The Modern Hotelier #131: #131: Reinventing Hotel Search Using Artificial Intelligence | with Harriet Brown
===
Steve Carran: Welcome to another episode of the modern hotelier. We're excited to release this episode with Harriet Brown from Mobi AI. David, what was one of your favorite takeaways from this?
David Millili: Well, I just think it's, it's amazing how more and more companies in hospitality are leveraging AI in ways that probably the listeners, you and I haven't thought about. So I thought it was really just intriguing to hear what they're doing at Mobi.
Steve Carran: Absolutely. And I think the the travel planning as well, where it's not only you're not only planning the hotel, uh, But the flights, activities around the area, really kind of creating that unique hyper local experience for, you know, any type of traveler out there. So excited to see this technology in the works and want to try it someday too.
So, uh, enjoy that episode and let us know what you think.
David Millili: Enjoy. 
Welcome to the Modern Hotelier Hospitality's Most Engaged Podcast. I'm David Millili
Steve Carran: I'm Steve Carran.
Jon Bumhoffer: And I'm Jon Bumhoffer.
David Millili: Steve, who do we have on the program today?
Steve Carran: Yeah, David. Today we have on Harriet Brown. Harriet's the VP of product at Mobi AI. Harriet has an impressive background in AI and machine learning for the travel industry, and Mobi is a collaborative AI tool transforming productivity, and decision making. Welcome to the show, Harriet.
Harriet Brown: Thanks so much for having me.
Steve Carran: Absolutely.
David Millili: Great. So we're going to go through a quick lightning round of questions. we're going to go through your background, your career, and then we're going to dive into some industry topics. Sound good?
Harriet Brown: Perfect.
David Millili: Okay, here we go. What is something that you wish you were better at?
Harriet Brown: Oh, so many things, but, um, drawing. I can't do more than boxes and stick figures. I wish I could.
David Millili: What's the most used app on your phone?
Harriet Brown: definitely Google Calendar. I always tell my husband if it's not on the calendar, it doesn't exist.
David Millili: That's good. What's a luxury you can't live without?
Harriet Brown: house cleaner. With two little boys, if we didn't have some help there, we would be in a lot of trouble.
David Millili: If you could trade places with anyone for a day, who would it be?
Harriet Brown: Toto Wolf, who is the team principal for Mercedes F1 team. such a cool and impressive guy, and to run an F1 team for a day would be pretty fun.
David Millili: What's the best piece of advice you've received?
Harriet Brown: I would say come from a place of yes. I think some of the people that I have the hardest time working with are people who come from a place of no.
David Millili: Got it. And last, what's your favorite city and why?
Harriet Brown: London. Uh, I was born in the UK, so we go over fairly regularly to meet family, but it's a city where I'm always uncovering and discovering new things, so you can never get bored when you're there. That
Steve Carran: wife says the same thing about the calendar as well. So, uh, I can totally relate to that. But, uh, so that was great. Now we're going to get into your background a little bit. What makes you tick? So like you said, you were born in London, but you grew up in Courtland Manor, New York. Is that correct?
Harriet Brown: is correct. Yes.
Steve Carran: How did that shape you into who you are today?
Harriet Brown: so Cortland Manor is a relatively small town right by West Point, right on the Hudson River. A beautiful, um, area. So being outdoors was a big part of my childhood. My parents love hiking and camping. and, you know, until you, uh, move away from it, you don't really notice how lucky you are. You know, we were a very much a middle class family, you know, went to a high school that, you know, was good, but not great.
and, you know, had a wonderful mix of friends with all different kinds of, of backgrounds, ethnicity, socioeconomic, um, it was a great place to grow up.
David Millili: So you got your degree in computer science at Carnegie Mellon. How did you end up at that school and that degree, that major? It's 
Harriet Brown: Well, it's funny, uh, I can give my mother all of the credit for me being a computer science major, when I was trying to figure out what I wanted to do with my life, I was like, well, I really like math, I really like science, and I don't like writing in English. And she's like, well, you know, com Computers are kind of a big thing.
You should do computer science. And I was like, okay, yeah, that sounds good. And so, you know, looking at what were the schools that, you know, are great for computer science, Curtie Mellon is certainly up there on the list. For some reason, as a teenager, I didn't want to move to California. I think looking backwards, I would say, why not?
Like, what a great place to be. But I had always been someone who wanted to be on the East Coast. so, you know, applied was very, very fortunate to get in. It's really an interesting school because you actually have to apply to the specific program that you want to enter into. So you don't just kind of show up undecided and then say, this is what I want to declare a year or two later.
So I actually had to get into the School of Computer Science, which is, no small feat because I only at the time accepted about 130 or 140 students per year. But it's a really interesting university because it's very well known for computer science and engineering, but equally so musical theater, business, music, and so it's got a really interesting student body, and Pittsburgh is an amazing city.
I had a great time there.
Steve Carran: That's, that's great. So, uh, you've worked in a variety of industries. What really makes you passionate about travel and solving problems with AI?
Harriet Brown: Yeah, so, when I think about things that I really like to do, as a child I loved jigsaw puzzles. Like, there was something really satisfying about taking like a whole bunch of disparate things and figuring out how they kind of pieced together. And so really kind of problem solving and seeing the forest for the trees was something that I enjoyed.
travel, I kind of fell into it. I mean. having been born in England, we did travel a lot as kids, flying over back to visit family and stuff. My parents love to travel. Um, so I certainly enjoyed travel, growing up. it all happened because I was in an AI class of all things. And our professor, Andrew Moore, had invited one of his PhD students to come and give a guest lecture.
and it just so happened that they had ended up at a company called ITA software, which builds, QPX, which is, one of the leading airfare, pricing and shopping, engines. And so he came and gave this lecture all about the AI algorithms that were used in QPX and I was like. Wow, this is really, really fascinating.
and so, you know, when the stars aligned and, you know, I had the opportunity to go and work at ITA, it kind of felt like something that I don't know was all meant to be in some kind of way.
Steve Carran: That's great. So now we're going to dive into a little bit more about Your career and how you got to Mobi. So, after college, you were a consultant for Monitor Group. Then, like you said, you, uh, moved to A T A, I T A Software, to be an account and support manager. What did you learn in those early years that you still take with you today?
Harriet Brown: so yeah, I started my career in, consulting and I would say like learning to work with clients was really important. I mean, certainly now most of my. product management career has been B to B. So you have customers that you need to work with. So I certainly learned that I also learned how to make slide decks, right?
I mean, consultants, if you've ever seen them on planes, they're like magical, putting putting decks together at like lightning speed. And so doing my time and consulting, I definitely took those skills forward with me. and yeah, then I T A once I moved there. I actually started as kind of a person who worked hand in hand with the airlines.
It was kind of a role, everything from an account manager to a bug triage, or to a project manager, to a product specialist. And so kind of I got to take those client management skills and then use them working with airlines. So, you know, the first six months I spent a lot of time researching ATPCO and how fares work in airlines and how to triage tax bugs and things like that.
Like you got into the nitty gritty of problems, but then I got to spend lots of time with people who worked at airlines and really understood kind of their side of the business, which was fascinating.
Steve Carran: Very cool.
David Millili: So ITA got acquired, by Google and you stayed on for 12 years. most recent was role product UX team focused on building innovative solutions, for airlines, travel, just your distributors and integrating that with Google cloud AI. Is that kind of like the second kind of coming of your, your love of AI and getting into that and getting more passionate about it.
Harriet Brown: Yeah, I mean, we were so excited when Google bought ITA, you know, I think the opportunity to build something like Google Flight Search, right, like, take all of the, um, credible search technology that we had with ITA and really scale it and make it an amazing consumer facing product, um, was such a joy.
From a career standpoint though, like it opened up the entire world of Google, right? Like think of all of the products that Google has with a billion plus users. Like, you know, I stayed, I did some really interesting projects with the travel team, but then I kind of traveled around the alphabet universe for a few years.
I went and worked on a product called Android for Education, which was all about getting Android tablets into elementary schools. I worked on Android OS as part of that, like, which was pretty interesting. Like, you want to talk about learning skills, like Android ships once a year, you were on that train or you were off that train, right?
And so like learning that discipline of releasing was really helpful. And then like all Google does, right, or not all Google does, but, um, you know, as Google can do. They shut things down, right? And so they had Chromebooks for education that were massively successful. They had Android for work, which was all about managing devices.
We were kind of a niche use case in between the two of those things. so they said, go find another job, right? Like, go find something else to do here. so I ended up working on the Google Play Store for a while. at the time they were building a new homepage layout and, um, I got to build the onboarding experience for that, like a little very UI driven thing that was seen by a billion people.
Like what a cool thing to say that you built, you know, you built and designed something that was seen by that many people around the world. and then I went out on maternity leave with my first son, came back and I was like, you know. I don't know if selling more things in Google Play is, like, really my calling in life, and so I went to go work for Verilate, which is Alphabet's life sciences company, right after it had spun out of Google X with the glucose sensing contact lens, if you remember that.
And that was my first taste of a startup, right? Like, because it had just spun out. It was kind of a strange startup because it was backed by Google, so the money didn't really matter, right? Like, you weren't worried about, like, going out of business. Like, you had, Alphabet's pockets to back you.
But that was really the how do you get from zero to one, did a bunch of interesting stuff there. but then the pandemic happened, things in healthcare obviously changed a lot as, uh, healthcare systems were trying to figure out how to treat, COVID. And so I ended up coming back to Google Travel. And so it was a really interesting, you know, I think people talk about careers as ladders, like mine at Google was a jungle gym.
it was really fun to kind of go up and down and sideways across different teams. but travel was where my passion always was. And so, uh, yes, in my last role, they hired me back because Google Cloud had just signed their blockbuster deal with Sabre. and so Sabre was, you know, went all in on Google Cloud to modernize all of their infrastructure, but alongside that, they wanted to use Google Cloud and particularly all of the AI tools that Google Cloud had, to write kind of really, I should say reinvent or reinvigorate their product roadmap, and so I was a big part of that experience helping them build their travel AI platform and stuff that they're using for, AI enabled or AI powered retailing.
it kind of all brought it together at the end.
Steve Carran: Very cool. Was there ever anything that you saw or maybe looking back and you're just like, that's so Google, like that would only happen at Google.
Harriet Brown: I mean, definitely, yes, I mean, what was interesting is so, Google had acquired ITA at a time when. Google Cloud didn't exist, like Google was primarily a B2C company, and there wasn't really a B2B component, so we were somewhat of a strange entity within that, and so, you know, it's really interesting to think about the order of scale of things, right?
So, like, we're like, hey, we've got this amazing shopping engine. We've, you know, got amazing customers, American Delta United, think of how we could grow it in this way and that way. And you were always comparing it, you know, if I said, I need five engineers to build this thing, you would always have to equate it to how much additional ad revenue could five engineers get to us, right?
The scale was so different and that was definitely an only at Google, type of thing. I will say the other thing too that was amazing, the day that the deal closed when they bought ITA, I think within a couple of hours they already had people in our building like ripping out the old Wi Fi and putting in the Google Wi Fi, which is amazing.
Like their infrastructure and ability to like, you know, handle those types of things is incredible.
Steve Carran: very cool. So about a year ago, you came to Mobi AI. What made you come, come over to Mobi?
Harriet Brown: In some ways it was the people. So, uh, Jim Russell, who had run the engineering team for Google Flights and Google Hotels, he came to Mobi about six months before I did. I've known Jim for 15 years. He was the head of engineering at ITA, have always really respected him, as a leader and a wonderful colleague.
He, his role at Mubi is our chief product and engineering officer, and, uh, he, wanted a product person. So he kind of reached out to me and, you know, it's one of those things I don't often take recruiter calls, but when Jim was like, Hey, can we grab coffee? I was like, yes, of course I will take that meeting.
Right. and what he really shared, I think was an incredible vision, which was not only for me and my career to kind of get in, to come in and lead a product management and design team and really. There's a lot of stuff that we can do to help shape the way our startup is going to grow and the types of products that we're going to build.
But also the vision that we could take this core AI platform that Mubi has been building for over a decade and point it at a whole bunch of use cases that on the surface may seem quite disparate. But when you look at the size and the shape of the problems across these industries, they're really quite similar.
and so I was just really, you know, I love the people there. I love the opportunity. And it was time for a change. You know, Google had, even in the 12 years, Google had changed a lot, right? When I joined the company, it was, what, 30, 000 full time employees. Now it's probably at about 200, 000. Like, it just operates very differently.
and so I was really excited to get back to a smaller company. and kind of grow it.
David Millili: So for those who might not be familiar with Mobi, can you explain really what Mobi is and what you do there?
Harriet Brown: Sure. So, Mobi is what we call a Human Collaborative AI Company, and you may ask yourself, what on earth does that mean? the way that we describe it, is actually kind of a double meaning. So, on the On the collaborative part, when we think about how you build something powered by AI, there are many different forms, branches, types of AI, and so because of the deep expertise that we have at our company, people with PhDs from MIT in AI, we actually really understand what is the right AI tool for the right part of the problem?
Or how do you stitch together multiple AI techniques in order to come up with a system that really works? So that's kind of one form of collaboration. The other form is the interaction between the user and AI, right? I think people think a lot about AI as automating things that humans can do, right?
Here's a set of manual tasks that I have, the AI will just take care of it for me. We think about AI of how can we help a human do the things that are difficult for them? And that often is things like complex planning and reasoning, right? If I presented you three choices and I said, well, this one is optimized for this and this one is optimized for that and this is optimized for this.
You can look at those three options and say, oh, okay, this is what's the most important to me. I'll choose that option. But a human trying to do all of that rationalizing and understanding and scenario playing is really, really difficult. And so that's kind of where we have pointed our products and our platform is kind of helping people collaborate with the AI to make decision making better and easier, as opposed to just automating tasks, that can be done.
You know, fairly easily.
Steve Carran: I 
love that. AI not replacing humans, but making their life easier and Letting them do a better job themselves. 
So, now, uh, we're going to move on to the industry thoughts. So, we'll learn a little bit more about Mobi and, uh, about how it's going to affect the hospitality industry. So, Mobi received the People's Choice Award at the FocusRite conference and also was named the runner up as the Travel Innovator of the Year.
What did that mean for you and can you tell us a little bit more about that process?
Harriet Brown: Sure. Yeah, it was, uh, an incredible honor to be up on stage at FocusRite. Um, I remember early in my career at ITA going to FocusRite and kind of sitting there and be like, wow, maybe one day I'll be up on stage. So to actually have that full circle moment, um, was, um, really delightful. I would say, you know, what's so interesting.
Thing about the, um, innovator competition at, at Focus, right? Is kind of like Shark Tank, right? So you have six minutes to go up there and make your pitch. And so it was actually a really nice forcing function for you to think about how can I tell my product story in a compact and compelling way? And so, you know, the team spent a bunch of time like trying to figure out our narrative and explain something that's actually really quite complicated in ways that the rest, that the audience and the judges could understand.
Um, so we were up there talking about intent driven search, which is kind of, how do we think about reinventing hotel search? you know, going from the fixed boxes of, I want to go to this destination, check in on this date and check out on that date. How can you just ask for what it is that you're looking for and us help you find answers that way in a much less constrained manner?
David Millili: So what are the primary challenges facing the travel industry today and how are AI driven solutions like Mobi helping address those issues?
Harriet Brown: Yeah, I think that there's a lot of um, a lot of issues, a lot of opportunities, right? you know, one of, I think that there's a couple of, you know, call them holy grails, white whales, you know, whatever term you prefer. Ever since I've been in the industry, people have talked about personalization and travel, right?
Like, what does it mean to deliver personalization? And so much of the narrative that I have heard is trying to predict what it is that you might want. And I think in travel, that's really, really difficult to do, not only because, you
I think the average person flies on an airplane like just over one time per year, right? And so you don't actually have a lot of data points to be able to do prediction really well. but also who you are on any given trip may really change, right? Like me as a traveler is very different when I'm traveling for business than when I'm with my kids than when I'm with my partner.
And so trying to do any of that prediction. It's very, very challenging. And so, you know, where I think we have the opportunity now is actually to change the way that we think about delivering personalization. Instead of trying to be predictive, just let the person ask for what it is that they want. Let them ask a personal question and give them a highly relevant personalized answer to that, as opposed to having them log in and show up and trying to guess what it is that they're want, that they want.
So I think that like personalization is certainly one of them. you know, I think that one of the other things that has, has changed, like if you think about the arc over time, right, we used to call up travel agents because we couldn't book it ourselves, right? You couldn't, it was very difficult to buy a plane ticket.
I remember vividly my dad being on the phone, you know, with an airline trying to book our tickets to Europe for our next, you know, You know, summer vacation. And then dot com happened, right? The, you know, all of the airlines dot coms, the aggregators dot coms. Cool. Now I can all do it myself. But now I think that there's so much content out there, right?
All of that is there for you. And now you have the problem of how am I going to sift through it all, right? Like there's so many things and we all. You know, we don't want to have FOMO, right? Like we don't want to book something and be like, Oh, but there was just a much better option around the corner. And so now we have the challenge of how do we take all of this information and make it easy to digest and make decisions with.
And that's where I think that there's a tremendous amount of opportunity for AI.
David Millili: That's great.
Steve Carran: I love that. And your answer right there kind of leads into this next question as well. So how does Mobi's platform personalized travel experience to boost customer and not only customer engagement, but also loyalty as well?
Harriet Brown: Yeah. So, we have a couple of different ways that we think about it. Um, one, as I said, is, answering personal questions with really relevant answers. but doing it in a way that doesn't feel like a black box, but it feels like explainable, right? So if I say, you know, I want to go somewhere that is a nonstop flight from Boston, somewhere that's warm in January with a resort, with a kids club and a spa and somewhere that, you know, I can go ziplining with my kids, for example, not only can I find that thing in the world, but I can also show you.
the context behind that, right? Like, here is a flight that you could take. Here is the resort. Here is a picture of the spa. Here is the kids club and maybe some reviews where parents said how much fun their kids had. And here are some exact, you know, zip line experiences that you could go on. so we really think about, you know, how do we make sure that we have accurate answers that have all of the kind of proof that sits behind it so that you can trust in the answer.
We also think a lot about how do you create emotional loyalty. I think in the industry there's been a lot of focus on transactional loyalty, like I'm sure you have frequent, you know, you have loyalty accounts with all of the major brands, all of the major airlines. If you're really into it, you're getting all of the credit cards and opening them and closing them to get all of your bonus points and stuff like that, right?
Like the, the travel industry. has really figured out transactional loyalty. But how do you create emotional loyalty, right? Like how do you say, well, I'm always going to fly United because I love United or why am I always going to stay at Marriott because I always love Marriott, right? Like we have. these preferences that take away, the cost factor.
I mean, maybe not completely, but if all things are equal, I would rather stay with, choice A or choice B. And travel is so amazing because I think it really gives you opportunities to create. emotional loyalty, right? Um, you know, whether it's the person at the front desk who, you know, helped you out in a tricky situation, right?
Like, I'll never forget showing up in San Francisco one time and I booked the hotel for the wrong week, right? And there it was, it was 11 o'clock at night, right? I'd just flown in from, from Boston to San Francisco and I was like, what am I going to have to do now? They're like, don't worry, we'll just take care of it.
We've got a room, no big deal. And like, I love that hotel, right? Like, they helped me out in a point that was, like, could have been really challenging. They could have told me to take a hike or fork over an additional thousand dollars or whatever it was, but they didn't. They just took care of it in that moment.
Or similarly, You know, there was a time, where I was, uh, down in Washington, D. C., a snowstorm was rolling in, I had to get home, and, uh, someone at United was able to get me on, like, the last flight going out that night, you know, so I could get home, like, you, you, you become emotionally loyal when, people can kind of, Take what your needs and your constraints are and help find a way to say yes.
and I think those, that's really the types of things that we're trying to create at Mobi is opportunities to deliver emotional loyalty.
David Millili: That's great. So in your view, what does the future hold for travel technology and what role does AI play in that?
Harriet Brown: I think that AI is going to be weaved into so much of what we end up doing. I think we're still in the very early days. I mean, we talk to partners all the time. I think we're kind of transitioning from, I'm working on my AI strategy and I have lots of different use cases, to I am starting to explore with some of those use cases and if I can come up with something.
I think that. there will be opportunities to, um, make it, uh, really seamless, right? And actually, you know, it'll be there, but like, you don't think of as an example, like, all of the, like, server, uh, technology and stuff like that, that's sitting underneath all of the digital things that you use every day.
Like, you don't think of that. And I think, AI will kind of become, um, That it will start to do things that are helpful, to, uh, running businesses within travel. and, uh, you know, it'll just kind of grow upon itself. I think it's going to be an evolution. Like in all of my experience, like things in the travel industry don't change overnight, right?
Like I think it's going to be a long process and people are going to figure it out. Travel fundamentally is about getting people from place A to place B and, you know, having worked with the airlines for a long amount of time, like, their job is to fly airplanes and make sure that they're safe, right?
Like, that's the fundamental role of an airline. Um, and similarly, when you're in a hotel, right, their job is to, you know, make sure that you have a warm, safe place for you to sleep at night, right? And that should always be their focus. And I think AI will just help them do that better, faster, and easier.
Steve Carran: that's great. Do you have any advice for companies that might be a little lenient or hesitant to incorporate AI into maybe their operations or even that customer facing side as well?
Harriet Brown: Yes, um, when I was at Google Cloud and, you know, I, I was there, you know, when Vertex AI and Gemini, you know, chat GPT were all popping up and, you know, the entire focus of, of Google Cloud, like all ran and swarmed over to like, okay, what are we doing in AI? In my experience, I think that the biggest kind of challenge to avoid is analysis paralysis.
I think people are really hesitant and they're worried about, am I going to get it just right? As opposed to kind of try something in a safe, contained way, see what works, see what doesn't work, learn from it and iterate. And so, you know, my advice has always been like, just get going, right? Like you may not get it perfect, right?
That's totally okay. You know, do it in a way that is safe, right? You know, contained, that you can fail and it not be catastrophic. Absolutely. But you're going to learn so much more by doing than trying to think and plan and make it perfect. the first time around, and that's everything I learned at Google.
Like it was experiment, experiment, experiment, right? Like Google is running millions of experiments at any one time and they're making data driven decisions as opposed to kind of theoretical decisions, right, of what you think might happen. And I think that that's super important with AI is try it and see what happens and learn from it and, you know, iterate and keep going.
Steve Carran: That's great. That's great. And just some brownie points for me. I use Gemini every day. So it's, it's, it's great. so Harriet, can you tell me, um, one thing that is a hot topic around AI's language models? Can you talk about some of the challenges with models?
Harriet Brown: So, as we have been investing in kind of creating a better way to search, I think a lot of, people have started with either a chatbot or a search experience that's powered with a language model. language models, and I think everybody got really excited the first time that they used chat GPT or Gemini or, or whatnot.
is, those language models are, really good at doing certain things, right? So, uh, as an example, as a parent, if, you know, my kid's teacher comes to me and says, Hey, could you write a poem about your kid? Like, I am not a writer. I would struggle to do that. But if I could, Ask Chat GPT or Gemini and say, Hey, write a limerick about a kid who loves ice hockey and soccer and is really kind of funny and goofy, it would spit out something really wonderful very quickly.
I would be thrilled. My kid would be thrilled. But if you're asking a language model to do search, it's not really equipped to do that, right? It's a predictive model. It's trying to take words and predict what is the next best one to put into the sentence. And that's not really the job of what search is, right?
Like, search is trying to find things that meet specific constraints. It has to be factual. It has to be accurate. And it needs a whole bunch of live data sources to be able to do that really well. So I think as people have been trying to use LLMs for search. They kind of first said, well, let me just try the LLM.
And I'll never forget, I was working with somebody at Google, who said, Hey, I just asked to find me flights from, Houston to Dallas and look, look at the answers that it gave me. And I was like, well, the flight time. From Houston to Dallas is not 30 minutes, like there is no scheduled flight that is 30 minutes, right?
Like it just hallucinated and made this up. and so what you're seeing is, is, but it sounded really convincing, right? It had like carriers and flight numbers and times and like only when you scratch the surface. So you're like, wait a second, this isn't really true. So to compensate for that, everyone's now trying to kind of control the creative machine, right?
They're trying to augment it. They're trying to have it make call out and make, API calls and stuff into systems that can do that, which then becomes a really complicated problem of building that system. and so what we have done at Mobi is really two things. One is we use language models to do the thing that they're good at, which is, understanding what the person is asking for, like really good at understanding that.
But then we use more structured search with planning, reasoning, and optimization AI algorithms behind that to do the actual search. and then the other thing is, is what is the data that you're bringing to an LLM? And Mobi has really invested. So, um, I'm interested heavily in building a data store that has 40 million points of interest around the world that are touristically interesting, but enriching those things, taking lots of disparate data sources, bringing them together to be able to kind of create a fingerprint of that space.
Bye. Structured in a way that you can use it for really fast search. And so, you know, when I look at what people are trying to do with LLMs, you know, a lot of the answer is like, well, bring more data, data to the picture, right? The LLMs are trained on the internet. You need to bring in the things about what they're doing.
you know, your hotels, your world or whatever is part of that. But the things that make it interesting are actually data sets that a hotel might not have, right? So as example, hotels know a lot about the attributes of their own property, but they don't often have structured data about all of the things that are outside of the hotel.
And that's what you really care about, right? You care if there's a coffee street or a coffee shop down the street. You care if there's a steak restaurant. I care if there's a playground, right? Like that's what I really want to know. And that kind of data store is a huge part of what Boots does. Um, And plugged it all together, kind of on our customers behalf.
Steve Carran: Which is so smart, because hyperlocal experiences are kind of everything in travel right now, right? So,
Harriet Brown: Yeah, I mean experiential travel, right? Like people travel to do things, right? And if you could help people find their hotel or whatever in the context of all of the things that they want to do, again, that's how you build the emotional loyalty, right? You don't necessarily remember the pillow that you slept on, but you certainly remember like the look on your kid's face when they went on the merry go round.
right next to the hotel is an example.
Steve Carran: totally. Absolutely. So Harriet, we've been asking you the questions this whole time. This is where we turn the tables and we let you ask David and I a question.
Harriet Brown: Oh, that's, uh, that's fun. Let me think. what are the topics that, you think are most controversial to talk about on your podcast? Right? Like, what are, what are some of the, the things that, um, you think where there's a lot of like, spark and, and debate?
Steve Carran: I mean, I, I think we, David and I, we're very pro technology and, Sometimes in this industry, people aren't that pro technology. when I was in sales, I was on the PMS side and Mobile check in was a big thing. And a lot of the pushback I would get and, you know, we still get sometimes on this podcast is people love standing in line.
People love waiting for your room. They love that, you know, that experience at the front desk. And I'm thinking, I don't, I want to, especially if I'm traveling for, you said, like business or leisure, if I'm traveling for business, I want to get in, go to my room and either get ready for my meeting or. Get to sleep.
So I think, um, technology is something that is always kind of a little tricky to discuss because we're not that fast moving industry that can change and adopt technology real quick. So, me and David being pretty pro technology. Sometimes it's, uh, we have to walk a little delicately.
David Millili: Yeah. My, my feedback would be that hospitality and travel, it's very funny how this it's an industry that assumes they're doing a great job of taking care of their guests. And then they're scared that technology is going to take away the human element and remove certain things that like the Steve Swift, that's not needed.
And I was using the example, you know, I was on a trip, went to a hotel, they had a self service kiosk. That was great. I didn't talk to anybody. I got my key, got in my room and then I got to the bar and I had a great conversation with the staff and the technology totally improved. So Steve and I are aligned on this.
And I think, you know, I was just always here. It's funny with like, Oh, they're taking away the human element. It's like. Dude, how many times I have to go up to the front desk at nine o'clock with my bags and, and the person says, are you checking in? And you're like, wow, that was really a great interaction.
Very, very observant point by the front desk agent. So that's, I think technology is, is the biggest challenge or most controversial right now.
Harriet Brown: Hmm.
Steve Carran: Absolutely. Well, that was great. So our producer, John has been listening the whole time. We're going to kick it over to him for the last question, and then we're going to get you out of here.
Harriet Brown: All right.
Jon Bumhoffer: I'm curious when you were, um, kind of creating Mobi and even your work with clients, like what are some of your favorite, like aha moments that came about, unleashing this technology?
Harriet Brown: Yeah. one thing that was, so. Interesting and joyous about coming to Mobi is I feel like oftentimes at startup, there's a lot of really good ideas. There's a lot of good kind of business cases that have been built. And so there's a good pitch, but the actual technology that sits under it is, you know, it.
Vaporware maybe, right? Like where it doesn't really exist. What has been so fun about being at Mobi is it's the exact opposite in a way. That the tech is there, right? Like they've had incredible engineers and people building this platform. You know, I like to think about it as a toolbox with all of these different tools.
And so what's been so fun as a product person was coming in and saying, like, what can I build with all of these tools? So intent driven search was actually a perfect example of that. We've been working on a travel agent platform, so tools for travel agents to help find trips, generate itineraries, and things like that.
And when I looked at it, I was like, Actually, there's another product in here that we haven't thought about, and it's a search and discovery product all about helping people find hotels better, right? Like if all we did was take a natural language interface and stick it on top of the APIs that we already have, we could transform Hotel search.
and you know, great idea. What was magical was like, we built our first demo in two or three weeks and it actually worked. And that was because the tech was actually there. The API is already existed. The UI needed some work. And, you know, the, the kind of sketching of what that experience might look like required some effort, but the actual building of it, was fairly straightforward, and I think that's very, very unique.
David Millili: Well, that does it for another episode of The Modern Hotelier. Harriet, let people know how they can get in touch with you, find out more about Mobi.
Harriet Brown: Yeah, sure. So please, uh, check us out, um, our website is Mobi. ai. You can find us on LinkedIn. Our communications team does an amazing job of posting all kinds of interesting content there. Um, and you can reach out to me at Harriet at Mobi. ai.
David Millili: So that does it for another episode of The Modern Hotelier, Hospitality's most engaged podcast. Whether you're watching or listening, we appreciate you and hope to see you again soon. Thanks for joining us.