How Short Term Rental Hosts Should Actually Use AI in 2026

Mark. So, um, I think probably the main, main, uh, cut and thrust of this session is just going to be like picking your brain on AI, and specifically I think post-Scale AI, which is a really fantastic conference last week. There's a lot of, a lot of, I want to say, like disinformation in the market. There's a lot of people doing some really cool stuff in the market, and we heard from Graham Sykes last, last, I should say Forge, last week. Talking about, you know, 1,000+ AI agents and 150 different tools that he's using, really sort of, I'd say, innovating in the sector. But for a lot of the sort of businesses out here that are sort of a one-person band, so to speak, you know, I think everyone wants to know what we should be using, where we should be spending our time, and what's the best use of time and resource. And specifically, I think one of the things I want to pick your brain about is, is WordPress. Um, it's definitely something, um, I've got a ton of experience using. I think most people in the market use it. Um, uh, so yeah, I'd love you just to start off with your sort of WordPress journey, what you think is interesting in terms of developments there, and is it, is it here to stay? Uh, is anyone going to come and, and eat their lunch?

Yeah, so WordPress is very interesting at the moment. Um, and there's been a lot of cool discussions, and I've been on stage with Richard Vaughan and um, at, uh, various events over the last couple of months, and we've been on podcasts and webinars talking about it. And obviously Boris, uh, from OnSeason has joined in, and there's a lot of opinions at the moment. One of the things that caught my eye is, well, WordPress, for those who don't know, the CMS solution, content management solution for websites, it, uh, is the go-to, 43% of the entire market in January 2026. We're using WordPress websites. Uh, the reason being is the solutions like Wix and Squarespace, or even, you know, for smaller operators who are using a PMS solution, it's very walled garden. So, you know, you've got, uh, you've got a website, but you're very restricted in what you can and can't do. So WordPress is good because it's open source. Uh, search engines love it. It's like a blank canvas. It's like, you know, it's like having, uh, literally the world is your oyster. But what's really interesting over the last couple of months is that 43% has dropped down to 41%. Why is that? Well, as you've mentioned, there are so many AI solutions now. Replit, Lovable, you name it, there are AI solutions out there. Obviously, Boris Onsiesen is creating his own AI solution, etc. So there sort of has been a communication and discourse around this. But there's no way a company as big as WordPress and the overriding company that owns it is going to just, you know, give up their dominance on this. And they are doing a lot, like you mentioned, within the platform to be more AI ready, so to speak. For us, you know, I can't lie, you know, we have had our heads sort of turned, especially when people do speak about it so openly. And, you know, I respect Richard Vought massively. And you know, when he brings that— bridges that topic, that discussion, you have to pay attention to it. But of all the research that I've seen, of all the people that I've spoken to, including the likes of Gil Kahn from Crafted Stays and true experts on these topics, it is still the best option. I think where WordPress gets a bad rap is load speeds or potentially security. But for everybody that we've dealt with, and we've worked with over 2,000 host websites now all over the world, you know, we've generated $1 billion in bookings over the last 10 years. When WordPress websites go wrong, it's not the solution, it's the human behind it. So for example, you mentioned plugins. One of the reasons why WordPress websites could load so slowly is because it's very tempting to just dump and upload a ton of plugins to a website. When you do that, it naturally will just bulk up the site and slow it down. Security, Again, humans get in the way. They maybe take down a couple of the barriers that make it secure, and then obviously things get in. You know, luckily at Boostly, we've got everything covered for that. That's why we do sort of take control of these WordPress sites for our clients. We give them full, you know, proximity to do what they want. But obviously this is, this is like our bread and butter. But still, like WordPress, from all the research that I've seen, it is the best and it will continue to be the best. And for us as a business, one of the reasons why we got very popular over the last 10 years is because we've built these 27 integrations that go into the PMS. So you guessed it, Hostway, Logify, etc. And we've been able to build a very good booking flow from landing on the homepage, going to the listing and checking out, you know, which is again one of our sort of USPs here. And going back to these AI websites, Lovable, whatnot, I've not seen a very good booking flow on, you know, these AI sites that have been spun up. They do get disjointed and broken for them for the prime reason being is they haven't got a very good integration that goes into said PMS.

It's a really interesting point actually as well. So I was going to ask, there's a couple of things on that. So PMSs, they're all going, you know, very, very strong into AI. You know, they've obviously seen that there's a risk to their business model. There's a risk, I guess, that they could be replaced. But also there's a risk that people move to, to the next big shiny thing. At the Scale conference, Guesty were, were, you know, a ton of, ton of very, very good updates. They were sort of talking the talk and I think probably one of the leading PMSs out there now. But one of the things that was really obvious is that if you're not primed and not ready for, you know, the next evolution, you're going to get left behind. I'm never going to, going to knock, you know, rival PMSs in the sector. But there are some out there that are very, very late to this. So just interested in your thoughts, having seen and worked with them pretty much. Yeah, yeah, yeah.

I mean, there are a lot of PMSs, a lot of them. There's over 1,000 and there's lots of different solutions, which is a good and bad thing, right? When there's so many solutions, it helps keep costs down, etc., and there's a lot of choice, yada yada. There's some that are You know, brand-specific, niche-specific, you know, you've got location-specific. For example, in Dubai, there's some really standout PMSs like Mr. Alfred that is so good to the market. And like Australia, wherever, right? What I'm, what I'm noticing is that PMSs that are struggling isn't that they feel like somebody can just come and create a PMS. I don't think anybody can do that. I've seen people try. But it's not going to work. What they're struggling with is the perception of what a PMS is in the market, from hosts on a smaller scale to operators at a medium level and like the bigger operators at a large, larger scale. Um, and you know, it's very interesting to see what everybody's doing and how, how they're doing it. Some have really gone all in and created MCPs, uh, which is like basically, for those who don't know what an MCP is, it's a fantastic and easy way to create and to connect your PMS to your large language model of choice, Claude, ChatGPT, Manus, Perplexity, you name it, right? And you know, the ones that move first, they got spoken about a lot. And what's really fascinating is that people are going in and they're building tools that sit above their PMS or they're building tools that will help them within the PMS. One of my favorite examples of a Boostly client is they used the MCP from Hospitable, connected up to Claude, This person has got no technical prowess in any way, shape, or form. You know, if they were to rank themselves 1 out of 10 on their AI skills, they'd probably put themselves at a 0. But what they were able to do is they used the MCP from Claude into Hospitable. The Claude went in and they were to identify which blank dates there were in their calendar for the next 6 months. And he said, listen, I need bookings. So it came up with a plan, right? It said what we should do is we should reach out to people that stayed with us over the last couple of years that haven't yet rebooked. And it put the plan together. The host signed off on it, said, yep, go for it. And what it did is it went into it, identified the guests, sent them messages, and those messages either went direct to their email or even via through the Airbnb Messenger. And it generated $7,000 worth of bookings within 72 hours, which was great. He would never have done that. Which is amazing, right? And that's just one example. I've seen people on a larger scale that have built out their own messaging systems, their own version of operations and pricing and etc. So it's really cool. So I feel with the PMSs that are moving quick, Guesty obviously are doing something now with the AI agents that they've built in. It is fascinating to watch from like the outside and obviously when I get to go and have a play on the inside as well. We've talked about it a lot over the last couple of years, consolidation in the market. Uh, we've— I know Richard has been mentioning it a numerous of times. You know, you can just go look at the thought leaders in this industry to talk about lots of times. I think now it's probably happening. More and more people that are leaving their smaller ones, uh, moving to the bigger ones. And instead of being 1,000, I think we're looking at a case where we could probably drop down to a couple of hundreds, and then it'll go down to the top 100, and it'll probably drop down even further to like the top 60, 70, and those will be the ones that, that win.

Yeah, it's really interesting. I was going to say it's a very fragmented market, but there is definitely not just consolidation via sort of acquisition, but I think there's a lot of people who are going to drop out the market because just because they're going to be left behind, right?

Yeah.

Yeah. One of the questions I had for you was I see AI is basically affecting every part of every business in this sector. And one of the interesting bits, I think it was this week or last week, maybe there was like a threshold crossed where I think the amount of content being produced by, by robots on the internet was like over 50%. The, the interesting, I think, area that people are looking at now is like, where is AI going to be the most prevalent? Is it content production? Is it cutting costs out of your business? We heard Graham talk about you know, massive inroads being made at Sykes, or is it Discovery? Like, are people— I mean, everyone seems to think that Discovery is going to, you know, Google is going to suddenly disappear and, you know, everyone's going to be sort of buying via Claude and ChatGPT. But the reality of it is, you know, isn't it just that this is just one of, you know, multiple different avenues that you need to plow going forwards?

Yeah. So I can speak to what we're doing at Boostly. So I created an AI chief of staff, called him Icarus, right? It was built on, on OpenClaw because the AI experience that you have is very similar to your website experience, right? You can either have the walled garden approach where if you go into Claude or if you go into Perplexity or if you go on to Manus or, you know, ChatGPT, you can connect. So use these things called connectors to go into like your tools and solutions. So Um, like ChatGPT has a few, Claude has like more than them, Manus has got a ton, Perplexity have got tons. Like, and now I go into Perplexity and I go into Manus every single day, and there's more software and solutions that are being connected up. So like, for example, we use a lot of Airtable. I don't know if anybody else uses Airtable, but we use, um, Close CRM, we use ActiveCampaign. Obviously with Manus you've got Meta, because Meta kind of owns it. So you can connect in your Facebook, your Instagram, your ads, etc. So you can just one-click connect. One-click connect is very simple to do, very easy to do, but it is a walled garden. You know, you have got to sit inside their guidelines or guardrails of what they face out. So you can't do anything like cool and cheeky. I did a little bit of that, but then I've also gone like to the next level, which is using something like Open Claw, which is like an open project. It is very more DIY where the walled garden approach is very done with you. This is like DIY, so you have to figure it all out. It's a lot of chatting with, with Icarus, my, my AI chief of staff that I named him on, on Telegram, trying to figure all this out. But once you do build it, there's loads of cool things that you can do. So what, what I'm using my AI chief of staff to do is number one, he's identifying. So what I used to do, just to dial back a bit, what I used to do is every 3 months is I would do like look at all my costings and look at all the outgoings.. And I would try and identify ways that we could save money. I did it once every 3 months. It took a long time to do, but it was good because at the end of the day, we're cutting costs. Icarus now does it pretty much every month for me and I get it on a schedule. So you can say, hey, on this month or this week, do this and it will go and do it for us. So it's letting us know by looking at our transactions, our credit cards and yada yada, where we're, where we're using, where we're wasting money. Because what it does as well is it goes into our emails. It looks at all of our emails, it looks at our usage, and it tells us, hey, like, you're using this software, you're paying for this, but no, you're not using it. What are you doing for it? And that's been number one, that's been good. So that has been a cost-cutting, uh, thing that we have created with one of our agents. Um, one of the other things that we've been doing, which I talk about this a lot on, on Instagram, LinkedIn, I give updates now and again, is we've got it tapped into our CRM. So we use Close as our CRM, and we've got like 40,000 leads. And I spoke about this when I came in the first time before the Short Stay Summit, and I spoke about Icarus, and then it was like 45, 50 days in or something like that. And it generated like a couple of thousand. Now it's generated over $60,000 through these agents that we have built that live inside our Close CRM. And what they do, because I've been able to build it on, on open Clore and not in like a more of a walled garden, is that there's so many cool things that I can do. So one of the agents will see the details, so they'll see the website of our prospects. It'll be either a host or management company or whatever. So we'll see the website, we'll see where they're based, we'll see how many properties they've got, they'll see their listings, and it will go and do research. This is awesome. So it goes out by itself, does research on each prospect, and then what it'll do is it will come back and it'll tell another agent, right, this is all the research that we find. I want you to put together an email slash SMS to this prospect that's on, that's on Boothley's books but hasn't signed up about why they should work with us. And the research that it does and the message that it, it puts together is so customised like that person will have to read it instead of me just sending out like a mass email to 40,000 people and just sort of praying and hoping that someone's going to open it at the right time and then action on it. Imagine something that speaks so personally to you, like it would be impossible to not read it. And so from that alone, doing the research, sending the message out, we've had— and everything that we send them to as well so that the agent isn't calling the lead and not closing the lead, that still needs a human.. But all of that donkey work to get to that stage where a book gets called with a human is being done by AI. And every single agent that touches a lead, it's got a UTM link attached to it. So I've been able to like track everything that these are doing over the last couple of months. And now we're sort of like, say, over $50,000 has been tracked back to these. And, you know, I spent a lot of money last year on human SDRs, like sales development reps. Um, not one of them produced any revenue. And I, I used to think it was like my hiring, maybe I'm just like crap at hiring salespeople, uh, but what it were is, as humans, we're all guilty of this, right? We're all business owners, we all thrive on the new leads. Like, we've all watched Glengarry Glen Ross too much, you know, it's new leads, new customers all the time, when, when truthfully, like, the true income, the true revenue sits in the leads that are in your inbox. If somebody like says to you like 2 years ago, nah, don't contact me, they— no human will look at that lead, go, well, that's a good lead. But the agent and the AI doesn't care. It will literally not care. It will go in, do the research, and it's just how we've reached out to them has been wrong in the past. Sales, great. Content creation, I draw the line at content creation AI is crap, right? And I feel like this terminology AI slop has been banded around, and I think that it's been just— and like that stat that you said, over 50% of content's done by it, you can tell because as soon as I see AI written content or an AI image, my mind just naturally goes, next, right? I feel like with content, the cream will rise to the, to the top. Right? And, you know, I've obviously used AI to create content for me, but I feel like what will happen is that people will switch off so much that it'll just go back the other way and people will not write content with AI because people now can spot it a mile off. I mean, even ChatGPT, one of their big sort of pushes and promotions in their product releases that they got rid of the em dash in their content because everybody was complaining about em dashes. And yeah, I feel like content, no, not for me. Doesn't say that I haven't used it to quickly whack together an email. Right. But I think we're going to go more back to like proper using proper copywriters. If I was a copywriter or a good copywriter, I'd be putting my prices up because you're going to be inundated. But the other side of things, like I said, sales, marketing, cost cutting, AI chief of staff. 100%.

I'll come back to the AI Chief of Staff in a second, if that's all right, Mark. I was going to say, really interesting talk by Neelie and a lot of the work that she's doing around not just being discovered online, but being chosen online, which is really interesting. It's kind of almost like, you know, you've got to get your foot through the door, but once you've got your foot through the door, people have actually got to listen to you. And if you're just churning out garbage, you're just, you're going to just disappear. One of the things I think is really interesting about, about purely about content production and copywriting is, is ultimately you're paying for someone to, to write in a way that is, is human. It does sound different. It's very hard to replicate. A lot of the models seem to sort of just go back to basics and just churn out as much as possible. So One of the things that Neelie was talking about is the story and how important it is now to have a story and to be, you know, for as an individual entrepreneur. And Mark, maybe you'll be able to sort of give us a bit of your thoughts on this. But, you know, AI is producing or giving people the ability to run a team rather than have a team of 20 people, to have a team of 20 agents. But the real skill, the story, the storytelling, the, you know, the nuance of your business comes back down to the founder and what they do. So One of the key things I'd love to hear about from your perspective is how important that is going forwards to really have a story about your business and really get it out there.

Yeah, I piggyback off what Neely says and I agree. I mean, Neely obviously first and foremost is a story writer. She is a storyteller. She helped me write my first book, The Book Direct Playbook. Phenomenal, talented, talented writer. And I agree. The— and it's so bizarre, right, where I think the companies that are going to win, and I've noticed this a lot recently, is the companies that build a community. Now, I'm very lucky at Boostly, I built a community first before I built the business. So I did this not on purpose at all. I started in a Facebook group, build a community, build like a, like a following per se, built like a personal brand, and then Boostly just came naturally on the back of it. And the things that we've done contribute to that. So now, like, even though that we've got, you know, over 1,000 active clients right now and, you know, we've got people all over the world, like, we still do the community building aspect where we have twice a week group coaching calls, like similar to this, like, and we've got a private community, like similar to, similar to like the school group that you've got. Most companies don't do that. I think you— I'm thinking on top of my head, like other website designers, None of them do that. Uh, and that's sort of like that community aspect, that brand building, uh, is going to be huge moving forward because there are so many tools out there right now. Like, you literally just have to go into LinkedIn and every single week there's a new company or a new product or a new whatever that's been started because it has never been easier to start something right now. Um, and that is a, that is a benefit of AI, you know, like up until I think probably January, you know, there's like, I don't know about you, you two, but I've got like a notes in my Apple iPhone with ideas, you know, and just like things for like a rainy day. And those ideas are no longer just ideas on a note. You can just whack someone in on one of these AI tools with Replit, Lovable, whatever, and you can put something together like a first version, which is crazy. It's a blessing and a curse. We launched a product that we've been working on for nearly 2 years this year called Boostly Connect. And as soon as I put something out on Facebook about it, oh yeah, I built that. Like, what? And I saw what they built. They hadn't, because like we have put time and effort and like proper engineers onto it. We didn't just vibe code it over a weekend while watching the football, you know what I mean? So it's so hard to launch something now because like everybody else is launching it and people get like launch fatigue, product fatigue. But the ones that are going to win, and I come back to PMSs and I come back to every business, is like the ones that have got the story that Nili talks about. But I'd go on top of that, the one that's got the community behind it as well, like building superfans. So when, you know, somebody posts online about something new or whatever, you've got people that literally go out and fight for you in the comments going, well, no, you should be using this. This is like, this is the one you should be using. Or if somebody says something negative on like a forum or a group or a Trustpilot, you've got like an army of people who will jump in and defend you. It's like, that's the community. People that support each other. Um, and, uh, you know, that community can be offline, it can be online, it can be wherever. But I feel like that's combined with the story, which I agree, the personal brand, I agree, uh, the community. I feel like that's where companies are going to win. And I include property managers in this, like people who have got a business. Let's just say you're a, you're a management company, and let's just say you've got 50 owners Um, the OVO Group in France, so Louis Andrews and his crew, once a year they get all of the owners together and they have like a retreat for a couple of days. Like, that is community building 101.

I—

then it separates you from just being a place that sends out an invoice with a reporting once a month asking for your 10, 15, 20% to, uh, you know, a face, a name, a community. And that is key for moving forward.

100%. And I was going to say, actually, Graham talked about this, um, as well during the, uh, the Scale conference. He talked about his moat, and his moat being the owner, uh, relations team. And he just said, look, you know, we're cutting, um— he showed a really good graph on, on the screen of like, you know, areas where they were cutting cost out. And there was marketing, there was finance, there were all these areas. And he just said, look, that owner team, that's the people, that's the it's a people business and you need those people. And it's super interesting to hear him say that because Sykes is, you know, phenomenally large company. Well, so Forge is a huge company with, I don't know how many, tens of thousands, uh, now sort of 30,000+ properties under management. But it all comes back down to that whole thing of empowering the, the sort of physical and the human connections between owners and, and, and your company, which is, I think, so important. So I want to just really quickly go back onto your AI chief of staff because I don't think we've given that enough airtime yet. How did you set it up and what do you set it up on and what does it do for you?

Yeah, so I set it up with OpenClaw. Like I said, there's, there's two ways we can go about this. You can either do it with OpenClaw, which is obviously the Peter Steinberg open project that was put onto GitHub that went very viral, very famous. Now he's been hired at OpenAI because he got so famous and powerful and, you know, very cool story. I built it on OpenClaw. The reason why is that it is more DIY. So it's a lot of learning, you know, spent a lot of hours sort of screenshotting, asking, you know, ChatGPT, Claude, what's going on? How do I fix this? Yada yada. So it's a lot to set up. I did it on a server, so I didn't do it on like a Mac Mini that everybody was going out and buying in January and February and put that in place. So when you first get going, you sort of have your central agent that you talk to, and I call it my AI chief of staff. I plugged into there all of like the important things I wanted to plug into. Obviously, I didn't give it access to silly things like bank account details or like my own personal email. Did it, set it up like we were advised to do and did it right properly. But on the back of that, these agents are very powerful, very clever, and they spawn sub-agents, you know, which I find phenomenal. So Icarus, because I called it after the Greek mythology,. And it's a nice little reminder to me to not get too cocky. Icarus, obviously the Greek god who got the, got the wings on his feet, got really confident, got cocky, too close to the sun and burnt and died. Right. That's my daily reminder is don't be doing something silly with this AI. Don't go and give your bank details and all your cryptocurrency and let it go YOLO and all of that. So, but on the back of that, it started creating other agents. So Icarus created Daedalus and Hermes and all of these and they're all given assigned tasks because I would come to it every single day and I was like, oh, I want you to do this. Or if I'd see something cool on Twitter, I'd be like, oh, can you do that? And all I was literally doing is copying and pasting a tweet or copy and pasting a Reddit or taking a screenshot of a Reddit and I would feed it into it. And what was really crazy was that one day I just voice noted because I've talked to it on Telegram, so you can choose where you want to talk to it. I talked to it on Telegram and I hate typing on my phone because I've got these fat thumbs from playing PlayStation too much when I was younger. So I voice noted it. And it was bizarre because it said, oh, I don't know how to listen to this. So we went and created his own skill that was able to listen to voice notes and then voice note back to me, which I was like, I didn't even train it to do that. It just went, oh, you want to do this? Let's do that. And I thought that was pretty intense. And it just sort of expanded every day. And it's very addictive, right? Talking about video games, it is like playing a video game, but, uh, it's a video game that's just never-ending, you know. And it was, uh, it was very cool. Um, since January, um, around about April, May, June time, what— because of the popularity of his open clause now, like I mentioned with Claude, like I mentioned with Manus and Perplexity and whatnot, they've made it a lot more easier to connect up, you know, tools and APIs and MCPs. So if you don't need to go down the OpenClaw route, um, you can easily create agents now that don't have to be on OpenClaw. I just prefer it because, like I said, it is a blank canvas. You can do really what you want with it. I couldn't have done that closed CRM example with a Claude, for example, or a Manus or Perplexity. Um, so what does it do? Well, number one, it is like my, you know, AI chief of staff. It is my business overview. Every single day I wake up and I have these triggers and I've got these scheduled tasks set up where it tells me exactly where we are as a business, where I give it a goal of where we want to be by the end of this year. And it just lets me know every day where, how far we are off it. It looks at all of our projects, it looks at all of our inboxes from the customers and all of the things it's asking. And it gives me like a high-level business overview. It doesn't just give me data. It goes, right, this is what we're seeing. And it just, every single day, it's coming in and just giving me like a high-level overview of where we should spend our time and focus on. And it's, it's, it's pretty cool, you know. And again, all of this is for free unless you spend in tokens. So I probably spent a couple hundred dollars a month on tokens on this. But it is being so impactful, helping us as a company decide on where we go.

I was going to say, Mark, it's really interesting. I mean, I haven't done that. I've created some AI agents in Claude and very sort of low-level sort of, I want to say, sort of help. But it's really interesting to see like how much that can impact on your business. One of the things that I thought was really interesting is previous business, we We had HubSpot as our CRM system. We had a PMS called Supercontrol, which I think we all know isn't the best in terms of tech and enablement. We hooked that all up together with the WordPress website and a Stripe account just to get the right booking flow, and that cost us tens of thousands of pounds. But these days, that's something you can do very, very quickly and efficiently. But one of the other things I was going to say is now with Claude is You know, I've been able to hook that up to the backend of my PMS that I'm using for my new business. And it does all of the automated emails. It creates, you know, retargets. It does retargeting. So anyone that's been on the website, filled in their shopping cart, it will then take their email address. It will then create an email. I don't actually allow it to automatically send that email. I always check and have— I've got all the checks and balances that will check my emails and just make sure I haven't already emailed that person or, you know, all that sort of stuff, which Again, like it's super— it wasn't available to, you know, sort of mere mortals, you know, a year or two ago. Now it's super available and really helpful, right?

What's been interesting as well, and it's getting to that point of the discussion now, is do you build this yourself or do you just wait for your software solution to do it for you? Like we mentioned things like HubSpot and Close CRM, etc., Guesty and whatnot. They're now building their in-house versions. And, um, one of the, one of the negatives— this all sounds great, right? If you're watching this, listening to this, oh my God, I want to get crack on with this. It does sound good, but, um, it's getting to the point now, because these models and these tokens are getting more expensive, especially if you do it on like the latest releases, which can get very expensive. Uh, I know some people that have just burnt through money to the point where they've just tried to create a version of a CRM right? And they're spending like $1,000 a month where you could just go to, well, Boostly, but pays $97 a month and you've got a built-in CRM that does everything that you're talking about and, and more, like reaching out to guests, etc., automations. So there is like this sort of tipping point. Um, doesn't mean it can't be done, and it is exciting as long as you do it right. Like, there's loads of ways to do it wrong, but as long as you do it right, it's It is a lot of fun. You've got the ability to do it. It's cheaper than, you know, paying someone to do it for you. I would say it's like it's yours, but is it really? Because obviously building it on cloud, you're still building on someone else's land. But it is cool what you can do. And the way that I feel like for everybody to first go out and do it would be just create your AI chief of staff, just connect everything in, set up goals, like brain dump it in. I use a tool I mentioned it last time called Whisperflow. Like I say, not very good at typing, but I'm really good at talking in what I want to like these AIs. And Whisperflow is free, which is cool. And you can just voice note in and it, and it recognizes your tone, your voice, your accent way better than, you know, voice noting on your Apple phone or whatever. And, you know, you can really build something decent there. And then when you want to build like these tools that do things automatically for you, like I say, money-wise it can get crazy, but time-wise, this is the thing that a lot of people don't realize, like it can literally eat your day away if you just get so sucked into it to the point where you can just— it takes your eye off the ball, which is like the important parts of your business, which is, you know, bringing in money or whatever that thing should be. So you've always got to like make that balance. But again, do it right, do it well, don't go crazy, don't be like Icarus, you know. And, and there's a lot of rewards and I've seen it. I've seen it for me, like the rewards have been cool. You know, is it perfect system? No, but it's definitely made an impact and it's got me excited for what this next level is because it's only going to get crazier from here. It's not going to go backwards. It's not going to level out.

Yeah, it's really interesting as well. Graham was very sort of, I want to say, painted a cautionary tale about tokens, the cost of running AI, and specifically, you know, Um, you know, they've got to make money at some point. At the moment they're not. So there are, um, you know, there is a day of reckoning down there.

Yeah. And the context is with, with Graham and Forge, he's— he has got a lot of investors that have basically thrown a shit ton of money at him and just said spend it. So he's got a very advantageous position where he can just be a kid in a toy shop and live out his geek dreams, right? Anyone— and he can just do what he wants with all of this, uh, because it got to a point last year where investors were sort of looking at companies that weren't moving and just going, what are you doing? Do this. So Graham's got a phenomenal position here where he's like created thousands of things and he's showed me what he's created. It is insane. But those same investors that were just throwing like a ton of money around will come at some point and go, you know, we gave you like all this money, like what's actually been built? You know what I mean? So I feel for like us mere mortals here, like, you know, small startups, hobbyist lifestyle businesses, whatnot. Obviously, you've got to watch it. Don't go anything crazy. Make sure— I mean, for me, when I went down this rabbit hole, I wanted one thing and one thing only. I didn't want it to become an expensive hobby. It had to generate revenue. That was the first thing in my mind, is it had to generate revenue. You know, luckily for me, within 7 days it had done right. But, you know, if I'd have gone from another point of view without a goal in mind, I could have quite easily just wasted like £1,000 and been nothing better for it.

Is it giving us the ability, though, to compete with those larger businesses now? Do you think that, you know, you said the cost of entry of the market is a lot lower, you know, is speed and flexibility and now like, you know, AI going to give us a lot more, a lot more businesses, a lot more competition?

Uh, there will be a lot of competition, yes, but I come back to that cream will rise to the top. The first question was, do we have an advantage over the bigger businesses? I mean, no. I mean, these bigger businesses have got much deeper pockets. There's no way they're going to let us little mere mortals win because they'll just absolutely chuck a shit ton of money in, um, you know what I mean? But Like for what I talk about, direct bookings, right? And, you know, building a business that's not on Airbnb, Booking.com, Expedia, 100%. Yes, you can make a difference. I just take Sai's example that I gave earlier. You know, he just connected up Hospitable to Claude. He wanted bookings. Claude found these leads, messaged out, got bookings. Right. Brilliant. Not every single one was direct. But a large chunk, excuse me, were. So that's good, right? Because otherwise, this time last year, Si would have just gone on to Airbnb and lowered his rates, right? That's all he would have done. He wouldn't have thought about messaging leads. He wouldn't have thought about putting out like a messaging campaign or an email campaign because he would have gone, A, I haven't got the time. B, I haven't got the knowledge. C, I haven't got the money, right? All these free excuses that I've been sort of getting from hosts for the last 10 years when I'm talking about direct bookings, they just go out the window. Because A, time, literally took nothing. Claude did it mostly for him. Knowledge, doesn't need to know it because Claude does it for him. And then money, all it costs is the tokens, right? And so that's not a lot of money to do it. The only thing now that is going to get in our way from being able to make this a success is our imagination. I talked about those notes on our note app, right? They're gone. You can put together some very simple tools and whatnot. 100%. There's going to be a ton of competition because there's going to be everybody— I just go into Facebook groups, everybody's releasing tools. Yes, but what will stand the test of time, what will be here in 2028, is the businesses that build a community, that have got a good story, got a good brand, got a good product. And you know, it is very simple to create these AI tools, it is very hard to maintain them, right? They break a lot, right? And trust me, we know right? With everything that we've created, they break a lot. So you've got to be persistent. And people will, because it's such a low barrier to entry, people will spin up these things left, right and center. But they'll also stop very simply, right? Because again, they'll create this tool, they'll throw it out there to the world. And then nobody buys from them. They'll go, oh, that was a rubbish, like, experiment. See you later. Right? So as long as you're persistent, consistent, good quality product, build a good community, good personal brand, good story, you will be here in 2028 for sure.

Yeah, it does put everything into perspective. I mean, I do, you know, I don't know how many of the 1,000 or so AI agents and tools that Graham's created, how many have fallen by the wayside, haven't gone anywhere, and what sort of the overall, I want to say, like, use of resource looks like. But it does feel like size of business probably makes that much more complex in terms of what you need to have. One of the things I was going to ask actually, so I think it's probably relevant for a lot of people, particularly Gareth, your business, which is around discovery, because I think everyone is sort of sat there. I mean, I had this question in Scale, you know, should we all just be sort of giving up on Google and spending all our time optimizing for LLMs? The reality is, and my view around this is that, you know, it's a another and a new and, you know, a marketing channel or a channel to focus on. But you shouldn't just be dropping your tools on everything else. It should be part of the overall mix. But what are your sort of views around discovery and how important it is to be prepared for things like, you know, showing up in, on LLMs and how you adapt your business for that?

So that's for me, is it? That one?

Well, actually, it was for both of you because I know, Gareth, you're spending a lot of time focusing on that area. So should we go with you first, Gareth?

Yes.

Yeah.

Yes. So Discoverit, I mean, when AI first properly hit the headlines, I thought that AI was going to just change things so that the big companies no longer had the control. They couldn't gatekeep anymore because AI was going to find you that exact property in the exact place you wanted. Didn't matter how much you spent on AdWords or anything else. But it's going the same way as Google now. It's the vast majority of traffic is going to be those who pay for it. Because these AI platforms, they have to make money. And so, you know, just like on Google, I did a search the other day, there was half a page of AI with sponsored links in it, then there was 4 Google Ads, then there was a map with people connecting in or also paying for that, and then position 1 was way down the page. I think AI search results are going to go the same way. So I think the best thing you can do is just— this kind of stuff that worked with Google, make sure you've got the right information on the page for the bots to pick up so that you do surface in the results. But it's not going to be enough just on its own. Hopefully MCP connections can help at some point when somebody builds something that can connect into every individual host, you know, independent and smaller agencies. But I just think the discovery layer— I used to think AI had solved it. I'm now thinking AI is actually just going to continue the path that Google created. So for me, discovery is the thing that is going to be the hardest thing to sort of overcome.

Interesting. Mark, I've got a quick question for you on discovery. You know, I'm looking at creating a new brand in the market, for a very sort of niche area. But one of the things that I was looking at is, you know, does that produce, you know, it would be a I want to say, like, is AI making people think more in terms of niches, in terms of brands? Are they focusing more on working with a particular business because of their, their focus rather than just being, you know, across all of the sort of big guys? And yeah, does it— what's, what's, what, what do you see the market sort of developing into, if that makes sense?

Well, I can speak from personal. When I've not run a Google search in 2026, if I'm looking for something like a product or a service or a solution, I'm now going into Perplexity. And I'm asking that because it goes out and does, in my opinion, much better searches, and it can be more bespoke to me. And I don't care if it comes back with like, you know, whatever it comes back with. Because what I found with AI, and I've also seen this as well with customers that are booking calls for our website services, is that you trust the results more than just like a vague Google search. So like last year I'd run a Google search, I'd find a solution, you know, I'd maybe then have to go and watch some, some YouTube videos or check the social media content or see if I've got any blogs or see if anybody's talking about it or see if there's any YouTube videos about it. Now, like, you run the search with Perplexity or whatever you like to use, it goes and gets you the results. You ask it maybe a couple of quick follow-up questions, but the result that it gives you, you just trust it. Because these, you know, Perplexity knows everything about me. I've trained Perplexity to know, you know, I've got 4 kids, I've got a, got a wife, you know, we like to travel, yada yada yada. You know, we know it knows what we like because it knows about us and it remembers us. So when it gives me the answer, I'm like, sound, I'm gonna go for that. I'm not gonna dig around more. And we're noticing it with Boothly as well, like the ones that say, oh, I discovered you on ChatGPT or a large language model search. It doesn't take a lot of convincing on the sales call. It's just a case of, you know, you know, showing up and answering a few questions and you're good to go. Compared to if they just found us, you know, generically or wherever, it's a lot trickier. So in terms of discoverability, in terms of how people discover you, you know, more than ever now, you've got to be able to create good content. I stand by this. Video is key. Right? Good YouTube quality video. I think YouTube is going to just absolutely boom, uh, most importantly because that is where a lot of these large language models will get their data from. They're, you know, every single YouTube video now, the transcription is being read and it's being fed into them, etc. I don't think Google are going to disappear anytime quickly. I think Gemini and obviously what they build and stuff around that's going to massively help. Uh, I think YouTube is going to be key, having good quality videos that speaks specifically to the niche you're talking to is going to, is going to go crazy. And this is like the final thing I'll say before I have to go is when I first started Boostly 10 years ago, somebody said to me that the riches are in the niches. And I was like, what does that mean? He said, well, you want to bury yourself so far down in a niche there's not room for anybody else. And I was like, right, stop talking in tongues. What does that actually mean? And he said, but it basically means is that when I started Boostly, I could have quite easily said we're a marketing agency or a website design agency. But then I went down in that niche further. I was like, right, well, we're going to work with like short-term accommodation owners, right? We're going to, we're going to, we're going to help you with direct bookings. We're going to do website design for direct bookings. Right. And we just went right down that niche. So now we're in a niche so buried that there's not room for anybody else. So when people think of direct bookings and websites, they come to Boostly. And I feel like that will be key for a lot of startups moving forward is that don't just be very broad, be very niche, right? Have a good niche and then build your brand around, around that. It works on a number of levels, social media being one, but GEO is going to be massive because again, SEO is appearing for search terms that get searched for, right? And there are short-tail keywords and then there's long-tail keywords. With GEO, if you can be specifically all about this long-tail keyword search, as in like, you know, I'm, I'm me and my wife, my 4 kids, we're going to be traveling. I need a, I need a, let's just say a rental car, right? In America, I'm going to pick it up in New York, need to drop it off at Buffalo, right? Obviously, we're a larger family. So if there's a rental car company that is gearing up their whole content and their strategy around larger family cars, that's not the big ones like Enterprise or whatnot, then they will come up in the search more often than not. They will get recommended on these GEOs and SEOs and AEOs more than if you were just like a rental car company. So that's a very mad example because I was thinking about what I've got to search for next. Well, this is what I'm sort of talking about is truly now the riches are going to be in those niches.

It's really just that whole idea of being super hyper-personalised for every customer, knowing, really knowing them. And again, comes back down to, you know, AI feeds off data, right? So the more data you can give it, the more you can, you can make it personalised. The better it'll be. Yeah, that's really interesting. I appreciate you've got to go very quickly. One, I'll actually give Gareth— do you have any questions before Mark goes? I'll have one final one.

No, I'm just enjoying the conversation.

Oh, good. I hope it's helped.

I've got one final one for you just before you go, and that is, what's your prediction for the next 12 months in terms of AI? What's going to happen?

Uh, so at the start of this year, I said that 2026 and 2027 was going to be the year of the driverless car. I think that that is just going to not slow down. That is going to just go crazy. Uh, I was chatting with my 10-year-old yesterday, you know, we're driving to soccer training, football training. I was like, I can see a time where cars don't have wing mirrors because you don't— we don't need wing mirrors because it's going to be the robots that are driving your cars. And I feel that's just going to just massively kick on to another gear, and I've seen it already. And then further from that, I think 2028/2029 is when we're going to start seeing robots. And I think that these— it's going to change everything, especially for our industry, massively. Having like these robots or humanoid robots, when we see robots walking down the street, for example, it's just going to— the minds are going to get blown. And I feel like that's what's going to be on, on the horizon. What do I think with like ChatGPT, Claude, uh, Manus, whatever? I feel like they're just going to keep getting better and better and better. I feel like, um, you know, Claude at the moment is going for a bit of a— Anthropic's going for a bit of a thing with, with its latest models and whatnot. I think there's going to be a lot of, you know, a race to be number one, and it could get a little bit crazy at some point. I don't get involved in, in that. I just like using my AI for what I want to use it but it is fascinating. And I said this to somebody at the Short Stay Summit, you know, it's going to be massively interesting to see which companies come back a year later, which new ones pop up, what solutions are for our industry and whatnot. And, you know, it's— I would not be surprised at all if like next year or maybe the 2028 Short Stay Summit, we've got a booth that is showcasing humanoid robots for short-term rental properties, which will be fascinating.

Well, you heard it here first. Um, I really appreciate it. Um, it's been a really, really, uh, great, uh, opportunity to sort of pick your brain. And, um, yeah, uh, I guess we'll see what happens in the next 12 months.

Yeah, well, thanks for having me. I do enjoy jumping into the school group from time to time, but I do have to go and run out the door now for the school run. So thank you very much, chaps. Appreciate it.

Likewise. Thanks, Mark.