Heads In Beds Show - The "Do's" and "Don'ts" Of Using AI With Your Vacation Rental Marketing

Episode Date: October 8, 2025

In this episode, we dive into AI at a deep level: how SHOULD you use it and NOT use it to market your vacation rental business?Enjoy!⭐️ Links & Show NotesPaul Manzey Conrad O'ConnellConrad's Book: Mastering Vacation Rental MarketingConrad's Course: Mastering Vacation Rental Marketing 101🔗 Connect With BuildUp BookingsWebsiteFacebook PageInstagramTwitter🚀 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 vacation rental 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. All right, Paul, good afternoon. What has happening, my friend? It is, you know, it's getting to be that sports season. We got a little bit of football going on right now.

Basketball is just around the corner. But we get to really enjoy next week's competition. As we're recording this. No, you get to really enjoy it. Well, I'm really, really going to enjoy it. You can enjoy it. We'll all enjoy it together. But I will be on site at Bethpage for the Ryder Cup next week.

Have you been to New York, like recently? What was last time you were in New York? It's at least a decade. So it's going to be. it's going to be just fun to we do we i told you we're going to a practice round then we have kind of a day we have thursday to do a little light sightseeing well we're not going to go do anything crazy but uh yeah then friday morning uh it's it's day one and this is yeah i i remember going to hazeltine now 12 years ago or something like and that being like a crowning

like that was like one of those once-in-a-lifetime experiences and and we're lucky we've gotten a couple in the Midwest here. We're going to go do that page and maybe sometime we'll go over to Europe. So that's my goal. But how are you doing, sir? Sorry to stealing it. Not as good as you. That's for sure. But no, I'll get on my side right now. Just busy. Busy as can be, you know, as far as buildup stuff goes and just feels like a busy season right now for my boy, speaking of like sports and stuff like that. So I've won, my middle one's in soccer. Liam, he's in soccer. So that's like every, what is it, like Monday, Wednesday, I think for the most part. And then Julian's playing flag football this year. So he's kind of getting up into that.

But that's every Tuesday, Thursday. So literally Monday, Tuesday, Wednesday, Thursday, I have practices typically from like 530 to 630 or 630 to 730. Just they alternate days and stuff like that. So yeah, you know, got to get up there, get everything ready, drive all the way out to the sports park where they have these little games and matches and stuff. And it's fine, but it's just like, you know, this is where the soccer mom sort of thing

comes from. But it's like sometimes I'm going solo. My wife comes back, you know, here takes care of the baby. So it's just me sitting out there waiting from the boys to play their games and stuff. So it's fun. The fall weather here is excellent. So definitely happy we're doing this now and not in the summer because the kids are, you know, out there running around and they're not completely getting, getting washed up. And, you know, it's interesting to see my son get really into flag football.

And obviously, it's precursor. We'll see if he wants to play actual football. But he can juke a kid. He's not bigger than any of them. Maybe one or two of them, he's bigger then. But he can make a miss. So I think that's a good thing for flag football. We'll see how it all progresses, you know, in his, in his career, as it were.

I love it. But I'm having a good time for sure. it's funny his coach you know uh paul gives him tips and things to do and things not to do i feel like we need to be the coach we need to get our whistle on and our you know visor on straight or hat on straight we need to give some of these vacation managers some advice on what to do and what not to do so that is the topic of today's episode uh what to do and what to do i guess with like AI in general in general i think we're going to niche it down or narrow it down a little bit more specifically to probably focus a little bit more on the SEO side because it seems like obviously all the rage on

LinkedIn and in general, marketing media right now is this idea that SEO is a stupid word. It's old and outdated, you know, crusty and we should be put out to pasture. And it's all about answer engine optimization or generative search optimization. We've already done an acronym episode. You can go back in the feed seven or eight and you'll find something. I think it was like AIO, GEO, SEO, oh my, I think was how I titled it. I was in my bag on that one. Yeah, as a Jones would say, Mark Jones would say, I was deep in the bag on that one,

get my french fries out of there. So I'm happy with that one. But now, today we're going to narrow it down, talk most about SEO. Because I think these tools that you have available now with all the AI platforms out there do absolutely help you do better SEO if use correctly. But you can also almost like a knife or a gun, you can use it the wrong way and really hurt yourself or harm what you're trying to do SEO-wise. So this was your topic. I'll let you take it away. But that's the premise, I think, what we're going on.

Yeah, I mean, I think this is kind of the next natural step, the part two of, okay, well, yeah, we've now talked about AEO, NLMs and all these. other fun things like that. But I think when I do searches, like real-time searches with people on calls and things like that and just like do comparisons of what they get from a chat cheapy to response or what they get from a Gemini response. And it is just so drastically different from user and user. And it does. I think, I mean, that's kind of what got my brain thinking about just this episode today

and just kind of thinking like at a certain point, the LLMs kind of want to serve up the right answer for you. And that's what Google wants to do, too. But they want to serve the greater search ecosystem. They're not just looking at your results. And chat GPT, I think that's LMs. I think that's how the concept is supposed to be. But the reality is that your individual user, your model, that's yours. That's your personalized results. And at a certain point, you are, you're going to kind of create your own echo chamber getting the results that you want to get to. So do we want to show up in LLMs? Absolutely. This is always a this is this is this is a good thing. And I think for most websites that I'm looking at, I do. I see 80% growth, 90% growth. Now that gets us up to 30 or 40 sessions

to be clear, but but those are fun numbers to look at and kind of get excited about a booking here, booking there. Those those are good. But but that's what we're talking. about like when i think that that's always something anytime we're talking about lLMs and anything relating to that the scale does seem so vast oh my gosh three thousand percent increase in this and that and well yeah we're comparing to something that didn't exist you know 36 months 24 months ago things like that so yes all of this growth is going to be dramatic and and i do i think that that's something where if we're talking about sustainability of the business chasing LLM results is not going to be sustainable for the business, continuing to focus on

SEO and using LLMs or AI to improve your SEO practices, strategies, all those things, which we're going to talk about. Absolutely. But yeah, I just, that was one of those things that I think I've heard, I've seen, I've read, I've watched enough that it's just, you hit a breaking point, and you're just like, no, we need to just parse this down a little bit. Bring it back to, again, the baseline level of what do you want to use, what don't you want to use, and how can you avoid some of these, hopefully, mistakes that we do see people make when they're trying to incorporate AI and SEO. I'll make one comment, and then we can dive into the twos and dose specifically here,

which is I'm still waiting for, I haven't got an email or DM yet to this effect. Show me a site that's doing phenomenally well in so-called LLM search, or chat chis-a-b-T. So it's a site that's recommended frequently when you ask about certain topics or ask about certain bits of information or it's recommended frequently in chatch-a-tetee the like, Gemini, et cetera. But it's doing horribly SEO-wise. So in quote-unquote regular traditional search,

it's nowhere there, but it's in one of these platforms. Because certainly if there was some technique or tactic that was very proprietary and very specific to chat chvety, again, I'm not saying there's not things you don't need to be aware of. There it is, and we've talked about that. We can talk about that today a little bit about things that you need to be aware of. an obvious one being like, don't block the open AI crawler from visiting your website. That's one that, you know, we cleaned up the other day for a client.

So we'll talk about that a second. But that's what I've been seen yet. So my point is there's a lot of overlapping things. So speaking of visualizations, that you mentioned the fact of the growth traffic, don't use percentages to show the traffic referrals from chat-tbt versus Google. Use a pie chart. And in the pie chart, you will see a pie chart that is 80% Google, if we were just talking about search traffic coming in.

90% Google, maybe depending on the site and the brand. You might actually see a lot more referrals from Bing, from Yahoo, from duck, go. I have a client that gets 10 times more traffic from duck, dot, go than they do chat TVT. And then you'll see a tiny sliver, and yes, it's growing, but a tiny sliver of traffic

coming in from chat TVT and the like right now. So should you ignore it? Absolutely not. But again, is there something that is, you know, so concretely actionable or so specific beyond just some of the basics that you need to check on that would require you reengineering

your complete outbound content strategy, your marketing strategy. That's where you and I seem to be on the same page, which is like, no, that is not the case. So, yeah. Which one you want to take? Do you want to take a do? Do you want to be more optimistic and start with the do-first?

Do you start with the don't first. Boy, I got to do it. You know, me, I got to do it, don't. Okay, take a don't. That's how it all started. I'm going to start with the top one that I put, like, don't build a strategy to increase your presence in LLM search results. That's, again, I just talked about it.

Don't, like, you can build a strategy that's going to improve your overall user experience on the website. You can build one to improve your search results in Google, in Bing, things that are actionable, things that are, I think, report, like you can, you can do reporting. You can actually understand how you're improving Google Search Console gives you some of these tools. That's the other thing. We don't really have a whole lot there. So just focusing on improving your LLM results, even if you're able to do that, you're probably just more or less gaming the system. And at some point, much like SEO did back in the day, it's going to revert. all that work, all that time and effort and strategy that you put in is going to be a wasted

cost at a certain point. So instead of doing that, build your content to answer the questions that your travelers or owners are going to ask the LLMs. That's your, that's the answer. Here's the don't. Here's the do. Do continue to do what we've been talking about for the last two years plus-ish, $100,000. Take a look.

We've been talking about it, but that is. Like, if you're thinking about how people are using the LLMs, start asking it, what questions are people asking you about areas in, you know, things in my area, about accommodations, lodging, things like that. That's how you can start building some answers to these questions and have a better understanding of how others potentially are using the systems. But just going about it and saying, I'm going to just show up the LMs. Good luck. Have fun. and come talk to us again in six months when it hasn't really moved the needle for you.

Yeah, I guess, again, you know, me, I love my analogies. I guess it's just this idea that like there's one specific workout that you can do, you know, that will make your chest massive or your arms huge, right? If you're a guy and you go to the gym, it's like, it's really always a combination of things, right? There may be a few different exercises that you do. And again, it's not that different, right?

So this idea of like, I'm going to have a certain style of, you know, SEO strategy or whatever LLM strategy that's going to be different. It really isn't. Like, I'm working on improving. our SEO audit right now. And I'm adding a section now for like LLM checks and it's things like making sure that it's indexable with the LLMs, which is like looking in there, make sure you're not blocking the user agent. Maybe it is like double checking headers, but we already double checked headers. That was already something we were doing for SEO. Make sure we had an H1,

H2, H3. That's already best practice. Okay. There's been some things that I have research and looked up. It seemed to make a difference. So don't refer to it as like, come visit us. Like in the copy, for example, it may say, you may want to say, come visit us in North Middle Beach, South Carolina. That's where we're located. It's a little things like this that I'm kind of looking through and saying, okay, maybe these are areas that we can improve upon. I think that's all somewhat tangible. And one of that's more done, by the way, I'm happy to send the SEO audit to people if they're interested in that

because it'll kind of be a combo like SEO plus LLM search audit. Yeah, I couldn't agree more with your don'ts and your due there. So I'll kind of take the next one and go from there. I'll rip off from the other list and then I go back to your list there, for sure. I love AI, the dues. Use AI for a first draft. Use AI to help you brainstorm. That is the number one use case that I have for AI. I'm not often publishing anything

with either buildup bookings marketing or with client marketing that is straight out of AI and that's something that you kind of mentioned there. Don't just copy and paste straight from AI and put it in the website. But what I think you can do so well is like, for example, write your list of like the 10 best seafood restaurants

in North Myrtle Beach, copy and paste it, then go put it into chat GPT and be like, hey, here's my 10, did I miss any? Like, is there anything you would have included that's not, you know, currently in my list. They may say, oh, you didn't mention this place. Like, that's something. Oh, you may be like, oh, I haven't gone there yet.

You know, you don't know everything. And the LLM kind of does in a way know everything, certainly it feels that way lately as far as like fact-based stuff. So I love using the AI for first draft or checking me, like is my logic sound here? Like, I'll just ask it almost like I would ask. I've said this now for a while. An AI tool like JATGBT is the best junior employee you've ever had. It's 20 bucks a month. It's never sick.

It never takes a day off. It's so great in that respect. But it's also not what these proponents would have you say it's not AGI. It's not like this Uber intelligent, you know, 190 IQ thing. But what it does really, I think, a good job of. is like fact-checking, you were checking certain things. So, like, I did that recently, too. I wrote a blog post.

Is there anything you would correct in this blog post? Pasted it in there. I wrote it, but then AI kind of fact-checked for me. Hey, you mentioned these two things, which kind of seemed contradictory to one another. And I was like, oh, I just didn't give you enough information. I fixed that paragraph, you know, explained it better. I said, what do you think now does not make more sense? And Jack TBT's like, yeah, absolutely.

So I'm a big fan of that. You shouldn't have writer's block. You shouldn't have that, like, blank cursor feeling when it comes to writing a blog post. Like, obviously, I'm a big fan of using, like, AI-specific blog post creation tools if you're going to do that. So we have a few clients now where we use Kappa, CUPPA. I think it's .AI now. It used to be .SH, but I think it's Kappa.

coma. And that's a blog post article that gives you a heck of a first draft. You shouldn't just take that and post it on the website and publish it. I will say, though, if you're doing that, you're probably better than like 80% of people that are doing nothing. So if your option is do an AI blog post or do no blog post, I feel like there actually is more opportunity or upside in doing an AI blog post if you create a prompt that makes sense.

But yeah, that's just, I think, the biggest unlock for anyone doing content creation or whatever on the AI tools is like just let it give you first draft and then you tweak it from there and eventually you may find over time that it ends up working really well for you so that'd be my do and my don't on my side of um i do i'm just jumping right off of that i mean i think that that's like where the don't i'll double we'll redouble it back to the don't don't prioritize quantity over quality is that it is you that i think that's the the myth is that you can start to generate a ton of content and you can but is it of going to be quality because Because I think that's something that we've harped upon a lot is that 500 words does not make a blog post.

Three paragraphs does not like, there's not a standard set for what a blog post looks like. And the answers that we have to provide are not, cannot be always relayed in a very short amount of messaging there. So I think it is. It's you want to do 12, you know, do you want to do 12 really, really good blog posts throughout the year? month, or do you want to do 240 pieces that are just garbage? Like this, at a certain point, first of all, that would be, I'll take my eyes on the 12th. That'd be way too much. Google's going to say, no, this is this can't be. At a certain point, the algorithm is going to understand that you haven't put that much content on the site previously. Now you've got a whole bunch of content. Something's

probably not right there. Something looks a little funky and you're still, again, you're going to go backwards as opposed to going forward as a result of overdoing it. So I think, I think making sure that you're continuing to focus. Yes, using the tools to improve that content overall. Absolutely. And on the due side of things, using those tools to analyze large amounts of data. Like that's something that I think can be very difficult because, again, you go after large amounts of quantity. On the content side of things, well, I have to have keywords and I have to do all these different searches.

I have to understand what's happening. Well, with all the tools out there, my Sem Rush, your A.H. Reps and what other, I mean, since that time, there's probably been more tools that are becoming some good options for your SEO toolkit as well. But they've got a lot of data. We are humans. That is something that we can process the data at a high level. But LLMs are really, really good at that. Those are the tools, again. That's that junior employee that they can understand those trends. They can start to identify those. things so that you can focus more on doing the human element of creating the actual experience and then doing more of the, I don't know, adding the humanity in because there is a difference. I mean, I think still, when you look at, and then maybe that's a good question. When you look at content, do you have a good idea, do you think you have a good idea of whether or not it's AI generated? I think I'm okay at that.

You know, probably there's been some situations where I haven't nailed it. I'll be honest, I started listening. I listened to a lot of like non-mirical music in the background when I have the chance to do like work and I want to do it. And there's certain YouTube channels I listen to over the years. Some of there are now doing like AI music, which people in the comments hate. And I'm like, this is fantastic. You know, I'm just like this music.

But I didn't know that. I listened to some of these videos for a while without realizing that they were like AI generated music. But it's like all not, either nonsense lyrics or not word lyrics. It's just like sounds. And I'm like, this is fantastic. Like I love the focus. If you Google or if you go on YouTube and search like focus music, it comes up.

I love that kind of stuff. So that's a good example of like before that was. someone in like a, I don't even know these audio tools. I'm going to peek out of my tail here and I don't know what I'm talking about. But they were generating these beats, I'm sure, one by one and using some of these advanced audio production tools. And now an AI can just like click a button and like kind of train off that and use things. So like, is that a good thing for the artist?

Probably not, if I'm being honest, but like the music is what I'm after when I'm trying to, you know, work on a project. So I guess I feel the same way about, you know, anything, AI content creation in terms of like, it's all about how you present it, I think, to some degree too, right? Like a lot of AI content I see is just like handled very poorly on the actual website itself. I know it sounds very simple, but just like putting a blog post up there with like no images, no internal linking, no additional information, no like personal tips, no links to the websites. It's just like, here's 10 good seafood restaurants at Destin Florida. I think that's also

not going to rank well. But if you take the same AI article with the same text and just like put an image in there for each heading, put some internal linking recommendations to actual restaurants, talk about like, hey, the best dish at this restaurant that most people seem to like crawl all the Google reviews or something and then say, hey, people mention the swordfish more than anything at this restaurant, that sort of thing. Like those are ways I think we can take AI and make it more useful to the user and that would perform a lot better in search. So yeah, one thing I was going to say, I didn't want to interrupt you, though, talking about AI ingesting large amounts of data.

You probably have a large amount of data with your guest messages or things that you send in like guidebooks or people ask about. That's probably an amazing training source on AI, you know, information. So if you were trying to do like a guidebook for your destination and you could export all of like your emails or messages or something like that on a platform like Airbnb or Verbo or your just email inbox and then dump that into chat TPT or Gemini, Gemini is better at this because there are context windows way bigger, by the way, million tokens, so you can put a ton of stuff in there.

And then, yeah, be like, generate a list based on my specific recommendations using something like Gemini or even better. I would argue notebook because notebook is not going to go out to external sources. It's only going to use the training model that you upload. So if you uploaded a thousand email messages, you know, and say what local recommendations that I give to guess the most often, you know, cite your sources, etc. It's going to tell you like, oh, you recommended Cooper's 117 times to guess. Like, that's probably your favorite restaurant or whatever the case may be.

So, yeah, big, big fan of that. And I think, I guess to be honest with you, I'm at the point where I, maybe I don't care as much as other people. I will say there's people who abuse their AI privileges to use these tools on platforms like LinkedIn that I've just just unfollowed because it's like, you know, no, you no longer have an original thought, you know, and that's something that I do find quite troublesome, you know, long term. But in the short term, it's something that I think we just have to deal with for sure. Yeah. I think do use AI to pivoting over to like what you can use it for for any sort of like, like use it to research things that seem, you know, obvious that will least point you in the right direction. Like, I'll give me an example.

I was building a Google Ads campaign for a new client yesterday that has no historical data. So I don't know exactly where most of their guests are going to come from. Typically, what I would do with Google Ads is I would target it. Let's say at the top like 10, 12 states or something like that that are traveling to that destination to that area. And I would ask the client, hey, do you have any address data I can look at so that I can, you know, build off there? So instead, I just went to Gemini. And I said, based on what you know, based on like search trend data, most people coming to this

destination or coming from what area? What states did they live in when they're traveling here? And I just put the top 10 states into Google odds. Now, maybe wrong. Maybe there's a state there that shouldn't have been or I missed a state. That's plausible. But like it suggested very logical states. Like that's a good starting point. And then once we get more data, we can come back in there and modify it and, you know, get more information once we have actual conversion data. But right now we don't. So I just kind of have to essentially guess, you know, educated guessing, but I'm educated guessing using my knowledge plus the knowledge of AI. And, you know, the client agreed with the AI anyways as far as like what states he gets most visitors from. So that's, I think, a do.

which is like, again, have it help you a little bit, have it give you a little bit more information or knowledge. It knows a lot of type, you know, type search trend data that I think you can build off of. I think what you don't want to do is just like let go the wheel once you do that build. So like I said a reminder two weeks later to be like, go back and look at all the click through rate data by state to see like, is there certain states that are much better than others.

Maybe I go back and, you know, change it or modify. So I think we're at the stage now where it's like, yes, as much as we kind of let go the wheel and Google take the wheel is the joke that Matt and I have on my team on the ads team is like, we do that a lot now with bidding strategy, with PMAX, with all the stuff, but you can't just let it go completely. Like, you can let go bits and pieces of it, but you've got to go back there and audit it. A funny, ironic idea here on the PMAX campaigns, I've seen many of these, is go dump out your search terms data for PMAX, like do the last 30 days, dump out your search terms,

explain to Google Gemini or, again, I like to use a Gemini for this because the context window never is an issue on chat GPD sometimes it is. And I say, these are all the keywords that are showing for my performance max campaign on Google ads, but I only want them to show for topics closely related to whatever my actual asset group is supposed to be. So for example, it may be an asset group where it's like Destin, Florida Vacation Reynolds. So I'll tell, you know, Gemini, I'll be like, only to suggest keywords that are close to this, otherwise make a list of all the negative keyword suggestions that you would say I could

do analyze each keyword. You click and Gemini will take a while because it'll go through like a thousand lines of data from search terms, go get a coffee, come back, go get a soda, come back. And it's like, here's your list of negative keywords. And I'm like, wait, so Google has the knowledge to add a negative keywords that they wanted to, but it's up to you to go in there and actually tell Google what you want as negative keywords. And then I'll say, okay, find the core themes that are like the root terms for each of these. And then I upload those and I did a bunch of negative keywording the past week

or two from PMAX campaigns with that. So that's where it's like, you can trust Google in a way or you can trust some of these AI tools in a way, but you've got to go back and be double-checking all your assumptions. And you got to go back and have that human layer of like, all right, how I'd make this better on that side. I think that's, we have the tools available to us. All we have to do is understand how to get the most out of them. Yeah, I'm jumping to another dude. Well, real quick, Paul, how long would it take you, for the listener? How long would it take you for, to review a list of 2,000 keywords in a search campaign,

like search terms reports, two hours maybe? Probably. I mean, that's, to really review it, to really, at least an hour. Like, and, and it takes, it takes 60 seconds now, maybe, maybe two minutes, you know. to do that. It's incredible. So yeah, use that to your advantage, right? Like you can make, if you know what you're doing, you're going to make better decisions,

but you have to know, you have to know what to look for. And that's still just this piece where we're not really there yet. And hey, maybe I'm, maybe I'm wrong, but you got to think that like Google is incentivized to show your ads as much as possible. You're incentivized to only show your ads in relevancy. There's always going to be this mismatch of, you know, outcomes there. And I think that it's hard to believe that Google would just be like, oh, yeah, sorry, we accidentally showed you for a few keywords we didn't intend to because

they just make too much money for them to be completely worried about that. That won your ads to perform well. Don't get me wrong. But they're going to let 20% of it slide and be like, oh, yeah, we'll show for that. Even though it's relevant, you have to go back there and, like, crack the whip on Google ads. And it's funny. I do it so well with Google, which I just find some of the ironic. Connect the dots, Google.

Come on. At some point, they will connect the dots. And then we have to find another way to show value. So that's, well, we'll show the value right now as is and use the tools. Nice. But what's on your list next? What would be your next Tuesday? I think this is something that, that I've been.

been harping on a lot the last few months and just a lot of conversations it's a structure just structuring your data the way it needs to be structuring your content this is and not just for AI but for humans too like the reality is we've all lost our attention spans we have about two to three minutes so being able to highlight the key areas like that's fundamental to creating content already so just putting in proper h1s h3s bullet points making sure that human can find the key information on your page, making sure that AI can find the key information on your page, that is, I think that's becoming more important. It's, I don't like the bullet pointing of everything, but at the same time, if that's what people are looking for, that's what you have to

deliver. Like this is, you don't need to write, same thing, just like you don't have to write 500 words with this. You don't have to write a thousand words worth of content. You need to provide the right information so that people can find it and use it. Hiding it in a lot of content doesn't do us that good. So I think how you're structuring the content is just as important as what you're putting in the content so that, again, everybody can understand it, find it, use it to get those answers and results that we're all looking for there. Yeah.

It's funny. I think we talked about like things to do pages before as kind of like one of the keywords that it has unfortunately just been crushed by Google over the years. AIA overviews, all these additional snippets and so on and so forth. So that traffic has kind of gone away a lot, even if you rank well in Google. But when you go look at what Google provides you, it is very surface level or it's very monetized. Like they're trying to sell you tickets or they're trying to sell you ads or something like

that to a specific show or it's just very surface level. But I think that most people don't want that. Like most people want more of a deeper dive into something. If they're planning a trip, they don't just want to know. Like you and I were talking about golf stuff as we often do before you hit record. And like you had mentioned like, oh, there's all these courses nearby in this and the Cheboygan area. You got to understand each one of them are.

So, yeah, at a high level, I just want to know what each one is. That's like my initial Google search result. But if I'm doing a deep dive, I don't know everything about each course. I want to know the, I'm going to go look at the scorecard. I'm a weirdo, right? I'm going to go through and look at stuff that, like, most people wouldn't look at maybe at the surface level. But like, that's kind of the guess that you're making content for.

Like, assume the person you're making content for is the person that wants to research a lot of it. And sure, many will not. But for the ones that do, that's the ones you're going to win over, I think, from a content perspective, too. It's like, oh, this person, like, not only told me where to go for the best ice cream in my destination, they told me where to park, you know, here's the place to park. They told me all this data. That's like the stuff that AI can't quite do yet, or if they do, the AI is just ripping off you, which you've got to kind of accept at this point as like a good thing.

Like the AI ripping from your blog article and then hopefully giving you some citation traffic, it's going to happen whether you like it or not. You may as well just take the best of it and, you know, build off that. Well, I think just like AI is using your content, don't use AI to accidentally plagiarize someone else's content. something that I, any time you are, if you are directly using, and again, don't copy paste, but if you are going to use a large chunk of what AI is creating for you, the LLM is creating for you, run it through a plagiarizer checker. Like that is something that, especially if you're in a competitive market or something like that. And again, you should know that you shouldn't be doing something like that. But that's the reality is that you don't know where

sources are being pulled from or check your sources. Make sure. You're checking you the sources for the content at the very least because if content is being pulled directly from someone else's blog post, local competitor's blog post, they're probably not going to be super happy about that. And it's reality. Like I've seen these types of things happen in markets like the Destin 30A PCB area. So, you know, these are the types of things that I think we just need to be very careful with how we're using anything that is. is truly AI generated, like that hasn't had some type of human pass over top of it. Yeah, that outcome is always out there. And there are plenty of plagiarizer checkers that are out there just to make sure that you're in scope and a lot of them are free.

So that's something that don't, don't just trust the AI. But if you are, please, at least run a plagiarizer checker so you don't get sued for that, too. Yeah, yeah, I think that could be a, we could just say that after every block here. don't trust the AI completely. Because, yeah, even though it's much better, I mean, think of like the OG version of chat TPT that we all kind of started exploring. I mean, it would just make up the most absurd things. You know, I mean, just the most absurd things.

So it's gotten better at that respect in my opinion or in my usage of it. But by no means is a perfect. I mean, at the bottom of I'm open, I have chat TPT open right now. At the bottom, the first thing under the little chat box is chat TB can make mistakes. You know, double check all responses, basically, which is I'm sure they're a little legally required disclaimer, you know, to understand there. So it couldn't be more of a big thing there. Yeah, I mean, I think I've hit all the main ones on my side.

I think if I could put one, maybe or two more do's and don'ts in here, I think it's just like still understand like your, like the most important thing is that you understand your customer and your, and your client, whether it be the homeowner or whether it be the guest side of things. What's going to come out of chat, TBT in general from a content perspective is going to be the most like milk toast for generic version of whatever it is that you're trying to write or create if you just do a general prompt. So I actually did one night, an advisory client I was working with today.

And it was, um, I went to chatty, and did like give me a fall guide for like fall things to do in their destination and then I did the same thing but I copy and pasted like the actual bullet points that she'd written and I put it in chat TPT and either the difference we knows to start then I went to cup and did the same thing and like cup a spit out like a way better article like way more detailed version of that article so like the version of like you know doing the basics with chat TBT you get very like general high level information that's not really that useful give it a little bit more context than it will at least get to like maybe a C minus level if you give it the right context and use a right tool for the job that's actually made to write like blog content like cup is it went from that to you know level like b plus you know where then you're building off of something there and the output was like 2 000 words which the client had asked hey is it better if it's longer and i'm honestly at a gut level my quick answer to you was like yes it's probably better if it's longer because it's going to give you more things to like trim back if that makes sense you know like if you're a haircuter i imagine you'd rather like start with long hair and then cut it back to short

hair versus like someone that doesn't have much hair and then it's like how do i cut this and not ruin their hair i don't know um never cut hair before but i think that that's like the logic that I would have is like, yes, you don't just want to pump out high, quantity, low quality. I hear you got that and I agree there. But I think it's better to start from like, give me more of a detailed output with your prompt and then your output. Give me a little bit more. And then I can come back and maybe in trim what I don't want. And I think that's a good way to think about it. Because I think most people who early on were saying like this AI stuff is useless, just were using the worst prompts. They were

just like, write me an article on how to save for retirement. And it was just like four bullet points. And they were just like, this is stupid. But if you're like, write me a blog post. And then here's heading one, two, three, four, five, six, seven. Here's the four bullet points that I want you to cover in each thing. It may take you 30, 40 minutes to write that outline, but then you're saving two hours of writing, you know, whether, you know, but you have to understand what you're putting in that outline, or you have to understand what you're putting in that prompt for it to work well.

So I guess that's kind of the, you know, again, the back, the backside of all those pieces, which I see a lot of people making that mistake still. It's just like a D minus prompt will give you a F response, you know, a plus prompt will give you a maybe an a minus response. And that gap is very wide in my experience. It's not CRM rule, bad daddy and bad dad out applies to just about everything there. I think it still comes down to the human element. Like, don't, one of those don'ts, don't abandon the human element.

Like, there are plenty of places where you can use AI to automate your business to make it, make it a lot better to improve efficiencies, do all those things. I think there are some areas where you don't want to remove that human element. And I think it's important to understand. I think maybe sometimes we think that we're automating out things that maybe are hard for us. But that's one of those things that guests your owners really appreciate. So I think that that's one of those things where not just in the SEO side of things, but I think we have to understand where the human element is still important and where AI can play a better role. Because, yeah, to not use AI, to not leverage these tools, I mean, you're doing a disservice to yourself.

you're doing a disservice to your business, you're doing a disservice to up and down the line. So I think that it's really important to do that while still understanding there's got to be someone behind the scenes driving the ship there. There's got to be some type of captain that is at the helm so that AI does not run amok because, you know, we've all seen Terminator or if we haven't, go to the 90s movies. Look, it's great times. Yeah, that's it.

You got to wonder. Yeah, what's Arnold takes on all this? It's like, get to the chat to BT now. Like, it's just like, I can imagine, you know, all the prompting. Well, good. Well, I think we need to put a bow in this one. We're coming up against the time wise. Any parting thoughts that we should say other than like, we are very pro LLM.

I think sometimes people maybe don't know that context about us because we do. But I think what you and I both dislike very strongly is this idea that like there's a one button solution. And there's a lot of folks out there that kind of market this one button solution. And the truth is, you know, for many things, for most things that you're talking about in a practical day-to-day basis. basis. This is like, again, is it a horse versus a tractor debate? I think that's more so what it is. Like, the horse could only do you plant so many rows of, you know, whatever in the garden, right? And the tractor can do it all day long. Like, but there's also a lot of downsides to the tractors that we have currently. And just this idea that like everything's just one button and it's not there. And you know this intuitively. Because if you go look at any of the top companies in your market, you know, you see, you can see what their marketing looks like. And it's not as simple as just clicking one button. And if anything, you might almost need like. like more different people to manage all the AI outputs and prompting and all these things now. So it's like this idea that like everybody's job is toast. I just I just don't believe that in any possible stretch of the imagination is going to be the case for most small businesses.

So it's still there. But yeah, any other parting thoughts before we put a bow on this one? I think we got it. We covered this one pretty good. Yeah. All good. All right. The do's and don'ts of AI and LLM with your SEO strategy.

You know, you know what's coming, folks. There's one more thing you need to do. Actually, two things you need to do. You need to go to the show notes. so you can click on links and you can reach out to Paul if you want to chat with Paul. You can reach out to myself on the Buildup Bookings website. Paul, you'd reach out to him on LinkedIn if you want to go that way. So that's one thing you need to do.

One more thing you need to do is leave us a review. So go to your podcast, have a choice. Click five stars. We appreciate that. Spotify iTunes listeners who've made it all the way to the end. You are our most common downloads. So you're being drafted into the Buildup Bookings heads and beds army, army, if you will.

And your first task, you know, out of boot camp here is to go leave us a review and Spotify or iTunes. If you're listening on another platform, we appreciate our review anyway. We're not going to be that picky. But we super appreciate the reviews on Spotify and iTunes because that's where we get the most downloads. But that's all I got for you. If you have any questions, send us a message. Let us know what you're thinking. And Paul, at the last minute here, you're going to the writer a cup in person.

This is recorded beforehand. So if he nails it, he deserves credit. What is the final? What is the final score? How many points will the U.S. win by in Bethpage in New York? The U.S. ends up with 17 and a half points. However, let's see here, we'll lose 17 and a half to, what is that?

13 and a half something like that you could just pick the 17 I think it's 28 points so that's or whatever 70 or 20 or whatever something there's 12 singles matches then there's 10 other matches yeah yeah that's the case yeah 30

so there's 22 points available in total or 32 points available in total is that right well we're struggling yeah it's 17 we're at the US 17 and a half that's that's my number that's it so okay all right we'll see where it lands up if you're listening later Google Ryder Cup 2025 results and see how close Paul got it. It'll be fun. He's usually pretty good

this stuff. So all good. All right. We'll put a bow on this one. Thank you, dear listener. Have an awesome day.