Heads In Beds Show - Are You Making Real or Fake Progress With AI?

Episode Date: April 1, 2026

In this episode Conrad and Paul talk about the realities of AI slop, how you know if you’re making real progress or fake progress, being fooled by hallucinations in AI data, and a lot more! Enjoy!⭐️ Links & Show NotesPaul Manzey Conrad O'ConnellConrad's Book: Mastering Vacation Rental MarketingConrad's Course: Mastering Vacation Rental Marketing 101🔗 Connect With BuildUp BookingsWebsiteBook A Call With Us🚀 About BuildUp BookingsBuildUp Bookings is a team of creative, problem solvers made to drive you more traffic, direct bookings and results for your accommodations brand. Reach out to us for help on search, social and email marketing for your vacation rental brand.

Welcome to the Heads of Med Show presented by Buildup Bookings. We teach you how to get more vacational properties, earn more revenue per property, master marketing, and increase your occupancy. Take your vacation rental marketing game to the next level by listening in. I'm your co-host Conrad. And I'm your co-host, Paul. Hey, hey, Paul, what's going on? What's Apple? Well, you know, it's, uh, I, I don't know.

It's, it's that time of year that we got a little March madness. Baseball opening night is not opening. day. It's opening night. Someone can fight me in that somewhere down the future if they really want to do that. But there's baseball starting. I'm going on a little vacate here. So I'm going to head down to Florida. What's Florida Paul like?

What are the differences between up North Paul and Florida Paul? I feel like the listeners need to know. I hope it's that I can turn my devices off. That's what I'm really hoping that that's going to be the big difference. There's no difference, really. It's just that I wear, I can't even say I wear different clothes because I wear shorts around here all the time. You guys just don't get to see it most of the time. So I think that's maybe in my more natural setting is that I can wear it without having to shiver half the time.

But yeah, it's just, it's a great time for the kids to be with Grandma and Grandpa, and Grandpa, I spend some time by the pool and me to enjoy those times as well. So it should be good. But, you know, how? Do we have an over under on the first? score you're going to shoot when you get out to the golf course? Oh. You want to set it right now?

I, it's going to, the over under number, I'm going to put at 91 and I'm begging for it to be under there. Okay. I'll take the under. It's, I hope it's, I think it is. We talked about it. I know these courses at least. So this should be just re-learning the swing a little bit for 2026.

So that's, yeah, yeah. It'll be a good time. That's, that's all. I'm jealous. It sounds like you're going to a cool spot in Florida, so it should be fun. You know, it's funny. You want to make progress in your scores this year, of course, as we all do, whatever hobby or interest we have a choice. So too does every vacational manager. They want to make progress on their marketing, on their business, on their growth goals. And obviously, you know, for the record, right, I read this in the record right away, Paul and I are big AI fans. We love

using these AI tools. I feel like Claude's my best friend now. Like there's so many things there that we're doing there to the other basis. But we can all fall in this trap, I think, no matter how skilled you are with some of these AI tools, they can lead you down the path, I think, of making what I'm kind of calling like fake progress, like this idea that you're actually not making as much progress as you think because you're clicking a button and it's producing reports, it's producing information, it's doing things for you. And of course, you see viral threads on Twitter and you see content on LinkedIn every day, but I have a team of agents, it's doing 100 things. And then at the end of the day, it has to come back to like a practical application

of what these things actually do in your business. And we have to let the hype maybe machine die down a little bit or wear off a little bit. And we've got to find like, what are the things that are actually, these tools are actually doing. How are they making me progress? How are I making progress with these tools? And how do I make sure that I use those appropriately in my business? So anyways, we're going to be focused mostly in the marketing side, but we may touch on a few other topics that are marketing adjacent as you were.

And ultimately build systems that really work well. So this idea kind of came to us because it's a trend right now, right? Like again, you lock on all the time and you see everything, you know, like the one thing I say in Twitter that does crack me up a little bit is people that are kind of becoming these like open claw influencers to some agree. and they'll post about how they're burning a billion tokens a month. And they're doing these things. And then everyone's like, cool, like, what are the products that you're building

with these billion tokens that you're burning every month? And then sometimes you learn what they're building is sometimes not that interesting or not that good or frankly, it doesn't really isn't a better solution than other products that exist in the marketplace. So then they kind of get a little heat for that because they're like, okay, well, you're doing stuff and these machines are constantly working. But it's kind of producing what some people are calling to, like AI slop, like is producing AI content or it's producing an app or whatever the case may be.

but it's like, congrats, you spend a billion tokens recreating, you know, an app that's already out there that's free that works well. Like, congratulations. You rebuild Apple reminders on your phone. Like, is that really going to move the needles? Is that some drastic thing? Or is it more just a bit of like a loop going around in circles? So anyways, we don't have a great title for this. We think the title we've come up with is decent. But this is the concept we're trying to achieve. So what's your thoughts on this, Paul? I mean, like I said, we're AI friendly focused here. But we also have to acknowledge this is a problem with some of these tools out there for sure. I mean, even the way you were just describing that, it takes me back to a pre-AI time to like time tracking and time management. And I remember at different times having

different companies put a real focus on time management, time utilization. And again, when that happens, I think there is a tendency to make yourself look feel bigger, busier than you actually are and you understand some things there. Are you saying you submitted a complete not truthful time log, Paul? Are you admitting that now in front of the court? I am, that's not even, no, no, no, no, no. I had to review the time logs. That was the downside of it. I wish I would have gotten to fluff my numbers there a little bit.

Are you saying that your team submitted, not true. I am saying that I saw some things that I clearly understood as not being actually moving the needle. But a lot of the solutions that are being created right now are two problems that are not problems. And it's just, again, that's cool stuff. I'm not going to deny that there's some amazing things coming out, but they're not solving problems. They're not contributing to the ultimate end goal. They're not contributing to the KPIs that we're measured. I think talking about goals, talking about what we always say measurable, make it measurable, track it back to something that is tangible, a goal, you know, something that is a cheap goal, that is all those things. But if it's not

moving the needle for your business, then, then yeah, are you really making any progress? And admittedly, you know, I've been on the end of that where it feels good to be writing a lot of this content and getting a lot of this content ready to go. And, and then you get ready to post and you're like, well, but this is, this is really not moving the needle anymore than anything else is. Like, are there some other ways that I can actually improve the experience for the booking experience and things like that? And I think that's where, We don't have a lot of those options. So is it good to be doing some of this stuff?

Sure. It's, there's some value, I think. But putting that slop on the internet, putting this slop on your website, I don't think it's going to have a positive impact on the long term. And in the short term, we may absolutely be seeing some impact of maybe how many people have been putting some slop on their websites because just yesterday, today, In this frame, we're in the process of the March spam update of 2026 being rolled out across Google. And much like core updates previously, helpful content updates, these impact the overall search ecosystem.

This is why it's just so much fun to be in the SEO space is that at any given time, and Google's always updating the algorithm probably 100 times a day that we don't realize in a very small scale. But the large-scale changes really do tend to have pretty major impacts. I mean, you talk about the helpful content update that took out some small business websites. Like, there's still a bigger fuffle over recipe sites and things like that where AI is just taking their content. I mean, and in some cases, just getting rid of the organic results. So it is. All this is to say that you can make progress.

You can move the needle with AI. I just haven't seen a whole lot of people doing it. I think that when you're really starting to harness, I mean, when I talk to you, these are the types of things where strategically operationally behind the scenes, you've got some stuff that's moving. Again, I still think like the case rental industry is an AI chatbot industry.

That's about how far we've gone. Maybe let's assess some sentiment. Maybe let's understand this. Maybe let's understand a little bit more about our guests or maybe our homeowners, but not a lot more. So yeah, I think that that's really moving the needle of AI progress. We have to find something that's measurable, something that you can set a goal at and really try to put a strategy in place to reach that end goal.

But what are you seeing, I guess, on the front lines as you're talking to people and seeing AI being put into use? Yeah, I mean, I think we had Jack Zappa on Art of Hospitality, and I think he had a really solid thing where we talked about the fact of like, okay, I have specific problems I'm trying to solve. mostly it seemed to be around guest communication and like in particular kind of problem communication hey like this roof is leaking or there's a pipe leaking or something

how does they get routed to the right person very quickly AI can be such a great middle middleman middle character whatever you want to call it that would get that issue from the guest complaining about it over to the person that can actually fix that problem very quickly so it's not like an anti or it's not like a we need less people type of thing at all is the way I think about it necessarily it's a oh gosh this was great because they're the latency between a guest complaining about an issue

and someone being addressed that can actually fix that issue was minutes instead of potentially hours or days or whatever the case maybe if a guest complained about it later. So that's kind of what I get excited about when I think of the use cases of AI is like our clients' success team as great as they are. And I think I try to give them a very reasonable number of clients to work on. There's always latency. There's always that delay between they ask for something and then it has to get kind of filtered back to the right people on the team. Operationally, that's like the challenges that we have as an agency. How could we reduce that latency with respect to like they put a request in? Could have

go in there and be like, I think you're asking this. Let me pre-write a message. That's one thing we're playing with right now. And like, I think this is what this is the case. But then the CS could quickly review it and be like, yep, correct or no, not correct, or I need to make some tweaks. But then you kind of have a draft waiting for you. I like those ideas a lot too is things like email responding to emails where it's like, give me a draft. I want to send it. I think it's still need that I think needs to kind of be in control of the car or be in control of the boat or whatever you want to call it. But I like the idea a lot of AI being that little helpful assistant that's like, I think this is what you need. Confirm it and then go ahead and send it or do it

for me, but this is probably what it is. And I think that the further we get along what practical AI looks like in most businesses, it's going to get smarter. It's going to get better. So I think the human loop stuff will actually go down a little bit over time from what we compared to right now. But then the guest expectations will be, well, I sent the message. I expect a response back right away. You know, if I asked for the Wi-Fi code and I dealt with a property manager that was using AI messaging, which like you said, the AI chatbot stuff is not like revolutionary ground baking in some ways. But in other ways, it is really helpful. Like it is a great tool. Like, I don't know, it's like a lot of things aren't, we don't feel them as revolutionary

groundbreaking now, but we also be really mad if they didn't work, you know, like really fast internet when you're on your phone in the middle of nowhere is kind of incredible. You know, like, if when internet, like, if you're ever on like these really slow old connections and they're not working well, you're like, oh, this phone isn't working. It's working fine. It's just you're used to internet being super fast when you're on your phone and what it doesn't work. You get angry at it, even though it's actually like completely reasonable for seven years ago, let's say for the internet to be that fast on your phone. We just get so used to things. So that's how I feel about things like AI chat bots are messaging for property management companies is if I have stayed with one of those companies, I send the message and then they respond in six seconds with the Wi-Fi code because I lost it. I'm like, that's good hospitality. I like that's like, you know, if it's like, oh, yeah, sorry, Susan was out of office and then Susan returned tomorrow at 10 a. And meanwhile, I didn't have the Wi-Fi last night to watch my Netflix before I went to sleep. Like maybe that's going to make me a little perturbed. It's like, okay, it's not a big deal, but it's enough where I'm not happy and it's a problem. And I always use Wi-Fit code as the example because it's like universal. I always use Wi-FETCO is the example of how this go.

I mean, I think I don't have a lot of good comparisons for this. I think because the AI industry and our application of it is so new, you know, I would say like, I do feel like there's examples of this though in like real life. So one thing I've referenced before, I think on the podcast is going to the gym. And if when I do my treadmill days and I watch people maybe because I'm like, all right, I'm here for 40, 50 minutes. Like there's not really a lot to do. It's like I listen to music or podcast or whatever, but it's just kind of like boring.

You're just kind of looking ahead is what it is. And you watch people go to the gym and they walk around for 50 minutes and they don't do any exercise and they go home or like very little exercise. And I'm like, that's kind of how I feel sometimes, like people that are doing this, this AI, so-called AI, you know, all this building. And then, like, they walk home and I'm like, okay, like, you burn like 17 calories while you were here. That doesn't seemingly move the needle that much, right? And your fitness goals or you lifted two weights, like, that's not really going to make much of a difference. I think that's kind of what I worry about a little bit with some of the AI stuff is that it can create the scenario where you are lifting two weights, you're walking 17 steps and you're not actually doing what needs to be done when you're at the gym, which is hopefully lifting a ton of weights or, you know, doing a lot of cardio or something that would actually help you, you know, potentially get a little. little healthier. So that's kind of the analogy that I could do in my head. But I think because it is honestly, if you look at like the interfaces of chat GPT and Claude and Gemini, all these interfaces, they're so fast. And it's almost like addicting to like put a message in there and then get a response back right away that gives you something to like work on. We'll do our marketing strategy example here in a second. But it's kind of fun to do that. So like it just kind of makes you want to do more of it.

It's like this fun loop of like asked a question and get a response back, et cetera. Is it a good response? Sometimes no. But again, I don't know. it's like that feeling like you're making progress than you may not be. So I think there's a lot to be said for that part of it as well. I think there are a few people, not as much in this space, but I think there are a few people for whom it is more important that people know they're working with AI than it is that they're actually moving the needle with AI. And that's, I mean, again, that's fine. If you're if you're able to sell products, if you can do whatever. But truly that the AI is one of those things right now that I mean, it's the AI. It's the buzzword. You got to have it in your podcast. You got to have it in your conference presentation. You got to have it everywhere. To a point where it's more important to say you're using AI and doing AI and tell everybody about all the cool things you're doing with AI

than to practically use AI in the scenes. And I think that's where some people want to share the cool things that they're doing. That's fine. But are we getting to that completion? Are we doing it? Or is it just, I have to be seen? been doing AI because I don't know. It's that influencer type of thing where we don't need AI influencers. We need people that use AI more effectively is what I would say. But I digress.

That is just what you know, let's talk about a little experiment that we were talking about there with using AI and what someone could potentially do, a potential use case down the road for, oh, let's say marketing, putting together a marketing strategy via AI. Yeah. Well, I was going to say let's maybe. Maybe. Let's go back to that classics, right? So we talk, I mean, see previous 180 episodes on search, organic and paid, social, organic and paid, and email or SMS or outbound messaging. If we just kind of focus on those things for a minute.

I think an example of fake AI progress with organic search is I've generated 100 blog posts and they're sort of thin, 300-word blog posts. They don't have much substance. They don't have much unique content. But yeah, you can click a button and generate 300 unique blog posts, you know, unique blog posts with AI. And that is a thing that you can certainly do. I really, highly, highly, highly doubt those will provide any. meaningful organic search traffic to your website, to your blog over the next 12 months.

Maybe one or two might do a little bit of traffic and then it'll quickly fall out. I feel like every few weeks we see someone on Twitter who's talking about the fact that they published 10,000 pages on their website using some AI platform and they see this initial spike and they kind of don't realize that like, you know, that is like the rise before the fall, that sort of thing. And then, you know, they never talk about it four months later because that traffic's all gone away. So again, I'm not saying don't use AI to help you create content. That is very logical and smart thing to do.

Please do that if it makes sense. but I also believe that nothing unique, like just copy and pasting a pure AI written blog post, particularly from like a free chatubt or free Gemini, paste that on your website is a path to nowhere. Like you could publish 10,000 of those things. And I think they would garner you very, very, very little traffic, nothing meaningful. In fact, I would wager that 10 human written articles following like best SEO practices will drive you more traffic than a thousand, very short, you know, not well optimized AI articles that you might work on with, you know, kind of a cheaper model or a free model.

Again, a lot of nuance on that of like how to do it the right way and stuff like, that, which we've worked on a lot over the last few years, but that is certainly something that I think would fit into that fake SEO progress category. Maybe a positive example on the SEO side. Some of these tools on crawling have gotten so good. And the SEO technical like auditing capabilities of AI tools have gotten really impressive. Like the AI will not give up, quad code in particular. If I give it a 10,000 page website, it'll be like, cool, I'll need to do this in chunks, but I'll do this and I'll come back to when I'm done. And like, it just sits there and reviews every single page if you want to look at this stuff. That's unbelievable. Because if you want a human to go and audit 10,000

unique pages and let's say, let's say spell check a whole website. I used to do that by hand. You know, back in the day, you can now use Cloud Code, crawl the whole website and say, look for any typos on every single page. And they'll be like, sure, I'll do that. And I'll come back to you in 25 minutes if there's any typos or broken links or whatever the case may be. Some of these things you could obviously do with, um, we're screaming frog as well.

But, you know, Claude is getting so much smarter. Like that's the thing about Screaming Frog. A great way to display information. Never was an interface to fix anything. I could see a future with Claude where like you started on your website. You say, find me any typos, flag them for me. You know what? actually just go ahead and fix them for me.

And then Claude's like, I got you. I'll log in the back end of the website. Just fix it for you. It's like, oh my God, that's amazing. Right. Like,

you can kind of see where those things are going. So that would be an example in my mind. If we just maybe went one at a time here of SEO fake progress, generating a thousand AI generate articles that don't really move the deal. And real progress, which is like Claude, go crawl my website. Or find me all my typos or find me all my broken links and then log in my back in

my website and fix it or use this command line interface, log into my WordPress admin area and make these changes in the database. And that to me is like, oh, that's exciting. Like that's one thing that we don't worry about anymore or one thing that we can fix now in 10 minutes. They usually take 10 hours that we can provide a better experience. The Google, I would argue, as well by fixing things like that.

So that's what if you were to do like, you could do an SEO example or paid out's example, but curious what you would kind of feel on your side. And that's the SEO example. I think that's the perfect SEO example is that, I mean, I have. I've seen people really start to try to get more granular with their content. And again, going down to city levels and, market levels and all these, but at a certain point in time, are you providing any additional value? And that's, that is something that in an SEO world, that's what we were focused on.

We were focused on putting more content out there, volume of content out there. And now we've used AI to, to once again, you put volume of content out there because we think that's what Google wants. The end user does not need that answer. The end user is, I think that that That's where John Mueller has been saying it for me ever, right? But everybody's, all these SEO people have been saying it forever. But because we see the tools, the tricks, the tips, you know, all these things that we think move the needle. And we track it in search console. We track analytics and all these things.

But then you can't. You go out there and you put content out. Content is not what the people are looking for. Putting more content out is not going to get you in more LLM results. Answering more questions. We'll put you in LLM results. I think you hit the nail on the head there. If you're putting out a piece of content that has answers to the top 15 questions that

your travelers or your homeowners have right now, they're going to rank. Your website is going to rank. It might not rank 15 different times, but it's going to rank. It's going to show up an LLM results. So I do. I think that that's something where just putting the volume of content out there, I have seen people try to do that. And I have seen, I'm not going to see their rankings tank, but they have not seen the

growth that they anticipated by creating 30 pages worth of content. That's something that you have to drive an end user case. I think that like a 10,000 page website may not drive any additional value than a 75 page website will, just based on the content that's on those pages. So I do. I think the hard part about our space already, the vacation rental space is that we already have bloated websites. We got all these API, we have all these rental listings that are coming in that could be 5, 10, 15, 60, 70, 200.

So there's already bloat. We don't need to add AI to add more bloat to the website. And I think it's the same, like, the same thing you can go into the paid ad side of things. Where I've seen people try to do that, generate a lot of campaigns, single ad word campaigns. Oh boy, that's something we tossed aside. in like earlier iterations of understanding of Google Ads.

And now we think that that's a good thing because we can do it. Just because I think there's a rule out there just because you can do something. Doesn't mean you should do something. And it is. So I mean, I think that ultimately we can't forget about the best practices of marketing that you've used for the last. And for some people they haven't. The five for us, it's a decade of experience. knowing that what has worked and what hasn't worked.

And just because AI says, oh, yeah, this is going to do something. Let's just check ourselves. Let's do a little reality check and say, you know, this is why this isn't going to work. And this is the little shout-up for this is not to say people can't grow into really great marketers. But those people who are specialists who have a year or two of experience, who have shown some results over the last couple years, can they do it is i don't i don't know that they have the long day it is i think there are there are some people who are showing results now that aren't going to be able to show results in a certain period of time

so it is it's there's something to be said for AI cannot replicate experience of having run and failed with your campaigns so that you can improve them moving forward they they don't know your failures you know your failures that's what you learn that's what you put into these updated renovated optimized, enhanced campaign strategies. Yeah, yeah. And I mean, I think you said this a minute ago, or we said this before rate record maybe, but this idea of like being X percent better than the AI or that last 10 percent or that last, you said 1 percent, I think even two before the call on something my paid ads

would be, which I'll give you an example, again, on the good side for this conversation. We have a Google Ads script that we built and it goes and looks at all the search terms and it does matching analysis. It just kind of says basically briefly, does this search term seem like it makes sense for this ad group for this, you know, kind of ad concept that we're going after. And what's absolutely hilarious about it is that Google is showing all these ads erroneously, one would argue, you know, to some degree on broad match keywords or even phrase match nowadays. It's kind of the same almost as broad match. There's not a massive difference. And Gemini detects it unbelievably well and Google

ad still serves that click. So, so when I take the search term and I go ask Gemini, which is Google's API product, should this, should this keyword show for this type of ad that we're running, you know, Gemini is very good at detecting like, nope, nope, nope, nope, okay, yeah, that one makes sense, you know, et cetera. But I'm like, wait a second. Google is the one that served up this ad. You know, so like Google's telling me now and Google's telling me yes at the same time, which is funny, by the way. I think I've talked about this before with like, if you have an AdSense website, Google's like hammer that thing with ads. But add ads after ads after ads in your own website.

And then the Google Oriani team is like, we don't send traffic to websites with a lot of ads on it. And it's like, wait a second. One side of Google is saying one thing, one side of Google is saying something different. And, you know, same thing with Open AI, completely reversing their stance on ads as an example. We'll never have ads. And then, you know, Altman, same Altman, Melton five months later, six months later, ads, here's ads in the free version of chat dbt, which makes complete sense, by the way. I'm not saying that I'm anti-ads and the free version of chat dbt actually is the only logical thing. He should have never said that he wasn't going to run ads because that's stupid, but he also has some of the changes mind of law. So,

yeah, I think to go back to like, again, the good version of it and the bad version of it, if you're using AI on PPC, I think that's a good example of where it would make sense. I don't know if I have a lot of great examples for, you know, poor use of AI in PPC, other than just like just accepting the autowritten copy. Like I saw an ad that, you know, Gemini wrote the other day and we just kind of clicked to prove on it. And it wasn't horribly off, but it was just, there was enough off there where it kind of went back and said, okay, we can improve our documentation of writing ads now. Yes, Gemini will write the ad for you.

Is it actually good compared to some small tweaks? No, but honestly, like, it wasn't terrible. So like, I think that you're probably okay if you do that, but just be cautious with where it's coming from. We didn't think of it as AI, but I mean, we, ads has already had the negative aspects of AI, air quotes there with recommendations. That's something. It's, I think that that's, it's not.

AI per se and I mean it's it's algorithmically generated but ultimately it's it's something of the same thing now they're not producing true insights I don't think I think it was you know for that case even though they're looking at your individual and they're trying to pull out specific keywords it's the same script it's watch for its repeat across there so I think we've been maybe we've just been like ingrained that on the ad side where our skepticism's a little stronger. So then when you see the AI, we're much less likely, those who really focus on the Google ad side of things are much less likely to do some of the negative opportunities in Google ads because you've already seen it not work with recommendations and hopefully are making

those corrections along the way as well. Do you think that's it? You just have to get burned by the AI a few times. And they're like, oh boy, I can't trust this entirely. Or yeah, burned by Google Ads recommendations. Yeah. I had an email recently from a client who said that, oh, your recommendations is not 100% therefore bad. And I'm like, buddy, if I could only show you how bad an account we do, if it was 100% optimization, it would blow your mind. But yeah, to be fair, like within these systems, they trick you. It's like they're mentally tricking you, right, into thinking, you know, what it is. I'm like, have you ever used T-Mu before? There's when we were with the app T-Map. I am familiar with it. I have not made that purchase, but whole have I seen the horror stories

that have come out. It's wild. Like, when you download the T-Mu app, everything inside that app is essentially like a sort of gamification or a nudge into, you know, what basically like to get you to buy more things on Tmu, of course. But there's that there's an element of it where it's like this is designed almost by like a sicko. Like the amount of things that they've layered in there. It's not like one or two things. Like booking.com has like urgency, right? On booking dot com. And it's like, we have two rooms left at this price. And it's like, okay, that's true. Because they could change the price by a penny. And there's like little tricks that they're using on you. The Tumu stuff is wild because like they'll be like, this was the one I saw it in there

yesterday. And now I'm just opening the app out of curiosity. I might be buying anything. I bought one thing out there. And I'm just like looking at it because I'm just like, how does this work? They had one yesterday. This was fascinating where you go on the app, you would put something in your car and it would say, we'll give you $100 of cash bonus if you buy additional things. And I'm like, okay, maybe that makes sense. Maybe it doesn't. How do you unlock this additional cash bonus? So you had to put extra items in your cart. And what they were doing is they gave you like $80 of credit for free. But to unlock it, you had to get to $100. There was a $20 gap. And I'm like, okay, $20 gap, that makes sense.

How does one obtain this $20 gap? Well, you get the cash back, but it's like 1% or less than 1% per purchase, you get towards that credit. So like there'd be products on a page. It would like position you to like buy these products, but you would have to buy. I did the math, like a thousand dollars of products in order to unlock this $100 bonus. So I'm like, okay, if you plan on buying $1,000 with the products from T-Mood, maybe that makes sense.

Because like one thing I did buy, for example, it makes sense. We're shipping out like a lot of cards right now, like cards that my son is selling on eBay and stuff. And like my old collection were selling. So I do need a lot of those little sleeves to put the cards in. So it made complete sense of why I would buy a good quantity of those because they're selling a lot of them. But again, not a thousand dollars worth, maybe $50 worth, maybe would make sense. But it was just so funny how I was in there.

And I was like, this is totally tricking you. And again, I think the AI interfaces can do the same thing. They're totally tricking you into thinking that like there's a there. There's like some amazing insight on the surface. And EGino and my support team talks about this all time. AI will always give you an answer no matter what. So like if you keep asking it a question, assuming it's not like a trust and safety thing or something like that. But if you ask it for it for.

an opinion on a website, you could put in Amazon.com into any LLM platform and say, give me advice on how Amazon.com can convert better. And it'll give you ideas. Now, there may be bad ideas. That's kind of my point here, is that it can give you bad ideas or it'll give you ideas, they may or may not be good ideas, but it'll keep responding. Like, you could just say, oh, give me more ideas, oh, give me more ideas. It'll just, it will never stop. It'll be a loop in perpetuity. So if you're thinking that the AI is a perfect sparring partner to kind of disagree with you or say like, hey, here's how this thing should work. I think you'll be sorely mistaken. And I think that comes down to copy,

on your website. I think that comes down to how you think about website design elements, asking AI for feedback on those things. It'll just give you ideas forever. And that's where this kind of idea of fake AI progress can come into play because you might keep changing things, not realizing that you're basically just spitting your wheels. Like you're not actually making any progress. So that's kind of the core thing there. So anyways, a lot of topics there. But maybe we can talk about like email stuff. We can talk about outbound stuff or if you have any thoughts on that before we go to that next chapter. That's okay too. Yeah. I think, I mean, the, the, I think it's the same type of potential issue that I've seen for, you know, the single keyword ad groups and stuff like that

for Google ads, you do have people who are really excited about getting down to a really granular level with how they're communicating with their potential travelers, potential guests, or potential homeowners. But they create so many of these emails and these customized email sequences and none of them ever go out or none of them are ever activated. You know, that's something where there's, I, working with someone who does, who's working well in HubSpot, has all of these automations in place. But unfortunately, we built out all these intense audience journeys through the software and all these other pieces. But nobody's actually following that journey.

So none of these automations, none of these really cool email campaigns are actually coming to light there. So I do. I think that that's something where you can do so much and you can create all these new pathways. But if people aren't using the pathways, if you're not getting people to those pathways, again, we're just creating a problem or creating a solution for a problem that doesn't really exist. So we are. We're big email fans, I think. And in general, there's not a whole lot of negativity there.

And I have seen really customize emails be effective. and increase conversion rates and do things like that. But that only works if we're truly understanding the underlying issue, the underlying traveler journey, homeowner journey, and making sure that they're making it from point A to point B, not point X to point three, six. I don't know, like something that we're creating a pathway that they're never going to get to. I think that that's where ultimately, when it's a recurring theme here is,

AI doesn't know any better than what you know of your own thing. You can help it. You can feed it some of your information. You can feed it some of your data sources. You can feed it your PMS information. But ultimately, I don't think you ever want to delegate that decision to an AI or let them make a decision for you. I think that ultimately you have to have that understanding of your business.

You can ask for insights. You can ask for, you know, what should. I be targeting? Who should I be targeting? What type of people? What type of traveler? Is it families? Is it corporate? Is it this? Is it that? That's great. Giving those insights. But again, having them actually make a decision about what type of campaign, what type of strategy you're running, there's just nothing. You're going to get a vanilla, a vanilla option, which might work. And that's going to scare you even more because then you're going to ask for more insights. And you're going to ask for, like, if you see results from your first AI strategy, campaign, anything like that, I would be petrified because now you're going to keep listening.

You're going to keep relying on that. And at some point along the lines, you're not going to give a desired result. And you're going to go back and you're going to ask it again to refine it and make it better. They're not going to know how to do that. And this is the ultimate trap, I think, of AI, is that once you do see success, you're going to continue to anticipate. that success. And as a marketer, we should all know that is not going to happen. It's not copy, paste, get the results you get, just turn it around, do it somewhere else. It's not how it works in space. It's not how it works in marketing. So I do. I, I'm concerned about that. Like, you will

have someone do this next exercise that we're going to wrap up on and just roll it out. And then they see a little bit of success. Not maybe great success, but a little bit. And that's enough. It's a taste. It's just to bring them through and then they're going to be hooked. And we all know that sad story. So, yeah, I don't know. It's a bit of a journey. But I am seeing it more or what I think is that and I'm interpreting as that. So I don't know. It's just something to consider as you're implementing AI into your daily lives and your daily campaigns and your strategies is you're still, you are still the decision maker of your business. You are still the decision maker of your strategies.

make sure that is the case. Yeah, yeah. No, 100%. I mean, I think so the example that you're talking about, the thing we talked about is I went into Gemini, and I said, build me a marketing plan for a vacational company. That was my prompt.

So anyone who's into like prompt engineering would say, that's a stable prompt. Yes, I know. That was the point to some degree. So I will say, it started decently. So this is kind of where, you know, again, like it's giving me ideas.

It's giving me things to kind of build off of. So it said, we need to focus on guest acquisition, which is filling calendars, and owner acquisition growing portfolio. Agree there 100%. This is what it said, though. The 2026 market is defined by, quote, micro trips, which is last minute booking windows.

So within 29 days, and AI-driven search filters that prioritize guest favorites and wellness amenities. Okay. I mean, I don't know if that's accurate or not. I mean, yes, in theory, a property being a guest favorite on Airbnb is better than it not being a guest favorite. One would assume being a super host, but it certainly doesn't, like, limit your ability to create bookings. If you, you know, if you don't have those things, those are nice things. So the first thing, and this was a fun one, Paul of course, for this exercise. The first thing I told me is that the new SEO is AI search optimizations.

But here's what it says. It says platforms like Airbnb and Verbo, now use AI to cementingly match properties to guest moods. And share your first two properties are just room shots, but emotional anchors. So for example, a steaming coffee cup next to a foggy window. That was the recommendation from AI. He's shaking set folks for those listening. For image. So don't have a structure shot.

Don't have a nice sunset. don't have an amenity shot, do a quote-unquote emotional anchor shot, according to AI, which is a steaming coffee cup next to a foggy window. I would love to if a client would let me put that as their lead photo and see what happened to their clickly rights in Airbnb. That'd be fun. This one I didn't hate to be honestly. The next thing was tagging for discovery.

So include things in your listing, like mention that it's sleep optimized, which can be a blackout curtain sound machine, et cetera, comfortable bed, of course. Call it a digital nomad hub, have the right connections or like a desk set up for people that might be traveling or include an EV charger in your property. That can potentially help you get more people that are searching. I actually kind of like those ideas. So I don't know anything negative say about that. It says this one, though, use AI tools to rewrite titles weekly based in local events.

So for example, three minutes to festival name plus free parking. Rewrite your titles weekly, weekly, according to Gemini, based on local events. I don't imagine a property is doing really well and you go in there and screw the title for no reason. Like imagine that would be like advice you give. Like, oh yeah, it's property doing a lot. So, yeah. I mean, a lot of things here. Like at one point, it said, build a direct booking website, and then you have to integrate Google Pay and Apple Pay.

Everyone expects to be able to book on those platforms. I don't have a single vacational website, direct booking website that has Apple Pay on it, to my knowledge. And we have a client that does, oh, I don't know, 15 million your author website with no Apple Pay. So could Apple Pay help? Again, not saying that it can, but is that like the most important thing for you to have in your website to get direct bookings? Also, no, you need traffic, which kind of begs the whole point here, what we're talking about today. It's a recommend using quote-to-a-blog the billboard effect. So always list your properties as broadly as possible on Airbnb, verbobbooking.com, et cetera.

even if you're feeding some of those platforms higher rates. In effect, some people may find you off an OTA platform and then find your website directly. I love that. I can agree with that. It's funny. Like, I agree with like one third of what this AI plan did, which is kind of fun. Deerat targeting on email sequences for people who went to start the booking process but didn't finish the process. Again, I kind of like that idea too.

Yeah. And it just kind of gives other social advice. So it says, how about this one? Offer your guests a $25 dollar digital gift card to guests who tag your property in a high quality reel or story. So we're relying on the guests to create the content. Maybe. Like, I don't think you'll get a lot of takers on that, to be honest with you, but there's something there that you could do there.

Then it said, you know, have a Google My Business profile, right, pain points, focus blog content for owners. So for homeowner acquisition, like create that kind of content. Can't disagree with that. Work on real estate agent programs, offer a referral bounty, and then build owner benefit packages. So here's like the product that we offer owners is this owner benefit package. When you sign with us, you get benefit ABC. Actually, don't disagree with that.

See, they're a homeowner side isn't that bad. Yeah. Yeah. Some of them are decent. Some of them are okay. Some of them are pretty bad, to be honest with you. But it's like, it's very incomplete,

is my point. Like, there's a lot of things here that, like, if you try to implement this plan, I feel like you'd be so confused, you know, about what you actually need to do, that you would end up being,

you know, more confused than anything of what you need to build from there. So that's kind of our whole point here as we come to a close, which is that like, AI can give you ideas and it can give you helpful things. The quality of your prompt matters, by the way, the same prompt and clawed,

and it gave me way better result because it was using my skills that I built and like putting context in from previous plans that I built. And I joked with Paul Bore record. Oh no, damn it. Claude's too good. Like it's like it's good.

You know, based on fact that I had prompted it before and it has a lot of context about what I do and my approaches to things. So it produced a 10 page report saying, here's the summary. Here's our market opportunity. Here's what we're working on. Target audience.

And even before Claude let me get the marketing plan back, it asked me questions. So it's like, tell me about your market. Tell me what location. Tell me what properties you have. Tell me your revenue, et cetera. So Claude is, you know, obviously, leagues,

leagues ahead of Gemini in this respect. But I will say this. I did use the free version of Gemini or like the cheap version, just to put the point a little bit, right, of like, this is not the same quality. I suspect even if I reran that with Gemini Pro, I'd get a better output for sure. But here's the, I mean, but who's to say that anybody,

I mean, I do assume that most people are using a paid version of the AI. I wouldn't assume that. If you go look at the data, it's, you know, 90-something percent. Yeah, are using the free version. It's so, so is it that far off to think that, like, one of the people who is going to run this, you know, if you are, if you're a small business, you're not necessarily investing in that, right? Maybe you're right. Well, certainly I will say this. Even if you had a Cloud Pro account, I can be very confident in saying this. I highly doubt

your Cloud Pro account has more contacts than my Cloud Pro account about vacational marketing, right? Mine has 100 pages, well, it has my whole book in there, actually. So like literally 300 pages plus everything else I've done. And all my granola transcripts from actual real client calls, which is all, like unique data, real data based on client interactions. There's no way that anybody would have that, right? Except for me and a handful of other people. So that's an example, too, of like the AI, to some degree, is as smart as you fine tune it to be.

I feel like I heard that word a lot previously. Fine tune your AI to be appropriate to. I feel like I haven't heard that as much over the past a little bit, but I heard that for a while, you know, fine tune your model. And that's kind of, I think, what you're doing and what you're building from. I mean, I'll say this. I have some people like to use the voice mode on these platforms. I never did it until like a few days ago.

I use the voice mode on Claude. It's kind of dumb, too. with like, it's not as smart as the regular version of co-work. Yeah, like, when I'm asking questions, like, it can't do a lot of things that I can do on my computer. And part of that's the whole Claude functionality that's on co-work and code versus what's on chat. But I asked it to, like, check my emails for today. I was like, hey, Claude, check my emails for today.

And it could do it. It's like, I don't have access to your emails. I can read only your calendar and some other things. So, like, even it said in the voice mode, there's certain things that aren't enabled yet. So that's like, even if you're using it, like a so-called assistant, I think you'd be somewhat let down by the capabilities there. But that's it. We'll put a bow on this one.

real progress versus kind of this fake AI drop progress that doesn't actually do the video. Just, just, I don't know. To use AI in best judgment, as we have done for the last 18 months, I guess, and hope for the best. Yeah, yeah. It's a tool. You have to use the tool in a logical way. If, uh, what's, what's the, what's the Gary Vaynerchuk line about like, what's

a basketball worth to Paul? 15 bucks. What's the basketball worth to Conrad? 15 bucks. What's worth to Michael Jordan? A few billion, right? He took that tool and he turned it into something different. So like the AI as a tool, absolutely use it.

You know, use it like Jordan, use the basketball. Don't use it like I would use basketball and just miss every time. So that's not good. All good. We'll put a bow on this one. Leave us a review to me this far. Always curious and looking for more reviews. Again, thousands of you download.

Tens of thousands of you download according to our analytics on the show since the inception. And we don't certainly have tens of thousands of reviews. That's for darn sure, isn't it, Paul? So we love some more reviews. Go leave five stars, Apple Podcasts, iTunes, Spotify. That's where we get a lot of downloads. So appreciate that. Have an awesome rest of your day.

We'll catch you on the next one. Peace.