Glenn Fogel - President, CEO David Goulden - Executive VP & CFO

Mark Mahaney - RBC Lloyd Walmsley - Deutsche Bank Mark May - Citi Justin Post - Bank of America Merrill Lynch Kevin Kopelman - Cowen and Company Brian Nowak - Morgan Stanley Douglas Anmuth - JPMorgan Naved Khan - SunTrust Mike Olson - Piper Jaffray Anthony DiClemente - Evercore James Hardiman - Wedbush Deepak Mathivanan - Barclays

Welcome to Booking Holdings' Third Quarter 2018 Conference Call. Booking Holdings would like to remind everyone that this call may contain forward-looking statements, which are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. These forward-looking statements are not guarantees of future performance and are subject to certain risks, uncertainties and assumptions that are difficult to predict. Therefore, actual results may differ materially from those expressed, implied or forecasted in any such forward-looking statements. Expressions of future goals or expectations and similar expressions reflecting something other than historical facts are intended to identify forward-looking statements. For a list of factors that could cause Booking Holdings' actual results to differ materially from those described in the forward-looking statements, please refer to the safe harbor statements at the end of Booking Holdings' earnings press release as well as Booking Holdings' most recent filings with the Securities and Exchange Commission. Unless required by law, Booking Holdings undertakes no obligation to update publicly any forward-looking statements, whether as a result of new information, future events or otherwise. A copy of Booking Holdings' earnings press release, together with an accompanying financial and statistical supplement, is available in the For Investors section of Booking Holdings' website, www.bookingholdings.com. And now I'd like to introduce Booking Holdings speakers for this afternoon, Glenn Fogel and David Goulden. Please go ahead, gentlemen.

Thank you, and welcome to Booking Holdings' third quarter conference call. I am joined this afternoon by our CFO, David Goulden. We had solid execution in our busiest quarter of the year and passed a new milestone of 200 million room nights booked in a single quarter, reporting 201 million worldwide room nights booked. This is up 13% year-over-year and exceeded the high end of our guidance range. Consolidated gross bookings were up 12% year-over-year in U.S. dollars or about 14% on a constant currency basis. Our revenue increased 11% year-over-year in U.S. dollars or about 13% on a constant currency basis, and adjusted EBITDA grew 8% in U.S. dollars with a similar currency impact on the year-over-year growth rate. Our strong room night growth rate was helped by our later-than-normal summer booking system. In addition, we have higher growth coming from our performance marketing channels, which benefited in part from our lapping the start of last year's optimization efforts. Our guidance for Q4 room night growth of 12% reflects our expectation that we will have more stability in our top line growth rate and reflects the continuation of our marketing spend. And we note that we continue to grow significantly faster than the overall global accommodations market. I am pleased with this strategic approach we have taken with our paid channels this year. Going forward, we expect these markets will remain very dynamic, which is why we'll continue to employ our data-driven approach to achieve the right balance of growth and acceptable ROIs. We will look to spend in the paid channels when we see opportunities to acquire high-quality traffic at attractive ROIs. And as I said before and want to emphasize, we work closely with those advertising partners that help us build our brand and maintain our competitive strengths while working in a manner that is best for our customers. Conversely, we will reduce our participation with those who do not work in a way that we believe best serves our customers. As you can see in our results, we're beginning to ramp up our brand advertising. As we've discussed in the past, we expect to increase our brand advertising efforts to further our goal of bringing more customers directly to our platforms and to increase product awareness, particularly in our home segment. We are increasing our spend in digital channels as they provide efficient reach and allow us to better measure the effectiveness of our spend. This is part of our long-term strategy. And though we expect to increase our investment in this area, we are mindful to spend as efficiently as possible, letting results and data help direct the pace of the investment. We have progressed in many of the investment areas that we have discussed over the current year. In particular, we've expanded our product breadth and customer experience in the areas of alternative accommodations and experiences. We continue to build supply in alternative accommodations. And as of September 30, Booking.com had over 5.7 million reported listings in homes, apartments and other unique places to stay, which is an increase of 21% year-over-year. Booking.com's total reported listings were about 29 million of September 30. Our room night growth in alternative accommodation remains robust and is higher than our consolidated growth rate. We're also pleased to see that the number of customers whose first-ever booking with us through our home product is growing nicely. We have developed new services specifically for our booking home host. These services include streamlined property onboarding, enhanced guest management tools and property profiling capabilities, which allows host to highlight their property's unique aspects. We believe we are building a leading platform to search and discover truly unique accommodations in the same frictionless path that our customers have come to expect from us. And we continue to believe that the best customer benefit comes from offering both hotel and home accommodations on one unified platform. We recognize that building a great home platform is only part of the challenge to building a great home business. And customer awareness of our capabilities, particularly in certain geographies like the United States, is low and will require a higher marketing investment in the coming quarters. However, we believe we are on the right path and, in the long run, will achieve a leading global home business. In the area of experiences, we are building out our product capabilities, creating a seamless integrated way to offer more choices for our accommodations customers. This quarter, Booking.com made progress integrating FareHarbor's products, giving Booking.com access to new local attractions in the U.S. and the ability to leverage FareHarbor's technology to help even more tours and attractions around the world come online. Booking.com's experiences product is scaling well, and we now offer experiences in approximately 70 cities worldwide. In addition, we are making progress with experience in other areas such as offering at-hotel services and restaurant booking options that we note these are nascent efforts. As mobile becomes an increasingly integral part of the travel ecosystem, we continue to invest in our mobile platform, so we can offer all of our products, accommodations, ground transport and experiences, whenever and wherever our customers may be. We are happy with our progress in this area as today, over 50% of our accommodation booking transactions come from mobile devices. We continue to seek ways to promote cross-brand initiatives that allow us to share best practices throughout our company and get further benefits from our multi-brand platform. Priceline and Agoda are now working more closely together in certain overlap market such as the U.S., and we are seeing some efficiencies there. With greater technical and managerial cooperation between KAYAK and OpenTable, we're excited about the pace of innovation happening at both companies. OpenTable has been growing nicely, and we are confident about its large opportunity. We believe through tighter interaction with our other brands in the future, OpenTable will provide benefits to all of our customers and create additional value to our entire enterprise. The number of restaurants on the OpenTable platform has grown by 55% since its acquisition in 2014. And over the prior 12 months, OpenTable has seated over 330 million diners via its online reservation system, which is 69% higher than during 2014. OpenTable is now increasing its development velocity, and we look forward to some exciting improvements in the platform in the quarters to come. Finally, I would like to note our $200 million investment in Grab, the leading on-demand transportation and mobile services platform in Southeast Asia, that we announced last week. This investment and strategic partnership is consistent with our long-term strategy: to create the optimal worldwide travel platform, which will make all aspects of travel easier for our customers by having an integrated seamless way to experience the world at the lowest friction possible. I have mentioned this long-term strategy before and will always be a work in progress as technology changes and advances. But I am pleased to see the initial foundations being laid out. In sum, I am very pleased with our execution in the third quarter. The pace of innovation remains robust, and I am confident in the long-term growth prospects of the company. I want to thank our approximately 24,000 employees for their hard work and dedication during this busy quarter, providing unparalleled service to both our customers and property partners around the world. I will now turn the call over to our CFO, David Goulden, for the detailed financial review.

Thank you, Glenn, and good afternoon. I'll discuss our operating results and cash flows for the third quarter and then provide guidance for the fourth quarter of 2018. All growth rates are relative to the prior year comparable period, unless otherwise indicated. As we discussed last quarter, all year-over-year growth rates referenced in my remarks in Q4 guidance will compare the current year income statement under the new revenue accounting standard to the prior year under the previous accounting standard. Gross bookings and other metrics like room night reservations are not impacted by the new revenue accounting standard. Our non-GAAP financial results and forecast include stock-based compensation and are reconciled to our GAAP results in our earnings release. Now on to our results for the quarter. We are pleased with our booked room night growth of 13% in Q3, which is well ahead of the high end of our guidance range and accelerated from Q2. When we provided our guidance for the third quarter in early August, we were in a period of relatively high uncertainty given the impact that the World Cup and an unusual weather pattern in Europe we're having on the summer travel season. Following our earnings call, we saw an acceleration in our room night growth rate that continued through the end of the quarter driven by the late summer travel peak and improved results from some of our performance marketing channels. We continue to show progress in growing bookings through our direct channel, which remains our largest single source of new customers. The direct channel continues to represent about half of our booked room nights and continued to grow faster than the overall growth rate. Average daily rates for accommodations, or ADRs, were up over 1% in Q3 versus the prior year on a constant currency basis, which is better than our forecast of flat. Strong ADRs in some of our core travel markets more than offset the negative impact from mix. Changes in foreign exchange rates reduced Q3 growth rates in U.S. dollars by approximately 3 percentage points versus last year and by about 1 percentage point versus guidance. We estimate that changes in FX rates impacted gross bookings, revenue and EBITDA growth rates by a similar amount. Q3 gross bookings grew by 12% expressed in U.S. dollars and grew by 14% on a constant currency basis, coming in about 6 percentage points above the high end of our guidance range. Consolidated revenue for the third quarter was $4.8 billion and grew by 11% in U.S. dollars and by about 13% in constant currency. Revenue for the third quarter of 2018 under our current revenue standard was approximately 1% lower than it would have been if reported under the previous revenue standard. Advertising and other revenue, which is mainly comprised of non-intercompany revenue from KAYAK and OpenTable, grew by 14% in Q3 compared to prior year, including revenue from Momondo, an acquisition we closed in July 2017. In the third quarter, we started lapping our strategy to optimize performance marketing ROIs, which began in mid-Q3 2017. As a result, we continue to see leverage from performance marketing in the quarter, however, at a lower rate than the first half of the year. The leverage from performance marketing in the quarter was more than offset by the deleverage in sales and other expense, which continues to be driven by the growth of our merchant business in Booking.com. We will see pressure on the sales and other expense line as we continue to ramp our merchant business. However, a significant portion of these expenses are offset in revenue. As part of our effort to drive more direct traffic to our websites, we increased our spending on brand marketing in the quarter by 27% versus Q3 last year, which contributed about 40 bps of deleverage. As expected, the year-on-year margin pressure from nonmarketing OpEx, which includes sales and other, diminished in Q3 relative to the first half of 2018. GAAP operating income grew by 7%, and GAAP operating margin decreased by about 170 bps compared to Q3 last year. GAAP operating income in Q3 is negatively impacted by $23 million pretax related to -- sorry, by $27 million -- $23 million pretax related to a net travel transaction tax charge from prior periods that is recorded in our G&A expense line. GAAP net income amounted to $1.8 billion or $37.02 per share, which grew by 8%. Our GAAP net income includes a $31 million pretax net benefit related to unrealized gains and losses on our equity investments in Meituan and Ctrip. GAAP net income was also negatively impacted by the net travel transaction charge I just mentioned. We excluded the unrealized gain and the net travel transaction charge from our non-GAAP results. Our non-GAAP tax rate for the quarter was 21%, which was in line with our forecast. Adjusted EBITDA for Q3 amounted to $2.36 billion, which met the high end of our guidance range and was up 8% year-on-year. As I previously mentioned, we saw a negative impact on our growth rates from the changes in FX since providing our guidance. Our adjusted EBITDA margin of 49% was slightly below our forecast. This reflects higher spending in marketing, which helped our room night growth. Our non-GAAP EPS was $37.78, up 7% versus the prior year. Non-GAAP net income reflects a non-GAAP tax rate of 21.1% in Q3, which increased from the prior year due to the impacts of the U.S. Tax Acts and the high Innovation Box Tax rate in the Netherlands. Our 4% lower share count in Q3 versus last year offset the negative impact on EPS growth from the higher tax rates. Our cash and investments amounted to $16.2 billion at quarter end. In Q3, we generated $2 billion of operating cash flow, which grew by 4% for the quarter and by 22% on a year-to-date basis compared to prior year. Our free cash flow for Q3 was $1.8 billion, which grew by 2% for the quarter and by 19% on a year-to-date basis compared to prior year. We returned about $2.2 billion during the third quarter to our shareholders through share buybacks. Since the start of the year, we reduced our fully diluted share count by approximately 4%. And on September 30, we had approximately $6.4 billion remaining on our share repurchase authorization. We will continue to be both programmatic and opportunistic with regard to our repurchases. And under stable business and market conditions, we expect to complete this authorization within the remainder of the 2- to 3-year time period we talked about in May. Turning to Q4. Our guidance reflects our quarter-to-date results and assumes our growth rates will decelerate over the remainder of the quarter mainly due to the size of our business and consistent with long-term trends. Our approach to guidance has not changed. Foreign exchange rates are expected to be an approximately 4 percentage point headwind to year-over-year growth rates in Q4, which we estimate will impact gross bookings revenue and EBITDA growth rates by similar amounts. We used a dollar-to-euro exchange rate of 1.145 when setting our Q4 guidance. We are forecasting booked room nights to grow by 9% to 12% and gross bookings to grow by 6% to 9% in U.S. dollars and by 10% to 13% on a constant currency basis. Our Q4 forecast assumes that constant currency ADRs for the company will be up by about 2% compared to prior year. We forecast Q4 revenue to grow by 13% to 16% in U.S. dollars and by 17% to 20% on a constant currency basis. We estimate that our Q4 revenue as reported on the new revenue accounting standard will be slightly less than -- 5% higher than it would have been under previous standard. Note that our EBITDA margins and growth rates versus prior year will also benefit from the impact of the change in the revenue accounting standard. Q4 adjusted EBITDA is expected to range between $1.19 billion and $1.22 billion, which represents 11% to 14% growth versus prior year. We forecast that adjusted EBITDA margin will be down modestly versus Q4 last year. After adjusting for the impact of the revenue accounting change, forecast EBITDA will be approximately flat versus Q4 of last year, at the midpoint of our guidance range. Note that this growth rate was negatively impacted by unfavorable year-over-year FX change I mentioned earlier. To help you understand the underlying drivers, leverage and deleverage in the business in Q4, I will talk about these on a like-for-like revenue basis to eliminate the impact of the Q4 benefit from revenue accounting change. We are forecasting deleverage from the performance marketing expense line in Q4. Our Q4 forecast reflects the full impact of lapping our ROI optimization efforts we started in the middle of Q3 last year. We also expect to continue to lean into certain performance channels where we see high-quality traffic. We expect to grow our brand marketing expense in the quarter at a faster rate than we did in Q3, which will contribute additional deleverage to the P&L. We believe prudent investment in brand marketing is important to drive more direct traffic to our websites and build better profit awareness. Sales and other expenses continue -- is expected to continue to pressure margins primarily due to the ramp-up of our merchant business. We continue to push the expansion on our merchant platform to help build a home business and for other beneficial transaction effects. Finally, we expect continued deleverage in personnel from the investments we made in the second half of 2017 and earlier this year. We forecast GAAP EPS between $18.05 and $18.55 for Q4. Our EPS guidance assumes a fully diluted share count of about 46.8 million shares, reflecting the beneficial impacts of the common stock repurchase we've made to date, which will reduce our share count by 6% versus Q4 last year. Our non-GAAP EPS guidance for Q4 assumes tax rate of approximately 20%. Note that our GAAP tax rate in Q4 last year was negatively impacted by a onetime expense item related to the U.S. Tax Act and is, therefore, not comparable. We are forecasting Q4 non-GAAP EPS of approximately $18.90 to $19.40, which represents 12% to 15% growth versus prior year. Our non-GAAP EPS forecast assumes an estimated income tax rate of approximately 20%, which is higher than the prior rate of 17% due to the impacts from the U.S. tax act as well as the increased rate of the Innovation Box Tax in the Netherlands. The negative impact from the higher tax rate is more than offset by a lower share count versus Q4 last year. We have hedge contracts in place to substantially shield our fourth quarter EBITDA and net income from any further fluctuation in currencies versus the dollar between now and the end of the quarter, but the hedges do not protect our gross bookings, revenue or operating profit from the impact of foreign currency fluctuations. Our forecast does not assume any significant change in macroeconomic conditions in general or in travel market in particular. With that, we'll now take your questions.

[Operator Instructions] Our first question comes from Mark Mahaney with RBC.

Okay. I want to ask about the marketing strategy a little bit. I know you've gone through this kind of over a year of this kind of toggle, I think, away from performance and more towards brand advertising. Sounds like some of the upside, the acceleration you saw this quarter, it happened at -- came through the lapping of performance marketing. But maybe just address a little bit broader this overall shift towards brand advertising. Do you feel like you're at an optimal mix now? There's going to be constant experimentation. I assume the answer to that is yes. But any more color on whether after this year of experimentation you feel like you've gotten a better, more sustainable growth profile setup because of this toggle in marketing shift.

Mark, it's Glenn speaking. So I know you'd like me to be able to tell you it's all fine, stable, and it's going to be the same going forward. But the fact of the matter, these markets, as you all know, are extremely dynamic. So when we find good opportunities, we find high-quality traffic at the ROIs that we like, we step in, and we're going to buy. Similarly, when we think that there's something that we don't like, we're going to step away, whether that's because the platform's acting the way that we don't believe is in the best interest of our customers or for some other reason, we're going to vary how we're going to spend in those pay-for-performance channels. But what I said a year ago and I continue to say it now, is the importance of increasing our brand spend. And we're going to continue to do that, and you saw some of the increase already, and we here said we're going to spend more. And the reason is we believe, in the long run, we want to try and drive as much traffic as we can directly, and part of that is creating a brand awareness. And we talked about this in the past several times. So I can't give you any sort of specifics about how much it is we're going to do in performance, is it up or down. I can say there's probably going to be dynamic. There's going to be movements up and down going forward. But I absolutely believe, in the long run, the right thing is we'll try to increase that brand advertising to get the people to come direct. David, if you want to add any color, go ahead.

Oh, Glenn, thank you. I think just consistent with Glenn's comments, we did reinforce in my remarks that we're very focused upon the growth and the balance of that direct business, again, which remains our largest single source of new customers and represents about half of our booked room nights. So it's a very strategic part, not the only part. But the spending on brand obviously supports that and also supports our longer-term strategy to expand our portfolios to provide more value to our customers.

And our next question comes from Lloyd Walmsley from Deutsche Bank.

I had a couple, if I can. I wanted to first, I guess, follow up on the marketing questions. Can you give us any sense for how we should think about the return timing on the brand spend? And are you planning you're able to scale that up with more confidence around measuring a specific payback? Or is that more of a strategic decision still to spend on brand? And then, I guess, the second one, you mentioned a number of customers making their first booking is growing nicely. So any specific growth kind of rate or relative growth rate to your room nights? Is it faster or slower than overall room nights? Or maybe you can give us an update on that customer count metric given it's been a couple of years since you've updated that. Anything further you can share there?

Lloyd, so talking about marketing and the brand, what we're trying to do here? I can't give you any sort of specifics of when we think the ROI is going to be. I will say that it's still very early. If you recall last year, we talked a little bit about how we increased the number of people in terms of our brand marketing, how we're increasing the ability to develop technology tools to be able to measure, and we're continuing to do that. And we do like what we're seeing, and we are increasing the spend. This is a difficult thing as we continue to go forward in terms of understanding the attribution and how much of a factor you're getting for your brand that's going over to where your pay-for-performance is now performing better than it would have in the past and many different ways to understand what's the true ROI. It's going to be something we're going to continue to experiment, but I do believe we are going to continue to ramp up that spend, as I said in the last question, and it is something that we do believe is very important. Regarding the second thing, I want to make sure you heard what I said perfectly. What I was talking about is we are having people who are coming to our first time-ever, we call it, book time with Booking, and the first thing they're using is the home product. That's something we really like to see growing because that's showing that there's awareness of our booking home product beginning to grow. In terms of specifics, I'm not -- we're not going to give out any numbers on that area.

And our next question comes from Mark May with Citi.

Sorry, again, on the marketing side, but up until the last year or so, the company's marketing spend as a percent of revenue is, I think, pretty consistently growing a couple hundred basis points a year. I guess, now it looks as though you're kind of leaned back into marketing. Is that kind of the pace that we should be thinking about? I'm just trying to understand a little bit how much of the past year or so has been a bit of an anomaly. And would you -- should we be thinking about kind of pre-2017 to think more about the marketing leverage in the business?

Yes. Mark, this is David. Let me just start off with that. I understand where the question is going. Let me just -- first of all, I think you do have to look at the year in 2 halves. As you mentioned, the first half, broadly speaking, we have the benefit from our ROI authorization activities comparing optimized -- less optimized a year ago, and in the second half, we start to lap it. So you start seeing the two halves start to play out in a slight different way. I think one way to look at it, if you just look at the full year, and I know this is a projection based upon the top end of our guidance range, Q4, but if you look at the full year and you add everything together, you see 13% room night growth, 17% revenue growth, 17% EBITDA growth, 16% EPS growth. So I think it represents a fairly healthy year. Obviously, there's been significant stories during that year. And I know it's been a little bit of challenge sometimes for us to walk everybody through the moving parts. So one -- and one way to look at it is just kind of step back and look at the bigger picture for the year. The other part of your question is kind of what will things look like going forward. And I'm not going to give you a specific amount of leverage or deleverage, but I mean, our longer-term view of the business hasn't changed. And as we put our -- there's always going to be a bit of an ROI reset going on with -- being certain of our performance channels over the course of the last 18 months, but our longer-term view hasn't changed. And as we think about our 2019 plan, we think about our longer-term plan, what we plan to do is to invest in the business, to grow at above-market growth rates, expect to have and tolerate some modest deleverage in our EBITDA margin rates to achieve that build-out of the platform and that above-market position but have a close eye on the EBITDA dollar growth. So that gives you a little bit direction. But I don't want to get into the specifics yet as to how much, and we're still obviously working through that. We'll give you more as we finish our 2019 plan.

And our next question comes from Justin Post with Bank of America Merrill Lynch.

One big-picture question and one housekeeping. On big picture, I think you spent about $100 million more than we thought on marketing, just our estimates, and generated about 6 million more room nights than we were thinking. Did something shift interquarter where you found pockets of better ROI? Or something changed in the market that made you more aggressive on the marketing spend? That's the first question versus your original outlook. And then secondly, just the strength in merchant bookings growth versus agency. Can you just remind us of the dynamics of that?

Sure. I'll take the first one, Justin. I'll let Dave take the second one. So as I remarked earlier, when we see good opportunities, we're going to spend. When we see areas that we don't think is appropriate, we're going to lean back in terms of marketing spend. The markets are extremely dynamic. And as David mentioned, there were some interesting things happening towards the end of Q2, running into Q3 that made things a little bit harder to try and foresee the future. So in each one of those things where we did spend more at certain times, and we did end up with more room nights. That's clearly in the math there, and we're pleased about it. And we hope going forward that we'll continue to find these type of opportunities that are going to help us build the franchise.

Yes. Thanks, Glenn. And then on the -- on housekeeping, actually, it's a fairly important point. You can really see the transition occurring -- starting to occur in Q4 '17 when the growth in our merchant bookings started to significantly exceed the growth in our agency bookings as we are building out and deploying a global payments platform at Booking.com. That does a number of great things for us, for our customers and our partners. For our customers, it gives them many more choices to how they may want to pay for their transactions either in advance or closer to their stay. It gives them more opportunities to pay with the payment product of their choice. It may not necessarily be a credit card. It could be something like an Alipay, for example. For us, it lets us basically provide our customers with a more consistent service because we're in charge of exactly how our payment flow works. And then for our partners, again, we offer them more ability to accept different payment forms from different customers in different parts of the world because we can basically pay them, the partner, in the form of whatever they like to take even though we may have taken the payment in on the front end. There are different payment mechanisms. So it is the build-out of that global payments platform principally rolling through Booking.com having the impacts. And you see that mixed shift occurring. It's been occurring for quite some time as we see merchant revenues -- merchant bookings, rather, outstripping the growth of the agency bookings.

Okay. And one follow-up, is Agoda simply like growing faster than your other platforms? Can you tell us about that?

We don't usually talk about individual brands, which is growing faster, which ones are not. But I will say that Asia is a fast-growing area. It's also a very competitive area. It's been that way for some time, and I suspect it's going to continue to be that way for some time in the future.

And our next question comes from Kevin Kopelman with Cowen and Company.

Just a follow-up on the marketing questions. Can you give us an update on your relationship with Google? And given your comments in the prepared remarks, whether you feel they're helping you best serve your customers.

So as you know, we don't generally talk about any individual advertising platform. And we also talked about over many, many years, we've had a wonderful relationship with people at Google, basically helping build out our business together with them, a very symbiotic relationship, helping improve each other's capabilities. And we're very pleased with that. And we hope that continues for a very long time. I can't say much beyond that, that would be helpful to you, I don't think.

Okay, understood. And then just another question on brand marketing. You talked about wanting to grow that. Right now, you're about $500 million annually. How are you thinking about what the right amount is to spend on brand to make sure you're getting out there around the world today?

So as you know, one of the things that we're very proud about is our history of doing experimentation and not trying to make guesses or just projections without data. So what we're going to is continue what we've always done, and that is spend, test; test, spend, see how it goes. It's complicated. It's going to take a long time. And over that -- over this long period of time, we will find out what the optimal amount to spend is. But I can't give you a number now given that it's going to take significant amount of experimentation for a very long time before we are able to come up with what we think the optimal is. And the interesting thing about that is because this world is so dynamic, things change so rapidly, I don't expect we're ever going to come up with a set numbers. It's always going to be moving. One thing, though, I will add that I am happy to see is more and more of the brand marketing is going to digital video. And one of the good things about that is the measurement of that spend is better than it used to be when you're doing over-the-air TV. So at least we're getting better data, the measure against.

Glenn, and lastly, can I just -- can you give us quarter-to-date share repurchase?

I believe that's David to say no.

That -- Kevin, we'll just update you every quarter on how we've done. Yes, I think I saw a new policy. I think it's the right way to go.

And our next question comes from Brian Nowak with Morgan Stanley.

I have two. The first one, just trying to a little bit better understand the 3Q and the 4Q guidance. The room night acceleration and the guidance deceleration, can you just talk to any specific regions or products that were the biggest contributors to the overall acceleration and growth? Then the second one, just to kind of, again, better understand the quarter. As you said, direct traffic, it sounds like, grew faster than the overall rate, so I think direct grew in the mix, yet performance probably drove more upside than expected. How did that happen? We're you able to use data to sort of test and deliver more volume at lower ROIs? Or were ROIs more stable than what you thought? Just how did performance come in better than expected? Was it that lower ROI or not?

Brian, let me start, and then I'm sure maybe pass it over to Glenn for a couple of comments. So yes, you asked a number of things there. So yes, just to confirm, direct did increase mix in the quarter and also our performance growth in the performance channels, again, they did in the early part of the year, specifically in the Q2, to just kind of confirm that. And why did we see some more opportunities there? Well, we always test the amount of kind of elasticity and what the potential is to increase growth rates to return for investment. We're testing constantly in those channels, as I mentioned, a very dynamic and real-time testing kind of what we see as opportunity. And we saw a little bit more opportunity to drive growth without dropping ROI very much, if that makes sense. The prior elasticity test we had during the year would say that if you were wanting to drive a lot of growth, you have a high reduction in ROI. We saw some opportunities in some of our key channels to drive some more growth with less of a drop in ROI. We thought that was a smart thing to actually do. So that accounts for a number of things we've already talked about. It accounts for kind of why our performance in marketing spend was perhaps a little bit higher than many people have estimated. But as we commented upon the performance in room night, that was part of the reason why we're able to do better, and we know that's not the only reason why we're able to do better in the room nights. And that's why we lent into those channels a little bit more in Q3 than we expected to when we gave our guidance. There were some other factors, of course, that we're very much at the top of our mind when we gave that guidance at the start of the quarter. I mentioned there are other unusual patterns we were seeing in July. In August, those started to get better. Particularly, the whole weather situation in Europe got better, both in terms of the heat wave breaking in the north and the scorch wave breaking in the south. So those are all contributing factors but specifically related to the channels, I think, that gives you some color.

I would just probably just add that -- and I understand where people are coming for all these questions and looking at the numbers for sort of guidance was. But what I continue to try and do is have our investors and people that are looking at our company is focus on the long term, what we're trying to grow, what we're trying to build, what's going to be great for customers and partners. So in the long run, we're able to achieve the mission, which really is to help people experience the world no -- to experience the world with less friction and more profits for us. That's what we're trying to do here. And some quarters are going to be better than expected. Some quarters are perhaps not going to be as good. And there's so much -- and again, I'll use the word dynamic. We got a lot of players in a market that are also bidding against us. Bidding their own strategies, brand advertising. We are trying to figure out what works, what doesn't work, lots of factors happening. But as long as we continue to build our products, make things better for the customers, create what I talked on the past about this holistic system that really makes it easier for people to travel. As long as we focus on that and continue to do that, I believe, in the long run, we will continue to grow this company at a reasonable rate. And as we said so many times, we're still a single digit in accommodations in the whole industry. So that's a lot of ramp left.

And our next question comes from Douglas Anmuth with JPMorgan.

Maybe a good segue into something that's not marketing-related. But Glenn, just hoping you can talk a little bit more about nonhotel accommodations just pass through the busiest part of the season. How do you sum up where you are here in terms of supply, the value proposition to consumers and also awareness in how that ties into your key priorities as you look into '19?

Yes. Look, it's incredibly important. We believe it's a great opportunity. We reported listing is up 21% year-over-year. You got to have the actual supply to actually book anything. If you don't have that, then you're not going to be able to move business, so that's a good start. But we've also talked in the past about the need to spend money to improve all the different things that need to be done to make a host, the person who's -- owns that property, be willing to rent it out to somebody. And we need to get the awareness to the demand side to get them to know, hey, we got this great product, come and get it. That's a lot of effort. And we were starting from behind from some of our competitors, no doubt about that. That being said, we talked about things that we built, and I talked a little bit about it in my prepared remarks, about some of the stuff that we've been building. And it's really a long-term project. This is not going to be done in a quarter or 2 quarters, even a year. But it's something that is a long-term thing that we absolutely believe in, and we believe that we're going to be able to continue to do it in the way that we think is the best way, and that's having both hotels and home properties on the same site. So the customer sees both products right there, sees the reviews right there. And we believe that you don't hit somebody with a traveler's fee at the end. And we believe incredibly important is the ability to do it on an instant-booking basis, so that when a customer is ready to purchase, it's done. It's not going back and forth, back and forth. And one of the things in that, though, is that there's an issue of trust. And that some of the tools we talked about for the host to be able to build some of the tools to make those hosts feel comfortable in an instant-booking area. So lots of things to be done, but it's an important part of what we're trying to build here. And I've seen -- as I mentioned earlier, seen people the first time they ever booked with us is booking a home product, loving that.

And our next question comes from Naved Khan with SunTrust.

It's Naved Khan from SunTrust. Maybe two questions. So can you just maybe talk a little bit about the China outbound demand and maybe plans for HotelsCombined now that you've gotten approval from Europe for acquiring that business? And then I had a follow-up on metasearch.

So question -- first question, HotelsCombined, and I may have missed it. I'm not -- I don't believe we actually got the final approval yet. So we're going to pass on that question. So your question about meta in general?

Yes. And then just, along with HotelsCombined now, so wanted to know if you're willing to provide some color on China outbound demand, what are the -- any kind of trends you might be willing to share.

China outbound. Yes, look, incredibly important. I love saying it, so I'll say it again. I love China see how the locomotive, the travel train. I just saw some interesting statistics that I thought were indicative of how important this can be. So 2014, there were 55 million Chinese passports they could use to go outbound. 2017, it's 120 million passports. That shows kind of the growth. People don't get a passport unless you intend to actually go out of the country. China is also making things, making it easier for a Chinese tourist to go out. So in 2015, there were 46 countries that were visa-free. Now there are 74 countries, makes it a lot easier to go. As the -- and this is anywhere in the world. When the economy is doing well, as people get more money, they want to travel more. And in China, when the person is able to afford to travel outside China, they want to go. If you look at some of the social stuff going on, social media, people putting out their pictures of visiting London, in Paris and Rome and all that. But it's more than that. It's all over the world, and I've talked about the times being in Alaska seeing Chinese tourists, being in Iceland seeing Chinese tourists. It is so important for us to be able to get a good share of that product, and that's why we're doing a lot of effort. We have over 1,000 people in China working for us there. I think we have either a dozen or 14 offices in China. We have made -- as you know, we've made some significant investments to have good partners in China. We invested in Ctrip some time ago. We invested in Meituan, and we recently made an investment in DiDi. By making these investments, we are getting different ways to reach the Chinese tourists, so that we can show them all the great things that we offer to them. So when they want to travel, we've got what they need. And we're excited about it. We're pleased about it. And we think it's going to have a long-term benefit to our company.

And Glenn, if I can just pick up the point on meta because that was in the first question. For obvious reason, we're not in the so-called HotelsCombined because we have not got the final approvals yet. So we've been working down that path nicely. But what I would say is that we think the meta segment is very important. And it is a segment because I think are a group of people who will always want to go a meta first before choosing their travel, and there are other people who are prepared to build their longest-term relationships with the OTA platforms. And we think that winning model is a global multi-platform model, which we are building in the meta space. So we think it's a very important part of our strategy, and people shouldn't get confused. We don't think it's necessarily competitive with our parts of the business. It's complementary because it addresses a segment of the marketplace that just prefers to go there first, and we want to have a great solution for those customers as well.

That's helpful. And just on the -- your commentary on meta. So now that you are sort of lapping the changes you made last year, the ROI changes, do you see more room for improving ROI there? Or do you think that they're at their appropriate level for now?

Well, in terms of meta as a channel now for our businesses, I think we are now will be fully lapping in the Q4. The areas where there was the biggest share shift in terms of the changes that we made when we started our ROI authorization effort was, in fact, in meta, so we should get benefits in their Q4 over Q4. But we like our ROI approach. We think that what we've done is -- got things to a level that works nicely for us and for our business. So now we're lapping those, and we're looking incrementally opportunities just like I mentioned we did with other channels in the Q3.

And our next question comes from Mike Olson with Piper Jaffray.

One follow-up on homes. Have you been able to decipher whether or not vacation homes are mostly additive or cannibalistic to traditional bookings? And then separately, on experiences. Should we think about this as being a significant area of investment in the near future? Or is it fairly minor relative to your spend in other categories? Just wondering how that stacks up on the priority list.

I'm sorry, could you read that second one again, Mike?

Experiences, just wondering where it stacks up on the priority list and how much the near-term investments need to be there.

All right. So in terms of cannibalistic in terms of homes, I don't usually think of it that way, really. I think about providing the choices. On what the consumer decide, it's fine by us. It's not our decision to try and direct them one way or the other. We do recognize, though, if we don't have the home, we may lose somebody who wanted a home, even though they may come to us and want to check out the hotels and we don't have the home. That may be a negative. We've always said that it's important for us to try and get every single type of accommodation on that platform. It's homes, apartments, it's -- we say a unique place to stay. We really mean unique. We're talking about yurts. We're talking igloos in the winter. We're talking everything there. That way, by providing that incredible breadth, really makes it more appealing to the consumer that when they want to travel or thinking on what they're going to do, they come to us. They know they're going to be able to get what they want on our side. Now in terms of experiences, in the long run, it is important. It's not something that we're going to just go 1 million miles an hour right now and crash the income statement by just throwing money at it to build out the thing, but it's something that is very, very important. And the reason is not only the money that it will bring to the bottom line, the incremental EBITDA, but we also believe, in the long run, it goes to that thing that I've talked about so many times, is providing that system that makes it easier and better for the person to travel. And I've experienced it myself as a consumer. I use it. And in cities where we have these things like experiences, it's just been a wonderful thing to be able just to show that QR code off the Booking.com app, skip the line to go on to the [indiscernible] or go up to the castle at the top of the mountain and celebrate. I did that. And I said, this is great. And that's the type of thing we do, so to build loyalty, it'll make it easier. So there's a lot of compound besides just getting the incremental amount of money out of that one transaction.

And our next question comes from Anthony DiClemente with Evercore.

I have two. Glenn, the hotel suppliers themselves have been out there talking about the success of their own direct-to-consumer initiatives. They're competing to grow direct relationships with their guests as you are. So I guess, the question is, how is that going? How much of your spend on brand is really to actually compete with what amounts of your suppliers were trying to build those same direct relationships in the channel? And then, second question would be for, I guess, either David or Glenn. What's the right capital structure for Booking Holdings is? Is the business overcapitalized relative to what really is steady cash flows over time? So maybe just talking about the puts and takes of stock buybacks versus a strategic M&A versus the releases of your capital would be really helpful.

So our business is based on creating a great relationship both with people who want to stay in accommodation and people who are supplying the accommodation. And whether it be a large multinational, international chain or a single little hotel owner in the middle of some small village in France, it's always trying to provide value to them, as we provide value to the demand side. And sure, I've seen the edge for some of the large international chains, and nothing wrong with that. They want to attract people to come to them directly. They think it may be cheaper for them to do that. There's nothing wrong with that. We have a great relationship. I'd point out that the international large chain is actually a small portion of our business. We've talked about in the past, about 15% of our business. But we want to do everything that we can to be helpful and help make their business better. That's how we win in the long run. And we do provide a lot of value. We have the largest demand platform in the world for accommodations. So anybody who has a hotel and wants to fill a bed with a body and make some money, we're the people to come to. And some of the stuff that we do doing over 40 languages, over 40 languages, and doing customer service in these languages, we paid for that. That was [indiscernible] That's our cost. Doing all of the marketing that we do, those things are done more efficiently than almost any hotel could do on their own. And they'll find that efficiency, we're providing that value to these hotels. I want to -- I read about this and I hear about people saying the fight between the people like us and the hotels, and I really wish people would stop thinking this is a sporting event, a boxing thing or something like that. This is a thing where, sure, people want to attract people to their own site to say, look, why don't you come to us direct, too? I don't wanted to pay money for marketing. But in the end, I really want to change the -- just the tone of everything, so that we see that we're really helping each other. And this is really a symbiotic relationship. And I'll let David talk about capital structure.

Yes. Anthony, obviously, capital structure could be a very long topic, and we've only got a couple of minutes for it. So I'll try and give you the kind of a higher view. First of all, I think you have to step back and look at we try to do with the business. There is a huge travel market opportunity out there, and we only have -- we have a single-digit share of the combination segment where we are the strongest. Therefore, we have a lower single-digit share of the broader travel marketplace. And our mission is to build out the best online global travel marketplace on the planet, offering choice, value, service and seamlessness for our customers. And that price is a huge price. And that is a price, I think, everybody wants us to go forward and try to develop. So that is our primary mission in life, and we are fortunate to have financial resources to be able to do that and, of course, be able to return significant amounts of money to our shareholders. Now that ability to return to shareholders as aggressively as we can right now has been aided by the Tax Act that is relatively new. I think that the commitment to return $10 billion over the next 2 to 3 years is a very important step forward. But I do want to put it in the context that broader opportunity to invest both organically and inorganically in building out this incredibly large opportunity we have in front of us where I think we have the ability to generate something, which is of significant value for our customers, our shareholders, for our partners, which is kind of our primary mission in life. So if that helps, that's how we think about it.

And our next question comes from James Hardiman with Wedbush.

You've touched on this a little bit but maybe clarify a little bit. You had a strong finish to the third quarter and then better-than-expected guidance for the fourth quarter. I'm trying to figure out how much of that is momentum versus coincidence. Obviously, you don't give us a fourth quarter guide until now. But you spoke to a delayed summer travel booking scenario. I wouldn't think that, that would affect the fourth quarter very much. But I guess, at the end of the day, I'm trying to get at, are you more optimistic about the fourth quarter today than you were 3 months ago? And then maybe as somewhat of a side, maybe talk a little bit geographically. Europe was called out as maybe a headwind to overall growth last time we talked. Obviously, the mix of Europe changes in the fourth quarter. But I'm just trying to figure out if that's sort of resolved itself and whether or not Europe is a headwind or a tailwind as we look into the fourth quarter and beyond.

Yes, James, let me start with that. So to answer the direct question, are we more optimistic about Q4 now than we were when we did our guidance last time? Yes. I think we're pretty clear last time when we were guiding, we were sitting looking at June and the July period, which were impacted by some macro facets that were out of our control, Europe being biggest in the marketplace was going through the most ridiculous weather pattern we've ever seen where the south was -- sorry, where the north was kind of warm and dry, and people didn't want to go anywhere vacation; and the south was basically in the 40s and raining, and people didn't want to go there. So that really affected our business. We couldn't predict when that was going to change, and that was also compounded by the World Cup going on, which, again, had big impacts on Europe, with big countries doing well in the tournament. So we were seeing that looking at a high degree of uncertainty. As I mentioned, as we developed through the quarter, things got better. Things got better than we expected in most of our geographies. And Europe was certainly the biggest one, and things did improve in Europe. And obviously, we ended the quarter at a higher growth rate than the average for the quarter because we commented that we expect a deceleration back on our August call. So we are looking at Q4 from a stronger lens than we expected to be, if things hadn't improved. And as I mentioned, our guidance for Q4 would reflect some deceleration from what we saw from a growth rate in October. So those are the factors, which we are looking at. And hopefully, that clarifies the puts and takes. But the short answer to the question is, yes, we are in a stronger position about our view of Q4 now than we would have been, if things have not improved in Q3, which is what we've built into our Q3 guidance.

Got it. And the geographic question, it sounds like Europe, the concerns there have largely been alleviated. Is Europe now a tailwind to growth going forward?

I'm not going to get into the relative growth rate of different segments. But obviously, a lot of the things that were out of our control, we talked how it impacted our view of the world in early Q3 were happening in Europe. And yes, we talked about some level of acceleration from Q2 into Q3, and we saw that across most of our major geographies.

And our last question comes from Deepak Mathivanan with Barclays.

Just a question on the 50% transaction mix from mobile. As mobile ramped over the last several quarters, has there been a meaningful change on repeat behavior cost of customer acquisition that could potentially be an incremental positive to unit economics going forward as -- given that we have now crossed our significant treasure hold in terms of 50%? And is the mix between paid and organic significantly the front on mobile devices?

So we don't -- we're not going to get into talking about the relative costs for mobile or desktop. I would, of course, caution the obvious thing, but some people seem to sometimes forget this that mobile does not mean app and does not mean free. So I just want to make sure everybody is on the same page on that one because it's not. We are very pleased now because we believe, as people continue to move towards the mobile device and doing all of their -- doing their travel with a mobile, it gives us a great opportunity to continue to work with the customer throughout the trip, providing them with more services, more things and making them better experience, make their travel better, so that they are then more loyal to come back to us again because we're offering always great things on the mobile. That's the good part about mobile. That's where it's really at. And I wouldn't get too much to wrap around whether or not the cost is higher or lower. Again, these are competitive markets. And over time, I imagine they're probably going to end up in an area where the equilibrium is not that different in desktop, who knows. That's my point about mobile. Dave, if you want to add anything to that.

No, Glenn. I think that's the key point. The key point is that when we talk about the trip and be able to interact with the customer during their trip, their mobile device is always with them. It's always on. The desktop, they can go back to it occasionally, but they tend not to take the desktop with them. So if they open their laptops at night, maybe they [indiscernible] may want to try today. But the mobile device is there the entire time. So it's a very strategic shift in the business and particularly as we build out all the aspects of the trip. We like our customers interacting with us much more often on mobile than they can on a desktop device.

So thank you for joining us for the call today. Very pleased with our quarterly results, and we remain very excited about the future prospects of the company. I look forward to updating you after our fourth quarter call. Thank you very much, and good night.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude today's program, and you may all disconnect. Everyone, have a wonderful day.