Highgate Technology Ventures on the Agentic Hotel Stack

It's a consolidated brain and it knows you as a customer, not only the property information, but between you, what you asked for, how you talked to them, what you got back, what you request taken care of, imagine the power of that information. That's the key. From Hotel Tech Report, it's Hotel Tech Insider, a show about the future of hotels and the technology that powers them. Today on the show we have Korean Jacob, also known as KJ, from Highgate Technology Ventures. I'd argue that KJ has the single best seat in all of Hotel Technology. KJ spent 15 years building Highgate's legendary revenue management practice, ultimately has their chief revenue officer before launching their technology investment arm about a decade ago. In this episode, KJ explains why every single category of hotel software is up for disruption, which vendors are sugar-coating what AI actually means for hoteliers, and where the next big winners will get built. KJ, thanks so much for coming on the show today. Thanks, Jonathan. Good to be with you. California boys. KJ, I have so many hotel companies that reach out to me and ask me, how should we be playing this? Do we hire a head of AI? Do we start building software in-house? Most of the time I tell them, you should do what Highgate is doing. You should have a separate technology investing arm so that these incentives don't get blurred. You have this great portfolio of assets to test products on and scale and get a good preferential valuation. Since I tell everybody that they should be more like Highgate, I want you to tell everybody what Highgate is and how Highgate Technology Ventures plays into the broader management company thesis. No wonder we're getting all these leads coming from all over the place. I think it's interesting and from Highgate's perspective, Highgate obviously started off as a realistic site, which is the Highgate capital or investment site of the business. Then we have Highgate Hotels, which is the operating center of the business. We started HTV, Highgate Technology Ventures, the technology investment site of the business almost 10 years back, actually pretty much a little over 10 years ago. I think the rational that time was, we will clearly think even then by 2016, we'd seen a number of companies come through pitch staff and then go on to become reasonably large in scale, even from the OTS. We thought the two ways to do is two beneficiaries. One is can we see very good companies that are solving something very important and can drive tremendous value for the adults at the same time have a team and the ability and the scale to actually become large and commercially successful and strong. Then we can achieve two things at once. We can get the best technologies for the hotels. At the same time, we can achieve great outcomes on the investment. Ultimately, the investment TCs and the investment outcome is actually important as a separate thing for HTV. It's not linked categorically to drive more technology to the Highgate Hotels, but we actually have cross relationships there that they send us over leads and we send them good stuff that we do and we almost, before investing in any company, we if it's a hotel space, we would definitely want Highgate also to diligence and live to see what they would use the products. Today, the team is in good shape and they would continue forever. So that was probably the port of HTV, but clearly, like you said, we are not investing off a balance sheet and we're not corporate VC for the sake of it. It is absolutely fundamentally a different set of business. HTV operates in its own. Not a fund, but a portfolio of investments overall. Do you raise in multiple funds or is it evergreen that's coming from the founders of highviator? How does the capital structure work at LP guys? So the beauty here with HTV, Highgate Tech Ventures is it's not a fund. So we're not incentivized to deploy capital at the management fees on the or on big hits that we've made for using other people's money. It's all part of capital money. So all of us have to invest the capital in every single deal. And it's all opportunistic, right? So you could do typically, we normally could do maybe two, maybe three for most in a year. So you do two exclusively or they could be cases in 2021, 22 where we did zero, which was actually pretty good. R01, maybe, and they'd have to be highly important. And our TCS is a partner led and not a fund. We have to be very included in ensuring that it is a success or at least it's not a failure too. And how is the landscape of investments evolved since 2016 when you guys started? What kind of deals were you looking at that and excited about and how is that investment profile shifted today? I think that's interesting. I think in 2016, when we started, first, there were a few people playing in hospitality and tech. And we've literally seen over the last 10 years, you've seen more and more VC, more and more private equity growth equity come into the space. I think the potential realization that travel is 10% of the world's GDP. But maybe at the same time, there were a lot more deals and scalable assets in other industries. For some more reason, I don't think it's an odd reason actually. It's because of the fragmentation of our industry. It's very difficult to see scalable assets. Unlike other industries where we healthcare or finance or insurance, it's more scalable assets. So that's where you see more of that venture and growth equity and P money flowing to. And because of the lack of scale assets, probably this was an as attractive for people. But we've seen in the last 10 years that changed dramatically. And I still remember how I've also seen the multiples change. I mean, when we were in 16, we were paying five, six times revenue for companies that were growing 50 to 100%. And then suddenly the game 10, 10, 12 times revenue before COVID and then right after the big phase there. And then now when you go to the eye, well, now those multiples are starting to adjust again. And then it's a re-adjustment of like, it's no longer the rule of 40 for for you to be getting good multiple. It's like the rule of 60, right? If you're performing. So we're seeing that change over a lot. But one thing I can say is I've dramatically seen a lot of good capital come into the space. Even if people might not be happy about it, but you can clearly see where some of the skilled companies that you see today outside of opera or I'm an article or Amadeus or save where any of these players you see a lot more new folks would not be five, 10 years ago, which is great. And as you think about where the opportunities are in the market, obviously multiples for traditional SaaS are compressing a little bit with like SaaSocalypse. What's your view on the state of the market today? On SaaSocalypse, everything that went down from that Citroen research article in January that we all love. How do you view the market on a go-forward basis? And how do you think about the market in terms of how you approach investments differently today than maybe had I asked you just a year ago? Yeah, I think we've all seen if you'd look at the market by itself, it swings way more like to the left or the right. You could see post 2020 market skyrocket and everything else than 2021-22 is disaster the other way around. And then it adjusted off that, but the market's right in a way, right? So the market's going to be right and saying based on what we see today, companies that are SaaS-based, that are just going to be selling software tools to help users be more efficient. And people say per se and they're focused on per se and per user and all that is fine. I mean, that's just a concept, right? That's the payment mechanism, but it's like the underlying thesis is software companies are software helping people be more efficient in their job. What AI has changed is saying, no, I want to take out the inefficiency, which is the humans. As crazy as it's phones, I'm going to say it. The AI based companies are not saying, I need to make you be better. They're saying, I think I can do what you're doing better, but a lot of our tech companies, every one of our tech companies, including probably some of ours, are sure, according to saying, oh no, no, no, no, no, you're in the driver's seat. We're helping you do better. We're just there to help you. Don't fall for that trap. That's clearly the first phase, but ultimately we all know the capabilities of AI and people say, it's not as good today, but it's going to be, that's true, but it's getting better every single day. Go back to when Chad GPD launched, what, 2023, November? It's so different from what you see from then, right? Where we've got agents doing stuff for us, we've got people writing codes of all that's changing. So I think if you look at what's happening is at that point of time, or at the point of time, I think that within the SaaS companies, there are going to be huge winners and losers. The winners are going to be those same SaaS companies that adapt and move into an AI based structure, where every everything they do is AI. It's not just the products that they use, but including their own operations and how they run it. And that's called AI needed. It's AI needed. It was just not a product. AI first is everything about what you do is done that way. And I think they are going to be successful because they're going to have low cost structures. They're going to have highly efficient products that are meant to reduce inefficiencies of human beings and perform those better. And certain task humans will be better. And in some cases, I can argue and say, yes, the humans will become the really good human, like the people can also elevate because they can become more strategic. The ones that want to do the same mundane work that somebody else is going to be inferior. So the same way in the SaaS companies, as the successful ones we've talked about, the failures are the ones that don't adapt, don't adopt that. And then the other thing is you see about AI first companies which are coming out of the woodworks. Like every day, they're like five of them coming out like mushrooms. And say, what are they doing? They're just like, I'll connect your data. I'll give you all the information. They're just doing GPT or cloud or open source market for wrapper and agency. They're trying to make a fool out of like hoteliers. Wow, this is so good. But what they don't have is they don't have the core foundational infrastructure of what some of these software companies have built. So in my opinion, if I classified into three categories, is AI native first companies who haven't built integrations enough, haven't built the pipeline of flow of information. Yes, I know it's easier to build integrations today, but there's still the complexity around people who won't give you that access easily. Yes, you can use robotic cross automation, you can do all of that stuff. And then there's the other structure of SAS companies, which have great in depth built this over the years. The ones that have that foundational stuff that actually build and move to an AI based approach first will win. And they will have that more. Are there any categories in particular that you think are easiest to disrupt? I've always kind of said, I feel and I know you may have some skin in the game on some of this, but I felt like the CRM space, it still feels like this fancy database with an email marketing wrapper like send grid on top of it. And it's just now, it doesn't feel like it's evolved as much as it could in AI. I feel like that's one category that's pretty right for disruption. Are there any other areas where you feel like the legacy vendors haven't adopted AI enough that you feel like there's this opening to have a wedge for an AI native startup? Yeah, yeah, no, I think for one is, I can start with the final with the other end, the financial side absolutely can be disrupt because AI is very good at math. And you've given all the content information, you'll see a lot of that coming in and whether it's going to be horizontal players coming into hospitality and driving that financial side of reporting. And the financial side, I mean, the reporting side is one part. I think that's totally disruptive. I think if you see the other side being a pure, yeah, you want to see RM absolutely. I mean, imagine sending monthly emails from a hotel anymore. It's no longer the case. AI can be personalized knowing when to send, who to send, what to send, how to send, like all that stuff is the kid. Like that for whole different. And all Sierra companies are doing the same shitty job the day of like, oh, we give you the option to send once a month. And I'm sure if you build AI morals around it to understand the business, then understand the guest profiles that you have, understand the guest experience, understand the scenario, it's going to do the planning, the content creation, the execution, the distribution, everything for you. I mean, absolutely, that's a disruption place for the Sierra perspective. I think we're seeing more and more of which I think is, do we need the CRS in the end, right? It was a question when we had a lot of the PMS is connecting with the channel managers. Do we need the channel managers in the edge? Why do we need each other managers? Why can you do each integrations or agentic right and read between one system and the other? Yes, people talk about MCP is one, not everybody would be on that, right? There'll be people who won't do that. Yeah, it's almost faster to have a direct integration from the PMS to the channels and then programmatically booking.com, updates of feature, just automatically do that versus rely on an intermediary to plan it into their roadmap. Correct, correct. RMS is purely by itself, absolutely disruptive. I mean, it shouldn't be. And they say that the CRS piece, what does CRS do? It does rate management. And then it pushes the series, pushing the rate as a channel manager. The rate management, if it's doing intelligently, is be can be done by an RMS which moves into both running pricing as well as driving the distribution mix together. Why do you need two different systems for that? So then you eliminate one and then you have another one which is just a pipe for sending information to different sources, right? So that's the, you talk about, I think the PMS is going to be disruptive because what does a PMS today? Why do we need to have a human being, press a key to say, I check you in Jordan, Melinda. Why? Why couldn't it be like, you've completed your formality, you've gone ahead, you've got an email or text message, hey, you're coming in tomorrow. Check it. This is you. We've got everything to fulfill, you can just clarify information, just a quick snapshot of your picture to confirm your identity. Yes, we have all your payment details. The file can be charged for you. If you want to just you can either charge for you or you can charge, we can charge when you check out, but you just want to verify something. Yeah, verify you take care of that. You go to the hotel, everything is taken care of. Why then should I go to a PMS and have a human being check it in? It's a trigger, right then. Does the check in for you? Agentic and it's done. In fact, it's technically the rumors that it must be done. So then you're starting to mix the whole digital experience with agentic operations from when a guest is booked to when a guest is left or checked out. I think the PMS is going into all these different categories as kind of an implicit admission that they think that everything's up for disruption. And so I think what they're trying to do is basically provide as many of those agentic workflows on top of the database. So we have they're so important to their customers that if the PMS, the core database functionality becomes monetized that they could say, hey, we increased your rent part by this through our revenue management system and we drove this. And you know where I think an agentic first company will warn you is run all the workflows and get a database and sell it for one fifth of the cost and then this industry is going to be pro. It's going to be good for hotels. So I think about it. I think that we're an AI first company would do okay. Where do you see the commercial model of hotel software going? I mean, there's the token side. There's percentage of transaction volume, which obviously was the legacy model. Then there's the SaaS model that kind of sits in between and it feels like we're pendulum swinging between those three models. Are there any other models that you think could work in the space? I think the best model is clearly the subscription base SaaS model structure, which is recurring revenue coming through everything. The difference is not that payment model. The key in SaaS is not it was you need to blame the model that says per user. So it's going to get stuck to this. Again, I go back to the same thing. If you choose to try to improve the efficiency of the worker versus improving the efficiency of what the job needs to be done or execute it, that's the key. So it's almost like AI based outcomes which SaaS pricing is, for me, is like as an investor, you know, you think about it and you say, hey, because I don't think people are ready no matter what they say. Can you share the value that is being created by AI between you and the software? I mean, you can pitch that. In my opinion, you're going to get some success. And I think nothing wrong with transactional pricing for a while. You're right. But the industry always liked safety and security knowing that you aren't getting this no matter what up or down. And so there was always a premium price to that. And just going back to payments, actually, payments is a valuable business. I don't know where it came up saying it wasn't. After you take the difference between what is processed to you versus what's your rent revenue, right? And then on top of that, yes, you have to obviously pay credit card providers, which is why they're talking about your 50% whatever. But the difference with payments is, once you connect the pipe, which is the same thing as like as more and more goes to the pipe, you're just going to keep taking your toll on it. So the concept of the total stuff is really, really good. Like who wouldn't take toll-based pricing anywhere else? It's like one of the most attractive markets. As far as companies that you're excited about at scale today, or you think are having really interesting strategies, one of the themes that I think is really interesting, still super early in terms of how it plays out is this idea of having a data set and having a software. And so two examples that come to mind are do I know buying hot stats is really interesting because now you have full PNL benchmarking data. And then actable with profit sword and Alice. And so it's like, oh, you could change your housekeeping schedule and now see how it impacts your housekeeping line in your cost-proc pipe room. Those kinds of solving of problem and then measuring it in the PNL kind of plays are really interesting. Are there any scaled companies that you've just seen a move or just like, yeah, I don't know if this is going to work. I don't know if you're going to be the one to execute it. Yeah. But there again, Jordan, I don't think you actually need to buy the company. So you could buy it with me to profit so that the housekeeping are a do it all and cost that. Imagine if you did one core thing well and then we were able to interact with the others around it, then the addressable market is far larger, right? So I'll give you a plastic case. I think two companies of ours won't really scale as lighthouse, which is obviously brought all this the breach of thing, the business intelligence, they did all that stuff for people. And now they're bringing it all together and also probably interacting with other potential providers out there. Not if we don't need to buy everything, but then they become this whole commercial platform AIB's commercial platform that is there that's available and obviously the launch of onus that they're there, which is actually saying, I can do all this for you to the users. So they're getting into that AIB, so that's scale in a different way. Another company not scale of ours, logic on the other types of RMS, but it's actually moving the other way and say, hey, we'll take all the data from whether it's from Amidea or from Estia, from marketing, from CRM, all the stuff that's there, sales, all of it, together connected all together and say, hey, we can solve the commercial structure with you because we can connect all the stuff easily and bring it to you. And then we have none, which is another classic example of financial side of it, to start doing the group level automation, move from group level automation to actually hiring the payments for groups, saying, hey, we're automating all this for you. These are classic work stuff done. And then moving from there to actually saying, let's look at the entire ledger, the financial side and saying, are you posting correctly? Are you collecting correctly? Are you doing all of that stuff, so making sure all of that stuff is done, including the point of handling AP and AR and AR, if we need to, that point of time. So it becomes the whole financial structure layer that's there. I'll give you a classic example today. There are companies that have mushroomed across the world, across the country, which says, we will reconcile how much you have got paid by OTAs or others and what you should have got paid, what you should have got and what you got. And if there's a difference, we'll go collect that from you. And then we will share in that. But do you know why that happens in the first place? The reason why it happens is because you were supposed to check in today and you don't check in today and you check in tomorrow. Now, the hotel for some reason, it's a prepaid reservation, you already need an Expedia or booking. They post it from tomorrow on Wednesday, they forget to night three. And it sits there and somebody does a reconciliation later. The key is posting that revenue then. And now all of that is possible here with AI. So you handling that in power structure here to make it more efficient while collecting the right revenue. So that's why I was talking about the first one, I was saying commercial, we feel that commercial agentic operations or commercial structure that's going to move in a way that can handle, we're going to see a lot more coming together, revenue, sales, marketing, all of that stuff. In some cases, it doesn't have to be together as one company, but it could be working together very, very well to get that in structure done. And then we have the financial infrastructure site, which is our reach, which is moving totally into saying, hey, everything in financial payments, we can handle it. And obviously, the news is doing its payments and that stuff. But this could handle it across any different platform. And it's just not limited to that single structure. It's not just payment, but it's the handle of financial infrastructure for you, which including agentic execution layers, right, that's there. And then I told you about, that's one side, I think for the marketing side, we're definitely going to see that 100% disrupted. I'm still surprised that we haven't seen any fast level launch of any major AI beast stuff on the market in front of us, yet to see something there, which could be really good. And then we have inspired actually a company that we looked at very clearly. It's owning the entire operational layer from once the booking to check out. And it hasn't breamed. It's where all the information of the property sits, all the information of the guests comes in, whether they text, whether they WhatsApp, whether they call by AI voice, whether they do all of this in just communication points of view. They all come in the single brain and then it guides the customer from the time of booking to check in, to instate and say, have a problem with towels, it sends it to task management system. I have an issue with my light bulb, the center of engineering. It controls that entire stuff for the guests, ready to check, here it is, link pay, you're out. Then you say, you know the guest has had a great experience, you post that review, you say, hey, post this review for me, and it's controlled that experience. So that's not an area, yeah, yeah, yeah, they moved. So they can now, it was based on a suite and we just got in earlier, so but they were in the TV business before, then they did apps. But now they got inspired one and people talk with the conscience, but the conscience is just the out of self AI conscience, but it's the brain that controls everything else today. And people are saying, oh, I'll do AI voice, then I'll do guest messaging. But then what's the point of two brains? One brain from guest messaging, AI voice, and that is the next sense you call it, you're going to get a different thing from AI voice and then you text it, you're going to get it from the different messaging. So this one says, it's a consolidated brain and it knows you as a customer, not only the property information, but going to you, what you asked for, how you talk to them, what you got back, what you request taken care of, imagine the power of that information. That's the key. So we're seeing all this stuff. So we're seeing literally across the space that there is movement happening across our companies. Hotel Trader again is another company. I think you know this. They're in the B2B distribution space. Now they're perfect for being optimized distribution and saying, hey, this is how you're going to drive optimized level of distribution to your thing at the highest profitable basis because of when you need it or when you don't need it the most, you know. So at a high level outside the portfolio companies, we are super interested across and we see dislocation happening all across. It's almost like a moving board that's happening and it's the race from everybody running together. We see a lot of AI first companies coming in, coming very quickly to the market, some raising lots of capital, some just raising a little bit and building for us. And then we're going to see movements happening of existing companies moving from one space to the other, not just through acquisitions, but because innovation. And then we're going to see some level of consolidation happening. So I think there's going to be a lot of consolidation happening in industry pretty quickly overall. And how do you think about the nature of that? That's more like consolidation, PMS starts building all these other features like we're talking about, or is it M&A or all of the above? I think PMS is definitely a couple of the PMS is actually going to build a lot of these features. The thing is they've spread two things because I think it's impossible to say, I can do all of the stuff on operations and get it packed out. I mean, PMS's today have not even solved room assignment, for God's sake. It's the number one thing you solve room assignment. You've taken care of a huge amount of time spent on figuring out which room to give to who. Well, that's the reason why check-in takes four long is because they're hunting for that bloody room. So literally, they don't know what you want to give you, but imagine all of that taking channels. But it's chasing the money, right? It's like, ah, easy money, I can just chase distribution and put a channel stuff and connect using that, that's easy stuff. So and I know, like as an investor, if I was the same thing, I'd say, where are we going to make the fastest buck the biggest one sooner than later? And each of those are too big to be ignored till, but I think you're going to start seeing something very good in different areas. You're going to have excellent operating platforms, you're going to have excellent distribution platforms, excellent marketing platforms. So I think you're going to see consolidation of all which will be fine in its ease of use. I have one platform. I can just choose it. So I think you're going to also see excellent choice of best in breeds. Not the way we had before of 30 different systems or 20 different systems, but probably four or five. So that's what I think is going to happen. And if it could also happen, I think the best of breeds through M&A could be something super interesting, which could create a massive winner on scale too. There was a lot of hardware excitement, I feel like in the last few years, especially with AI, some people say hardware is more defensible, I've seen motion sensing cameras in the ceilings of lobbies that have data platforms. I know you guys invested like a smoke detector company that got up charges when there's smoke. Are there any interesting hardware plays that you see in hotels right now and have any been particularly successful that you've seen? Yeah, actually there was one, because it kind of went with the name when I shop selling high tech was actually interesting. First of all, actually with this one, you're right, rest is a smoke sensing device. That's a simple thing. There is no way any housekeeper, security manager, anybody can detect smoking, happy hotel. Putting a 500 dollar plug out in a room type thing, you will be charged as much for smoking and not re-entering the amount, it doesn't make sense. This one is works for you 24/7, it's linked. their 24 hours a day, like literally there. And they used data science to figure out exactly what's happened and trigger an event. And data science is a holiday on any of these. So that's working really, really well. So automated. This company's actually pretty cool. I met the guy earlier and they were doing a lot of image and camera stuff that they were doing for certain governments and saying for security reasons, I think they said that they would turn all the cameras in a hotel, whether being the corridor, at the pool area, on the lobby, all of that stuff. They would take that and bring that data into a localized server because it's too expensive to take it and put it in the cloud. And then on that localized server, they would run AI queries or AI-based patent recognition. And they would figure out, like, if for example, a tree was left next to room 914, and the first time the housekeeper just passed us by, but this tree was left, the camera detects it, then you send a messaging back to somebody saying next to 914 tree left, it goes into the FMB system or pick the tree. That messaging is sent out and that gets picked up that it doesn't sit there all day long. Towel sitting where it shouldn't be sitting, picks it up, using AI message, the messaging sets. So it converts now real life stuff that's happening into a task that needs to be executed by somebody human, but it's all task managed through AI or patent recognition. That was actually interesting, very, very simple use case. And is that through their own proprietary cameras, do you remember? Or is it? No, no, no. It's basically taking use of an asset that's on premium and reutilizing or utilizing it, using which is actually brilliant to you. I was thinking about those. I remember, I mean, it's still a lot of hotels have these little buttons that you put on your room service, still get it taken away, but this kind of technology could do away with that category. Exactly. It replaces all those microchips that you have to put everywhere, all those chips that you have to put on the bloody tray. You don't need any of that stuff. And use your own camera stuff, too. And then it controls from security point of view, it controls everything. So yeah, thanks so much for coming on KJ. This conversation's been awesome and I know our audience has been alive. Thank you, Jordan. Loved it. It was always great to talk to you and hopefully see you in LA or New York. Exactly. That's all for today's episode. Thanks for listening to Hotel Tech Insider, produced by hoteltechreport.com. Our goal of this podcast is to show you how the best in the business are leveraging technology to grow their properties and outperform the concept by using innovative digital tools and strategies. I encourage all of our listeners to go try at least one of these strategies or tools that you learn from today's episode. Successful digital transformation is all about consistent small experiments over a long period of time. So don't wait until tomorrow to try something new. Do you know a hotelier who would be great to feature on this show or do you think that your story would bring a lot of value to our audience? Reach out to me directly on LinkedIn by searching for Jordan Hollander. For more episodes like this, follow Hotel Tech Insider on all major streaming platforms like Spotify and Apple Music.