Advertising in the Age of AI with Sean Sweeney Part 1 (283)
In today’s conversation, I sit down with Sean Sweeney, the “Blues Brother of PPC” from Chicago, to cut through the AI hype and talk about what’s actually changing in digital advertising. We get candid about Google Performance Max and Meta’s algorithmic buying: why “just give us your credit card and we’ll do the rest” can erode brand control, underfund top‑of‑funnel, and starve discovery.
We dig into the real work that still moves the needle: full‑funnel strategy, first‑party data, programmatic done right, and organic discovery that feeds paid performance. We unpack AEO/GEO for AI chatbots, why no tool can definitively track AI recommendations, and how to increase your odds of inclusion with human‑helpful, machine‑readable content (schema, FAQs, glossaries), plus platform plays like Reddit and Pinterest. We also talk creative in the AI era: how a smart mix of real assets with AI enhancements can outperform big‑budget shoots, and why model selection matters, using the right AI for the right task is a competitive edge.
Finally, we call out the incentives baked into ad platforms, the shrinking visibility into data, and why mid‑market brands can’t afford to hand over the steering wheel. If your SEO, paid, content, and “AI optimization” still live in silos, this episode is your nudge to integrate your discovery ecosystem end‑to‑end.
https://www.firstpositiondigitalsolutions.com/
Looking for a podcast guest? Author Matt Rouse
Hook Digital Marketing | Hook Digital Marketing Canada
Market your local business on autopilot: SMB Autopilot
Your hosts,
Matt Wiles and Mae Heckenham,
welcome you to the show.
The digital marketing masters podcast.
Talk about advertising and SEO.
In the age of AI
with our special guests.
The blues brother of paper click.
The Chicago search man,
Sean Sweeney.
Hello, everyone. This is Mahek Yur, and we are back.
Today, Matt is sitting down with Sean Sweeney who runs a paid media agency and clearly
has a lot of feelings about what Google
and Meta are quietly doing to our ad budget. Same here, Sean. Same.
This one gets into performance max, GEO, SEO,
first party data, and why the fundamentals of marketing are somehow both old news
and the most important thing you can do right now.
So let's dive right into it.
Welcome back. Today, my guest is Sean Sweeney.
Sean, how you doing? Hello, Matt. From from beautiful sunny Chicago?
Not today, but, yes, it's getting there. Everybody's always hyped about the AI stuff. Right? Yeah.
But in the background,
there's lots of changes happening in the digital ad world.
And Yeah. I think most people are unaware of any of those changes.
Yeah. I I mean, there's been
some public more public changes. Like, if you look at some of the buying media, you know, we focus on paid digital media. So we when you look at the the buying platforms that most brands use like Google and and Meta,
they've implemented
AI based,
buying and and algorithmic
optimization.
So the idea being, hey. Hey. Just you know, I think Zuckerberg even said it once, and I kinda railed on it when when he said it. He's just like, just give us your credit card, and we'll do the rest. And I just, like, shook my head like, oh my gosh. That's the worst thing you can do as a brand.
But the problem is most brands are doing that. Like, they're they're you know, I think the percentage was, like, 91%
are testing it.
That goes for Google too with Performance Max. Right? So you see it suddenly coming out in the buying platforms.
I think middle market brands who we focus on working with,
they're testing out things on their operational side more than anything
to be obviously be as efficient as they can on their side. I think it mainly centers around creative and creative testing,
that process flow, and how much they can save by trying to create these things on their own. What I have not seen yet is a
desire from a lot of these brands to
have us
only use AI optimization for media spend. Right? Right. That that's what that's the part I'm a little worried about only because
I think that's where things will go wrong. I think, ultimately, it might get to that point regardless of what I care about and what I want, but
I think there will always you know, humans will always need to have oversight. I mean, with these things like Performance Max and and Meta right now, it's like you you're making a you know, you're you're making, like, a deal with the devil in some ways where it's like it's a trust for performance thing, where you're basically
trusting Google and you're trusting Meta that if I give you my credit card and I tell you what targets I want you to or I tell you what my outcome is that I want, you'll figure it out, and you'll drive it for me. Right? So it's like that trust for performance,
you know, promise that you're making there.
And,
you know, from what we've seen in trying those out for clients, they just don't perform well. And so
I do think that it we're in a big test and learn phase with a lot of this as it relates to digital advertising.
I do think creative's getting hurt creative types are getting hurt more and more right now because it's becoming easier and easier to create the ads and and replicate them quickly,
iterate them, you know, do AB testing. And I think that's what lot of brands are focused on right now is how do they become more efficient on the creative side. They'll turn towards media more,
on the programmatic side and and moving forward, I think, shortly. But right now, it seems to be mostly on the creative side. And then just when they use it with Google and Meta, they're testing it out as well. Just give us your credit card, and we'll do the rest. Now that sounds a lot like what Zuckerberg actually meant when he said,
we're going to get to a point where if you're a business, you basically come to us and tell us your objective.
You connect your bank account, and you don't need any creative. You don't need any targeting demographic.
You don't need any measurement except to be able to read the results that we spit out.
I think about it more than I should because the pitch sounds incredibly
perfect on the surface. Right?
You hand over the budget, the algorithm does the work, and you go home.
Except what you're actually doing is handing the steering wheel to a system whose entire incentive is to spend your money,
not protect it. And you see Sean's point about Performance Max is exactly this.
Like, the AI optimizes for it what it can measure, and what it can measure is almost never the full picture. But when you're talking about Performance Max ads, which is Google, and then also
in the meta ad system,
you've got, Andromeda,
and Gem is what they're running right now. They don't label it that in the system, but that's the AI in the background. No problem. Yeah. I think that
in our experience running ads for our clients
that
you can't just run the AI based ads yet because it doesn't seem to be able to give it the proper
mix of, like,
brand awareness
versus,
like, the steps to conversion.
It seems to really miss the brand awareness piece.
It's not
building enough audience
at the top of the funnel, in my opinion. I've had this conversation a lot lately.
A lot, actually. And
where we've had clients turn off the top of the funnel, and they're like, no. We're just gonna stick with Google. We think we can do it. I I had to tell a client a couple days ago, listen. In about two months after the effects of our CTV campaign or our YouTube video campaign have worn off now that we've stopped them,
my concern is that you're gonna have performance dip drastically in the next month and a half, two months because that that recall will die, and we're we're not replenishing it. We're just focusing on that lower funnel
activity, and it's just gonna run out at some point. And so
that that's starting to happen a lot where we're not focusing on, you know, the discovery part and brand awareness part as you say.
And I get it. Right? I I don't blame brands for that because it's so the attribution thing is still a problem. I mean,
I'm sitting here waiting for any AI company to come out and solve the attribution problem because I will spend all my money on that solution if it works. Right. You know? But there's nothing out yet, and that's probably by design. No one in that tech wants to fix that. Right? Because then then nothing is is left to chance and nothing's left to the imagination that, like, oh, this could work. But knowing that, like, CTV, you can't really track it that well in terms of tracking it all the way to unless you're not, like, a CBG or QSR and running those QR codes that do decently well, you know, when you're running those on CTV ads.
You know, you're not really getting a lot of performance out of that and same with YouTube. So
but they're brand builders. You know? They're discovery builders. And and that's the thing that I I think is going away quickly, and that's a concern for me
because I'm a I'm a traditional like, we gotta do full funnel marketing.
Yeah. Yeah. We had over attribution until several years ago.
If somebody clicked on anything on the Internet and went to your site and bought something, you knew what happened.
Yeah. That's just not the case anymore,
but people still think that's the case.
Well, you
could still go to, like, you know, GA 4 and see what your referring website is. But, again, that's, you know, last click attribution that fit perfectly into what Google wants you to do, you know, and and how they want you to think because that helps their search business. So
it it's hard to break that mold, and
it's it you know,
I sound if sometimes I feel like when I'm talking to clients, I sound very old fashioned
when I talk about full funnel marketing.
It it seems like that's becoming more old fashioned to people, which
I don't understand it because I'm not even I'm not even I don't even have a marketing degree, and I understand
the value of you have to constantly get that discovery part right.
And especially on the organic side now with, you know, with the SEO changing, you know, lot of people are, oh, SEO is dead. Not It's dead. It's just changing a little bit. You just gotta be on top of it. You know, I I yeah. Had a dollar for every time I heard that in my life. Okay. I will personally start funding a trophy for every time someone announces that SEO is dead and is immediately proven wrong.
As our experts have already established,
it is not dead.
It has never been dead. It just keeps changing shape and people keep mistaking that shape change for a funeral.
What is different now though is that your content needs to work for humans
and for the machines that are increasingly answering questions on humans behalf.
Now that is a different brief, but it's still an SEO grief.
Oh my gosh. It it's it's such a big thing every I I think it's just changing, and you just gotta keep up with it. And especially now lately, Google's changing a lot of things quickly lately.
But with these new LLMs and people getting used to it, there's a stat that I just read that, like, 50% of US consumers are now using
the LLMs regularly,
which that's a huge jump from last year. I think that was probably less than 12% last year, and now it's 50%.
At least try it once a week like I see it. It's
it's obvious to me. And now it's a new reality for these brands. Like, they have to now
fix and make sure that they're ready for the website's ready for this new LLM traffic
that's coming at part of their organic search process. And, you know, in most brands, that's all siloed. Right? Like, SEO teams over here. Big media teams over here. Creative teams over here. Like,
they kind they now have to work together
to really think the same about the way that they think about discovery. And so
we're pushing a lot of conversation about discovery because I think that's getting lost in all this AI junk,
you know. And and believe me, I'm I'm the one that I I wanna find the next great AI product. Right. I wanna bring it to my clients. I'm looking all the time, like, at this stuff and trying to figure out, like, is this real value or is this just bullshit, you know. And
I'm not seeing it yet. Like, I see stuff for me in the operational side, and we've implemented a lot of these things, but
nothing that I can bring to a client that I'm like, this this is what you need. And we we look for that and we just there's nothing yet. I mean, I don't and again, I'm talking to the for the middle market brands that can't they don't have a lot of r and d budget. You know what I mean? Like, we can't just, like, let you know, put them in a position where they have to go fishing. Like, we have to, like, test this product a little bit, make sure it's something that's really valuable for them. If I was calling on Target,
I'd throw them more ideas that are kinda like, you know, let's go fishing. Right? Let's figure out if this could work. But with middle market brands, they don't have the budget for that. They don't have the time for that or the team for that. So, you know, we're try we try to be thoughtful of, like, if we're gonna bring them a new product or new solution,
it's gotta work and it's gotta, you know, it's gotta be a value to them to drive their outcomes because that's the end of the game anyway. It doesn't matter about clicks or any of these other crowd or media metrics. It's about their outcomes. We're working with a brand right now that expanded
from Canada and The USA
over into the EU.
And
the EU has essentially, no one has ever heard of this product.
Interesting.
And so
it's a real
reminder of back to basics.
How do we educate people on not only what the brand is and what the product is and get some exposure to it? Back to this idea of is the end user that we're trying to reach or the end client, customer, whatever you wanna call them, the lead that we're trying to reach,
are they problem aware?
And then are they solution aware? And are they our solution aware? Yeah. Right? Now we can get them into a sales funnel
where we can say we're gonna give them exposure to more ideas and stuff around our product.
And then
once we do that,
what's the kind of order or process of things that they're gonna see
from an advertising perspective. Or if they go and they Google that brand or they type it into ChatGPT or whatever, Gemini or something. Right? But what's
gonna come out about that product or that brand? Right. And,
know, these are all,
like,
questions you don't normally have to answer if you're an existing established company.
Right. But I think you should go through that process and say, what if somebody's never heard of this product at all? Right. Right?
What would you do differently? And I think you'll find a lot of holes in your marketing funnel if you're doing that.
Well, we've definitely focused more lately on the discovery end, like, you know, to your point about not just for brands that are not known in certain areas, but also just for brands that, like, just are looking for new to file customers. Right? They people that maybe have an interest in the product or solution, but they just haven't found them yet.
If they don't wanna talk about it on the paid side through through the full funnel marketing,
we need to focus on the organic side at least to make sure that end is shored up to make sure that it gives us
an optimal chance on the paid side
to capture them. So my point is that their content is updated frequently and it's readable by the machines as well as consumers. Right? So that h ones have to be right. That those kind of things. The schema has to be set up right. We're even talking about clients like add glossary terms if you're kind of a sophisticated business.
And,
you know, add a add a really robust FAQ page.
That's the stuff that'll get you listed and cited in these answers because people are looking for these answers.
And Yeah. You know, if if we're not gonna talk about it as much or focus on it as much on the paid side,
we wanna at least help brands on the organic side where they don't have to invest as much money to fix this or to update their site,
to a point where it will drive more results organically,
not only through SEO but through the AIO stuff too, through their generative
content.
And I think that's where that's why we're kind of focusing on that a little bit more lately because it just seems like we're not gonna do it on the paid side. You don't wanna invest on the on the awareness side,
you know, for CTV, which I know is it's expensive. I mean, when you're talking, like, if you're trying to run on Hulu, which is a safer,
you know, area to run on, it's, like, $45.50 dollar CPM sometimes,
and it's expensive. But but at the end of the day, it does work in terms of, like, reaching the right people.
And so but if some brands don't have that, it's like, let's focus on your organic and make sure it's really tight, and you're maximizing
at least that opportunity to get the word out where you're not paying additional for for paid media.
And that's been a a more receptive conversation lately. I've been to
I've been to digital marketing,
summit in in Chicago recently,
which was fantastic, and then I just did the brand smart event in Chicago with EMA,
Chicago. And no one was talking about media. Everybody was talking about how can you make sure my site is being indexed properly. So I see my competition there,
but I don't see Right. Us there the way that we see them, and we gotta fix that. And so,
you know, that's really been the main conversation lately, which is great for us. Like, that's exactly what we wanna help them with. I think there's also an education component to the
kind of AEO, GEO stuff, which is essentially, it's how do you get your brand to show up in a chatbot
or an AI system when somebody's asking a question about it? Right? Yeah. So
they might call that, you know, AEO is like artificial intelligence
optimization.
Right? And then your GEO is your generative,
which is the same thing. It's just different words. All the EOs, the EIEIO,
I see that a lot. People talking about it. So
there's one thing on the education side that I think
is super valuable to companies to know,
but that
they're being sold kind of or or missold
this bill of goods that there is some sort of tracking system
that will definitively
tell them what's happening in AI chatbots,
and that is absolutely
bullshit. Yeah. There's nothing about that that works. Yeah. A recent study that was done with SparkTuro,
the important part is
they asked for recommendations
with thousands of people distributed
across different geographies
and areas and different demographics,
and they asked them for recommendations about the same types of products.
And to get the same top three results was a one in fourteen hundred chance.
So,
however,
that does not mean that you should not put effort into this. Yeah. Because what you want to do is improve the chance that you are one of those three.
Okay. This is the one.
And Matt said it as clearly as it can be said. There is no tool that can definitely track
where your brand shows up in the AI chatbot recommendations,
and anyone selling you that is overselling it.
The one in 1,400 stat is, like, genuinely wild.
The outputs are the variable. Right?
So, no, there is no dashboard that definitely tells you what's happening in there. Right? And, honestly, working with Matt, I've seen this firsthand
because clients come in having been sold some version of this tracking promise,
and it just doesn't hold up. Like, this is one of those things you can only really get when you're in the work every day,
And Matt's been saying this long before it became a talking point. So what you can do is put in the work to improve your odds of being included.
And that work, good content,
PR,
strong SEO fundamentals
is stuff you should be doing anyway.
Correct. Yeah. Right.
So
the somebody telling you, yes, we could definitively track this thing is not true.
Yeah. But
the kind of caveat to that is
you should work
to be included in those results.
And that work is about, like you were saying, top of funnel type SEO work. It's content marketing.
Right. It's PR.
Right? It's all these things that happen at the top of the funnel that you should be doing anyway that are great for your SEO.
Yeah. Right? And they're great for discovery, and they're great for your ads.
Yeah.
So how do you think, like and and this is another thing. I I don't mean to get too sideways on this. But No. No. A lot of people like to use
in Google, they like to, like, put targeting, and they wanna say, okay. I want somebody who's in the market for this type of thing. Mhmm. And if you're not familiar, folks at home,
you know,
in market audience means that Google has a decision where they've said, okay. Someone is looking for, let's say, patio furniture.
Yeah. Right? So there's an in market audience for patio furniture that you could advertise to. Well, how do you think Google figures out that somebody's in the market for patio furniture?
It's because their people are searching for it. Mhmm. Right? And so they're getting that search data, and they're saying, okay. We're gonna put them in the in market audience for patio furniture, and now you can run ads to them.
So
and when they're searching the sites that they find
are found because of SEO and content marketing.
Yep.
So to be able to advertise
to someone who's in market for that thing,
you also need to be doing content marketing for that thing.
Yeah. The challenge now that I found is I'm getting mixed signals of things I'm reading about
how these AI platforms
are
ingesting or grading your content. You know, if I've read one
POV that says they're not they're looking for long form now because they want more authoritative,
thoughtful pieces of content now and human written
versus
a lot of AI generated slop, which is happening a lot. I mean,
you know and then I've heard the opposite. I I just went to an event, and people were saying the opposite. No. Like, no. The AI slop still
index as well with a lot of these
LLMs. So I'm getting mixed reviews, so it's hard for me to, you know, tell a client, like, yes. You you do this or no. Don't do that. Right. One thing that's been somewhat universal
that we have not delved into yet,
not only for ourselves, but for our clients is Reddit.
I've had a lot of anecdotal
conversations
with marketers
that talk about how Reddit has done very well for them
in terms of driving more interest through these AI channel.
And
And Pinterest.
So I have not explored Pinterest as much, you know, which is interesting because some of our brands like, we have a we have a high end home remodeler out of Orlando, and they're like, their visuals are just stunning. You know? Whoever's
I always compliment them on their creative agency or whoever's taking those videos and photos. It's just beautiful.
And
they they we've never even talked about Pinterest. Even even to run as part of their paid ad campaigns, they just don't seem to have as much interest because there's house.
You know, there's that that publishing
house Yeah. The HOS. And something like that. Yeah. That's very similar. So I think that they find that that's more their audience. But Yeah. Well, HOS, you can find
essentially home services contractors?
Yeah. Right? That's what it's designed for. Yeah.
But it doesn't hurt you have a Pinterest. You know? Yeah.
Because, I mean, there's a lot of looky loos. Like like, we say that they're we think that their sales cycle is about six to eight months at least
for people doing looky loo stuff, and then they finally make the decision to to contact them.
But Reddit, you know, I've heard a lot I have not delved into it as much, but I've heard Reddit
it's treacherous. You gotta be careful because you don't wanna get downgraded or downvoted,
and you can't oversell or you really can't sell at all.
You know, we don't handle that for clients, but I've I've been very interested in in hearing these stories about how some brands really have taken advantage of it. And they've navigated that that fine line of, like, not selling, but also being helpful
and and trying to add to the the conversation. I think that that's that's an interesting area that we need to delve into a little bit more. Depends on the type of client, I think. What's difficult if you're just selling widgets kinda thing. Yeah. But if you're a service, then it's a lot easier.
Because you're being totally Yeah. You can just go out and yeah. If you're a plumber and you help a 100 people answer plumbing questions,
you're more likely to get quoted
as an expert in plumbing. Right? Right.
Because
and I I think this is something that
people fail to recognize is that the AI is not just
regurgitating
the information it saw on the Internet. It will quote stuff for you. Yeah. But
what it's actually doing is going it's analyzing the information that it finds,
and it's trying to determine
expertise,
let's call it. Right. If it's in if it's, like, a trades person or something like that, or it could be a clinician or whatever. Right? Right. Well, how do you determine
if a clinician
is better or more versed than another one? Or how can you tell if one electrician is better than another electrician? Skill. Yeah. Well, you got you got reviews, which everybody knows about because everybody's always trying to get reviews. Yep.
But
then
what's the next? You look at their website,
you don't really have a whole lot of options. Right? Right. And so if there's a bunch of information showing that that person is knowledgeable and helpful on the Internet and maybe they're answering stuff on Reddit, maybe they got a video series on YouTube,
They got a podcast about being an electrician or whatever, their home uses of electricity or solar or whatever.
Now there's reference material to show that that person is an expert,
and that makes them
more correlated
to expertise.
And this AI system runs on correlation.
Yeah.
Okay. The electrician example Matt just used is so good. I wanna make sure it lands because
AI isn't just pulling the most popular answer.
It's trying to figure out who actually knows what they're talking about,
and it does that the same way a person would using reviews,
website depth, presence on trusted platforms
like YouTube,
podcasts,
Reddit answers.
And the more places you show up as genuinely helpful and knowledgeable,
the more the system correlates you with that expertise.
So that means that your Reddit answers
or comments or
your YouTube video and your podcast are all quietly feeding the same machine.
And that's not a coincidence.
It's it's a strategy.
So I think anytime you can prove expertise, you're gonna have a better outcome. Yeah. And it seems to be working. Mean, I've had several people from different brands tell me the same thing. Like, hey. If you really it takes they're know, they all agree that it takes a lot of work, a lot of time. Mhmm. You gotta be on it a lot.
So I'm not you know, for me personally, it's like, don't have that time for me. And I'm sure there's thousands of agencies on there doing it too. We do there is some stuff you can do. We have a clinician who works in in oncology,
and
we have a video out that essentially is saying
that
getting
a late stage cancer like pancreatic cancer
Yeah. Is not necessarily a death sentence. Right? There's options,
and then it has hundreds of comments.
But we can't discuss people's medical information on YouTube. Right. Right. So kind of the short version is just,
hey. We'd love to discuss your case. Here's the phone number. Here's the website. Contact us. If you want a consultation, we can chat kind of thing. We can just copy and paste that kind of in many times
and take all the people who are saying, like, oh, just eat Invermectin and not eat sugar and delete those comments and ban those people. You know? Right. Because there's all kinds of crazy crap on the Internet too. So Yeah. That's the other thing with Reddit is it is a discussion.
You would think the self leasing would kinda weed out the fact checking needed, but sometimes it just doesn't. Sometimes it feeds itself, which
is dangerous. But Yeah. I think that's why a lot of brands don't step into it because they're just like because of that part of it, but it's like, that's on YouTube too. I mean, go on YouTube and
that crap everywhere. But not on your YouTube ads, though. Right?
Well, they're not supposed to be.
Every couple of years, the feds, like, go to Google and be like, uh-uh.
You know, you got child porn on here. Like, you're putting ads where child porn is or, you know, stuff like that. Yeah. There's always that, like, how do you block all the stuff you need to block?
Yeah. And you can't block it fast enough.
So,
I mean, you pretty much just gotta
you gotta draw a line in the sand at some point. Right? You gotta be like, we're doing our best to make sure that safety and Well, our privacy and whatever, right, are being handled.
Yeah. But also, we have to
be honest
about
the the truth of the Internet.
There are people who can do things faster than you can remove them, and there is no way to combat that. That's right. I think if they had honest conversations about that, it would be
people would understand understand it. Because I think I think most people are reasonable people. Right? And if they if Luckerberg went up to the senate and said, yeah. You know what? Or or, you know, the house in those committee meetings and just said, look. We're doing everything we can, but these guys are faster sometimes than we are. And we're just every time it pops it's like a whack a mole thing.
You know? Every time it pops up, we're gonna do our best to to get rid of it because we don't want that crap on our site. Like,
you know, if they if you I think the other problem is that they it's not in their best interest to do that,
which seems like it would be, but the problem is that it costs money to remove
stuff.
And
Yeah. If people find things that are not necessarily
illegal but also should not be on there,
then more people interact with those and it brings more people to the platform more often.
And they have a financial stake in that,
which is unfortunate, but that's, you know, that's the way the business world works sometimes.
So I think kind of to get back on the, you know,
honestly promoting, you know, your good business, your solid service or products. Right?
I think the idea of getting back to kind of fundamentals
of marketing
is where people should be focusing.
And it's not just like it's not like a backlash against AI or fake kind of thing. It's about
how do you
go through the steps of how somebody,
you know, gets trust in your brand or your product or your service and identifies that it's a thing that they need and then takes the steps,
you know, to make their purchase or to convert or whatever that end goal is.
Yeah. It just seems like that, you know, the the discovery
part is
being left behind a little bit in these conversations. That's all. You know? And we kind of identified that early on,
and we just decided we gotta make this a thing because it's it's important. And it it will decide whether our paid strategies
will pay off or not. It'll be it'll play a part in understanding if the paid strategies will work,
and how much we spend on those paid strategies. And so, you know, we we you can pay less in paid if you do well in organic. Right? And so
they feed each other and
we've never kind of focused on it that way, at least in the since the inception of our company almost six years ago.
It's only in the last, like, year in in understanding the power of of these LLMs and and tying into the SEO part of the game
that we're like, you know what? This really does all go together. I mean, we just have to, you know, we have to look at it that way.
And I'm hopeful that I mean, you know, with with creative being tested, I mean, I I love these tools. Like, I love diving into all these AI tools and testing them out. They're not all as intuitive as people would make you think. And I'm not a developer, you know, I'll admit that. Like, you know, I wish I was a developer because then I would probably be able to use some of these tools like an a then and something and create these agents and,
you know, do things a little bit faster, but I'm a little bit slower at it. But I figured it out. I mean, I I'm starting to figure it out,
because I wanna keep building things that are of value to the client. And that's where we are in our kind of evolution as an agency,
is we're trying to figure out, like, what are the things we can build that actually are of value to the client.
And, you know, beyond just, like, twenty four seven reporting dashboards and,
you know, things of that nature or just optimization auto optimizations if they wanna turn it on in the in the platforms, those kind of things, which nobody has thankfully to date, knock on wood.
Said, you know, turn on the AI auto optimization. It's like, really? No. You don't want We tested them. We tested them. Did you? Did you do, like, a run of e maybe against, like, human versus AI?
Honestly,
we had two
separate markets for the same product,
and so we're setting up a new one for that product in a different region.
But, hypothetically,
the system is supposed to just remove things like words and percentages and stuff that might be on the photos. Yeah. But, also, it will brighten them, change kind of, like, luminosity and and things like that in the images.
It's not like taking your product and, like, sticking rainbows and mermaids and shit on it or whatever. Right? It's slight modifications.
So we try to And the next band name is Rainbows and Unicorns. That's right. Rainbows and Unicorns and shit. But, yeah, we found it worked it worked reasonably well. Right?
You know, we had a better conversion rate. We can't a 100% attribute it to that fact.
Sure. But also and this marketing. Right? Right. This bears out historically
in that if Facebook
is
constantly
asking you to turn on a feature
that they
are rewarding you for using that feature.
And On themselves. So there there may be some artificial
increase in engagement or reach because of using those features.
But like I said, we don't have a definitive answer.
But also and this comes up all the time on, like, LinkedIn and stuff, is people say I can almost repeat it because I heard it so many times now, and everybody sounds the same. People buy from people.
They want real photos
of real humans using your product in real life. They don't want AI photos. They don't want AI videos.
They want human to human real emotional connection.
Right? Which,
in theory sounds a 100% true.
However,
in our experience,
a mix of real and AI together
is the thing that seems to sell the most.
I don't know if it's just
yeah. Our clients don't have the budget to spend $30,000
on a video shoot,
and so we're using the AI to make up for that fact.
However, we are always
putting
the real product photos with
AI kinda mixed in. Usually, we're using the AI to, like, animate stuff or
voice overs or that kind of thing.
Okay. This people buy from people line is one of those things that sound so true that nobody questions it, and I think Matt is essentially questioning it with actual data.
So real versus AI may not be the right frame. The better question is,
does it feel authentic,
and does it solve the creative brief?
And as a creative lead at Hook Digital Marketing,
working with SMBs every day, this is something I've seen play out, like, so often.
Like, a well used AI enhancement on a real product can outperform a fully human shoot that
cost nearly, what, $30,000.
And it's honestly been a game changer for the clients we work with because it kinda also levels the playing field.
These businesses can now compete in the market even when they lack the full resources.
So I think I agree that a smart mix isn't compromised.
It might just be the smarter creative strategy.
It's interesting. I I I remember the big backlash at Google for the Super Bowl ad where it was all AI, and you can definitely tell it had that, like, glow in the ad. Yeah. You can just tell it was AI.
I didn't have a problem with it. I don't know why people got so crazy about it. It's like, it's Google people. Like, we're using it every day. We're still gonna use it tomorrow. Who gives a they didn't even have to run an ad. Why they ran an ad? I don't even know. Because we weren't it sounds like we're not gonna use them. We're gonna use them less because of this or more because of this, and we're certainly not gonna use them less because they ran an AI ad. I will say, though, I I agree with you that I think a good mix
people just wanna know that there's authenticity somewhere. And if they see it an often enough, I feel like and I haven't tested this. It's just a thought my thought. But,
like, authenticity will will will reign supreme still eventually.
But I think in the meantime, people are okay with some fun AI stuff that you're doing. The one thing I would say, and I think our governing body of of advertising digital advertising should like, IAB here in The United States should,
you know, make this make this a standard is, like, if it's being used, call it out. Right? Like, make us call it out. Because I think if people
understand that, like, we're using it's kinda like ad choices, right, for digital, like, your voice is thing,
which I know kinda suck because it's like, for creative people, they're like, that thing's annoying because it looks bad, whatever. It messes up creative. Not really anymore. No one cares. You know what I mean? As long as but as long as you put it something out there that lets people know, like, hey. AI might have been using some of this. TV shows are doing that now. I watch the History Channel
a lot because I'm a history dork. And a lot of the recreations of battles and stuff, they say in the beginning, like, some of this is simulated with AI. Great. Love it. And I love the visuals. Like, I think it's fantastic.
I think if people are if there's, like, that conversation quickly and that notification quickly, I think people will buy into it, and there won't be any backlash. It's as long as there's some
authenticity
behind the brand.
There it cannot be an all AI thing where you're communicating just through AI generated content,
visuals,
videos, etcetera. Funny story. I I I've been trying to get a thirty second commercial for my own brand because I had a sponsorship where I could run it,
and I didn't get it done because I was starting for like, same thing. Like, I was starting from scratch trying to figure out what tool to use. We'll chat.
Yeah. It's hard
to, like, not use AI, especially if you're a middle marketer or smaller business.
Right. You've gotta take advantage of stuff like that because you don't have the capital to not
to do it the, you know, the other way. And so make it really polished and spent for you. Yeah.
Well, I think one of those along those lines, one of the things that
kind of I I identified
early because I was writing, you know, well, AI take my job
was that we need to kind of shift
some of the team members that we have on staff.
So we hired a software developer
Mhmm.
You know, couple two and a half years ago now.
And,
also, our creative director is a trained filmmaker.
Oh, interesting.
Yeah.
Which is I would never have hired a filmmaker
before AI.
So the people who are like, AI has taken all the film jobs. I'm like, well, in this case, there's no problem. Right? Yeah. Because I need somebody who understands structure.
They understand story.
Right? And I also need somebody who can, you know like, Scott runs our internal n a n server for our automations.
And,
you know, we have a software developer on staff. We need people who understand servers. We need people who understand GitHub and all this kinda crap. Right? And
so
these are kind of positions that we have in our company that allow us to take advantage of the
of, you know, some of the AI stuff that's out there that's technical and it's difficult to use, honestly. Right? Yeah.
So I think
kinda moving forward, though,
I think we're definitely at a space now where,
and I think you would agree that
kind of paid ads,
SEO,
and the AEO, GEO,
you know,
space as well as kind of the content marketing space
are kind of all turning into one job,
basically. And
that job is,
you know, how can I use an
old school kind of system of marketing
but modernize it with the tools and and the things that we have available?
Yeah.
It makes it feel like it's old school to me, especially when I talk about it. You know? I some people just roll their eyes and
you know? But, again, I'm I'm using all these new tools too. I'm trying them too. It's not like we're stuck in our ways of, like, it's gotta be a full funnel strategy. But I've just seen those work. And and not only that, like, the the you know what the you don't talk about anymore as much? First party data.
Why? Because you have no control over it. If you give your if you give your credit card to Google and Meta, you're not getting first party data.
Right? Or you're not getting, you know, great insight into your audiences because they don't share that in the reporting as much as they should. Right? When you turn it over to them in, a performance max situation, for example. They limit what you can pull out of the insights
in terms of insights. And so Yeah. You can't see the data. And so Do you remember ten years ago when it was actually called webmaster tools,
and they would tell you everything that someone typed into Google to find your website? Everything. Incredible. Yeah. Can you imagine that today? It'd be amazing.
I can use that all day long. They wouldn't need us. They wouldn't need us anymore. But I think the other thing, like, I do think there that. But Well, I do think there is a there is a
there is a,
not psychological,
but
I don't know what the word is I'm trying to think of. But there is I do feel that Google and Meta are
attacking agencies. They're trying to get rid of agencies,
and they're just making they're just slowly doing it. Right? And they're like, look. You don't need them. We can automate all of this. And I think
I do think for small businesses, smaller businesses,
yes. I think these tools are fantastic. I'm talking about businesses who spend
maybe 1,000 a month, maybe,
you know, at any kind of paid advertising for their business. I think that Meta and the the tools that Google and Meta offer right now are fantastic.
Those businesses aren't really focused on first party data. Just trying to get people in the barbershop chair
Right. You know, to grab the flowers when it's time, you know, that kind of stuff. Right?
But I do feel like for these middle market brands and these even larger brands I mean, larger brands can make mistakes and just, like, write it off. Right? These middle market brands don't have that kind of budget, and so you can't you know, to turn the strategy
over
you're really turning
you're really turning two different strategies over to two different platforms
wholly
in the trust that they're gonna, like, be good stewards of your cash and not overspend and not spend on bots that they either coming in as a as a click.
Which historically,
they have not had a great track record. Okay. This is the thesis of the whole episode right here, and I don't want it to get buried.
SEO,
paid ads, geo,
content marketing, they're not four separate lanes anymore.
They're one integrated function,
and the brands that keep treating them as siloed departments are going to keep being confused about why nothing is working
because discovery ecosystem
is one thing.
The sooner your team operates like it is, the better your outcomes are going to be across all of it. Like, what trust have they built with you over the years? Yeah. And the other thing is each of those ad platforms has, like, a thousand buttons in it.
Yeah. Right. You gonna know which which button to click or not click? I mean, it's unlikely. It's like me trying to fix my own car. It's not gonna fucking happen. Well, I mean, even with the AI, it's not gonna happen. I don't know how to fix that thing.
Well, not only that, you're not gonna get support either. Like, good luck trying to get anybody from Google on the phone to help you with something. They'll sell you all day long. Yeah. I get five emails from five different reps saying they're my Google rep, and and they're like, we'll help you. Well, you're only gonna help me try to spend. I'm not Right. Because they're all commission based. Doesn't work for
People don't know that the that the ad rep who calls you on Facebook or Google is commission based. They don't make more money unless you spend more. That's right. Oh, wow. Two things back to back that I feel every business owner needs to hear.
I think Sean's point about Google and Meta slowly making agencies seem unnecessary is something the industry doesn't talk about honestly enough. And I think he's right. For small businesses, it may work fine, but for mid market brands, that's like a dangerous game.
And what Matt said about your Google rep being commission based, like, their goal is not just to optimize your account,
it's to get you to spend more.
And I think that's not cynicism.
That's just how the incentive structure works.
So next time when they call with recommendations,
don't forget to run those through your own strategy first.
So their strategy is always to spend more. Yeah. And
I think the other thing
that you wanna focus on is,
like, there is a lot of behind the scenes technology
pieces
that businesses,
for the most part, have never even heard of that have a massive impact on your advertising. And I like a perfect example
would be Google Merchant.
Almost no one knows what Google Merchant is
or how Google Merchant works or what Google Merchant connects to or why you would wanna use it,
but it's free advertising,
just setting it up. Right? Yeah. Your products can go out and be shown on the Internet, and then you can connect it up to your ads, and you can run ads for products from a feed from your website. It's amazing. Yeah. Almost no one knows what it is. Right? Unless you're an agency.
Yeah.
Shit. Half the agencies I talked to don't even know what it is. Yeah. It's it's it's it's definitely a herd mentality. Right? Like, I I Yeah. You know, people read it,
Facebook and or Meta and Google make up 65
of every dollar spent on digital advertising.
So it was like, well, I should probably spend more money there. You know, it's like, okay.
You know, and then Amazon comes along, kinda breaks that up a little bit, but still it's now really those top three that a lot of people are spending on. But, you know, there's so many like, that's where I think programmatic stands out. And the programmatic has a lot of issues, don't get me wrong. And if you if you don't have somebody that knows how to,
navigate and, like, set it up in a brand safe way where you're limiting the bots and you're not doing mobile apps that are game based necessarily because sometimes those are just, you know, fake clicks and all that. You know, as long as you have someone in mind in that store, like, programmatic is the channel to be really valuable because you're you're you're hitting people, the same people on the open web,
not just on Google and Meta, and they spend, you know, where they're spending other time,
and you still can get that message out and frankly in a lot less expensive way.
And that's why, you know, we we we talk a lot about programmatic for that reason. It's like, look, Google and Meta are fine, you know, but at the end of the day, if you want a little more independence, wanna build up your first party data, like, let's do this with programmatic and let's run CTV and what we're targeting and that kind of thing, and that's worked almost every time.
So we talk a lot about that a lot. But what's interesting now that's happening in the age of AI is
we're starting to see clients come to us. Like, they'll run, like, our report of our performance through, like, a Claude or ChattyPT,
and we'll be like, tell us what's wrong with this or where we where we're missing the boat and all this stuff. And we had that happen recently with one of our clients.
And it really demoralized our team because we've been doing really well for this client.
And
it kind of hit us all like off guard. We it caught us off guard because we're like, what what what's going on? And it was it's simply the the the owner is just trying to, like, operationalize
more efficient ways of doing business and to be more profitable. Right? Which makes total sense. Like, I think he he had them all down with, like, Quad,
co work.
I think it's called co work. Right? On their Right.
On their desktops.
And he's like he's actually going out and teaching the employees, like, how to use this best for their particular area of
Tell him to hire a IT security company.
He's like, you know any in mind? Do you have any in mind? I do actually. You know what? North Shield IT is a good one. That was is that the guy that that does it in Canada in The US that we we Yeah. Our groups? Yeah. I really Ryan from Systemic Digital runs North Shield IT as a wild security company.
Man, all the I all the AI, like like bots and agents? Yeah. Good lord. It is a security nightmare. Yeah. Like,
here's the thing. I worked on servers and in IT as well as marketing, but I started in IT in the nineties,
and I'm not using it. Like Yeah. I'm still using on demand bots and our internal automation system because I don't trust those things.
And
I and it could just go install a repo
for a that is full of malware and put it on your machine or into your software and run it. You have no idea. Yeah. It's just it's a security nightmare.
What's one I let me ask you, like, what's one what's one AI solution or AI based solution that you've implemented for your business
that's made a massive difference?
And you feel like it's really moved the needle either up either
production wise or efficiency wise or, you know, what's really, like, moving you before you go, wow. That was a great idea to be doing that.
I think probably one of the biggest ones
was
a program that we use called Higgs Field.
And
Higgs Field includes
all the image generators,
editors,
and video generators into one application.
Wow. And so if you generate an image, you can then use that image that you generated.
So let me back up a step. We can upload like a product photo
Yeah. Generate an image with that product photo in it,
and then use that as the reference for a video generator all in the same app. Where it'll generate the video from scratch. Right? Yep.
So I didn't do that. Right. I just built that agent in in Mind Studio. So that was part of my level one certification. I built that exact
solution. Yeah.
Well, I think one of the the interesting things that we found by using apps
that have more than one AI in them.
So, like, something that has more than one image generator,
more than one you know, rather than just, we use chat GBG in this. Right?
Finding out which models are good at which things
is a game changer.
And
training our staff
on that fact and how to kind of use different models for different things and start to figure that out Yeah.
Makes us way more efficient and way more capable. We get a much better output in a shorter amount of time.
Interesting.
And that is part one.
What I'm taking away from this conversation is pretty simple.
The platforms want you to hand everything over and trust the automation,
but the market still requires you to think.
Organic and paid feed each other and discovery is being left behind and it shouldn't be.
And also knowing which AI tool does which job well is its own skill worth developing.
Alright folks, part two with Sean and Matt is dropping next week. So make sure you're subscribed so you don't miss it. And if this one got you thinking, share it with someone who needs these insights, and I will see you all in the next one.