STR Global Unlocked with Simon Lehmann: Unfiltered knowledge for the short term rental industry

033: Why Booking Trips Is About to Change FOREVER (CEO Explains) | Amirali Mohajer

• Powered by AJL Atelier • Season 1 • Episode 33

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 55:06

AI is changing how people search, plan and book travel and this shift could reshape the future of direct bookings, guest relationships and platform dependency in the short-term rental industry.

In this episode of STR Global Unlocked, Simon Lehmann speaks with Amirali Mohajer, Co-Founder & CEO of Host.AI, about the future of travel search, the rise of AI assistants, and why operators may finally have a real opportunity to reduce their reliance on OTAs like Booking.com, Airbnb and Expedia.

They discuss why direct bookings have been so difficult to scale, how AI could change guest acquisition, why owning guest data matters more than ever, and what short-term rental operators need to understand as travel discovery moves beyond traditional Google search.

If you want to understand how AI could transform travel distribution, direct bookings and the future of hospitality, this conversation is a must-watch.


Questions answered in the episode:

  1. How is AI changing the way people search, plan and book travel?
  2. Why could AI become a major turning point for the travel industry?
  3. Why are short-term rental operators so dependent on OTAs like Airbnb, Booking.com and Expedia?
  4. Why have direct bookings been so difficult to scale until now?
  5. Why is having a direct booking website alone not enough?
  6. How can operators build more trust and stronger guest relationships outside of OTAs?
  7. How does guest data influence future bookings and platform independence?
  8. What should operators do now to prepare for AI-powered travel discovery?

Resources

Stay connected:

SPEAKER_01

Traveling is changing fast. For a long time, short-term rental operators had to follow the rules of the big platform. The platforms control the visibility, the bookings, the guest data, and often the relationship with the guests as well. But now AI is starting to change how people search, plan book trips. And this could become a big turning point for our industry. That is why I'm really excited for today's conversation. I'm joined by Amirali Mohacher, co-founder and CEO of Host AI. He understands travel technology, platforms, and how AI can help hospitality operators build stronger guest relationships. So today we're not talking about AI as a trend, we're talking about what this could mean for direct bookings, guest ownership, and the future of short-term rentals. I'm Simon Lehman. I've spent the last 20 years in short-term rental and property management working with operators, platforms, hospitality businesses across many markets. For operators listening who are thinking about their own growth, their direct booking strategy, or how to depend less on platforms, this is also the kind of work I help clients with consultation. You'll find the link in the description below.

SPEAKER_02

Simon, it's a pleasure to be here. Thank you so much for having me. I'm a big fan of the show and glad to be here.

SPEAKER_01

Awesome. Tell us where you're at right now and give us a little bit of background on your surname, if you don't mind.

SPEAKER_02

Yeah, absolutely. So uh I'm currently actually in Mexico City. I'm normally based in San Francisco, recording from Mexico City, though it's a beautiful city out here. Uh, my last name's Moharger. It actually means the one who travels in Farsi. So there's some you know normative uh determinism in there.

SPEAKER_01

That uh you live and breathe your name, basically, right? So that's right, that's right.

SPEAKER_02

Yeah, I love that. So I love this industry.

SPEAKER_01

Yeah, we'll find about that, we'll find out more about that in our conversation, and I'm really looking forward to it. And once again, great to have you on the show. Before we get kicked this off, we're gonna do a quick rapid fire together uh to set the scene. Um, quick statement, quick answer from your side. Are you ready for this?

SPEAKER_02

Let's do it.

SPEAKER_00

OTAs, necessary or vulnerable? Vulnerable direct booking myth or missed opportunity?

SPEAKER_01

Missed opportunity AI in travel, evolutionary or disruptive? Disruptive biggest misconception operators have about distribution today.

SPEAKER_00

Um that it's determined.

SPEAKER_01

And one prediction what changes most in travel booking behavior over the next five years?

SPEAKER_02

Um use of AI assistants.

SPEAKER_01

Wonderful. Uh I love your misconception as well. I would have said uh cost of distribution. Uh interesting. We'll probably talk about that in our conversation as well. When we do consulting with property managers, I think you know the the to understand what is the true cost of distribution and customer acquisition is still an absolute black box for most of the operators, right? So anyway, so let's get into your background and and entry point. You've spent a significant part of your career uh in travel tech, including time again as well, and building platforms in different markets. And before we get into host AI, what did you actually learn from that experience about how distribution actually works in travel?

SPEAKER_02

Yeah, absolutely. Well, um, so I built a small regional OTA and then I worked at a very large one. And I think one of the key learnings from that was that playbooks are very similar, regardless of the OTA that you're working at. They just become much more sophisticated in how they're deployed, right? And so at the heart of it, an online travel agency is a marketplace, right? So what they they they built their uh moat and defensibility of their business by having unique inventory and unique guests, right? Um and so what does that mean? That means that their business model is built upon kind of maximizing their share of revenue at a brand so that they can leverage that relationship and build the best possible deal that they can get from their guests so that they can maintain that guest relationship and keep bringing them back, right? So effectively, on the supply side, uh the all online travel agencies are after unique inventory. That means the sources of supply that don't exist anywhere else. Yeah. And then the type of deals with that supply, the type of pricing, promotions, and kind of uh the structure around that offering, that's not easy to find on other online travel agencies. So that's what happens on the supply side. And then by nature of having that unique offering, it helps them essentially attract and retain guests more effectively, right? Um that makes sense at an OTA level. When you're sitting on the other side of that and you're an operator, you end up having a lot of repetitive conversations that you know, you if you drop your rates, if you activate this promotion, if you change your cancellation policy, we'll give you more impressions on our platform. Um and so uh there's this dynamic that has been going on really since the dawn of Expedia in the market, particularly on the hotel side, and it's carried over now to the short-term rental side as the inventory is becoming cross-listed on multiple platforms, and we're approaching that uh kind of um level of having all options available on all uh marketplaces.

SPEAKER_01

So, from all the experience you've gathered and what you have seen and what you just alluded to, where do you think, or where do you did you start to identify the limitations of that distribution model, if you wish?

SPEAKER_02

Well, there's that in that quest to compete, the a lot of um complexity ends up being embedded in the in the model. Um so you'll see the inventory being passed around like two to two or three layers, and that makes the the kind of experience suboptimal for the guest and also the the operator. And then this dynamic that I mentioned around repetitive conversations around decreasing your rates or kind of activating promotions that ends up building some level of frustration when when it comes to the operator, because uh they feel they are beholden to third-party distribution, right? If a large share of your revenue is coming from a single source, when they come to you with these asks, it's very difficult to say no, right? So you have to come. And so you would have to comply. And you know, to your point about distribution costs, that oh that's a hidden cost of distribution because essentially you dropping your rates, um, it's not necessarily captured in the in the commission you're paying. Definitely not.

SPEAKER_01

And it's not a it's not capturing your true cost of acquisition either.

SPEAKER_02

That's right, that's right. So that in itself is essentially costing you money to drop rates to get that extra impression, but it's not an explicitly kind of captured. And so what I noticed in you know hundreds of conversations I had with different operators is that they would essentially at the end of the day agree with the demands most of the time, but they feel frustrated about this dynamic, right? So I noticed a strong motivation to grow direct bookings, but growing their direct bookings is a form of essentially making themselves more independent, making themselves more less reliant on kind of uh third-party distribution. But it's a very difficult problem to solve. And you know, it's it's it's uh it's almost like opening a combination log. There's a lot of pieces to the puzzle that you need to get right for the whole thing to work, right? It's not just about having a strong-looking website, right? So it's about you know uh solving acquisition, solving pricing, solving persuasion, trust building, conversion optimization. All of these pieces need to need should work in order for you to be able to make direct bookings a meaningful channel uh for your brand, right? And historically that's been very difficult to do for independent operators because it takes a lot of time and attention effort to get all these pieces right. And you would have to basically put together all different pieces of software and third-party services and orchestrate them all together, and you would have to invest a lot to make all of that work. So oftentimes the the upfront investment would be high, and then the return on that investment would have been difficult to kind of make uh make work because things would fall through the craft, the orchestration wouldn't quite uh work as effectively, and and uh and so the that brings us to where we are today, where the motivation to grow direct bookings has always been there, but the ability to to actually make that work hasn't uh hasn't kind of proven out over the past 20, 30 years. And so, you know, uh that's been a challenge in the industry.

SPEAKER_01

So let's unpack that a little bit further. And you already alluded to that. The problem with direct bookings in general. I mean, we've seen it's been around a recurring theme for years. You know, some call it the devil, some call it, you know, uh whatever you want to call it, and some call it, you know, this is the future of wherever we need to go. And we're very obvious and clear about this that certain PMCs have already historically a lot higher direct bookings versus others, but especially the newcomers uh rely on on OTA or third-party bookings a lot more than well-established brands that have been around for decades and build a customer base and everything else. So, you know, this conversation is is never ending, you know, and and we still see more initiatives. And you just said it, you know, is it a di is it is it a website? No, it's not. It's a lot more. But as with evolving technology, things are ultimately going to change. So diary booking, um, everyone wants it at the end of the day, but very few have made it work at scale, with the exception for the ones who have been around for a long time, have been leveraged to uh able to leverage their brands and and their distribution captive area, I guess, in some particular markets, right? So why does direct booking be so difficult to solve? I mean, next to the points you've already mentioned, I mean, you need to connect a lot of different dots and and uh and and it's still very, very hard because especially when you fundraise, for example, I'm having currently some conversations with people who are out there in the marketplace thinking about hey, I want to build a direct brand, and then the investor is like, oh my God, we we we don't want to waste money just on marketing. We you know, are are can we just do this on uh third-party distribution where you know distribution costs are high today, they could be even higher in the future. And one thing we can clearly say is nobody uh really has been ever able to build a strong consumer-facing brand in this industry. And and now it seems is this gonna unravel? So back to the question, what why has direct booking been so difficult next to all the things we've just talked about?

SPEAKER_02

Yeah, absolutely. So if you think about the sources of direct bookings that you can get, right? The I'll talk about the different sources and level of difficulties, and I and I'll talk a little bit about actually what has been the main challenge up until today, right? So the the the simplest one to get direct bookings from is repeat guests, folks that have come, stayed with you again before. They know you, they know your brand. And so what what we see there is actually repeat guests is an area that small operators work kind of um very well in terms of driving repeat guests um because directly because you know they built that personal relationship. Um and as operators scale, that that that becomes a little bit more difficult. So they kind of they have to reacquire that same guest from the OTAs, right? So getting that repeat guest come back directly is the small, the easiest part. And historically that hasn't kind of really borne out. Um the second one is um uh the billboard effects. If folks find you on OTAs, they know booking directly is cheaper, so they seek you out. That's been something that's going on and kind of it's gradually building out the awareness in the in the consumer base, although I would say kind of majority of travelers still don't know that there's this dynamic of direct bookings offering better value. Um there is then the third kind of bucket, which is acquiring new guests for the same first time at that point of actually discovery and convincing them to book directly with you. Historically, that's been very hard because you know you would have to do that through search, right? Essentially, the the starting point for most travel is uh is SIL search. Google is the biggest travel company there is, right? And so that there's a reason for that. And and that search has been dominated by big platforms, right? And and um it it's been difficult for operators to be able to present themselves on search in a way that they can carve out a niche that's specific enough to be able to actually uh compete head on with uh with uh with the online travel agencies and do that in a in a in a repeatable way. And so the aggregation as a value proposition that uh online travel agencies bring to the table helps them essentially win um win on the search uh competition because they you know if you're searching for something like I'm looking for an Airbnb in Mexico City, um regardless of how big the brand is, say you have operate 600 listings or a thousand listings, you know, Airbnb, verbal and booking would just have you know an order of magnitude, more uh options available to the guest. And there's not enough of a kind of an intent signal from that broad search that would enable you to determine that you have something to offer where it's kind of more competitive than the OTAs. And so the OTAs will always be able to kind of outbid you on the on that search and kind of win uh win that battle, right? Um that's changing as search becomes more specific, and then we can kind of dive into that.

SPEAKER_01

But so being the devil's advocate here, you just said if somebody searches for an Airbnb in Mexico City, you'll that's like you already alluded to the term, and I think I just want to reiterate short-term rental is not Airbnb, short-term rental uh is uh a category, Airbnb is a channel and a marketplace. Uh, you know, once again, you know, I have never seen an Airbnb executive uh clean a toilet. Uh it's still the property manager who does that or the host at the end of the day. So we need to be very outspoken about that because then it's already narrowed down. And a lot of our listeners are actually doing the hardy arts and providing good guest experience and everything else to the industry. And but where do they get it wrong at the end of the day when we talk about operators in terms of the direct uh uh booking strategy? And we will obviously talk a little bit more about the opportunities they have now because ultimately we can say that the opportunities to drive more direct bookings have definitely uh improved, uh let's say.

SPEAKER_02

So there's um I think I can talk about the different steps uh in the journey and where there's opportunities for operators to improve what they're doing today, right? So the first one is around how specific are you around what your value proposition is, who's your guest profile, and are you actually able to articulate that in a kind of specific enough way to be able to capture that guest rather than a broad kind of definition of a guest who's looking for accommodation in this destination, right? So being more specific in terms of your value proposition is an improvement opportunity. I think most operators that I see too are are kind of more on the generic side, almost um kind of uh repeating the OTA playbook. The the second one is on pricing, and this is one that's not as um kind of talked about. There is um you need to be a very careful about how you price your direct listings versus what you show on OTAs. How much of a difference are you demonstrating there and how are you articulating that difference, right? Um what is the price elasticity in that decision? Oh, should I book on the OTA versus book direct, right? Um and is that clearly defined and and shown to the guests when they're making a decision, so they make that decision in an informed way. The third one is that how do you build trust outside of OTAs, right? So they come to you, they found the guests, they have they're they like the accommodation, they see that the value offering, they see it's a better uh price point, but they feel a little bit more concerned around booking directly with you because there's no intermediary in between to mediate, right? How do you feel make them feel comfortable? So that's I think there's a lot of opportunity to build portable reputation that's attached to your brand rather than reputation that's built onto on OTAs and kind of uh secured there. Um and then the fourth one is what are the how are you kind of building relationships with guests that are coming to stay with you from whatever town? Are you able to kind of uh collect um their contact information and build a relationship with them uh that builds your builds your brand affinity, your brand recall, and kind of gets them to remember uh you as a brand, not just a listing as a brand the next time they're trying to travel to your destination, and importantly have enough of an incentive to consider booking directly with you rather than going back to the platform, which is the default behavior, right? Um, and so those are all areas that I think um have a lot of room to improvement in terms of um how even large operators today are performing um their direct booking strategies.

SPEAKER_01

Um Yeah, I'm glad you alluded to those, and you haven't even mentioned AI yet. And you know, one thing that I I it even it reminds me when I was running Interhome as one of the largest property management companies in the world, uh, how little we did to leverage our brand within destination and property. And, you know, we did surveys back way then with uh Focus Ride to ask our travelers if they remember the brand who they booked with, who they stayed with. And to my massive surprise at the time, and I think that hasn't changed a lot in 10 years or even more, it was less than 70% of the bookers actually uh knew which property manager they stayed with. They remembered maybe the channel, but you know, I said the channel didn't do your laundry and didn't put your bed sheets on and didn't clean your property, that was your marketplace. But the guy who actually delivered the show on the in the destination was a property manager, and and they couldn't remember that. So that's a it's a perfect uh uh analysis from your side, and and I could only support that very strongly. One thing you have not mentioned in your four points is obviously talking about AI. AI as a as a leveling force now, as a probably fifth point in your conscious not to go there, I assume, to say, okay, now we have something else at our availability. And interesting enough, you're still we're still talking about the same crap that we have talked about 20 years, excuse me. We talked about pictures, content, you know, what's relevant, being being focused on our persona. Yeah, we take everybody, whatever, our messaging, our positioning, pricing. I mean, that becomes standard now in the conversation, right? But the next is, I mean, well, how am I going to be found on the new way of people how to search, right? And my children do not even know what Google is anymore. They just use something else uh to do whatever they have to do. And interesting enough, I get just before we started our conversation today, I got an email uh saying, you know, I entered consulting in ChatGPT and your brand didn't come up. Sending a consulting, like a sales email to me. I'm receiving uh pipeline sales emails, probably 10 a day for people who want to help me to increase my consulting business. But let's talk about property management. And everybody else thinks the same, right? And how do I increase my visibility on these new channels? And one, of course, thesis is that. That AI changes the capability gap between large prep platforms and smaller operators. And that's a strong statement, right? I mean, that's massive because we always argued in the past that Expedia Expedia when and homeway and and booking when they were still independent, they're spending like 10 billion with Google. How would you compete to build a direct business? But that has now changed. So what exactly is changing right now in your viewpoint?

SPEAKER_02

Well, a lot. I think AI, uh, from our point of view has two tailwinds for direct distribution, right? I think one is that the pick guest behavior and how that's changing. And I will we will talk about that in detail. And the second one is that its ability to essentially upskill operators in terms of their distribution capability. So level up their execution of their go-to-market, right? So that uh um combination lock that I mentioned before, AI can actually provide a cohesive answer to that so that it makes it more effective to solve. And then the second thing is also uh much more affordable to kind of uh invest into direct distribution versus what it was possible before. So there's that element in terms of given if the distribution same stays exactly as it is today, by leveling up your capability, you can shift share. Right? So that's one. The second one, though, is about how guest behavior is changing. And so let's talk about that one because I think that has the disruptive potential in the next two to five years, right? So before we kind of get into the impact on travel, let's talk about actually the AI itself and how that's changed since ChatGPT launched in November 2022, right? So it started with a conversational interface that kind of captured all our imagination that we now have a different technology, right? But it wasn't super useful other than the conversations at that point, right? What it then what it's been doing in the past uh four years is adding capabilities in terms of retrieving information, live information from the web and public sources and also private sources, and an ability to reason based on that information, right? So this is now moving from a chatbot to an assistant. So an assistant can help you.

SPEAKER_01

So that is what has fundamentally changed, and excuse me to interrupt, but that's important for our listeners. Before this was entirely based on LLMs, which were basically just massive data dumps on text, but now it actually accesses the live content that is available on the internet at the same time as well. And I think that's where still people think it the wrong way because we talk about LLMs all the time. Um, but now it's actually it's it's curating also what is available on the net. I think that's very important to understand as well.

SPEAKER_02

Yes. So curating the information and then reasoning on top to be able to solve complicated problems more effectively, right? And so that's the assistant world now. And then what we're entering now is an agentec world where not only they're able to retrieve information and can reason about the problem, but they're able to execute and actually act on the problem and solve it, right? And we're seeing in you know very strong progress in some areas, like engineering. So, Simon, I don't know if you've tried vibe coding, but basically now, now you can have a software engineer essentially build up products for you just by having a in uh you know conversation with it, and it's able to you know use its kind of training material, live information, tools available to it to then essentially complete the task, right? So in some areas we're seeing very rapid progress about uh on this agentic front, and then there are other areas that that progress has been slower to make, right? And a lot of researchers in the field call this the jagged edge of intelligence, right? So essentially we're making rapid progress towards what they call AGI, but it's not equal on all fronts, right? And one of the hypotheses that's in the industry around how this progress is actually evolving is about how verifiable is a task. So imagine if you build a piece of software, it's relatively easy to verify that that piece of software is actually built to the specs that you want it to, because there would be you define test cases, and those cases, if they pass a test, that's fine. And so these this verification is how they build reinforcement learning on top of the platform. That is not available in some areas, and I would argue a big chunk of travel planning and choosing accommodation falls into that category where it's very difficult to verify um whether AI actually presented the right option or not. And so agentic uh kind of progress in travel, my uh my hypothesis is that it's gonna be slower uh than it is in some other fits, right? And so if we break down what that means in travel, um so today we have conversational uh assistance that actually are empowered by live inventory on the web, live content on the web that they go and search, bring back options to you, and you have this conversational assisted uh kind of uh travel planning and discovery that's replacing what used to be Google search. Right? Yeah. So that is where we are today. We are seeing more live feed of availability and rates being worked on, where companies like bookend.com or Xpedia are building apps to basically pass that data onto Chat GPT. Google has the obviously the wealth of data they have from Google Hotels and Flights, etc., that they're trying to kind of weave into Gemini. Um so that is the next kind of evolution, but it's also been uh slow to drive adoption because the user experience is not there yet, right? And so the conversational like uh the interface with the pure text and some images embedded by force, it just doesn't seem like the on-lock yet. Um, and so what we're seeing is essentially it's more in an experimentational phase, right? And and an agentic um travel planning, I think it's as I said, it's gonna be uh from my point of view, it's gonna be some time before we get to a point that we would feel comfortable delegating the entire task of planning a trip for us or choosing an accommodation for us, choosing a destination for us. We may get there earlier on some of the more commoditized decisions around choosing um, you know, if I'm going on a business trip and I want to book a flight, I might feel a little bit more comfortable actually allocate delegating that decision to an agent because it's a lot more verifiable, you know. Like, I I don't want a red eye, you know, this is kind of luggage, hand luggage, aisle seat.

SPEAKER_01

I mean, airlines are airlines, right?

SPEAKER_02

Right. And so, like uh, you know, there's not that and you know, and it's not, I think, as much of a subjective dynamic to it. There's not that much of a taste element to selecting your airline as it would be the the destination you want to choose for your travel or the accommodation that you want to stay in, or the activities you would want to do while you're there. Right. So the the higher the taste element, the diff more difficult to verify. Because even if we when we book our accommodation, we not necessarily know ourselves if we made the right decision until we actually go and see the thing. Right? And yeah.

SPEAKER_01

But let's talk about that. Uh I mean that's a great segue for me to think about you know, search versus conversation, right? I mean, we're you sort of still, you know, maybe my wrong understanding, but you're still sort of narrowing it down to traditional search. But I think with us now using these tools, you know, perplexity, claude, chat GPT, whatever, on a daily basis, we're becoming a far more conversation, right? So so we're come we're becoming more conversational with these technologies in terms of, hey, um, I have this problem to solve. I'm currently at work and I want to do this and that and the other. So I'm writing a text to get the best possible answer out of it. For travel, that's going to be the same in terms of search, right? I mean, in the way back when OTAs, destination, date, number of people, that's it. No filters, nothing at all. Obviously, we came a long way. Uh, but now I think the future of search, and this is my view myself, and I'd love to hear you your point of view now as distinction, is different because now I'm a family of five, I have certain ages, so I can write, hey, next summer we want to go to this property next to nature, not too far away from the beaches, close to restaurants. It needs to be cozy. We want, you know, uh not a modern interior, we want a more local style feel. That's our budget. And uh we just want to have a wonderful time, give us a result. I mean, you can't search like that on a booking.com or or or or an OTA. How is this going to transition? You know, we're moving from traditional search, as I just alluded to, to conversational interfaces. I mean, in the past, you know, I remember the conversation we had at Focus Ride where everybody said voice is gonna, you know, revolutionize uh travel search. It hasn't, right? Um, for all the reasons we know, it's still challenging, it's getting better and better now. But sort of AI has now pushed voice out entirely because you can just either still use a voice interface or or just type it. So potentially to AI assist or even agent-driven bookings, how is that going to change this conversational search from the traditional search?

SPEAKER_02

Yeah, absolutely. So you're absolutely right. So basically, we're now sharing a lot more context than it would we would do in, let's say in Google search, you would say, Hey, I'm looking for accommodation in City X, and then you would do your research, right? In uh in an AI interface, you're sharing a lot more context about who you are. And so, but then what happens behind the scenes? The the your AI assistant takes that query, breaks down into chunks. Okay, so here's the elements that this guest is actually looking for, and then goes up, conducts um search on your behalf. And these are long tail searches, right? So, you know, like looking for accommodation in CDX for uh a family um of with a like I don't know, two bedrooms, looking for accommodation for a family that has a hot tub, looking for accommodation for family in CDX that is close to this desk uh point of interest that could be actually kind of interesting for them, right? So it would take your context, break that up down into some specific parts, go out, and conduct effectively like an equivalent of a Google search, right? So it would query the web, find relevant options, and here I think there's a significant difference because the attention aperture is much wider than our us humans. So for human searching, you need to be on the first page. For AI searching, it's not as important. So you doesn't matter. And then it would basically match the context of uh what it found on your website or domain uh with the query that the user actually uh gave it and reprioritize uh those options and basically cite one or two sources as the most relevant ones and the most reliable, trustworthy sources, right? So in addition to kind of looking at your domain, it's also kind of having uh we see that it looks at third-party sources like you know, conversations on Reddit or uh content from YouTube, etc., to kind of inform that whole process, right? So effectively, it's almost like in the old days you would have a assistant that would do a very thorough research rather than just one Google search that you were doing, right? So, what does that do for uh you as an independent brand? It means that that broader aperture opens up an opportunity for you to be discovered by this AI assistant along with a big brand. And so if you have the relevant uh uh kind of offering structured and packaged and showcased in that way, you there is an opportunity for you to be essentially prioritized by these AI assistants and be shown to the guest at the same time that options from Airbnb and booking are being demonstrated, right?

SPEAKER_01

So my previous podcast is not too wrong, no more Google. So it basically stops uh spending any money on SEO and just manage your content as good as you can uh in order to to improve your uh visibility.

SPEAKER_02

So uh the the work that you've done on SEO is actually the foundation that would help you be discovered on AI as a lot of it.

SPEAKER_01

It's important, yeah.

SPEAKER_02

Right? So it's I would not say to throw out your SEO work. In fact, I would say probably double down on it, right? There is there is uh there's a lot of the the the the kind of similarities between how Google crawl the web and how um OpenAI, Claw, Gemini uh crawl the web, right? Because essentially the the the mechanism this is the same. It's just that there's one layer on top of it that does essentially ingest the content uh and then reprioritize it based on relevance to the guest, right?

SPEAKER_01

So that's an in a very interesting point. I was just trying to be the devil's advocate to say, hey, we don't need it anymore, but you put that to the point because the mechanisms are still the same, because the biggest source of data next to the LLMs is still coming directly from the web, and then obviously Google has the most the most amount of content. So I think everybody without having an engineering and other or an online marketing background can understand that that uh you know giving up on your SEO strategy would not be that smart just because you found on uh on other uh um AI uh channels is not going to work because they ultimately search the web the same way. And I find that very often when I do some deep research and what kind of articles they pull out first, where I think, hey, that's from my knowledge, not as relevant as I know stuff that is a lot older or you know, whatever, and they just give me the top line stuff when you do the do do deep research on all these AI tools. So let's move a step forward, and we already touched a little bit on Google, but now we talk about the OTAs. Very recently I had uh the CEO of one of the largest OTAs on the pod and uh on STR Global unlocked, and I asked him, Are you gonna be the only employee in five years? And uh he laughed at me and said, And he said, No, yeah. Yeah, he said, no, I'm definitely not gonna be. I was like, okay, I trust you on that one. So aggregation has been their core value, and you said that at the beginning of our conversation, where you said, hey, you know, this is what this is about, you know, you were with a goddamn in aggregation, unique content. This is this is what this game is about, to core value proposition for decades, right? But if AI starts to surface, um options and to surface the options that are surfacing are different, right? Uh, across both platforms and direct channels, does that weaken the advantage of this aggregation because you create more visibility across a broader spectrum of being found?

SPEAKER_02

I would argue yes, although there's still a lot to be determined, right? So let me let me kind of uh uh break that down. So we talked about you know the OTA business model being built on having unique guess and unique inventory, right? So what happens if the process of actually discovering the type of accommodation that you want to book moves from the OTA to the AI assistant, right? So that because of the knowledge that these assistants have about you, your context, your motivations, and your limitations, because of you know all the conversations that you've had about with them, right? Um they are able to then determine what your requirements are more effectively. And they would, as we discussed, they would be able to kind of lean on information available across the web to then curate um choices for you. And then you would have that conversation. And you can see that as the UX and the UI of these platforms evolve, um, there is an opportunity that that becomes a more compelling experience in terms of planning and selecting the accommodation, right? So as if as the that shift happens, what means is that that as guests or as travelers, we end up spending more time on the AI interface than we would on the OTA, right? And then so then the question becomes okay, what happens to the business model of the online travel agency if that side of the platform, not entirely, but a portion of it gets funneled off to exact an abstraction layer upstream, right? So that essentially puts their core value proposition at a little bit of a challenge, and there then there's a decision point to be made if you're kind of building an OTA, right? Um there's a there's one decision is to say, okay, well, it's in order to compete, we need to build a better interface for travel planning, right? So essentially compete with uh with AI assistants and builds an experience that you're you go to the OTA to plan your trip using AI, using their inventory, and potentially, you know, as uh Glenn and the team at Booken have been planning this for a long time, move towards a connected trip rather than a point solution, right? Offer the whole trip as a exactly right, and so that would be, I think my view is that it probably bits double down on that front and say, Okay, well, listen, AI assistants are gonna move towards offering a better discovery interface. We need to do the same, and we need to basically win that race, otherwise, you know, move users would move to the AI interface. And then what that means is that um inventory is gonna get fed from different sources and the the OTA uh offering becomes more of a commodized solution, right? Um so that's kind of one potential direction. OTAs would move um uh towards uh uh building the connected trip and the interface. And we're seeing all hinting towards their intention to move that move into that direction. So Airbnb is moving to kind of offering multiple travel kind of products so that that they have the basically ingredients in place. They acquired an AI company a couple of years ago. I've might I don't have an inside knowledge, but my guess would be that they're this would be one of the areas they would want to work on. They recently hired the senior AI executive CTO from Meta, right? That's right. So I think that they're making uh my guess is that they would move towards kind of that direction. Um there's an alternative option that some OTAs may want may choose is to essentially um uh it's a bit of an innovator's dilemma. So disrupt yourself before others disrupt you. So they would say, okay, well, listen, it's uh the big research houses are better placed to win this battle. So what we would what we were gonna do is essentially kind of be this inventory layer and pass on our inventory and and you know our margins might might shrink, but we'll still own the inventory, and that's that would be our play. Um, and then there could be emergence of pure play inventory offerings like Shopify for travel, which is kind of some thesis I have about this space. Uh and so yeah, sorry, go on.

SPEAKER_01

Yeah, I just you know, just to jump in here. I mean, first the big question is still merchant or record needs to be resolved. I mean, somebody needs to take the money and be responsible. Platforms won't do that. So, you know, uh, that's one thing. And and and and and and uh and and and Second, you know, the OTAs won't let this go without a battle either. Um, so that they'll define their strategies at the same time. But I I I I I love your viewpoint on on how this is now being impacted in in what they're ultimately delivering uh to the consumer uh as well. And the seamless point uh is definitely going to be uh a very interesting point. But let's become a little bit more relevant before we wrap this up because that's what my audience is all about, is what app operators should do now. I mean, it you gave a lot of tactical and strategic views in how the market is evolving and how you see things uh developing. But most operators that listening to STR Global Unlocked, uh, you know, they're not building AI products. They're trying to run their businesses. Uh so the practical question: if you're a property manager today, and let's say you're 100% Airbnb dependent, what should operators actually do?

SPEAKER_02

Absolutely. So listen, when you're having a conversation or trying to book your travel accommodations, there are really two decisions that you're making, right? Which accommodation should I stay in and where should I book that accommodation? Right. So let's talk about what should operators do to kind of be surfaced when the guest is actually making the first decision, right? So what you want to do is have a very granular definition of the guest profiles that you're serving. Okay, so you know, across the seasons and these seasons, we actually serve more business travelers, but now we move to families, and then actually there's a segment of kind of young couples that are coming and they they're doing this and that. So essentially being very art specific and kind of proactive about who our guest profiles are, and then what would they want to do in this destination and what type of accommodation we might be looking for, and then articulate that on their uh domains so that they increase the likelihood that they get picked up by these AI assistants in their context matching, right? So essentially much more granular definition and expression of their value proposition, clearly articulated on the website and structured in a way that's easy for AI assistants to read, right? So this is about being able to be discovered more, right? The second piece then is okay, say you're increasingly get discovered. How do you actually convince people to then book directly with you? So then you need to be very clear around the value proposition of booking directly, which is you need to build trust, so portable reviews, right? So build your profile, trust profile on Google Business Profile, and TrustPilot. We've seen AI assistants actually kind of lean a lot into these uh review sources as a way to understand that this is a reliable uh domain that I'm I can't. Source and reference. Yes, because they can't what they're what these assistants are probably artistic right now is the utility of the user, right? And so in that in that decision making, there's a combination of relevancy and trust, right? So relevancy is what I was mentioning before in terms of content. Trust is to be able to actually articulate in a reliable way that you're uh you know you're a trustworthy source. So having reviews on third-party platforms that are outside OTAs, Google Business Profiles, Trust Pilots, um, etc., is very important, being referenced in kind of um uh industry magazines uh or kind of tourism boards and that, and those sources are also kind of um strong indicators of trust. And actually positive conversations on Reddit and other user-generated content about your brand is something else that helps you build that trust that actually you know convinces the assistant to recommend you. Um and then there is about articulation of the price proposition. So being very clear and rip almost repetitive, um, that you know, booking directly is uh more affordable. There you will save X percent on platform fees by booking directly um across your uh experience and continuously pitching that to guests. So it could be the uh the price differences, could be perks that you offer direct bookers, um, you know, more flexible cancellation policy. These are the type of offerings that would then essentially package your direct booking value proposition um uh uh to uh to um the AI assistants and kind of basically nudge them to to move that forward. Now, one of the kind of dynamics that I think actually would uh could potentially help drive direct um in this new world is that the awareness of direct booking value proposition was a big challenge in driving direct bookings historically, right? Individuals, travelers, just a lot of them don't know that there's this dynamic going on, they're used to OTAs, right? And there was no big entity to build that awareness and kind of market it to the end user. Now what would be interesting to see is that if we have this all-knowing assistant that's helping us plan, is that knowledge already embedded in this decision making? Right? So it's based on all the knowledge that's I already aggregated from the web, are there actually signals to say AI knows that book indirect is cheaper and would factor that in in making recommendations to to um to its audience? And so that could be an interesting tail one.

SPEAKER_01

So let us let the cat out of the back. We know SCR Global Unlocked is not a marketing platform. You've been very patient on not positioning your own business, but we still would like to know it. Um, you know, you've been very kind to us and sharing our your your expertise and your views in a very neutral way, but I allow you uh to make a slight touch on on host AI and and I think it summarizes it, right? So, you know, you described host AI as something like a growth function in a box, and I think you just summarized all that for operators. So without going into the product detail, what is the core problem um that host AI, you as a platform are actually solving?

SPEAKER_02

Absolutely. So we know direct bookings has been a big problem, and we know why that's been a big problem, that combination lock. And so our belief is that solving that combination lock requires an end-to-end integrated solution, right? That helps you with acquisition, helps you with conversion, and helps you with retention. And so that's what we're building at Hosei. So a cohesive platform you can use to drive direct bookings.

SPEAKER_01

Wow, very modest. Once again, I had other people on my podcast who tried to take half of their time to uh describe their product. But this is super interesting. And I think it was a nice segue to get to this conversation, uh, Amiraali, to think about you know, what is host AI about? Because we just basically created that storyline for your thoughts and and the challenges that the industry has in terms of distribution and and everything else. So this has been uh extremely inspiring. So, in you know, in the amount of time, um, let's talk about the next five years. Let's zoom out for a second. And if we look five years ahead, and maybe we we can reduce it to one year or two years, uh, because the compounding rate of AI is obviously significantly higher to anything we have seen in technology ever before, as long as I've been alive and many other people before. What does the travel distribution landscape look like in the let's say distant future, whatever that distinct between one and two or three years?

SPEAKER_02

Yeah, absolutely. So I think you know, I think it would remain um a rather complex dynamic um because the decision itself is a complex one, right? But the change uh that I see come in is that um AI assistants is gonna own more of the share of the decision-making process. Um I would see real-time inventory being passed by OTAs as well as a direct equivalent to OTAs, say the Shopify for Travel, being passed on to the being passed on to AI assistants in shaping that experience. I would see these platforms actually taking part in shaping that experience to some sort of a um you know a generative interface at the time of conversation. I think that would be the onlock for AI assistants actually really owning travel planning. As you're conversing with them, the interface adapts to your conversation and becomes um much more pleasant. Um, and you know, the booking, Airbnb, you know, uh Experia and kind of the Shopify would travel, ton of equivalent, host the eyes of the world, would uh be present there and help shape that experience along the assist along uh with the assistants. And the user would be presented with choices of accommodations to book. And this decision would um, you know, there there will be times where the user would select to book directly because the the price difference makes more sense. And then there would be times that they would uh would want to lean on a third party because you know they offer better payment options or they offer um um you know a little bit of a trust layer that makes sense for that context and the nature of that trip. So I would see these two actually coexisting alongside each other. Um I think from an operator point of view, I would expect a healthier distribution kind of portfolio where uh third parties make up maybe 50% or um belongs that ballpark of your um revenue generation and the rest are your own channel. That's a kind of a healthier place for uh an operator to be. Um but I don't see you know OTAs going anywhere uh from uh from a business model point of view.

SPEAKER_01

There will be many operators out there scratching their heads to get to 50%, but I think you've given a lot of you've given a lot of thought and uh and shared your views, also together with your own technology. I mean, Raleigh, what I find particularly interesting about this conversation is that it's not just about technology, uh, it's about how power uh distribution and control may shift in the industry, which you just alluded to. And for operators, that's a fundamental question that they need to work with. So I want to thank you very much for joining the show today. This was Global uh STR Global Unlock. Thank you for joining the show. Thank you for the show. And uh good luck in Mexico and uh safe travels.

SPEAKER_02

I appreciate you. Thank you so much.