk about this alittle bit.
So obviously there is a rise ofAI content across the spectrum
of the internet.
(01:16):
Now we know it's true.
Some of you may be producingcontent or articles using
ChatGPT.
Some of you may be using closedAI systems.
Some of you may be using moresophisticated tools.
But the point is that that toolin itself and that process in
itself or agencies who are doingthat on your behalf are not
(01:38):
doing you any favors.
Now, to understand it, let'sdive into something.
I want to talk to you aboutEEAT.
Like the word eat, but with oneextra letter E.
Now, that stands for expertise,experience, authoritativeness,
and trustworthiness.
Now, this is the gold standardof quality content.
(02:03):
That means that it has to betouched by a human being.
Because remember, Google's goaland all these search engines'
goals are to provide the mostrelevant and meaningful content
when someone searches forsomething.
And so generic content that'sspit out by artificial
(02:23):
intelligence, although valuable,is one of those things that
Google is getting better andbetter at sniffing out and
determining.
So let's talk about agencies fora little bit.
The one thing you need to knowas a business, and this is
absolutely critical.
If you don't hear anything else,you need to hear this and
understand this.
(02:45):
Agencies are out to producecontent and do things with the
minimal amount of overhead.
This means if they can churn outgeneric articles, post them to
your website and not involve youin the process, this is better
for them.
Because it's less work.
It's less communications.
(03:05):
It's pushing a button,generating content, sticking it
on your website, and saying,hey, look what I've produced for
you.
It's a fraction of the time,effort, and energy, and
completely negates possibly themost valuable thing you can have
when you create content.
That is the EEAT factor.
(03:26):
So it's good for the agency.
It's bad for you.
It's almost, you know,meaningless content in some
respects.
So let's talk about this alittle bit more.
Let's move into the risk ofinaccuracies in AI-generated
content, especially when itdoesn't pertain to your specific
(03:47):
business in your specific areas.
If it doesn't relate to thepeople you're trying to serve,
it's useless.
Okay, it doesn't matter what itsays.
If it doesn't relate to thepeople that you're trying to
service in your target area,that content is useless.
So what that means is that youneed to be involved in the
(04:09):
content.
You need to be involved, why?
Because you are the expert ofyour company.
The agency is not an expert ofyour company.
And so, therefore, they couldcreate the best content in the
world using artificialintelligence.
Although technically correct,it's going to miss the EEAT
spectrum.
(04:30):
Okay, so also, just to throwthis out, when you look at AI
content, look, it's good.
We use AI.
We use it in very specific ways,and it saves hours of time.
But it's also often inaccurate,right?
Sometimes it doesn't relate tothe business.
Sometimes it puts in things thatdon't even pertain to the
(04:52):
business, but rather theircompetition.
You see, AI content, althoughpowerful, if it's not carefully
reviewed and edited, it can leadto all kinds of issues, maybe
legal issues, reputational risk.
All kinds of different issuesare associated with just
churning and burning AI contentby giving it a subject, saying
(05:13):
work in a few of these keywords,and then spit the article that I
can put on a website.
You do not want to do this andyou certainly don't want
somebody that doesn't understandyour business doing this.
So what's the magic answer?
The magic answer is using thepower of artificial intelligence
combined with human sentiency.
(05:35):
And that human sentiency usuallyhas to be partly you as the
owner of a business.
And the reason being is thatDesign agencies are not someone
who understands everything aboutall businesses.
Even niche-based design ormarketing agencies don't
understand your business.
(05:57):
They may understand all thetechnicalities.
They may understand all therules, but they don't understand
you.
You see, that's the missingpiece here.
You're the authority element.
You're that person that's goingto bring in that trust factor by
incorporating you.
into the process.
So one thing that we do with ouragency is we'll use artificial
(06:21):
intelligence to create anoutline.
And then what we'll do is we'llcreate the scaffolding of an
article.
And we'll let artificialintelligence rough it in.
But between the outline and theactual development of the
article, We actually injectURLs.
We inject different things aboutthe business.
(06:42):
We inject different processes tomake sure that it's close.
And then we also remove anythingthat's not relevant to that
business to the best of ourunderstanding.
Then once we create thatcontent, we pass that on to the
client.
We ask them to humanize it, toput in their experience, to put
in their personal energy so thatthat article becomes more, well,
(07:06):
human.
And then what happens issomething really magic.
The EEAT factor starts to go up.
Now it becomes more originalcontent.
You didn't have to go all theway over to the idea of creating
an article from scratch, but youwere able to use the content
that was created as a basis foryour humanized content.
(07:31):
And then that's what wepublished.
So a quick side note, there arealso powerful tools that allow
you to go in and actually lookat the best articles that are in
your area that are beingsearched for the most.
So you should never go in tochat GPT or one of these things
and just ask it to write anarticle based on something that
(07:52):
you think is good.
you can actually find out whatpeople are looking for in terms
of content and then buildmeaningful content.
So a little bit of meaningfulcontent is going to far
outweigh, you know, just tonsand tons of generic content
you're spitting out andbasically just guessing on
whether or not that content isgood content.
(08:15):
Now, here's the sad part.
A lot of agencies, they guessand They just look at your
keywords.
They'll try to generate a fewarticles based on those and just
keep feeding the beast.
But that's like feeding cancerwith sugar.
You don't want to do that.
Meaningful is always better.
(08:36):
Now, just like, you know, inhistory, you know, we tend to
look at people in antiquity andthink they weren't as
intelligent as us.
But let's hit the pause buttonon that for a second.
And let's look at Machu Picchu.
Let's look at the pyramids ofEgypt.
(08:56):
Let's look at some of theseancient structures, which would
be extremely difficult, if notimpossible, for our modern
abilities to reproduce.
Now, there's a reason thatthere's these rocks that are
really, you know, so tightlyknit together still after
centuries.
(09:16):
thousands of years where youcan't put a piece of paper
between the two rocks or thegeometric rocks that are not
squares.
You know, the reason we havesquare blocks is because it's
easy to produce.
These asymmetrical elements thatare in some of these structures
are much more difficult toproduce.
(09:37):
So we would do well by learningfrom our ancestors and not
having this idea that justbecause we have different tools,
that somehow we're moreintelligent or more insightful
than people that have lived onthis earth before us.
And this is really important.
So when you look at somethinglike these cautionary tales,
(10:01):
such as Icarus, When you look atthese, I think we should take
those and then carefully thinkabout those within the construct
of our business.
Now, if we look at it from thatperspective, one of the modern
things that have come in issomething called the Habsburg
(10:21):
jaw of content decay is kind ofwhat I like to call this.
And basically, the Habsburgwanted to, there were a regal
group And what they wanted to dowas maintain their purity.
And so what they did was theykept inbreeding with themselves.
(10:42):
And rather than increasing theirroyal lineage, they actually
bred themselves out ofexistence.
And one of the visuals of thatgroup was this kind of like
extended jaw.
And that's where you get thisHabsburg jaw.
So just like the Habsburg keptmarrying their cousins to
(11:05):
maintain purity and ending upundermining their entire
destiny, some businesses andagencies are now inbreeding
their content by relying tooheavily on AI without fresh
human input.
Now, the content that lookspolished but lacks soul is
(11:26):
something that that AI looks at.
And if AI content is going touse AI content and get in this
feedback loop, then the contentbecomes more and more generic
without human input.
So search engines, although youthink they may want to play the
game of artificial intelligence,They do to an extent, but they
(11:50):
also understand that humansentient expertise is much, much
more important and more valuableto the reader than just this
churn-and-burn style content.
So the modern-day Habsburgtheory, which is a little
tongue-in-cheek, a little bit ofphilosophical debate— More just
(12:14):
fun stuff to talk about ispossibly a big warning sign
where, you know, the ancientsirens would call out to sailors
and lure them to the depths ofthe ocean to their death.
Perhaps this is something thatis a warning for small
businesses and businesses of allsizes.
(12:35):
that if you create a model ofartificial intelligence to whip
out content, then maybe you'repushing yourself into the depths
of your own demise for yourbusiness and the people that
trust you to keep your businessafloat.
So that's real strong when inreality, most businesses just
(12:55):
want to have a product, sell aproduct, and then push on.
Now, as far as gosh, I don'tknow, 10, 15 years ago, in our
industry, the marketing anddesign industry, it was already
being pushed that journalismdegrees would be huge.
They would be phenomenal.
And the reason being was thequality of content.
(13:18):
And the idea is that everybusiness kind of becomes their
own outlet for that content,which is true.
So if you're going to play thecontent game, you need to play
it right and not pay someonejust to kind of keep cranking
out content over and over again.
So if we really look at it, theexpertise and experience, it
(13:38):
just can't be fake artificialintelligence.
And it can't be faked byagencies pretending to know your
business.
I mean, I spend probably, Imean, with coaching and
retainer-based clients, I spendtime with them every other week,
once a month, and I truly try tounderstand and document things
(13:59):
about their business.
And I still...
and not the expert of theirbusiness.
And I don't pretend to be.
See, it's really liberating whenyou go in and you don't say, I
am an agency that specializes inelectricians, or I'm an agency
that understands you so muchthat I can create content and it
sounds and acts just like you.
(14:20):
You see, I'm putting myself in aplace where I can't do that
because the minute that businesswalks out, I'm talking to a
bookstore, or I'm talking to aplumber, or I'm talking to, you
get the point, I'm talkingtalking to different businesses
and to go through that deal withhundreds of businesses and then
still think that somehow I canfocus on your niche business in
(14:43):
your area is frankly impossible.
Now, we do our best, but we loveclients who participate.
Clients who participate alwaysdo better because their
resonance is sent out there.
Now, that sounds a little hokey,but it's true.
Because when people read yourword, they read your expertise
(15:03):
that goes into it, it helps.
And so part of what we do is wedon't really care if a client
understands the best grammaticalrules.
As a matter of fact, I workedwith a particular person.
I won't mention her name, butshe is huge in the copywriting
industry.
I mean, absolutely phenomenal inthe copywriting industry.
(15:25):
And I remember one time, I askedher about a sentence structure,
and I can't remember if it was acolon or a semicolon, but I
said, you know, you reallyshould have this here, I think,
in order for it to be accurate.
And I was kind of shockedbecause we're talking about a
person who$3,000 for a half hourof her time and has written
(15:47):
stuff for very largeorganizations.
And she basically, with theconfidence that she had, she
goes, I really don't care aboutthe grammar in some cases.
I care about the pace.
And she said a colon causessomeone to pause.
And I don't want them to pausein that instance.
(16:08):
She inserted an ellipse, youknow, that dot, dot, dot, and
then continued.
Now, the reason she did it wasbecause it kept the pace of the
content going in the right way.
You see, that's expertise.
That's something that's kind ofhard to fake.
Now, granted, I get it.
You're watching all these thingswith, you know, songs being
(16:30):
reproduced and all this otherstuff.
And you're thinking, well,surely they could reproduce her.
Yeah, but can they reproduce herexpertise?
Can they reproduce all thethings that go into that human
sentient experience to createthat content?
So don't let grammar or beingimperfect with grammar get in
your way.
Matter of fact, I tell ourclients, don't worry about it.
(16:51):
That's not your job to begrammatically sound.
Your job is to give me thathumanized factor and then let me
take it and I'll run it over thegoal and I'll polish it up a
little bit.
But you see, that's the process.
It's not a churn and burn.
Okay, so let's talk about thepeople that are working with
you.
Now, I get it.
(17:11):
Businesses, and I'm the sameway, I am jealously protective
of my clients.
And the thing about businesses,most of them, they're really
kind of protective of the peoplethat serve them, that agency,
that freelancer that comes inand sees them and shakes their
hands and smiles and talks aboutthe ballgame.
(17:33):
I get it.
You like your agency.
You like your freelancer.
That's okay.
There's nothing wrong with thatunless it inhibits your business
ability to be successful.
Remember, you're not running acountry club.
You're not running something tobring friends in and joke.
You're trying to grow yourbusiness and protect the people
(17:53):
you work with.
So let's talk about some flagswhen it comes to agencies that
may be using artificialintelligence-based content only.
The first one, every articlesounds the same.
The tone, the structure, and thevocabulary feel eerily similar
(18:14):
across blogs or pages, even fordifferent topics.
You see, the thing is, is thatartificial intelligence
generates content, but it alsokind of brings in that tone that
is similar.
Now, some of the more advancedtools that we use when we're
creating those outlines and thatscaffolding for an article, so
to speak, we can actually put inthe ability to alter the tone
(18:38):
that we give it.
But even at that, it's not goodenough.
And so don't let anyone tell youthat they can produce content,
create the right tone for thecontent, send you the content,
and then that content ismagically great.
No, it's not true.
There's no tool that replicateshuman sentiency.
(18:59):
The next one.
vague language with nospecifics.
You know, those articles thatsay things like it's important
to have a strategy withoutexplaining what that strategy is
or giving real examples.
So, for example, if I were tonot give you our example of how
we build content, then thatwould be more generic.
But I'm giving you a real worldexample of how we actually use
(19:22):
AI content with the client toproduce something meaningful.
There's no first-hand experienceor personalization in AI
content.
No mention of things like casestudies or customer feedback,
employee quotes, or thatbehind-the-scenes process.
Again, podcast is a goodexample.
(19:44):
I'm giving you content that I'mbasically rattling off based
upon what I run into each weekat our design agency.
And this is one of the abusesthat I've noticed this week, and
I'm simply sharing it with you.
But I'm doing it in more of ameaningful way that is bringing
my personal experience, my over30 years experience, into the
(20:07):
conversation and sharing thatwith you in hopes that you
properly harness it for yourbusiness.
Let's talk about emptycall-to-action sections.
You know, the call to actionsare generic.
Contact us today for more infowith no context or
personalization.
You see, the idea, and of courseyou're going to use some of that
(20:27):
stuff sometime, so that's alittle bit ambiguous.
But the idea is that, well, letme run you down this road real
quick.
One of our clients, they havemultiple locations.
And I got to be very, veryabstract about this.
And I apologize, but I just haveto, to protect their
intellectual property.
(20:47):
You see, I won't share with oneclient what another client is
doing because actually I thinkthat's a breach of trust.
And in this case, someone couldlearn some very powerful things
if they knew the company I wastalking about.
And so location number one,location number one was doing
really good out of the gate.
(21:09):
Um, When I say that, we'rerunning Google Ads, and through
Google Ads, they were gettingimmediate conversions.
Conversions equal meaningfulactions.
So clicks, although are goodactions, they're not meaningful
ultimately to businesses becauseit doesn't necessarily show what
they're converting to in termsof real sales.
(21:32):
Location number two was gettinga lot of activity, i.e.
clicks, but no conversions.
running the exact same businessin two locations, exact same
structure almost, minus a fewthings we did to basically
create the localization basedupon feedback and things like
(21:55):
that.
No success.
So what we did was we startedlooking at the demographics in
area number two versus thedemographics in area number one.
And what was interesting wasarea number two actually should
have been more successful thanarea number one, but it wasn't.
(22:16):
And the reason being was theproduct was actually good for a
demographic that they, thecompany didn't think that it was
good for.
So then we started engaging withlocation number two and through
their input, not our magic wandAI thing, but we were able to
(22:36):
work through using theirintelligence, not only
additional content, but thosecall to actions that address the
audience that was moreinterested in their product,
that the business wasn't awarethat that was more valuable to
them.
And so by doing that, theystarted getting conversions.
(22:57):
Now, that took a lot of work allthe way from the keywords we
were choosing to the ads we werebuilding to the landing page it
went to to how the landing pagewas built and rebuilt, the way
we used heat mapping to look atwhat was going on when a client
came to it.
(23:17):
adjusting the call to actionsand adjusting those messages.
So you can imagine that's quitea bit of work to get
conversions.
And someone who just says, well,we're going to run Google ads to
your landing page is one ofthose things where it's almost
comical when you get into thisbecause you realize it's like
saying, I'm going to work on thefirst two links of a chain and
(23:41):
then we're going to use rubberbands and then put a couple more
links in place.
It doesn't work like that.
So when you look at these callto actions and these empty call
to actions from an agency thatwe really try, when I say that,
I mean, we really, really put aneffort in understanding each
client we work with.
We don't call them by someautomated number.
(24:03):
Can I get your account number oranything like that?
We know our clients by name.
So we really put that effortinto doing that.
And when we do that, we can comeup with something that's good.
But even then, Even then, wehave to bring the client in to
help us figure out certainthings.
Now, when it comes to content,if I were to whip out content
(24:26):
using even the history of thatbusiness without bringing in
location number two, I wouldhave completely missed the mark.
If I use the main location, Imay have completely missed the
mark.
I had to use...
that one specific local place inorder to make this thing work
(24:47):
and put all the pieces together.
So you can see how importantthat all this stuff plays in
together.
And so generic call to actionsare just another one of those
things where, although we allhave the call now, the book now,
you know, all those buttons thatare there, But if the lead-up
(25:08):
point, if the crescendo is notbuilding toward the right
demographic at the right momentwith the right message, then
it's going to fall flat on itsface.
Another one, weird keywordstuffing.
All right, so keyword stuffing.
Just think of it as an articlewhere you start reading it and
(25:28):
let's say, for example, you'rean electrician in Hoboken.
And so what you do is you lookat the article and it says
Hoboken electricians,electricians and Hoboka, blah,
blah, blah, blah, blah, blah,blah, all the way through.
And so by the end, it looks likeit's just been like crammed full
of these words that arenonsensical or they don't flow
(25:51):
very well within the contentthat was created.
This is called keyword stuffing.
Now, Google's gotten really goodat looking at things like
keyword stuffing.
And a matter of fact, just abonus tip, the more you use a
keyword in an article, no, thatdoes not make your article more
effective.
And a matter of fact, when youlook at certain tools, It'll
(26:14):
actually look at the amount oftimes you should use it, and a
lot of times it's a lot lesspercentage than you think.
And so if you see weird keywordstuffing, that's an indicator of
a couple things.
One is AI content, but numbertwo, it's showing that whoever
you're using is old school, andthey don't quite understand the
way things have changed over theyears, and they're not keeping
(26:35):
up with Google's algorithms.
So no formatting or visual flow.
You know, I'm talking aboutthose long, long walls of
unbroken, no bullet points, anduse headings that are just at
the top of the page, things likethat.
Layout matters.
Content matters for thosearticles.
Using photos, using imagery,using infographics are really
(26:58):
important when creating content.
Now, it has been shown that youdon't want to be perfect in your
content.
You want to be better than thenext guy.
But one of the things that youcan obviously tell if it's AI
content just being pushedthrough is if you see these
weird generic images or you'reseeing these long flows of copy,
(27:22):
you're seeing a lack of...
you know, proper, just look atthe page.
If it looks like it's just beensplattered on the page, then our
odds are it's just been droppedin using an AI tool or just
copied and pasted with a coupleof generic modifications like
making the headers bigger,things like that.
So obviously that's not good foryour business.
(27:45):
Now, here's the big one.
It doesn't sound like you.
Okay.
It's got to sound like you.
So when you read it out loud,does it sound like something you
would say or your business wouldsay?
Or does it sound like somethingthat an agency would say or an
artificial intelligencemechanism would say without your
input?
(28:06):
Are you calling that person upand saying, well, this really
isn't that accurate.
This isn't that good.
And they're not really involvingyou in the process.
See, for us, it's so much betterto involve the client because
then we can say, Please lookover this.
Add your own input.
Strip out anything that's notgood.
(28:27):
Put your voice into part ofthis.
And it's simple.
It's really simple to do that.
And then when you do it, theproject just comes alive.
And that EEAT score starts toclimb.
And everything starts going inthe right direction.
So even if you're doingsomething technical, you need to
put the human sentiency spin onit.
(28:49):
So here's a fun one.
If you want to find out if youragency is good at it, what you
can do is ask them a question.
And ask follow-up questions.
You don't have to be an expertto ask questions.
You're paying people to dosomething, and they should be
able to give you good answersfor those questions.
And the question you could askis, can you walk me through your
process for ensuring thiscontent reflects our voice,
(29:13):
experience, and our goals as abusiness?
Now, if they can't provide youwith a good answer, all they can
provide is we use AI tools andwe plug this in and, you know,
just kind of churn and burncontent.
You can read through it and youcan tell if they're not using
real processes that aremeaningful to your business.
(29:36):
I mean, for example, you come tome with that.
If you're working with ourcompany, what I'll typically do
is when I start off, the firstthing we ask are, what are your
goals?
What are you trying to achieve?
And so right out of the gate, wewould have a good answer for
that question about how are youmeeting our goals?
They're baked into everythingthat we do.
(29:57):
If a client asks how we givetheir tone, we're constantly
asking for their tone.
Oftentimes with contentmarketing programs, we send them
the content that we outline andask them to help humanize it and
make sure that that's put intoit and make sure we're not
saying things that areinaccurate.
You see, If you bake it into theprocess, those are really easy
questions to ask.
(30:18):
If you're not baking into theprocess and all you're doing is
trying to create the illusion ofwork.
then those questions are goingto be a lot harder to answer.
And dive into it.
You don't have to acceptsomething that's ambiguous in
their answer.
They should be able to give yougood follow-ups.
I can't tell you how many times.
(30:39):
We've recently started workingwith a pretty large group, and
it's common for them to ask aquestion and then in real time
do a follow-up.
I mean, they want to know thatyou know.
The good companies, they don'tcare if they look foolish in
asking a question.
You know, me, when I talk tosomebody else, if it's something
I don't know about, I'll say,talk to me like I was a kid.
(31:01):
Don't worry about it.
Say whatever.
If I say something foolish, justtell me it's foolish, but
explain to me why it's foolish.
I want to learn.
These are not bad things, andyou should never be fearful of
someone who carries an air orpretends or maybe they do know a
ton of stuff about internetmarketing and design.
But if they do, they should beable to carry a conversation
(31:21):
and, frankly– be excited aboutsharing that content with you.
I mean, we get excited abouttalking to customers and
educating them and equippingthem and showing them exactly
what happens and look under thehood.
Because for me, what that doesis it takes the pressure off.
because I no longer have to beMr.
(31:44):
Authority guy on everything.
Now I can be that guy who says,well, this is what we offer, and
this is exactly what it does.
It's not a magic bullet.
And guess what?
There are no magic bullets.
Artificial intelligence is nomagic bullet.
So if we look at it from thatperspective, it's a lot easier
(32:04):
to kind of bring things intofocus.
Now, I will tell you this.
AI is phenomenal in a lot ofways.
I'll give you a couple of themright off the bat I've used it
for in just the last week.
I'll have a client send me a lotof historical data about their
business.
And then what I'll do is I willengage with artificial
(32:29):
intelligence and I'll ask it toparse data in certain ways.
And then it'll usually give me aresult and the result is not
what I'm looking for.
So then what I'll do is I'll askit to adjust the content and
then base that content on otherthings across the web.
And then I use that content toreally kind of go through
(32:53):
thousands of lines of code inorder to come up with this
boiled down, powerful, bit ofdata that can be used in a
meaningful way for a campaign.
And so we use it, for example,for isolating zip codes.
We can use it for isolatingdifferent cities, different
(33:14):
style of content, all kinds ofdifferent things you can use
with AI.
And then normally that wouldtake just a ton of time to build
pivot tables and sort throughdata and then collect the data,
you know, all those things.
And so it's really good for dataaggregation.
It's also good for brainstormingideas like this podcast.
(33:37):
I regularly communicate withartificial intelligence when I'm
driving down the road to go towork.
It's great because I can ask itfor analytical data and I can
ask it to organize and outlinethings and do things like that
and put it together in ameaningful way.
So when I get to the office,I've already got a lot of
(33:58):
content that I can sort throughas I'm driving.
So I save a lot of time in thatgoing from point A to point B.
And actually, ChatGPT is prettygood about that.
I mean, some of these, uh, AItools, you'll find that they,
uh, you know, the creations isonly as good as the creator,
(34:20):
right?
So the idea is if, um, an AImodel, uh, has a, it presents a
certain philosophical view.
It's because a philosophicalview is baked into the AI,
which, which is dangerous.
It's like George Orwell, 1984stuff.
Um, Yeah, I'll give you an ideaof that.
(34:41):
I have a friend of mine, a dearfriend of mine, and her
husband's going through someserious stuff.
And I was asking artificialintelligence to give me some
data based upon what's going on,based upon some recent
(35:02):
situations.
And it did not give me the data.
It gave me...
The data it presented wasactually generated by the places
where I was asking questions.
So let me try to be morespecific.
If I was asking, are cigarettesbad for you?
And all I got was case studiesthat was built by cigarette
(35:24):
manufacturers, then obviously Iwould know that that data is
skewed.
Well, this particular AI model,what it was doing was giving me
data that was funded by thepeople that I was questioning
what was going on with andreally had to do a lot of
dancing in order to get the datathat I wanted to get.
(35:46):
And AI, if I would have justtrusted it, it would have given
me this philosophical preceptthat was inaccurate.
Frankly, it was completely wrongbecause of information that was
actually released.
It showed the opposite of thesoftball pitch that AI was
(36:06):
giving me.
And so, you know, chat GPT, whenyou look at it.
Yeah, it's going to give someinaccurate stuff.
No, you shouldn't use it forjust generating articles.
None of that stuff.
But when you look at it for likedata aggregation and kind of
organizing things and thingslike that, if you get one of the
(36:26):
more advanced models withChatGPT, it can be quite
beneficial actually.
But not just to set it andforget it or just let it do
things on its own.
For example, regularly I willask it to organize this podcast.
As I'm talking, I'll say, well,this is what I want to do.
And then it'll pull in some dataand then I'll do it and we'll
(36:48):
massage the article using AI.
But then a lot of times thesubjects, the content is so
vanilla.
It's just so not us that we haveto go in and say, okay, really
this is what it is.
Really artificial intelligenceis a cancer for small
businesses.
(37:09):
It's not that it's bad practiceand this is the best practice.
You see the difference betweenthe two is one's going to kind
of give you this lukewarmversion of content and then the
other, which is human-basedcontent, is going to be more You
know, if you take the book ofRevelation, when it talks about
(37:34):
the seven churches, theLaodicean church is one of the
things that Christ had said was,I would rather you be, you know,
when he told John what to write,you know, let me try to clear
that up, was that he said, I'drather be hot or cold, but not
lukewarm.
So the meaning is that you don'twant to be lukewarm in your
(37:55):
content.
You want to be meaningful.
You want to provide somethingthat's going to help somebody
out.
I honestly won't when you'redone here because most of you
are never going to engage withour agency.
And the idea is I want you totake this and go back and build
a more powerful business becausein doing this for as long as
I've done it, here's what Iknow.
(38:15):
Businesses put everything on theline when they grow.
For small businesses, it seemslike everything's always against
them.
There's no lobbies, no realisticlobbying that's going on for the
small business.
Typically, they're the low guyon the totem pole.
It's very hard for a smallbusiness to make it.
(38:38):
If you're being attacked fromthe inside, Or if you've got
that cancer growing in yourbusiness and you don't know that
it's growing in your business,then you have no way to combat
it.
You have no one telling you,hey, quit eating sugar.
You know, that could be fuelingyour cancer.
You don't have that person andyou really need that person on
(38:58):
your side.
So frankly, God's been good tous.
And what I see is a way to kindof contribute and help you help
your family and help thosepeople that work with you.
All right, so let's wrap thisthing up.
EEAT is important.
EEAT is the gold standard.
(39:20):
You want authoritative humancontent based on expertise and
trust.
If your content does not haveyour voice, does not have your
tone, and it's generic, I don'tcare what it reads like, it's
not right for you.
And you need to make changes foryour business before you get too
deep in it.
(39:41):
And don't feel alone if you'redoing this.
I talked to this one company,and I'm 52, so I can officially
say this kid.
So you could tell he was young.
He was so excited aboutartificial intelligence.
And he was in charge of a mega,mega big project.
And I remember him talking to meabout how they're going to use
(40:03):
artificial intelligence to tospin up these localized websites
all across the country.
And they're going to rule theworld through AI generating over
100 pages per location oflocalized content.
And they're going to dominatethe world.
And I asked him, I said, well, Ihad questions.
One of them was about somethingcalled private blog networks and
(40:26):
PBNs based upon his interlinkingstrategy.
But the other one was EEAT.
So you see, even companies thatare big are making this mistake.
They're going too far into it,and they're not thinking about
quality over quantity.
They're thinking somehow theycan build the magic bullet.
(40:49):
So it's kind of like this.
If I gave you a million dollars,it would help you in your
business.
Now, if I gave everybody amillion dollars, Everybody.
Then how valuable is a milliondollars to you?
Not valuable.
Okay, so that's a challenge insearch engines' eyes because if
(41:12):
everybody uses AI content, guesswhat?
Everybody starts at that newbaseline.
And now search engines likeGoogle have to go, okay, what
makes the next group of contentmore valuable?
And that always goes back toEEAT.
And that goes back to theconcept of work.
You know, it seems like thepeople work so hard to avoid
(41:34):
work.
And in reality, work is whatmakes the difference.
Putting that effort in, workingthose extra hours when your
business is starting up, puttingin that thought, looking at it,
not being afraid to say, I'm notsure that this is right for our
business.
Putting money into it to giveyour agency time to do the stuff
(41:54):
that's right.
And never, ever, ever And thisis huge.
I use this all the time.
Never do things halfway.
Stay away from thosedo-it-yourself builders.
Stay away from that stuff thatjust sounds awesome and it's
going to save you all kinds ofmoney by not having to bring in
a professional.
That is a dream.
(42:15):
That is a fantasy.
Just like E-E-A-T is a fantasy.
No one ever enters a Pinto inthe Daytona 500 and says, okay,
if I finish in the top 10, I'mgoing to put more money into it.
No one ever does that.
And why is that?
Because they know by enteringthat car, they're never going to
break the top 10 to begin with.
(42:37):
They have to put their besteffort in, the best tools in,
the best people in, in order tohave a chance, not just for the
top 10, but just to make thequalifying field.
That's what the world is today.
And so make sure you work onyour content.
Make sure you check this out.
And don't get buried in thisavalanche of AI-generated
(43:00):
generic content that Google'sgoing to crawl and they're going
to recognize it because, yes,they are recognizing AI content
more and more every day.
So to the best of luck to youand to your success, I really
appreciate you tuning in andlistening.
You can subscribe and thenyou'll get notifications.
(43:21):
Just subscribe to our podcast.
We're on all major platforms.
If you go to integrisdesign.com,that is spelled I-N-T as in Tom,
E-G-R-I-S, design.com, you canalso sign up for there.
Just look for the Over the Bullpodcast.
Thank you again so much fortuning in.
To your success.
SPEAKER_00 (43:42):
Thanks for tuning in to Over the Bowl, brought to you
by Integris Design, afull-service design and
marketing agency out ofAsheville, North Carolina.
Until next time.
Popular Podcasts
New Heights with Jason & Travis Kelce
Football’s funniest family duo — Jason Kelce of the Philadelphia Eagles and Travis Kelce of the Kansas City Chiefs — team up to provide next-level access to life in the league as it unfolds. The two brothers and Super Bowl champions drop weekly insights about the weekly slate of games and share their INSIDE perspectives on trending NFL news and sports headlines. They also endlessly rag on each other as brothers do, chat the latest in pop culture and welcome some very popular and well-known friends to chat with them. Check out new episodes every Wednesday. Follow New Heights on the Wondery App, YouTube or wherever you get your podcasts. You can listen to new episodes early and ad-free, and get exclusive content on Wondery+. Join Wondery+ in the Wondery App, Apple Podcasts or Spotify. And join our new membership for a unique fan experience by going to the New Heights YouTube channel now!
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.