Booking Holdings Inc. ($BKNG)

I just got the thumbs up that we are good to go. So good morning, everyone. My name is Trevor Young. I am one of the Internet analysts here at Barclays, and I'm very pleased to be hosting the Booking Holdings team, Ewout Steenbergen, CFO. First time here at the conference. Thanks for joining us.

In this capacity.

In this capacity.

In previous capacities, I've been here before.

You've been here before. Very fair. So I'll start with a very easy question. As someone in the travel space, any upcoming travel plans that you're excited about besides joining this conference.

Yes, this is actually a happy travel month. For me, I had to go quickly last week to Europe, came back on Friday, last Sunday night here. I'm going tonight to Bangkok. So I spent a few days with Agoda, I have to make a stop on the way back in Paris for a Board meeting. And that's all business related. And then after that, this is actually the really fun part of the travel. Then my wife, my daughter and I, we have a trip planned for visiting a program with UNICEF. So we're a very big UNICEF supporters. And one of the programs that we have supported is in Tajikistan. So we might go to Tajikistan. It depends a little bit on the situation in the Middle East, if that's going to happen or not. Otherwise, we might end up somewhere in Florida, I think. But I hope I can go to Tajikistan at the end of May.

So all over. Excellent. Good stuff. Well, actually, that's probably a good segue to our first topic, which is on the Middle East situation. That seems to be most topical in terms of near-term impact on travel demand. You spoke to it a little bit last week in terms of quantifying the impact, but perhaps you could expand a little bit on the broader implications for travel demand. How does the conflict in Iran disrupt the EU to APAC corridor? Is it causing changes in destination choice, booking windows or flexibility preferences? And is there any kind of substitution benefits that are offsetting that?

Yes. So the Middle East is a very important element in travel globally. I think you all here in Europe realized that a lot. We had to explain it a lot in the U.S. because I don't think it was fully understand in the U.S. How much people travel from Europe to Asia through the Middle East, these very large carriers like Etihad or Emirates or Qatar Airways or the other way around, of course, people that live in Asia or Australia, how much they travel through the Middle East, for example, if they want to come to London. So very important travel corridor. About 4% of our room nights in 2025 were from Middle Eastern bookers. About 3% was Middle East as a destination. So it's about 7% if you combine that. By the way, Middle East includes Egypt and Turkey. So obviously, it's not the whole of the Middle East that is impacted by this conflict. But then also, you have the impact of the travel corridor. So all of that has been reflected in our results. Obviously, first quarter numbers were still quite strong, but that was because the impact was mostly in the month of March. So for the rest of the quarter was quite strong, but more reflected for the full second quarter guidance that we have provided. You asked specifically about some other indicators around the booking window or other elements. So what we saw is booking window definitely started to come in due to the situation in the Middle East in March. So that was one of the impacts. In our guidance, we have also assumed that ADRs will come down a bit. So some of those we have reflected in the outlook. But most importantly, I think what we are doing and how we look at the situation is this is ultimately going to be temporary. We are, as a business, because we are so global, because we bring travelers to and from 220 countries in the world. We do 1.25 billion room nights globally. If something happens across the world, if it is a geopolitical event, if it is a natural disaster, if it is a pandemic or anything, obviously, that impacts our business. But in the end, these situations are temporary. And what we've seen with situations in the past is usually travel comes out of this in a much stronger way. People like to travel. They like to have great experiences in their lives. They like to go to other places and explore the world, looking to other cultures and really spend their time there. So yes, the situation is temporary. But more importantly, we're looking at the structural elements in terms of performance of the company underneath. And actually, those looked really strong in the first quarter, and that's really encouraging, of course, for the medium- and long-term outlook of the company.

That makes a lot of sense. Just digging deeper on the actual impact in 1Q, you called out about a 2-point headwind on 1Q room nights and bookings, even though that was a full quarter impact. For 2Q, you actually guided to about a 3-point impact, which some might actually argue isn't all that bad considering because you're now contemplating a full quarter impact here versus just 1 month. So has some of that initial uncertainty or uptick in cancellations started to wane, which is why that 2Q impact isn't greater? Is that kind of what you're speaking to where folks are like, hey, we want to still travel, initial panic has subsided?

Yes. Two thoughts on that. The first is definitely in March, we saw an uptick in cancellations. So this is basically a catch-up past bookings that were made by travelers that are planning to travel to the Middle East or people in the Middle East are traveling to go outside. And suddenly, all those past bookings then get canceled because it's obvious that, that travel is not going to happen. So that's a bit of a spike and then it comes down. So we said that is March stand-alone growth was 1%. That impact was about 3 points. So if you normalize for that, it's about 4 points -- 4 percent point growth in March, which is in line with the top end of our guidance for the second quarter. So I think that is definitely one element. But more importantly, we're also looking at travel growth in other parts of the world that is not impacted by the Middle East. So if we look at Europeans traveling within Europe, it's still up high single digits in the first quarter, and that is basically in line with the growth we have seen at the end of 2025. If you look at Asians traveling within Asia, it was up low double-digit level, the same as what we have seen in the fourth quarter of 2025. We saw that growth in the U.S. was really strong. We saw that growth in Latin America was really strong. So again, the regions and parts of the world that are not being impacted by this, all look very healthy. And that's why I think when we look at the full year guidance, actually, I think it's still very robust guidance that we have provided for the whole year, particularly taking into account the assumption that this conflict will last for 4 months. Of course, anyone can put another assumption in. We can't predict better than anyone how long this is going to last. If people think it will end shorter, you can make your own adjustment or will take longer. But under the assumption that 1/3 of the year is being impacted by this conflict in the Middle East, we think that the guidance is actually still very solid. And that's driven by all those underlying positive trends in other regions, plus I think a lot of the structural elements in terms of the execution of our strategic initiatives.

Right. And let's actually hit on one of those other regions that you alluded to. The U.S. market actually was a bright spot for you in 1Q. I think you accelerated for 4 consecutive quarters on room nights there. You're now kind of in teens type territory. Some data would suggest the U.S. market is faring much better than other regions. How much of your strength in the U.S. is coming from kind of your own efforts to ramp versus just the broader demand backdrop being healthier than other places in the world?

So let me maybe first, Trevor, explain what we see in general in the U.S. market. And then I can explain then on top of it, what we have seen more, how we have performed relative to the market. So U.S. market by itself definitely continued positive signals. You recall that a number of quarters ago, we were one of the first to start to highlight the bifurcation in the U.S. consumer economy. We saw the top end doing very well. People that go and travel internationally, stay in 4- or 5-star hotels, sit in the front of the plane, all doing very well. But the lower end was really under pressure. We saw less travel, shorter booking windows, ADRs down, length of stay down. So a lot of more of those indicators. Fortunately, it looks like the lower end of the market is improving. So we saw ADRs now flat. That is definitely better because we saw down for a long period of time. You see the booking window now stable. So all of that is more encouraging and positive. So if we have the top end continuing to do well and the lower end improving, that's, I think, a very good overall setting for the U.S. travel market. Now how did we perform relative to the market? I think our third-party data sources are telling us U.S. market grew low single digits in the first quarter. So we grew low teens. So this is probably the widest gap we have seen in a very long period of time in terms of our relative outperformance to the U.S. market. That's a result of a number of reasons. The first is we see very strong domestic travel. The second is what is really encouraging, we see very strong direct booking growth. And direct booking growth, you can't change overnight. That's the result of many years of investments in product, in supply, in performance marketing, in brands. Definitely, I think the familiarity with our brand in the U.S. is much stronger than a few years ago. And then you get the effect of someone comes to you the first time through a paid channel. They have a great experience. They like it. They come the second time to you through a paid channel. They like it again. And then maybe the third time, they say, you know what, I opened up the app, I book just directly in the app and then they become a direct booker and then you get much more repeat to yourself. So that's an effect we're seeing in the U.S. So we think that's really positive. We also saw outbound doing well. We saw also our B2B business, which is smaller, but doing well. So it's a combination of factors in the U.S. where we see that growth so much above the market in general.

So strong backdrop, that lower-end consumer maybe firming up a little bit, but you're really seeing compounding benefits from all the work you've been doing in recent years, in particular, driving that direct mix.

Absolutely.

Got it. Okay. Shifting gears a little bit. We have to hit on AI. As agentic AI shifts more trip inspiration and planning into some of the horizontal platforms illustratively like a ChatGPT, what are the important factors or control points that preserve -- bookings direct customer relationship? What parts of the business are more defensible versus AI? Is it payments? Is it the customer service and accountability? Is it loyalty? Is it something else?

Yes. With 28 minutes still to go, I can probably talk for the next 28 minutes on this topic. But let me keep it high level and then we can dive deeper in whatever direction we can -- you want to go, Trevor. So let me really take this at the highest level of what is the strategy we are following. And the strategy has a couple of components. I'm leaving the whole internal efficiency benefits and all of that out of it. I'm speaking really about the go-to-market strategy here. One is we're partnering very closely with all the LLM developers. We're a launch partner with OpenAI. We're a partner with Microsoft, with Google. You probably have seen with the latest developments of Anthropic in their B2C efforts. We were a launch partner for Cloud for Life at 2 weeks ago. Why are we doing that? Because, first of all, we learn a lot from a technology perspective, what are they working on? What is the latest? We get an early heads up when they are having a product launch, and they ask us to be part of that. And we know -- so we know where that is heading from a technology perspective. And so that is great for us. We learn a lot from consumer behavior. We see how consumers are reacting on those latest products. That's great information and knowledge we have in the very early stage. Third is these channels will become future performance marketing channels for us. Like, for example, OpenAI is testing PPC now at this moment. We're part of a very small testing group with them on that. So this is playing to our strength, more performance marketing channels. We are -- I look at us as a very scientific company. We run very large optimization algorithms that is continuously determining the next incremental dollar or euro, where should we spend that? What is the highest incremental return and if we have many more performance marketing channels on top of the traditional gorilla that has been in traditional search for a very long period of time, and we have expanding already from that, of course, over time with social media and others. This is just additional channels, good for us, more diversification. Second part of the strategy is what do we do in our own environment because you were talking about direct. This is, of course, very important because as a company, over the last 15 years, we have moved from 100% dependency on performance marketing channels to 2/3 of our B2B traffic is coming direct to us today, 2/3, which is really incredible. Obviously, what we want to make sure is that, that is not moving backwards. So why are we -- how are we going to do that? We are working on a lot of top-of-funnel AI tools, that should give traveler customers the same experience as they can get in ChatGPT or Gemini. But then you can do the same kind of inspiration, the same kind of itinerary building, the same kind of planning and putting away and come back. By the way, we can make it very personalized because we know what you have done in the past and where you have traveled and what kind of hotel you like. If you like a small hotel with a great gym or a great bar and et cetera, we can curate all of the information for you around that. But then, of course, we can give that kind of an experience with a trip planning tool that is as good as the hyperscalers, but provided by a party that our travelers know, they trust, they have their loyalty program with. We have their credit card information. It's all taken care of. If something goes wrong, they know who to contact to fix it for them. So I think trust is a huge element today in AI, and we all have our experiences with AI hallucinations and very strange outcomes. So if we can provide the same experience, but in an environment that has that trust and then we're also working on expanded loyalty benefits of our Genius program, I think that's going to be a very strong proposition to the market. So those are the elements that we're thinking of and working on strategically in order to be ready for the future.

Yes. And I think an underappreciated aspect of that is all the work that you're doing to incorporate it for the consumer experience on your surfaces, right? It's perceived that you can only do this stuff on ChatGPT or Gemini, which just actually isn't correct, right? You're working on building this stuff out yourselves.

And actually, we already have a couple of tools out there that are really strong. I can, for example, point to Penny, which is an AI trip planner from the Priceline platform. It's generally seen by a lot of research as one of the best or maybe the best that is out there. And that is further advancing. You now can book directly through Penny. So that's actually in many of the trip planners an issue, you can't make that step to doing the booking that can be done. In Booking.com, there's a lot of AI tools that are being added to the existing UX. So we like that because, obviously, existing UX and Booking.com is optimized over 2 decades. So we -- it's more of a different strategy to really take customers gradually along the way with more and more AI tools that are embedded. Agoda is working on an AI trip planner. We have 2 start-ups, one in the West Coast, one on the East Coast that are working on something that is even more innovative from a concept perspective. So there will be a lot of new initiatives that will come to the market this year.

Okay. A lot of your comments there were actually focused more on the end consumer side of things. Let's hit on the supply side of things because this is often a worry we hear from investors about maybe some of the major hotel platforms, what are their aspirations on AI surfaces and the ChatGPTs of the world. Do you have any concerns about scaled partners trying to figure out their own integrations directly such that they may not need to rely on booking in some future state for a portion of their business?

Yes. That question came up a lot, I think, a couple of months ago. It seems to have gone away a lot. And the reason is I think the element that we have tried to explain and what was probably underappreciated at the time was the complexity that goes into the fulfillment for travel. What we do is in terms of 4.5 million properties, 4.5 million properties where we should have real-time data around availability, around pricing. So that connection should be really a like-to-like connection all the time. What we're doing around payments, we facilitate payments on over 100 local payment methods. This traveler from Asia wants to pay in his local payment method. That's not what this accommodation partner in Italy is accepting. We take care of that. By the way, it can be in different currencies as well. We facilitate over 50 different currencies. We do customer service in many different languages. We take care of a lot of modifications and cancellations and adjustments that have to be made along the way. There's a lot that goes there into the sausage making that isn't so much like, oh, yes, there's one agent that talks to another agent, and that is happening there. But more importantly, actually is the supplier value proposition. So think about this in the following way. So let me take Italy again as the example. So you have this small family hotel in Tuscany. So today, traditional search is already very hard for them. So how am I going to be more visible to, let's say, travelers from Asia? They want to really focus on travelers from Asia coming to their hotel. If you think about ChatGPT, how will a small family hotel ever show up on ChatGPT? How will they ever be visible? So what we do is we help them with global marketing, global reach, global promotion. They are very much in control on our platforms because they can say, it's a high season, I have a lot of regular customers coming to me, I only put 30% of my inventory on the platform, but in the low season, it's 60%. I can sit with Booking and say, you know what, I want to be more focused on getting Asian travelers. So from a pricing perspective or from a visibility perspective on the platform, they can make sure that, that is happening. So in a world where it's actually even becoming more unclear how you ever show up in an AI chatbot, I think that added value we have from a supply value proposition is, from my perspective, even becoming more important.

Yes, you'll solve a lot of pain points for them as the world gets more complex. Yes, makes a lot of sense. To wrap up on AI, let's hit on the internal usage a little bit. You saw some decent operating leverage last year by reducing customer service costs even as bookings continue to grow double digits. How much more opportunity is there in customer service specifically? And then beyond customer support and maybe developer productivity, where do you see the next step function improvement in cost structure or conversion efficiency from AI, again, internally facing?

Yes. In this field, the developments are going really fast. I would probably have given you another answer 2 months ago or 3 months ago than today. If I see how our developers are using cloud and how much benefit they take out of that, that's really a step change compared to, let's say, 2025. So really impressive how fast these developments are going and the benefits. I mean, we, of course, have to make a choice of how much of that is going to be converted into more productivity, so more development versus efficiencies, but the benefit is clearly there. I think in customer service, as you already mentioned, I think it's a fantastic example where, first of all, it's going to be much faster. Your resolution time as a traveler customer is much shorter. And you are having a much higher satisfaction. So we all have experience with customer service in general with any company where it sometimes can be very painful. You wait very long in line. And then in the end, you get someone on the phone and can't answer your question, you have to be connected to another department and so on. And in the end, 30 minutes later, 45 minutes later, you're still on the phone. So the fact that AI can give you a human-like experience, things can be resolved very quickly. Even if 100,000 people call at the same time, everyone can be supported at the same time only for very difficult cases, it really needs to escalate to a real human. Cost down, satisfaction scores up, I mean, it's the best combination. And there's much more that still can be done there in the future as well.

Okay. So more runway for potential savings there. Okay. Got it. Let's shift a little bit to supply as well as kind of growth verticals. First, we'll hit on accommodations. You're now at nearly 9 million active listings. How is Booking differentiating itself to attract high-quality independent host rather than just professional operators, particularly as competitors like maybe the Expedias of the world reinvest aggressively in their platform verbo?

Yes. I think this is a little bit a U.S.-centric question, Trevor, because I think if you think about the rest of the world, there is not so much of a question of both on the demand and on the supply side of people considering to be on our platform. Actually, we are a very large alternative accommodation platform in Europe. We're also one of the largest in Asia and the rest of the world. So in the U.S., our position is definitely smaller, particularly with the single homeowners. But that needs time. Again, we are relatively smaller. We're challenging in the U.S. market. We're growing. And in the end, this is the effect of creating more demand, create more supply interest. We're improving our product, have more listings on our platform. And then it's just a flywheel effect that will start and will turn faster and faster in the future. So this will come naturally over time. I think our differentiation in terms of alternative accommodation has nothing to do with exclusive supply. We never think about supply in an exclusive way, like it's only on our platform, and therefore, people have to come to us. By the way, that isn't a great model from an antitrust perspective as well. But in the end, we add value in a different way why people come to our platform. And particularly in alternative accommodation, it is that on our platform, you have traditional accommodations and alternative accommodations side by side. So we see so many times that people are looking for one type of accommodation and end up booking another type of accommodation. This family that goes on a trip, they need 2 hotel rooms, they prefer it to be connected, but the hotel doesn't give a guarantee. We show in the same city an apartment with 2 bedrooms -- a 2-bedroom apartments. They like it or they like, by the way, that they have a small kitchen with it, and that's ultimately what they book. So the fact that we show it all side by side, I think that is really a differentiator on our platform.

Okay. That makes a lot of sense. So let's stick with kind of some of the other pockets where you've had growth. Obviously, flights has been a big one for you, but probably more to come on some other verticals. You've ring-fenced $700 million in gross investments this year, a $300 million net EBITDA impact. Where are you reinvesting this year and why? And what are some of the areas of future growth as we look maybe beyond alternative accommodations and flights? What are the next big growth drivers for Booking?

Yes. I think the first category where we're investing is really the expansion of our proposition -- our core proposition. So think about other verticals like the advertising business, the attractions business, the flights business. We're expanding the international network of our restaurant business, OpenTable. We're expanding in business lines like B2B. So in B2B, we're bringing our B2B businesses that are part of our 3 OTA brands together in a new business in the future. So that is an element there. We're expanding internationally. Think about the U.S. and Asia, where a lot of investments there in localization, in product and supply, so to make those propositions better in Connected Trip. So that's all expansion of the proposition, both from a product and from a geographical perspective. The second category is in AI. So what I just explained, a lot of those top-of-the-funnel AI tools, some of the foundational AI investments, some of the investments around having our data ready and prepared for AI models and that we can train those models very, very quickly. Just to be clear, we are not the type of company that needs to spend a lot of CapEx on it. We're not developing foundational models ourselves. We're using models that are being developed by others that are training them and fine-tuning them for our use cases. And by the way, that gives us also, from an architectural perspective, the opportunity to swap models in and out, depending on what is the best model at any point in time. So those AI investments is the second category. And the third category is in fintech. We believe fintech is a really large opportunity for us in the future. So if you think about it, about -- now today, we said in the first quarter, low 70% of transactions, we facilitate the underlying payments of the transaction. Last year, we did $130 billion of payment transactions of those bookings. And that's already a huge number. If you compare it to the scale of some of the largest PSPs, that's already a huge number. People often don't realize that we are already quite a decent-sized PSP within our organization. And that's growing. That leads to a lot of benefits from an economics perspective. The most important way to think about it is it's the same transaction, but we are adding incremental contribution margin euros or dollars to that same transaction. So it's a net benefit for us as a company. But more importantly, fintech is really a strategic opportunity because we can expand in many different directions with additional products. We have started with paying their own currency on the traveler side in a number of currencies. We're expanding that. We're going to the supply side. There's many other products that we can think of more expansion of the insurance products, wallets. So there's a lot of things we can do there in the future around our fintech strategy. And so we have actually only just started with that. So strategically very important because it comes down to what we discussed at the beginning. This is an added value we provide for the traveler and for the supplier, but it is also something that is economically really good for us. So we're bringing actually all of that together.

Okay. You actually hit on one of my next payments. So just to recap on kind of fintech and payments, pretty high payment penetration already. It sounds like a lot of runway ahead across a suite of fintech type products, very attractive economics. Is that a fair way to summarize it?

Yes. And it's still going up. This is what we call merchant bookings. So merchant bookings is where we facilitate the payments. And it's still going up each and every period. It will be at some point an equilibrium. There will be always people that say, you know what, I just want to pay when I check in at a hotel. That's what we call the agency model. So that will be there. But I think we still see the merchant model still growing, still also this period.

Right. Okay. And to the earlier part of your prior answer, you hit on Connected Trip. This has been an initiative you've been working on for a number of years. I think trip volumes grew high 20s percent last year. It's now a double-digit percentage of mix. At what penetration rate does this kind of meaningfully change your ROAS, cancellation behavior, lifetime value of customers? And are you already seeing that contribution already? Or is there still a lot more to come?

Yes, we are already seeing that today. So the way to think about this is people that come more often to us. So our more repeat customers are more likely to book across multiple verticals, are more likely to combine many different elements of a trip in one booking are more likely to come direct and use the app, are more likely to be in the higher Genius loyalty program tiers. And so this is -- all of these elements are all interrelated and connected. So this is something we do deliberately, make sure that we have customers that become more repeat, more direct, more satisfied, come back more often. And the economics, of course, with that are also there. Think about this in the following way. If we have someone that makes one booking, let's say, okay, we know that someone goes to Barcelona because they book their flight ticket with us or an accommodation in Barcelona. Then we don't need to incur second time the acquisition cost for that customer to make a proposition of, hey, do you also want to have a tour there or do you want to have a rental car there or any other element to the trip. So the more we can do that, the more actually it adds to the overall economics of this particular one trip and this particular one traveler customer.

Right. Acquire them once and get as many of those modes of travel as you can as part of the trip.

Exactly.

Makes a lot of sense. Before we open it up to Q&A, I do want to hit on margin. You're guiding to '26 being potentially the sixth consecutive year of margin expansion. The last 3 years, you've averaged about 2 points expansion, but you're now starting to approach high 30s, 38%, which kind of puts you close to where Booking was pre-pandemic. And while the business is much larger now than kind of pre-pandemic, you've also layered in some areas such as flights and payments, which could be a bit of a headwind on overall EBITDA margin. Can you speak to any pockets of incremental savings for 2026? And how should we think about the path ahead on EBITDA margin maybe for the next 3 to 5 years? Can you surpass prior peak?

Yes. A lot went into that question. So let me try to unpack that a little bit. First, yes, our margins last year were -- EBITDA margins were close to 37%. I cannot emphasize enough, you know what I'm going to say there, that it is a fully loaded EBITDA margin. So we have very low CapEx. So we're not playing the game of let's CapEx everything under the sun and then it never shows up in EBITDA. It's fully loaded from a stock-based compensation perspective. We think stock-based compensation is a normal cost of running your business. If you wouldn't pay it, then you have to replace it with cash compensation. And by the way, you need the cash flow at the end if you have stock-based compensation to take the dilutive effect of it out. So it's a normal cost of running a business. I'm still flabbergasted that people think that it's a kind of free component in running a business. So fully loaded and therefore, on a like-for-like basis, higher than any of our direct competitors. But we have done very well with the expansion. And that is because although we are investing a lot in those strategic initiatives, we're also running transformation programs where we're looking for efficiencies and taking the benefits from that. So we have done that, for example, with the transformation program where we have said that we're going to take out $500 million to $550 million. We had in the run rate last year at the end of the year -- sorry, in year 2025, about $250 million. And this year, it will be fully in our numbers. So let's say, another $250 million to $300 million benefit that will come in this year. We've also said due to the situation in the Middle East, more short term, we're pulling back on some discretionary spend. That's about $100 million or so additional savings that we will achieve this year. And then the third element of this is the following. Travel is a scale business. If you're a small player, it's really hard. If you have scale, then you can run so much more volume over your scale, you can be more price competitive on the one hand, but that also will help you, I think, from an overall EBITDA margin perspective because the next incremental traveler that comes to us and book a hotel night on an airline ticket, we don't need new infrastructure costs, new office building, new layer of management. So all of that, the incremental contribution to the margin is high. So yes, you are right that we have been growing in verticals that are all stand-alone lower-margin verticals. Flights is lower margin, Asia growth is lower margin and several of that. But scale matters, plus I think the unit economic benefits of Connected Trip and combining those elements. So that's why we have been able to continue to expand margins over the last period. Even in a year like this, we said that up to 25 basis points, we still think that we can achieve EBITDA margins from the current level. You also asked if I'm -- unpacking the last part of your question, you also asked what that will mean for the future? Well, we haven't given specific guidance on it. But implicitly, we have said the following because we are following the framework, the algorithm, what we call 8-8-15, being at least 8% growth from a gross bookings and revenue perspective and then 15% -- at least 15% growth from an EPS perspective. The step from 8% revenues to 15% EPS, you need to get your EBITDA and then, let's say, a low double-digit level and then with the buyback benefit, get your EPS to the mid-teens. So which implicitly means from 8% revenue to low double-digit EBITDA, you need to have some kind of leverage, some kind of an efficiency benefit that goes in between there. So you can read implicitly something out of that.

Yes. That's actually probably a great place to pause, just being mindful of time. I do want to open it up to the audience for any Q&A. So we will start passing around the mic if anyone has any questions.

Given what you just laid out on the cash generation in the business, can you just explain how you're thinking about leverage going forward?

Yes. Our leverage is what we have said as a target, approximately 2x growth leverage through the cycle. So we don't want to be bound. It has to be always 2x. But we said through the cycle, we would like to be at that level. We were at the end of the first quarter at a leverage of 1.8x. So we have a lot of leverage capacity still. Actually, as we speak today, we're in the bond market with a new offering. So -- but we have also some debt maturing over the next few months. So all of that is happening at the same time. Not so much of a different story. We think actually stability, predictability in terms of our capital philosophy approach and targets is very much the same of what we have done in the past. We generate a lot of free cash flow last year, north of $9 billion. We are reinvesting a part of that back into our businesses because we think that's the most attractive way to really make sure that we continue to be a high-growth business in the future. M&A is always a possibility, but we have some regulatory restrictions there in certain businesses, in certain regions across the world. And then we're also actively returning capital to shareholders, both through dividends and buybacks. So I think it's a very repeatable model, very stable model. And I think nothing has changed there also with respect to the overall capital targets.

I think we have time for one more up here in the front.

In a world where I'm going to say to my AI agent, book me a hotel in New York City, how is it that ends up going to you over someone else?

Yes. So that can happen that people is saying that. By the way, this is where we not see a lot of traction today, the number of leads we get from those LLMs is relatively limited. It's not going up so much. And I think it still has to do with people that are not really sure because that's a big question, whereas that chatbot going to book you and through which platform and how do you know it's a decent hotel? And if you want to make a change, who do you have to call to make that update. So I think that's not happening a lot, but that might happen more in the future. So that's exactly why we are partnering with all of those LLM developers because we want to be the one where they are referring to at that moment, and that traffic is coming to us. But more importantly, I think is the other way. And we have said this with our fourth quarter earnings call, almost 90% of our supply base or independent hotels are alternative accommodations and are smaller chains. So smaller chains, meaning not the top 15. So leave out the Hiltons, the Hyatts, the Marriotts, the Wyndhams and so on. So almost 90%. These are the hotels that if you want to go to a Marriott, by the way, you're probably part of the Bonvoy program and you book it directly with them, and it's already the case. But the ones that are the independent ones, I think this is the one where we add the value that they are visible, that they get the traffic, that people know about them and that they do that through a platform that is a trusted platform for them. So I think that is the differentiation and the added value that we can deliver in that ecosystem, also very much, I think, if that's all driven by AI interfaces and agentic interfaces.

And I think we will wrap it there. We are up on time. Ewout, thanks so much.

Thanks.