Booking Holdings Inc (BKNG) 2021 Q2 法說會逐字稿

Welcome to Booking Holdings' Second Quarter 2021 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 guarantee 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 fact are intended by -- 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 would like to introduce Booking Holdings' first speaker for this afternoon, Mr. Glenn Fogel. You may begin your conference.

Thank you, and welcome to Booking Holdings' second quarter conference call. I'm joined this afternoon by our CFO, David Goulden.

I'm encouraged by another quarter of meaningful improvement in our accommodations business with Q2 room nights up sequentially 59% versus Q1. This compares very favorably to our pre-pandemic historical pattern of a slight decline in Q2 room nights versus Q1. Compared with 2019, Q2 room nights were down 26%, which was significantly better than the 43% decline we previously reported for the month of April and a 54% decline in Q1.

The acceleration in the second quarter was primarily driven by domestic and international booking trends in Europe following a ramp-up in vaccination rates and the relaxation of many travel restrictions in the region. The growth of international bookings in Europe was mainly from bookings within the European region. The very strong room night growth in the United States that we saw in April and highlighted on our last earnings call continued in May and June, resulting in very strong U.S. room night growth for the full quarter versus Q2 2019. And David will provide additional details on our second quarter results in his remarks.

We are, of course, closely monitoring the impact of the Delta variant on the rising COVID case counts around the world as well as some newly imposed travel restrictions which have led to a modest pullback in our booking trends in the month of July relative to June. However, the July booking trends were improved from our full Q2 results. While the rise of the Delta variant demonstrates the volatility and uncertainty around the exact timing and shape of the recovery for travel, we remain confident that we will eventually see a strong recovery in travel demand globally.

The sharp return to growth initially in the U.S. and many of the European markets that we have witnessed this year shows us clearly that leisure travelers are eager to get back to booking trips on our platform when restrictions are lifted and customers are able to travel. We expect to be much closer to our 2019 revenue levels in Q3 than we were in Q2, driven by the strong booking improvements we have seen in the last few months.

As we have done throughout the pandemic, we will continue to build on the strengths of our core accommodation business and support its long-term growth. The strength of our core business comes from the flywheel effect we get from our 2-sided marketplace, where we drive benefits to our traveler customers and our supply partners alike.

For our customers, we strive to deliver the best choice of accommodations, offer the most value and provide the easiest booking experience, all backed by excellent customer service and support. By addressing these critical needs of choice, value and ease, we create a superior booking experience and strengthen the relationships with our customers.

I'm pleased to report that at Booking.com, we are seeing pre-pandemic customers coming back to us to book their trips while also attracting new customers. And the current mix of prior and new customers is not significantly different than prior to the pandemic. One of the ways we drive value to our large customer base at Booking.com is through our Genius loyalty program.

This program provides discounts for Genius members at hundreds of thousands of properties on our platform and also offers value in other benefits like complimentary breakfast, free room upgrades and more recently, discounted airport taxes, just to name a few. We will continue innovating and adding to the ways we provide value to our Genius customers who have historically had a higher repeat rate and a higher mix of direct bookings when compared to non-Genius customers.

Our app is an important way we deliver an easier booking experience to our customers. Globally, in Q2, Booking.com was the #1 downloaded OTA app according to a third-party research firm. In the U.S. in Q2, we were the most downloaded major OTA app as downloads of the Booking.com app in the second quarter significantly increased sequentially. We also saw U.S. app users in Q2 meaningfully surpassed the prior peak observed before the COVID pandemic.

In the second quarter, we again saw a higher mix of our customers booking directly with us than in the comparable period in 2019. It is encouraging to see these gains even as we look for opportunities to lean into performance marketing channels where we see attractive ROIs. We have a long history of effectively managing our performance marketing channels to bring bookers to our platform profitably. We plan to continue with this proven approach in the future.

In addition, we're leveraging our marketing expertise and ROI focus as we test into other channels like social and digital media as well as when we deploy promotional campaigns like our Back to Travel campaign, which we ran first in the U.S. and then launched in the U.K. and across Europe. We will continue to expand the diversity of our marketing and customer acquisition channels as we aim to drive incremental traffic to our platform and increased consumer awareness of our brands.

While we will remain focused on our efforts to grow and retain our customer base, we believe we will continue to benefit from the secular tailwind of more people booking their trips online instead of off-line. Historically, the accommodation industry has seen a steady increase in online share each year. And looking ahead, we believe this trend will continue for the foreseeable future.

In April, McKinsey published a 24-country survey with results showing that across 11 major consumer-facing industries, travel had the greatest percentage of customers that plan to increase their usage of digital channels after the pandemic. On the other side of our marketplace, we are focused on helping our supply partners reach a broader audience of potential customers. Our scale and global reach allows us to connect our supply partners with a significant amount of demand from around the world, demonstrated by the 845 million room nights booked across our platforms in 2019.

In addition to being a large demand channel for our partners, we add value to our accommodation partners in other ways, by providing customer service support for travelers in over 40 languages, localized partner service support teams, market intelligence and data, product innovations in response to new traveler trends and fraud liability shift and access to alternative payment methods for payment-enabled transactions. Whether an alternative accommodation or an independent hotel or a large global hotel chain, we strive to be a valuable partner to all types of accommodations on our platform.

In the second quarter, we saw the first sequential increase in the number of properties on Booking.com and the lowest number of properties coming off of our platform in the quarter since the onset of the COVID-19 pandemic. As of June 30, we had over 28 million reported listings on Booking.com, of which 6.6 million were for alternative accommodation properties. Within alternative accommodations in the U.S., Booking.com continued to add targeted new properties in the quarter and also saw encouraging share gains with some of our larger professional managers.

While these are positive early developments, we recognize there is much work ahead to improve and grow our alternative accommodations product in the U.S. market. Our alternative accommodations business in Europe was strong in the quarter and represented an increasing share of our European accommodations business.

I want to move to our key strategic priorities of expanding Booking.com's payment platform and building the connected tradition, both of which we believe will further enhance the strength of our core accommodation business and support its continued growth. On our integrated payment platform at Booking.com, we have made continued progress with increased the adoption of payments by our supply partners in the U.S., including adding some major hotel chains in the second quarter. Around 24% of Booking.com's total gross bookings in Q2 were processed through its payment platform, which is up from about 22% for the full year 2020.

We recently announced the organization of all of our payments initiatives and efforts into a new fintech unit at Booking.com. First, the Fintech unit will be focused on a mainly bookings core business to run better, faster and more efficiently for both customers and our supply partners. In addition, we recognize that we have opportunities to better monetize our overall transaction flows. In 2019, we did almost $100 billion of transaction value and we believe setting up a separate fintech unit to better capitalize on these flows will benefit us in the long run.

On our connected trip vision, I mentioned on our last earnings call that the development of the connected trip this year will be focused on enabling travelers to book the major elements of their trip in one place on Booking.com. The top priority of this -- on this front has been to scale up a robust flight platform on Booking.com, which will give us the ability to engage with flight bookers fully in their travel journey and allow us an opportunity to cross-sell our accommodation and other services to these bookers. Since our last earnings call, we have launched our flight product in 6 new markets and they're now live in 24 countries.

Inter tickets booked through Booking's flight offering have continued to meaningfully exceed our expectations. However, they still represented a small portion of our total reported air tickets, which were up 120% in Q2 versus Q2 2019. Primarily, this was driven by Priceline. While it remains early days for Booking's flight product we are seeing positive data indicating we are getting entirely new customers for Booking.com.

In addition, we are seeing an encouraging attach rate of accommodation bookings from these new customers. These early data points help demonstrate that our flight offering creates a new funnel to bring incremental customers to the platform and they cross-sell an accommodation to these new customers. We expect to continue to build on the early success we are seeing with flight at Booking.com.

In conclusion, I am encouraged by the signs of recovery we are seeing in some parts of the world, and I'm confident that we will eventually see a strong recovery in travel demand globally. We continue with our most important work, to strengthen our company's position and execute against our strategic priorities and our teams are working hard to support the strong summer travel season this year in North America and Europe.

As I said before, we are thinking about our business beyond just getting back to 2019 levels of demand. and we are focused on building a larger and faster-growing business that generates more earnings after the full recovery and for the long run.

I will now turn the call over to our CFO, David Goulden. David?

Thank you, Glenn, and good afternoon. I'll review our operating results for the second quarter and provide some color on trends we've seen so far in the third quarter. To avoid comparison to pandemic impacted peers in 2020, all growth rates will be relative to comparable periods in 2019, unless otherwise indicated. Information regarding reconciliation of non-GAAP results to GAAP results can be found in our earnings release.

Now on to our results for the second quarter. On our last earnings call, we discussed the improvement in trends in Q1, which continued into April, driven by strong results in the U.S. and improvements in Europe. Following the earnings call, we saw the overall improvements in our trends accelerate in May and continue to get better in June, which resulted in our Q2 reported room nights declining 26% versus Q2 2019, which was significantly ahead of the 54% decline in Q1, but 43% decline we saw in April and our expectations in May.

The improvement in Q2 room nights growth rate versus Q1 was driven by Europe and the U.S. as well as better results in Rest of World. Europe showed the greatest level of recovery in the quarter and actually achieved slight room night growth versus 2019 in June. Booking trends in Europe clearly benefited from a notable improvement in vaccination rates as well as loosening travel restrictions.

The U.S. was again the strongest reporting major country in Q2 and had very strong room night growth versus 2019 for the full quarter. Asia partially offset the improvements in other regions with greater room night declines in Q2 than in Q1 due to the increase in COVID outbreaks with related travel restrictions.

In the month of June, our room nights were down 13%, and our monthly active unique customer accounts at Booking.com reached about 90% of the level we saw in June 2019. As Glenn mentioned, we are pleased to see the solid rebound in our customer base at Booking.com as well as a healthy mix of new customers, which is only a little lower than the mix of new customers in Q2 2019.

Mobile bookings, particularly through our apps, represented over 60% of our total room nights. Our app continues to represent an increased percentage of our mobile bookings. Our direct channel increased as a percentage of our room nights year-on-year and relative to Q2 2019. Domestic room nights grew in the mid-teens in Q2, while international room nights remained down significantly versus 2019, we saw a sequential improvement in our international bookings resulting in the international mix of our room nights increasing to about 25% in Q2 and from about 15% in Q1.

Our cancellation rates improved in Q1 and were in line with Q2 2019 levels in the quarter. The percentage of our Q2 2021 bookings made with flexible cancellation policies remained significantly higher than in Q2 2019. The booking window at Booking.com remained shorter than it was in the second quarter of 2019 as we continued to see a higher mix of near-term bookings. However, the booking window contracted less than it did in the prior 3 quarters.

The mix of alternative accommodation room nights on Booking.com in Q2 was 32%, which is 3 points higher than Q1. In June, our alternative accommodation room night growth was flat versus June 2019, the first time we've reached 2019 levels for this segment since the start of COVID. The sequential improvement from Q1 to Q2 was due primarily to the overall improvement in room night growth in Europe in the quarter.

As we noted last quarter, Europe is where we have our highest mix of alternative accommodations. Within Europe, -- Our mixed alternative accommodations remained about the same as Q1, and this represents a continued increase from 2019 to 2020 and into 2021.

Gross bookings declined 12% in Q2, which is less than the decline in reported room nights due to an increase in average day rates for accommodations of 11% versus 2019 on a constant currency basis and also due to a few points of changes in FX rates and strong performance in our flights business. Our accommodation constant currency ADR benefited by about 7% from an increased mix of business in North America, which is the high ADR region, and a decrease in mix of business in Asia, which is a lower ADR region.

Excluding regional mix effects, constant currency ADRs were up approximately 4%, driven mainly by rate increases in North America and in Europe. The increase in North America were driven by high levels of demand for beach-oriented leisure destinations. And in Europe, were driven by a higher mix of summer bookings, which have higher ADRs.

Airline tickets booked in the second quarter were up 120% versus 2019, driven by strong growth at Priceline and by flights bookings at Booking.com and Agoda, neither of which had flight crops in Q2 2019. We are encouraged to see another record freight quarter for air tickets booked through our flights business, which is a key component of our multiproduct connected trip strategy.

Consolidated revenue for the second quarter was $2.2 billion and decreased 44% versus 2019, which is better than our expectations. Revenue in the quarter declined meaningfully more than gross bookings due to bookings made in the quarter that are expected to check in, in future quarters at which point the revenue will be recognized.

Take rates in Q2 were about 10%, largely driven by these timing differences. As you'll recall, we discussed the impact of timing on take rates in Q1, Q2 and for the full year during our last call. We continue to expect these timing factors to impact full year take rates, although the second half of the year will be less negatively impacted than the first half of the year. Removing the impact of timing, our take rates on accommodation bookings in Q2 was stable versus Q2 2019.

The better-than-expected top line performance resulted in adjusted EBITDA of $48 million in the second quarter, which came in better than our expectations. With the exception of Q3 last year, this is the first EBITDA positive quarter since the first wave of COVID.

Marketing expense, which is a highly variable expense line, decreased 29% versus 2019. Marketing expense declined more than gross bookings due to higher ROIs in the pay channels and the increase in our direct mix. Sales and other expense in Q2 were significantly higher than they were in Q1 on a dollar basis. Sales and other expenses as a factor of revenue in Q2 was better than our expectations due to lower-than-expected bad debt and customer service-related expenses.

Personnel expenses in Q2 were higher than they were in Q1 on a dollar basis primarily due to the $136 million of expenses related to our decision to repay the government aid in the second quarter. Excluding this repayment, personnel expenses in Q2 would have been in line with our expectations. G&A and IT expenses were both higher in Q2 than they were in Q1 on a dollar basis and were in line with our expectations. We recorded a non-GAAP loss of $105 million in the quarter.

On a GAAP basis, we had an operating loss of $56 million in Q2. We recorded a GAAP net loss of $167 million in the quarter, which includes income tax expense of $126 million. On a GAAP and non-GAAP basis in Q2, we recorded a tax expense on a pretax loss due to higher earnings expectations for the full year relative to our expectations for Q1. For the full year, we expect our GAAP and non-GAAP tax rates to be slightly higher than in 2019.

Now on to our cash and liquidity position. Our Q2 ending cash and investment balance of $16.1 billion was down versus our Q1 ending balance of $16.4 billion. However, our Q1 ending balance benefited from the timing of the $2 billion raised in our Eurobond offering, which we completed in March and the subsequent redemptions of the 2 higher coupon senior notes occurring in April.

Adjusting our Q1 ending cash balance for the redemption of the 2 notes that happened in April would have resulted in an adjusted Q1 cash balance of $14.4 billion. Our Q2 ending balance was higher than this adjusted Q on balance, primarily due to operating cash flow of $1.2 billion and a $0.5 billion unrealized gain on our long-term investments.

$0.2 billion of operating cash flow in the quarter was driven almost entirely by a change in working capital. Change in working capital represented a source of cash of $1.2 billion in the quarter due to the increase in our deferred merchant bookings and other current liabilities, partially offset by the increase in our accounts receivable.

We will continue to focus on maintaining a strong liquidity position, given the continued uncertainty created by the COVID pandemic. Of the $16.1 billion of cash and investments at the end of Q2, $4.3 billion was related to our long-term strategic investments and $11.7 billion was cash and short-term investments. We ended the quarter with about $12.3 billion in debt, which is about $3.6 billion higher than our pre-pandemic levels. We have a $1 billion convertible note maturing in Q3.

While return of capital to shareholders will be an important component of our value creation strategy in the future, we remain on pause and we'll wait to reinitiate until we believe each of our 3 major regions is beyond the risk of a significant reversal in trends due to COVID. We're not there yet, given the current trends we're seeing in Asia and with our current close watch on how things are developing in Europe.

Now on to our force for the third quarter. With the recent rising case counts driven by Delta -- by the Delta variants in many countries, some governments around the world have responded with new travel and leisure restrictions as well as some stricter vaccination and testing requirements for tourists. However, there are indications that optimization rates are lagging the recent increases in case counts, particularly in countries with high vaccination rates, which could be an important factor in how governments plan their responses to the recent increase in COVID cases.

We're closely watching the U.K. where the vaccination rate is high and the government has moved forward with relaxing travel restriction despite rising case counts in the country, which are among the highest in Europe. We're encouraged by the recent decline in new case counts and by the continued lower level of hospitalizations in the U.K. compared with other outbreaks. We saw booking trends improve in the U.K. in July, leading up to and after travel restrictions were lifted on July 19.

Our July room nights declined about 22% versus 2019, which was a modest pullback from the 13% decline in June, primarily to softening booking trends in Europe. Looking within Europe, we saw reductions in room nights in July across several of our key countries, including Germany, France and Italy. But despite the recent pullback in these countries at the end of July, we had a higher amount of gross bookings on the books for the remaining summer period in Europe than we did at this same point in time in 2019.

Outside of Europe, the U.S. continued to have a very strong room night growth in July, although modestly below Q2 levels. While Asia and rest of world room night declines were about the same in July as they were in June. Asia continues to be the least recovered region in July and continues to be down significantly from 2019 levels.

The change in growth rates from June to July were similar for domestic and international room nights, with domestic remaining positive and international room nights remaining down significantly versus 2019. Given the recent additional uncertainty around COVID, driven primarily by the Delta variant, it's difficult to predict exactly how room nights in August and September will compare with the 22% reduction we saw in July.

Turning to the income statement, we expect Q3 gross bookings to decline several points less than room nights, driven by expected improvements in reported ADRs and by flight bookings. We expect that the Q3 revenue decline will significantly improve from Q2, reflecting the strong improvement in bookings in the last few months.

I just mentioned we have more gross bookings for the summer than at this time in 2019 for Europe. The same is also true for North America. We expect our Q3 revenue as a percentage of gross bookings will increase meaningfully from Q2 due to the high concentration of check-ins expected in the third quarter and will be about in line with Q3 2019. As a reminder, the exact relationship between revenue and gross bookings in Q3 will be impacted by how our bookings trend in August and September.

We expect marketing expenses in Q3 will decline several points less than gross bookings as we expect to invest in capturing demand and increasing awareness during the peak travel season and ahead of the continued global recovery of travel demand. We expect sales and other expenses in Q3 to be up significantly versus Q2 on dollar basis due to higher gross booking volumes in the third quarter as well as a mix -- as well as an increase in the mix of gross bookings process on a merchant basis. However, we expect sales and other as a percentage of revenue in Q3 will be a bit lower than in Q2.

We expect our more fixed expense categories in Q3 in aggregate to be about in line with Q2 on a dollar basis. We expect Q3 EBITDA will be the highest since Q3 2019. In conclusion, we are pleased with our better-than-expected results in Q2, which benefited from a recovery in travel demand and also reflects the strong fundamentals of our business and a good execution by our teams.

We remain confident in the eventual full recovery of travel demand globally and we're looking forward to a strong summer travel season this year in North America and Europe. We'll continue to responsibly invest in our business to ensure we're well positioned for the full recovery of travel and for building a larger and faster-growing business that generates more earnings than prior to the pandemic.

We'll now take your questions. Maria, if you could open the line for questions, please.

(Operator Instructions) And your first question will come from the line of Kevin Kopelman from Cowen.

Great. Could you talk more about the key drivers of your competitive share gains that it looks like you're seeing in the U.S.? And would you say growth in the U.S. over the past few months has been driven more by traditional hotel or more by alternative accommodations?

So Kevin, why don't I take a little of this and I'll let David add anything that he thinks I didn't put in that I should have said.

So we are very pleased with the strong results that we're seeing in the U.S., and that is a result of a lot of hard work by our team. And a lot of it is just the blocking and tackling, making sure that we're getting the right inventory, the right price, doing the right marketing, presenting the right offer to the customer at the right time and doing a bunch of things that we've been doing for so long in terms of improving conversion, a lot of AB testing, all the things we've always done, blocking and tackling.

There's no silver bullet. There's no sense of, oh, this is the magic key to unlock extra demand in the U.S. It's just a lot of very good work. Now in terms of alternative accommodations, I talked a little bit about how we're pleased to getting more inventory there. We're pleased with gaining some share with some of our professional managers, that's going well.

But overall, it's everything that we do to improve our business, provide a better service to both sides of that marketplace, both the customers, give them a great offer, and working with all of our suppliers from the biggest international chains, to the small alternative accommodations and everything in between to make sure we're providing them with the demand they need to make their business successful.

And David, anything specific you want to add to that?

Glenn, I think you summed it well. Just in response to your last part of the question, Kevin. As we said, our business in the U.S. is more heavily mixed to hotels and alternatives than our global average. So therefore, the growth rate has to be driven by both. We can't have levels of growth that we're seeing without seeing strong growth in the hotel business.

Your next question will come from the line of Mario Lu from Barclays.

I'll ask you on alternative accommodation. So I believe you mentioned you're gaining share on the managed properties globally. So can you provide some color onto what drove the share gain, whether it was just mostly geo-based or specific actions that you guys made on your end? And then, similarly, if you could share what percentage of the 6.6 million listings are actually managed properties versus by individuals.

I got a little backwards on that. So we don't disclose the breakdown of all the different categories of our alternative accommodations, how many are professionally managed, how many are single property owners and everything in between. I have said in the past, and I'll repeat it, that one of the areas that we do need to add more to is the single property owners. So we know that is an area we need to focus on.

In regards to your other questions about how we're doing adding share? I think we spoke a little bit about that when I did the prepared remarks and what I just said now, is a lot of this is just working hard with people who are the property owners or the managers, people who have inventory, they want to get it built. It's a wonderful thing in this business that everybody knows an empty bed at night is zero revenue. You fill that bed and you get an incredible margin on it.

And they want to fill up their properties. So we're there to make sure that they know we're there to help fill that and do all the things I mentioned before, making sure that we can bring in that demand. And I'll leave David anything you want to disclose to them, but if it's any different, I'm not sure if there's anything further to add.

No additional data points on the mix of 6.6 million. What I would just clarify, Mario, is that the comment that Glenn made about gaining share gains with our larger professional managers that was really related to how we're doing in the U.S. That was a specific U.S.-related comment. I'm not saying that's not the case worldwide, but I'm just saying that if you remember, our strategy is to really get a lot of additional properties in the U.S. was to target more of the professional managers where there's a lot of high-quality supply available through those margins.

Again, single property-owned properties that are available via managers. So the comment we were making was there relative to what we're doing in the U.S.

And your next question will come from Naved Khan from Truist Securities.

A couple of questions. I think, David, I think I heard you say marketing efficiency was higher versus 2019, and I did some rough math off of the bookings number. It does look -- it was higher in the last quarters. Can you just maybe talk about what are the drivers of efficiency in marketing?

And then maybe one for Glenn. Maybe, can you just talk about some reports of maybe flash deals being tested on the site maybe later this year? And what kind of interest you might be seeing from hotels in terms of [first visit]?

So David, do you want to take that first one?

Yes. Let me talk a little bit about marketing what was going on. I think I understood the question. But basically, what you see was in Q2, our marketing expense, the decline of our gross bookings, which is the way to look at it. Because if you look at it versus revenue, you've got all the different books to say timing and things in there. So look at it as a potential of growth bookings.

That ratio improved in the second quarter. So said differently, our marketing expenses declined more than our gross bookings did. Two factors, we did see higher ROIs across many of our paid channels that we store a couple of years ago, but we also saw a nice increase in mix towards direct as well. So the more we have direct mix and obviously then the less we're paying for marketing to attract those new customers.

So I'd say nothing particularly unique to call out in that. The market environment is built -- I'd say isn't fully stackable yet, clearly. We're still going through a stage of recovery in the industry. So not all factors are directly comparable with where we were with 2019, but we were pleased to see that the ROI did improve. Now bear in mind what I said about the Q3 implies that the opposite is likely to happen in Q3. Obviously, we're only a month into Q3, but we are stepping up our marketing spend in Q3. It is historically, the quarter where we spend the most is the peak season.

We want to really capture demand, the customers that are out there. We'll also be increasing our spend -- brand spend in Q3 relative to Q2. So that's one of the contributors to essentially what will be some deleverage in marketing spend relative to gross bookings in Q3.

So we'll see how the quarter developed, but that's what we expect to happen in the third quarter. But in the first 2 quarters, we did see that benefit in ROI helping us create some leverage on the marketing line.

And regarding your question about flash deals, if you read the same article that I did, I think there was something in there about there was no official response from Booking. So let me say that I have nothing to say specifically about this, but let me reemphasize that we are putting a lot of effort into working cooperatively with our supply partners to get the incremental demand through all different ways, being a better merchandise or providing more value to both sides of the marketplace is where we're going towards.

And your next question will come from Deepak Mathivanan from Wolfe Research.

Great. I wanted to ask a little bit about your experimentation with a more diversified marketing strategy. Historically, performance marketing has been the one that worked the best for the space. How should we think about your approach with some of the other channels? Are there specific kind of objectives that you look for as kind of travel comes back to leverage into those channels?

And then the second question, kind of related to that, but how should we think about your market share during this time? I know it's sort of still early and there's not a lot of data points yet. But any color that you can provide on how your market share is trending in some of the markets where travel is recovering? That would be great.

So regarding diversifying our marketing efforts, you are absolutely correct. We built this company on doing a great, great job with performance marketing channels. So we know that. But what a lot of people don't know is that Priceline.com was very, very successful in its early days in brand marketing. And of course, we, Booking.com, we brought it the U.S. We had lots of brand marketing.

The key thing in any type of marketing program is making sure that you're getting the ROI that you want to get. And we are going to continue to do that. I mentioned in my prepared remarks about how we are looking at new channels like social channels you've seen, I hope, some of the things that we've been putting out. But we're going to continue to experiment and do all different ways to make sure we are reaching out for every pocket of demand.

But we always do it with the knowledge that it's got to be cost effective. It's got to produce, in the long run, because it takes longer for a brand to actually produce results, so we recognize that. But in the long run, it's got to produce the results that we want. And we're going to keep on doing that.

And I believe -- I really believe that in the long run, we will have many different ways that we're going to be bringing in marketing. One of the key things, though, before you start spending a huge amount on a brand marketing campaign is making sure that you have the priorities the way you want it. And that's one of the things that we are continuing to do and perhaps the alternative accommodations area is really working to make sure in the U.S., for example, that we've got that product the way we watch.

In regards to market shares, Look, we're very pleased with how we're doing right now. I haven't seen results from some of our competitors, and I'm not really going to go down country by country in terms of what our shares are. But I am pleased with the results that we're achieving, and I'll let David say if he wants to give any specificity.

No, Glenn, I think you said it well. We are pleased. We'll have to wait and see when the dust settles. I think market share is a better measure over the course of a year than over a quarter or 2 quarters. We're only obviously 2 quarters into the year. But we're pleased with things -- with how things are going. Our growth rates relative to what we see happening out there. But we'll count that and talk about that if we can and more at the end of the year when we've got some more concrete data on how the market actually developed.

And your next question will come from the line of Justin Post from Bank of America.

A couple of questions. I guess, first, David, can you revisit your comment about July travel bookings being better than summer of '19 and what that means for revenues in the third quarter. Any thoughts on that?

And maybe, Glenn, I think you said June was back to 13% from 2019. how does that make you feel about the confidence of a full travel recovery? And then when we do recover, maybe any high-level thoughts on could your market share be higher? Could your margins be higher than '19? Any new thoughts you have on either of those items would be really helpful.

David, do you want to take that?

Yes. Why don't I start with just to kind of make sure that you know what we're saying? So to recalibrate or restate what we said. We have more gross bookings for the summer for the remaining summer months than at the same time in 2019, i.e., at the same time, ended July in 2019 than we had at that same time, for Europe and North America.

So assuming that cancellation rates stay the same, then that would potentially result in more revenue in those markets for the summer months, for the remaining months. Now I did notice -- I did also point out that we will have -- that we have a higher percentage of cancelable bookings out there or refundable bookings out there than we had at the same time in 2019 as well. So there's obviously some risk that more of those bookings cancel than they would have done in the same period of time.

So it means basically that our potential revenue base for the summer is higher now in those 2 markets than it was in 2019. Now obviously, the offset there is what's happening in Europe -- sorry, in Asia and what's happening in rest of world. We don't have that situation. The revenue or the bookings on the books, which will be a potential revenue in those markets are substantially below where they were at the same time at the end of July in 2019.

Just I would say that nobody knows when this full recovery is going to happen. And I think everybody is able to throw a dart, but it's really throwing darts. Given all the variations that are happening, these new variants come out, we've seen the impact there. However, the thing that I continue to say is how much we are very confident. I think everybody is that this will end at some point, and we'll come out of it strongly.

Now we've talked about this a bit in the past about how we want to come out of this. We want to have a bigger business, making more EBITDA growing faster. But we've also talked a bit about margins where we have a commitment -- we want to be a leader in the -- the leader in the industry in terms of our margin -- our EBITDA margin, but we recognize that a lot of things that we're doing nowadays can actually end up with a lower margin. Obviously, air, for example. I mean, it's wonderful when we say 120% increase over 2019 in air ticket, that's wonderful.

But we all know, those margins are nowhere near what they are in the combination business, and I can go on with different examples. The key thing for us and for our shareholders, I believe, is coming back with more EBITDA doubts and continue to grow that -- our business, so there's more of that. That's the way we're looking at it.

And your next question will come from Doug Anmuth from JPMorgan.

This is Dae Lee on for Doug. One for Glenn. In your prepared remarks, you talked about better monetize the transaction flow. And that's one of the drivers that led to the creation of this fintech unit. So just curious what the opportunities that you are seeing there? And is this something that will affect user experience as well or more focused around the back end?

And then second one for David. A follow-up question on the sales and marketing ROI. You guys had called out higher ROIs last quarter and again this quarter. So I was curious if this is a result of something that you're doing differently? Or is this just more of an outcome due to the competitive dynamic on the performance advertising channels?

So in terms of the fintech unit, there are many opportunities for us with a flow in 2019, $100 billion worth of transaction flowing through there. We know that there are ways that we can save money for the consumers and the suppliers on both sides, and we can make good money on it, too, in the long run.

So it's coming up with things like providing better FX tools. It's things, for example, making sure that if somebody wants to pay in an alternative payment method, that we're able to provide that. It means making sure that we can do a better job with different types of regulations in terms of making sure that any type of transaction is not fraudulent. It's all sorts of things.

As a player at scale -- we can do things that many, many on their partners, that our partners could not do, and it's also providing conveniences to our customers who want it, but our suppliers can't do it on their own. And there are lots of different things that we can do. Even things as simple as our Fintech unit setting something up like our e-Wallet, an electronic wallet that enables us to easily provide value to that customer, which can come from a supplier or it can be something that we're promoting ourselves, all different things.

So what I see is not only the basics that we need to make our connected trip work, which, of course, you have to have that if you want to create a connected trip with a single price point, one person pays one amount. But it's all these other types of things that we can provide that others really can't do on their own right now.

Yes. And then on the second quarter ROIs, our playbook really has not changed, and our strategy in the performance marketing channels have not changed. We continue to seek high-quality traffic at the right price. And to us, high quality means high converting traffic, lower probability cancellation, which we watch very carefully, high probability is coming back to us on a direct basis. And those are the things that are kind of going to our bidding strategy plus all the dynamics that occur in each of those marketplaces.

So we expect there'll be ROI volatility throughout the recovery. Of course, we also look at what the available demand is and when there are time to kind of lean into this and lean in more. We are certainly looking at this as an opportunity to win customers onto our platform and get new customers, and we mentioned a very healthy mix of new customers on -- in the business quite similar to the mixture it was in Q2 '19.

So we're pleased with how things are going. We like the way that we're bringing back existing customers. We like the way we're winning new customers. We're pleased to see the existing customers come back generally much more directly than the new customers, which you expect as part of the playbook.

So I'd say many factors go into those ROI calculations, including cost per click, conversion rates, cancellation rates, and we're still in a period of relatively high volatility in each of those compared to where we were in 2019.

And your next question will come from Brian Nowak from Morgan Stanley.

This is Alex Wang on for Brian. Two questions. One, I think you mentioned that your monthly traffic on Booking.com is back to about 90% of pre-pandemic, which is encouraging, particularly with some of the headwinds you're calling out in Asia. Just curious if you're able to segment out any differences in behavior on that new versus existing cohort? And going forward, any strategies to sort of grow those bases separately.

Second question around air. Like, it sounds like there's a lot of momentum there. Just curious on your views on some of the remaining execution hurdles you see for that initiative? And any plan to sort of grow consumer awareness for the new air product?

So I'll take the air first, and we'll see what David wants to say or not say about breakup of our customer base. So air obviously, Priceline has been doing it forever, since the company started, but in the U.S. only. In Booking.com only got going this very, very recently. And I mentioned 24 countries, nice, but that's a lot of countries that we haven't touched yet that have to be done.

And there are so many things that need to be done to really improve the air product at Booking.com. And I'm not going to list them all, but I see a lot of ways we can make that even better than it is. And that's a little bit why we haven't done any real large marketing effort on the air side yet. We got to have -- talked a little bit before, before you start really marketing, so make sure you got the product the way you want it.

So it's so encouraging that we're doing well with it right now, even though I see a lot of things that still need to be done to improve that product. And something that we're really pleased with is seeing the attachment rate, again, is something that we obviously, the reason that we want to do this is not just to sell the flight ticket, it's to actually get some that is higher margin, those accommodations and build out that connected trip.

So it's nice to see at the very early stage, and I urge you to listen carefully, I say very early stage, we're seeing good results. So I do see a lot of opportunity here in the future, but we won't be bringing out any sort of marketing on it until I feel that we have the product where we want it to be.

And Dave, I don't know if you want to talk about the other one.

Yes, Alex, there's not a lot more to say. I think we wanted to be helpful give you some data points to understand how our active customers, the people who are actually actively booking on the site, have recovered. And the mix -- a healthy mix of new customers only slightly below what we saw in Q2 '19. I think those are good signs.

So I don't want to get into the segmentation within our customers. Obviously, there some customers who book very often. There's some customers who don't book quite as much, and we kind of love them all, but we want to make sure we kind of see it differently. And I'd say that's part of the reason why we have our Genius program. Of course, the more often you work with us or book with us on Booking.com, the higher you move up the Genius ladder and you get more benefits from the Genius program, which is I think a nice way of driving loyalty.

So other than the data points we've given out, which we thought would just be helpful, states on the ground so people understand what's going on within the business. I don't want to get into any further details.

And your next question will come from Mark Mahaney from Evercore.

I guess I'll just follow up on -- keep sticking with the air. And I guess learnings to date and talk about this, is this a new customer acquisition tool? Is this is a cross-sell product? Do you notice greater engagement with the people in those 24 markets where you've rolled that with booking? Does this increase the overall spend frequency of purchase? What have you seen so far? And maybe it's all too early, so I guess if you're going to respond that way, then I'll ask, when do you think you'll have a decent read into what impact, if any, permanent -- if it's had a permanent impact on your Booking.com customer base?

So (inaudible) what you said, like too early. But I will add a little bit more -- I'll add a little more to that. In a sense, I'm very pleased because we haven't been out really marketing this, yet new customers are coming to the site. And then we're seeing them get an attractive cross-sell that I really like seeing that as a sign that this is the right direction that we're going.

Now I had a good sense that was going to happen anyway because we've been seeing that happen in Priceline for 2 decades where that's been happening. So I have good confidence that this would happen. But it's something that I believe we want to have a lot more data before we start coming back here, we start showing you, here's what the attachment rates are. Here as it compares to Priceline's. Here's what we see versus industry and how many new people are coming from the air funnel versus the others?

I would just lead everyone with the sense that we're very pleased this is bringing us new customers. New customers who are buying not just flight tickets, but some of them are buying hotels, too. this is proving out a little bit of our long-term vision of this connected trip.

Now do I expect that to happen with something like activities? I'm not really think a lot of people are going to come for an activity first, then they're going to buy a flight or then they're going to buy a hotel. It's -- that's going to be a lot more the other way in helping produce the loyalty and the repeat business that we talked about. And then goes into all the things I was talking about in terms of using a wallet, so we can get credits and have different suppliers be able to promote different offerings in different ways to different customers.

All that spins together, increasing that flywheel. And that's what we're trying to achieve, and it's just so great to see it start happening right now, even though it's very early.

And your next question will come from Vince Ciepiel from Cleveland Research.

I wanted to talk a little bit about your perspective on leisure versus corporate. I believe you historically see about 80% of the business in leisure. So first, there. Some markets are running 15%, 20% ahead of '19 levels. And I'm curious your perspective on the long-term trend line within leisure and how COVID has changed that maybe people's ability to work remote? You mentioned more leisure business shifting online with that McKinsey study.

But just curious how you think about leisure over the long run? And then the second part is on the corporate side. Any early indications of recovery there and how you think that evolves through the second half of this year?

Yes. Let me give generalities about what I think about leisure versus business and the core business in general, what I think is going to be happening in the future. And David go back and talk about what we've disclosed in the past about the mix.

So you're right in the sense that we are much more leisure oriented. And much of our business travel is small business people. These are people who are doing their own travel. This is not part of the big travel management company operation. So it's a little bit different when we have business travelers and they act somewhat similar to leisure sometimes.

The fact is and we've all seen is that leisure is out of the gate much faster than any business travel, which makes perfect sense. Because on the one hand, you have business people who say there's a risk factor when you put people in travel, et cetera. And there's also the issue of cost. Why have people travel if you don't have to?

That second part, I think, is very key for the future of the business. I believe that there's still going to be resistance by CFOs, other people who are cost conscious in their businesses about, gee, do we really have to have all the traveling we did in the past? Maybe not because with these new technology and such, we seem to be pretty effective without having to send somebody from New York to London at $15,000 for a 1-day meeting.

And I think that's going to somewhat change how the business of travel is done and hence there'll be fewer people upfront in the plane and spending a lot of money on those very high-cost 5-star hotels, et cetera, which will change things a little bit. For us, we're leisure. So it's not going to impact us negatively so much. In fact, they help us as there's more availability that needs to get filled up.

The other thing you mentioned, though, which usually how it plays out, though, with more people being able to work from home and deciding, gee, I think, Friday, I'm going to work from somewhere else and have a Friday, Saturday, Sunday or a Thursday, Friday, Saturday, Sunday mini holiday summer working on Thursday, Friday, but in a different location. How much is that going to actually build more travel? Uncertain at this time since everybody is still shifting around, what's the way to work in the future? How many days are we in offices? How many days are we not in offices? Nobody knows the answer to that yet, and it's going to take a long time to play out.

But I do hope that we'll, hopefully, build out more travel, we like more travel. So there's a lot of uncertainty about this and nobody really knows. And even as much as I've heard, the encouraging signs from some of the suppliers, particularly some of the airlines, I saw a Deloitte report came out a couple of days ago, maybe yesterday, in which the expectation of corporate travel was not as optimistic in the near term.

So I think we have to say we don't know. We'll find out as it rolls out. And David, I don't know if you want to give anything more to that in terms of the numbers for us in business travel because I heard a number he quoted, I'm not sure if that was right or not.

No. The number -- first of all, we haven't given that recently, but we're over 80% leisure. And as we said, our business -- first of all, it's a self-declared metric when you make a booking, so it's hard to be precise about. But the type of business travel, we do have is much more unmanaged business travel as Glenn explained. So we're heavily biased towards that leisure segment.

And our last question will come from Jed Kelly from Oppenheimer.

Great. Two, if I may. Just as you sort of -- you mentioned earlier in your prepared remarks that you're gaining share with professional property managers. Can you sort of talk about like your supply strategy heading into this winter in the U.S. trying to increase that single unit inventory? And then can you give us an update just on how APAC is trending, particularly with Agoda?

I'll let David talk what you want to say about Asia in general. And regarding this year, I just want to be a little careful. We said we are pleased with that, we are doing better. We're gaining share with the professional managers. And I like what we're seeing there. It's going to be a long haul in terms of building out the U.S. inventory for all types of alternative accommodations, whether it be professionally managed getting as many as we want there or single property, it's going to be a while to get to where we want to be in that.

I just want everybody to understand that this is a goal that we're working on hard by having the boots on the ground, talking to managers, sending out the right information, getting people to understand why we have a great proposition for them. And I'm confident we're going to do that. When you look at what we've achieved in Europe, you look at the share of our alternate accommodations in Europe, you say, boy, that's a goal to have in the U.S. too. We should be pushing for that. There's no reason we shouldn't.

Now customers are similar. The proposition is similar. There's nothing that we shouldn't be -- no reason we shouldn't be able to achieve that over time, but it's going to take time. And Dave, I don't know if you want to talk about Asia a little bit there.

Yes. So Asia, APAC, I mentioned that the room night growth was worse in Q2 than it was in Q1. So that region deteriorated, counteracting some of the benefits and strength we saw in Europe and North America. Of course, both Booking and Agoda are sizable businesses in APAC, although clearly that's the majority of Agoda's.

So the whole region is very depressed. As you know, vaccination rates are lagging in most parts of Asia, also response to COVID outbreaks tends to be more aggressive and restrictions are put in place more quickly based upon outbreaks in the Asia region across almost all countries. So travel level is very low, particularly in international travel levels exceptionally low and still a long way to go and no recovery in Q2. In fact, worse in Q2 versus Q1.

And that concludes our Q&A. I will turn the call over back to Glenn Fogel for closing remarks.

Thank you. So in closing, I want to reiterate our strong belief that our industry's full recovery will be hastened by everyone who can get a vaccine going out and getting it. We urge all people who are approved for and medically able to be vaccinated to do their part to make our society safer and go out and get a vaccine.

And as always, I want to thank our partners, our customers, dedicated employees and our shareholders. We appreciate your support as we continue to build on the long-term vision for our company. Thank you, and please be safe. Good night.

And that concludes our conference call. Thank you for participating. You may now disconnect.