Good morning, everyone. Welcome to the 2025 Nasdaq Morgan Stanley London TMT conference. We're thrilled today to have Ewout Steenbergen with us, the CFO of Booking Holdings. Good to see you.
Thank you so much, Brian.
Let me do the disclosures first, and then we will talk through everything going on in travel, in Booking specifically, and we probably will talk about agentic, I would imagine. Let's start with the disclosures. Please note that all important disclosures, including personal holdings disclosures and Morgan Stanley disclosures, appear on the Morgan Stanley public website at www.morganstanley.com/researchdisclosures. Some of the statements made today by Booking Holdings may be considered forward-looking. These statements involve a number of risks and uncertainties that could cause actual results to differ materially. Any forward-looking statements made today by the company are based on assumptions as of today, and Booking Holdings undertakes no obligation to update them. Please refer to Booking Holdings Form 10-K for discussion of the risk factors that may impact actual results. Good to see you.
Have you got the legal thing out of the way?
We're all good. We're all covered there. So let's sort of start with a reflection of your time at the company and what has changed. So next, we're coming up on two years, about a year and a half right now. Call it two years in about next year. I want to kind of start with what, in your mind, the biggest misperception or misunderstanding about Booking Holdings is now that you've had a year and a half kind of under the tent, having a lot of meetings with investors, analysts. What part is most misunderstood in your view?
Yeah. Good morning, by the way, everyone. Nice to see such a large crowd here in the room. If I think about the last one and a half years with the company, I think there are a couple of areas that I believe are a bit underappreciated still within the company. I think one element is clearly around the network effect. The number of customers that come to us, come direct to us, like our platform, see the value, come more frequently to us. It is more loyal customers, so it's usually Genius levels 2 and 3. They come more frequently. They book more Connected Trips with us. They move higher up in the loyalty program and come back over time. So there is that element that is self-stimulating. Because if you think about it, we're growing faster in flight tickets than any airline.
We're growing faster in rental car days than any rental car company. We are growing faster in attractions than specialized companies in those areas. We're growing fast our Connected Trip transactions and so on. So we are growing all of these things faster than the standalone category players, and so there is something there that these metrics are not just a coincidence. They are not standalone. They're all interrelated with each other, so I think that is one element where I think it's still sometimes not really understood how strong that particular element is within the company today. The second one I would like to mention is Asia. I think we have a really strong position in Asia. We are the largest technology travel platform outside of Mainland China by really a length. We have a really good team over there, very entrepreneurial, growing very fast.
So that is really helping future growth in the company. So those are the two I would like to point out.
Great. Okay. That's a good starting point. I think one of the things that you've done a really good job at the last year and a half has sort of been laying out the company's multi-year growth framework, how you think about driving 8% growth, the 8% growth leading to really healthy double-digit bottom line growth. Maybe just sort of remind everyone sort of what that framework is numerically. And as you kind of think about puts and takes of sources of upside to growth or potential risks to those growth frameworks, how do you think about both of those?
So the framework or the algorithm we call 8-8-15. So the eight and the eight stand for the growth in gross bookings and revenues. And then the 15 is the number that stands for the EPS growth in the future. So these are the metrics that we're focused on achieving. And we're confident that we will continue to achieve those metrics at those levels over the next few years. By the way, this year, on a constant currency basis, we will be exceeding the eight, eight, and the 15 according to our latest guidance. So maybe I should first focus on the top-line metrics. So why are we confident that we can hit the 8% for the top-line metrics? There's a few things that go there in the mix. So if you think about travel, usually what is happening with travel, it's growth faster than GDP.
Because you start with GDP, but then people get more income over time, and then when they have more discretionary income, what do they want to do? Fun things in life, travel, and explore the world, so you see that also in upcoming economies where there's much more travel than people did before, so that is why there is growth, growth above GDP. There's still a shift from offline to online, then we are growing in a lot of new verticals outside of accommodations, and we're growing that very fast. By the way, I should also have mentioned before, alternative accommodations, of course, that we are growing faster than the expert in that area, but growing in flights and attractions and rental cars and payments and all of these areas will help to drive additional growth.
But then there are two new elements that I would like to mention as well. One is you've heard us speaking this year about our investment program, and we're spending $170 million in new growth initiatives. So that is going to help with future top-line growth over the next few years. And then, of course, Generative AI. Generative AI is also a real commercial benefit for us, a commercial opportunity. So we also believe that will help with also further growth of the 8% top-line metrics. I'm pretty sure you have some follow-up question on that.
You know I'll have Gen AI questions. I want to start with the $170 million because I think it's interesting to sort of lay out these multi-year frameworks for consumer cyclically exposed companies. But 8-8-15 is pretty formulaic and actually helpful for investors. So maybe sort of walk us through over the course of the year, how does the budgeting work? Do you start the year saying, "We're going to spend more on Asia. We're going to spend this amount on alternatives. We're going to invest in this type of growth"? How does the actual sort of blocking and tackling of the annual budgeting process work just to ensure you're meeting or exceeding your targets?
Thank you for asking that question because I'm very passionate about that, and I have a very strong philosophy, and the company has a very strong philosophy around it, and the philosophy is that there are two elements there. On the one side, you have to be extremely disciplined from a planning perspective in terms of driving leverage, in terms of driving efficiencies. Because organizations always naturally would like to grow and expand, but you need to have a kind of healthy tension on an organization to say, "Each and every year, we would like to see the top line growing faster by a few points than our expense lines, variable and fixed expense lines," and large organizations can do that, and particularly in an industry where skill matters. We can do that. The next incremental traveler that comes to us, the incremental margins by definition should be higher.
We don't have a ton of fixed cost infrastructure, so growth should always add incremental margin opportunity, but you have to enforce that with healthy pressure during a planning cycle in order to get those outcomes. But then the other side is that creates room, that creates room to invest. And you want to be very explicit around where you want to invest because the areas where you find the efficiencies are not by definition the best areas where you can also do those reinvestments. And for the last two years now, we have a mechanism where we say, "Okay, what are the best growth opportunities within the company?" and make very explicit resource allocation decisions. Because you want the big opportunities to get significant dollars behind it in order to get good outcomes.
You don't want to just spend average dollars, mediocre levels in many different places because then you also probably get mediocre kind of results. So we're very explicit on where we will put those dollars behind for that reinvestment of the resources that we are freeing up.
What are the areas you start talking about the $170 million investment fund? What are some of the areas that we should think about that make up the biggest chunks of that $170 million? How are you allocating the capital?
Yeah. There are a couple of areas. There is an area around expanding some of our verticals, so investing in the attractions business, investing in our advertising business as examples. And you have already seen some growth in these areas this year. There's investing in expanding networks that we're having. An example of that is expanding the international network of OpenTable. By the way, OpenTable is doing really well and is seeing some very healthy growth this year. So we're very happy about the energy that we're seeing there in that business. Expanding internationally. So we're investing in product and tech, in supply, and many other areas in Asia because we really want to make sure that we keep that very strong position in Asia and benefit from the future growth that will come out of that region. So that is one area.
And then, of course, Generative AI is also an important area where we are investing really in several areas in product in terms of AI trip planning. And I can speak later on more about that. But we're testing and learning a lot of different tools within our products in Gen AI in order to know what really has traction in the future.
I don't think you mentioned alternatives in that at all. Can you sort of walk us through the areas you're investing on and alternatives, and how do you think about growing more supply and alternatives in the U.S.?
Yeah. So alternative accommodations, it isn't a specific new investment category, but of course, we have strategically been very focused on that over the last few years. We have now outgrown the largest player there for 17 out of the last 18 quarters. I think we should stop maybe talking about quarters and just talk about years. So let's say 4.5 years. We're about 70%+ of their size, so have become a really meaningful player in this space. How do we do that? For a couple of reasons. One is expanding supply. So we're still expanding supply. We started with really the professionally managed properties, then the semi-professionally managed properties, but then we're going more and more down to the individual properties. So there's still opportunity to expand supply, and that will deliver then ultimately growth.
Last quarter, we had supply growth of 10% year- over- year. So that's still meaningful, and that's a good indicator, I think, of what is more to come. I think our proposition that we present traditional hotels, motels next to alternative accommodations during search processes is really helpful for customers. So we see that they like that, compare it, and sometimes they come in with the intent of buying one type of property and they ultimately book another type of property. So I think that proposition really works. And just generally, there is a real demand for alternative accommodations. There's a lot of travelers like that type of property. So all of that combined, we believe that alternative accommodations for the near future will continue to outgrow our traditional accommodations. And that's a good thing because economically, we're really agnostic.
So we like that it is growing, and it gives us, I think, a great value proposition to our travelers.
Is the U.S. and sort of getting even more growth out of the U.S. and alternatives, is that really important to sort of maintain that outsized growth in the portfolio? How important is the U.S. kind of driving that growth from here?
Yeah, we see the U.S. as the opportunity because we are already very large in Europe and in Asia. But in the U.S., our position is still relatively modest. So modest for us is always a positive because that means upside potential to do more. We see the traction really coming in the U.S. We outgrew the U.S. in general from a room nights perspective in the third quarter by quite a wide margin. Alternative accommodations was a part of that. So we think there's really a positive traction in the U.S. And maybe if I may say one thing to that, because that was probably from all the results we published in the third quarter. And generally, I think you have seen very, very good performance in the third quarter.
There was one data point that for me was really outstanding and maybe the most important, and that related to the U.S., not specifically only to alternative accommodations, but in general, and that was the growth of our direct channel in the U.S. Because direct channel, you can't just grow overnight. You can't grow it from one quarter to the other. That is the result of a very long period of investments in brands, in product, in awareness, in familiarity. People come to us one time through a paid channel, second time through a paid channel. They like us. They have a good experience, and then at some point, they come direct to us, open up the app, and do the direct booking.
So that we saw an inflection point in the U.S., I think was for me really encouraging because that is really a very, very healthy indicator for the future.
Okay. It's been 15 minutes. Now we can talk about Gen AI. So about 20-25 years ago, the hotels had their online businesses, and online travel agencies came to them and said, "We're going to drive incremental growth. We're going to make your online business better. It'll all be incremental. It's going to drive faster. You'll have better unit economics, et cetera." And you and your competitors built incredible businesses. And I think to a certain extent, the hotels need you more than ever, my view. Now we have these new players, these agents sort of trying to build new places in the ecosystem. OpenAI is building out an agentic product. I think you're a partner with them. Gemini is going to try to build new products.
My worry is, is there a risk that this is sort of there's an analogy to what went on with the OTAs years ago where these agents could wedge themselves in between the consumer and the OTA? You still have a lot of value with your supply, but the extent to which you become more reliant on this new paid channel and the agent impacts the unit economics. What is your response to that, and how do you think about sort of preventing that disruption risk?
Yeah. So I'm always thinking about this topic more from an offensive perspective than a defensive perspective. Let me explain the reasons why we think this is an opportunity. The first is we believe this will create an expansion of the total addressable market. There will be a faster shift from offline to online travel. There's about still 30% or more than 30% of the market is still offline. So this will be an acceleration. Moreover, there's probably going to be a real benefit of keeping your travel within one agentic experience environment. So this gives us an opportunity to really create also more of a share of wallet, a bigger experience of connecting the dots of travel. So that is one. It will help with expanding the total addressable market.
The second is we think actually if you look at our channels, and by the way, we have always had a big mix of channels. And in the future, there will be again a huge mix of channels. Sometimes this discussion is too binary from my perspective, as if the world only moves to one model. That's not going to happen. There will be many different customers with many different preferences. In the paid channels, I think this is going to be a huge benefit for us because the reality is we have had one very large dominant player in traditional search over the last decade. And most likely, there are going to be four, five, six different winners from a large language model perspective, more diversification.
And where do we have, as a company, a real specialism? [It] is running very large optimization models, more traditional AI, machine learning, where we all the time can optimize our performance marketing spend across all these different channels. Where do we get the highest incremental return? And I think that's where you see our ROIs in marketing going up, period after period. So applying those models, those algorithms, and optimizing across more channels in the future is going to be a positive from our perspective. And the third is what we're doing is going to create an experience for customers that are coming direct to us that is even better than what they're getting today. Why are customers coming direct to us?
I was alluding to it a few minutes ago is because they know our platform, they like it, they have the loyalty, they get some loyalty benefits, and then they come more frequently back and book across more verticals. Now we can create an agentic experience in addition to that. The same as horizontal agents can do, we can do it as a vertical agent. We can inspire people about where to travel. We can build an itinerary. But then where these models are stopping and become more a lead generator in the future are handing it over because all of them have said, "We don't want to be a merchant of record. We don't want to be an OTA." If someone comes to our environment, the agentic experience continues. We will give an agentic experience pre-booking because things happen.
Maybe your flight gets delayed, and we automatically update your hotel, your rental car, your restaurant reservation, your attraction. Then we give you an agentic experience in trip. Hey, you are in London. You have an outside tour organized on Thursday, and you have The Louvre organized on Wednesday. But you know what? Thursday, it looks like it's really bad weather. Shouldn't we flip your itinerary around? Hey, we see that you don't have a restaurant reservation this evening. We know you like a sushi place. We can get you into a great sushi restaurant this evening at 7:00 P.M. Click here, and so on. So that's an experience, and then the post-trip experience. That is something that is so much better even than today, so much peace of mind. Plus, we will add some additional benefits from a loyalty perspective.
So we think we will have a proposition in the future for travelers to continue to come direct to us.
That third part, I think, is probably the most important part of the defense and the counterargument to the disruption concern because the extent to which you could make a better on-platform experience, in-app experience, you can actually drive even more direct traffic, higher conversion rates, bigger recurring user base. What are one or two products you've sort of tested on that front so far? Anything you can share on sort of signal you're getting and how do we think about the pipeline there?
Yeah. We're testing many different products. We're testing top-of-the-funnel AI trip planning tools. We have with Priceline a tool called Penny, which is more focused on voice. We have with Booking.com, the AI Trip Planner tool, which is more a standalone tool. We have some tools in other businesses that are in product, in UX. So there's many different versions there. And we do that deliberately to test and learn which ones have the best traction. And then we have deeper in-product kind of tools like natural language boxes that, for example, we call Smart Search.
You type in, "Hey, I want to have a hotel with a great bar and a good gym, and it needs to be in the city center, and I like it to be more boutique kind of hotel." You don't need to tick all the boxes in terms of filters, and that does the automatic selection by itself. So those are in-product tools that we're having. What are some of the early indicators that we are seeing in terms of outcomes? People that are using those tools, usually the conversion levels seem to be a bit better, and also the speed to booking is a bit faster. And the second what we are seeing is that people that use those tools, the cancellation rates are a bit lower. Again, these are early indicators. So of course, we need to ultimately see over larger volumes, but it's encouraging early indicators.
Why is that happening? Because people can find better directly what they are looking for. And therefore, I think they are more convinced to do the booking quicker and have a better conversion. But also in the past, maybe people find some accommodation, they do a booking, but they're not 100% convinced, and then they continue to search for something else. If they find something better, they cancel for free the first booking they made. If they find immediately what they're looking for, they're not so much focused on doing that second step. Cancellation rates are actually really important because all of the growth rates we report are on a net basis. So even a slight improvement of cancellation, slight reduction, meaning improvement of cancellation, increases the net growth of the company as well.
That's really helpful. That'll be an important signal to start.
I can also speak about customer service, by the way. We have great benefit.
Let's talk about that.
That's maybe less of a really commercial benefit, but indirectly it is, of course.
Yeah, I know it matters. I mean, I think, well, maybe let's talk about that. Let's sort of talk about some of the cost savings or the sources of efficiency you sort of see looking ahead to kind of maybe bend the cost curve using Gen AI or new types of tools. What are you doing now, and how do you think about potential savings of that?
Yeah. So in customer service, we are really encouraged because we have been leaning a lot into Gen AI tools with our software partners in that area. I know many companies talk a lot about Gen AI, but then if you ask where does it show up in the P&L, it's very hard to point at it. I think for us, this is one area where we have meaningful benefits in the P&L this year. So what is happening based on those Gen AI tools in customer service? We see that the contact rate has come down a lot. So contact rate means where travelers still need to speak to a human agent. The resolution time has really become much faster. And that means therefore that the average cost per booking is coming down very rapidly.
So just to give you an indication, customer service cost in the third quarter were in absolute dollar terms slightly down year- over- year over high single digits booking unit growth. So that's a meaningful benefit. But the best part is that then also our customer satisfaction scores have gone up, which also intuitively makes sense. If you don't need to wait long in line and speak to a human agent and maybe cannot immediately answer your question, if things get resolved very quickly and it isn't Gen AI human-like kind of experience, obviously that's really a nice outcome. So that is one. But in general, also, as you know, Brian, we are running these large transformation programs. The outcomes are now exceeding our original expectations. So we have increased the target to $500 million-$550 million.
So we're also very pleased with the progress we're making in general about the transformation.
Got it. Okay. I want to talk to you about marketing efficiency and kind of marketing channels. You mentioned before how there historically has been one big channel, Google. I'll call it organic. We've spent a lot. But then I think in the last, let's call it three to four quarters, you've talked more about social and sort of social being a pretty rapidly growing channel with hefty, pretty healthy ROIs. Talk to us about what's changed in social and sort of how do you think about social being a bigger part of the overall marketing mix as we go into 2026?
Yeah, and this is another great example of my point earlier that there's always a mix of channels, new channels open up, and that's a great development for us as a company. So last year, we were able to crack a little bit of code with social because what is very important for social, but in general for every performance marketing channel that we are using is we need to, based on the very scientific approach that we're having as a company, we need to be able to measure incrementality. We need to be able to measure results and returns, but also being able to measure it in a way that we know it's incremental. Meaning you don't want to just throw a lot of money at advertising and social media and hope that something good will come out of it.
You need to measure the actual return, but you also need to be able to measure that that customer would not have booked anyhow with us because then we're spending money on advertising that is unnecessary. So what happened last year, particularly with Meta, we were able to create a special environment where we had data that measurement was possible. And that gave us the confidence to say, "Okay, we know these incremental returns are real, and starting to lean into that." We are in the meantime continuing to expand with other social media channels. I'm not going to mention certain names because I don't want to make others smarter than they are. Some channels are really leaning in, making those co-investments. Some others still to follow. But it is an opportunity where we can continue to expand. By the way, it's not only to expand with others.
Also, you want to push the curve out with the existing channels. The more positive signals you get, the more demand you can create, the more you can push your ROI curves out. So we're, of course, doing that with the existing channels as well. So those are the elements that go into the mix there. But again, we're following a very scientific process, and therefore we really like this as an opportunity.
Anything that stands out in the characteristics of the incremental users, the incremental travelers you're capturing from social? Is it a different demographic? Is it younger? Is it older? Is it a certain geo? Anything you kind of share on what those incremental users look like that may be different than your current users?
Yeah, it's definitely a younger audience. We haven't disclosed a lot, so I'm a little careful there to say too much around it. But what I can say is the general philosophy we have with social, but by the way, with all paid channels, is the following. We like those channels because they are a source of new customers for us, and then over time, these customers come back a second time, a third time, and then at some point, we have an opportunity to make them more direct customers, make them more loyal customers, and bring them more actively into our ecosystem. So we will always continue to find new paid channels, expand those channels because we like that mechanism of we need to find new customers, younger customers, demographics we haven't tapped upon before, and then over time, they become more loyal customers and coming back more frequently.
All right. We have time for one more question. Just I want you to remind everybody on sort of the philosophy around capital allocation and share count shrink and sort of like stock-based companies. I know this is sort of a debate in the sector. Just how important is sort of delivering durable share count shrink to shareholders for Booking Holdings?
Yeah, if you go back to that 8-8-15, there's a few things you can read in that implicitly. So first of all, to go from 8% top-line growth to 15% or yeah, 15% mid-teens EPS growth, there's two things that are happening there. One, there needs to be continued efficiencies because you need to get EBITDA to low double-digit levels and then have a buyback program in order to get your EPS to the mid-teens level. So those are implicitly built in. We are, of course, doing very well from free cash flow generation. We have very good, strong cash balances today. We are actively returning capital to shareholders. We do that through dividends and buybacks. The buybacks, there is a structural element into it, a systematic element in it. And then there is also a little bit of a tactical element into it.
So we scale up and down by quarter based on a 10b5-1 grid. And there is price sensitivity in that grid. So you saw that the buyback levels were slowing down a little bit. We're at 700 million in the third quarter. But then when there is a dislocation of the share price according to the grid that we put in place, there is really quite a large increase in the daily buyback volume. So one can assume that that is happening now at this moment in the fourth quarter.
All right. Thank you very much, Ewout Appreciate your time as always.
Thank you.
Thank you.