Expedia Group Inc (EXPE) 2016 Q4 法說會逐字稿

Good day and welcome to Expedia's Q4 2016 earnings conference call. Today's conference is being recorded. At this time, I would like to turn the conference over to Alan Pickerill, Vice President, Investor Relations, at Expedia. Please go ahead, sir.

Thank you. Good afternoon, everybody. Welcome to Expedia, Inc.'s financial results conference call for the fourth quarter and full year ended December 31, 2016. I'm pleased to be joined on the call today by Dara Khosrowshahi, Expedia's CEO and President, and Mark Okerstrom, our CFO and EVP Operations.

The following discussion, including responses to your questions, reflects management's views as of today, February 9, 2017, only. We do not undertake any obligation to update or revise this information.

As always, some of the statements made on today's call are forward looking, typically preceded by words such as we expect, we believe, we anticipate, or similar statements. Please refer to today's press release and the Company's filings with the SEC for information about factors which could cause our actual results to differ materially from these forward-looking statements.

You'll find reconciliations of non-GAAP measures to the most comparable GAAP measures discussed today on our earnings release, which is posted on the Company's IR website at IR.ExpediaInc.com. I encourage you to periodically visit our Investor Relations site for important content, including today's earnings release and the updated investor deck.

Finally, unless otherwise stated, all references to cost of revenues, selling and marketing expense, general and administrative expense, and technology and content expense also excludes stock-based compensation and depreciation expense, and all comparisons on this call will be against our results for the comparable period of 2015.

As a quick reminder, we sold our 62.4% ownership stake in eLong on May 22, 2015, which was previously a consolidated entity of Expedia, Inc. For GAAP accounting purposes, the results of eLong are included in our results through the date of the sale. In order to allow investors to compare our full-year 2016 results on a like-for-like basis with our historical results, full-year commentary, and the earnings release, and on this call is principally focused on our results excluding eLong, which should be considered in addition to the GAAP results on a fully consolidated basis.

Fourth-quarter results do not contain any results related to eLong in either year.

With that, let me turn the call over to Dara.

Thanks, Alan.

2016 was a year of ups and downs for Expedia. We learned some valuable lessons on things we can do better, but we also chalked up great performances across the portfolio, including from some of our largest acquisitions, brands and businesses that will play a big part of our profits, cash flow, and overall growth moving forward. We exited 2016 with nice momentum and we start the new year from a position of strength.

For the full year 2016, our customers booked over $72 billion of travel products. We delivered full-year revenue of $8.8 billion, up 32% year over year; adjusted EBITDA over $1.6 billion, up 39%; and 246 million room nights, up 21%.

2016 also marked the completion of our most significant integration yet, that of the Orbitz family of consumer brands, as well as Orbitz for Business. I'm very proud of the entire team for their considerable efforts, which paid off in the form of $224 million of adjusted EBITDA in 2016, reflecting synergy realization meaningfully ahead of our expectations and overall profitability clearly better than what Orbitz could have delivered stand-alone.

As they say, however, no deed goes unpunished -- good deed goes unpunished, and we did pay a cost related to these migration efforts in the form of deceleration in room night growth in the middle of the year. Our teams are now completely refocused on our operational formula, high-velocity product experimentation driving conversion, aggressive optimization of our marketing channels on a global basis, expansion of our supply portfolio, and growth of our repeat user base, all in a continuous and self-reinforcing cycle.

We're seeing the output of our operational formula and improved room night growth particularly at Brand Expedia, Hotels.com, EAN, and Egencia, with the momentum continuing into January. In the meantime, we're optimizing our marketing formula for regional brands, including Travelocity, Orbitz, Hotwire, and Wotif with an eye towards improved marketing efficiencies and a bias towards the bottom line.

HomeAway has now been a member of the Expedia family for over a year, and our efforts to transition to a global e-commerce-enabled alternative lodging marketplace are on track. We've implemented the traveler service fee, announced the elimination of the tiered subscription model, and substantially increased our investments and capabilities in online marketing, product, and technology. We're on track financially with $163 million of adjusted EBITDA in 2016 and a very aggressive investment plan in 2017, on our way to the $350 million EBITDA target in 2018. The HomeAway team is hitting their marks, conversion growth is healthy, renewals and listings are on track and improving, and we're looking forward to continued success over the long term.

trivago had another great year in 2016, delivering standalone revenue of $836 million, up 53% year over year, and accelerating in both the third and fourth quarter. Team trivago didn't miss a beat in their execution cadence, even as they completed their IPO. We look forward to continued global expansion, market-share gains, and improving profitability in the years to come. I will also note that trivago plans to have its own conference call in a couple of weeks, where they will discuss performance and their outlook for their business in more detail.

Egencia delivered a solid 2016, with revenue up 16% and adjusted EBITDA up 18%, aided by the addition of Orbitz for Business. And we're particularly pleased with the strong Q4 EBITDA growth of 97%. Egencia is now a scaled corporate travel technology and service provider, with $6.4 billion in gross bookings, which the team is aiming to double over the next four years. We know that our corporate travel products and technology are best in class and we're working to dramatically improve our capabilities on the expense management side.

With integrations complete and our leading product offering in place, we plan to ramp up the Egencia sales force over the next few years to gain better global coverage and accelerate organic top line growth. And, of course, we'll continue to be opportunistic on the acquisition side.

Overall, I'm pleased with the momentum that our team built through Q4 and carried into the start of 2017. These are interesting times for all of us, but with a $1.3 trillion opportunity in the global travel market ahead of us, we're executing with a cadence and discipline that gives me confidence in our ability to deliver a great 2017 and beyond. Mark?

Thanks, Dara.

We finished the year with a solid fourth quarter that was largely in line with our expectations. Gross bookings were up 8%, revenue up 23%, and adjusted EBITDA up 58%. Excluding both Orbitz and HomeAway, these measures grew a healthy 12%, 17%, and 15%, respectively. Hotel revenue was up 13% for the quarter, driven by room-night growth of 15%. Excluding Orbitz, room nights grew 16%, with our global brands, Hotels.com, Expedia, EAN, and Egencia, each growing room nights at an even faster rate.

Average daily rates were down less than 1% in Q4, while revenue per room night decreased by less than 2%. The gap between the two was just over 110 basis points, which was better than we had expected as a result of favorable package margins and some favorability in our loyalty program costs in the quarter. Note that as we look forward into 2017, we do expect that gap to be a little bit bigger.

Advertising and media revenue, net of intercompany amounts, grew 43% year over year to a sizable $807 million for full-year 2016. For the fourth quarter, ad and media was up 36% year over year, driven by strong performance at trivago. On a standalone basis, trivago revenue grew 65% in the fourth quarter, accelerating from 57% growth in Q3. This is a big brand that is building impressive scale on a global basis.

For full-year 2016, HomeAway delivered approximately $6 billion of online gross bookings, up 46% year over year, driven by over 22 million online room nights. Note that we plan to disclose these metrics for HomeAway on a quarterly basis beginning in Q1 of 2017 and plan to disclose the 2016 comparable numbers ahead of our Q1 call.

In the fourth quarter, on a standalone basis, HomeAway revenue grew 30% and adjusted EBITDA grew 16% as we started to ramp additional marketing and technology investments.

To help you think about how to model HomeAway going forward, I did want to provide some additional color. For full-year 2016, subscription revenue comprised just over half of HomeAway's revenue, and we expect this to decline to just about 25% of the mix in 2017. As a reminder, we lapped the elimination of the tiered subscription model at the beginning of Q3 2017.

Transactional revenue grew just over 240% in the fourth quarter. As we look forward into 2017, we expect that revenue stream to grow nicely as we continue to benefit from the traveler service fee, although we do expect to see deceleration in this growth as we lap the initial rollout. Note also that about 5% of HomeAway revenue represents ancillary services, such as advertising and software, which we expect to grow much more moderately.

Our expense trends in Q4 moved closer to our target P&L as we began to lap over the impact of the Orbitz and HomeAway acquisitions. Cost of revenue grew nicely slower than revenue as we continued to drive for operational efficiency. Direct selling and marketing expenses grew faster than revenue as we continued to ramp up our variable marketing channels and push for global growth. Growth in technology and content expense remained ahead of revenue growth, but decelerated nicely from the growth seen in Q3 and a bit more than we had expected.

General and administrative expenses were down year over year as we lapped certain deal costs in the prior year.

From a capital deployment standpoint, we were pleased to be able to repurchase 4 million shares in 2016 for a total of $436 million and increased our quarterly dividend to $0.28 per share for payment in March.

Before I move to our guidance for 2017, I wanted to cover our efforts to migrate certain components of our technology infrastructure into the cloud computing environment. This is a significant and important effort that began in earnest in 2016 and resulted in total direct costs of approximately $40 million. We have aggressive plans for further migration in 2017 and currently expect nearly $110 million of direct cloud spend for the year.

We expect the incremental spend year over year to be largely offset by lower infrastructure CapEx requirements in 2017, and we expect this entire effort to be cash flow accretive within the next couple of years. Although the cloud migration is a profitability headwind in 2017, it makes strong financial sense and will result in better overall performance, resiliency, and consumer experiences.

Turning to our financial expectations for full-year 2017, on a consolidated basis, including the ramp-up in cloud spending, we are expecting adjusted EBITDA growth of 10% to 15%. Excluding cloud expenses, growth would be 14% to 19%.

In terms of the shape of the year, as usual, due to the seasonality of our business, we are expecting the majority of our adjusted EBITDA dollar growth to come in the second half of the year. As a reminder, we invest in selling and marketing to generate bookings ahead of the busy travel season, with the revenue recognition occurring at the time of travel, and as such, we usually see some pressure on earnings early in the year, with the upside coming disproportionately in Q3, the busiest time for travel.

In addition, the shift of Easter into Q2 this year will result in a negative impact on revenue and profitability in Q1.

With sizable integration work behind us, we're pleased to return to business as usual and expect to manage our expenses largely in line with our target P&L. We expect adjusted cost of revenue to grow nicely slower than revenue, while selling and marketing should grow faster as we continue to push for global growth. We expect technology and content expense to grow slightly faster than revenue, as leverage in our ordinary course tech spending will be more than offset by growth in our cloud spend, outlined earlier. And of course, general and administrative expenses should grow solidly slower than revenue in 2017.

Now I'd like to provide some additional color on trends in our capital expenditures. Excluding costs related to our new headquarters, for full-year 2016 approximately 60% of our CapEx was capitalized software development, 30% related to data center and infrastructure, and the remainder related to non-HQ global real estate projects.

For 2017, we expect this CapEx to be flat to slightly down, with a decrease in data center and infrastructure spend offset by growth in capitalized software development and non-HQ real estate CapEx.

Separately, in terms of our new headquarters, we now expect total billed costs of around $650 million, of which approximately $30 million was spent in 2016 and less than $100 million will be spent in 2017. The remainder will be split roughly evenly between 2018 and 2019.

Lastly, while there are many moving parts, we believe that an effective tax rate in the mid-20% range remains appropriate for forecasting purposes.

With that, let's turn to questions.

(Operator Instructions). Naved Khan, Cantor Fitzgerald.

On HomeAway, I think there is some talk about pulling in 20,000 or so of the rooms and deploying them into the core Expedia. What kind of pace do you expect for 2017? And of the 20,000 that you did pull into the core OTA, can you share some of the results in terms of what kind of performance you saw on the conversions and bookings?

Hi, Naved. As far as the 20,000 coming into the core Expedia, the overall production within our numbers was de minimus. It was a small number. We very much go forward on a test and learn basis. This is -- we are in very, very early days.

And as we move throughout 2017, we will add in more properties. We want to make sure that those properties are instant bookable, so to the extent that you are a supply partner of HomeAway's that is providing instant bookable properties, you will get more exposure not only on HomeAway because of their sorts, but also on Expedia.

Note that right now the properties are starting up on Expedia and then we'll roll them into Hotels.com and some of our other brands, but we will be rolling them in through the year. It's too soon to tell as to whether they will be a significant portion of our production this year, but we certainly think from a long-term perspective having this inventory as part of our marketplace is of very, very important strategic -- importance for us going forward.

Okay. That's helpful. And then quickly on the room nights side, I guess you had an easier comp because of the Paris attacks in Q4 of 2015. So if you sort of adjust for that, what kind of trends did you see? And I think on the last quarterly call, you did talk about trends you saw into October. Can you provide us of those trends you might be seeing in January (multiple speakers).

In general, certainly Paris was a tailwind for us this year, and when we look at our volumes in Paris and France, they're up pretty substantially on a room night basis. ADRs are down, so, if you noticed, our international ADRs are down on a Q4 basis, so Paris and London were negative on an ADR basis, but certainly positive on a volume basis.

But when we look at our portfolio broadly, the big brands, the Expedias, the Hotels.com, the EANs of the world, continued to perform more strongly in Q4 than Q3 and the trends that we see in January are constructive for us.

Naved, I'd also just add, too, that notwithstanding the easier comp from Paris, we actually had a 300 basis-point harder comp just on a sequential basis moving into Q4, so we view Paris and the harder comp as essentially offsetting each other.

Okay. That's very helpful. And any color commentary on the trends year to date?

Constructive trends. So, in general, we saw momentum get better Q4 relative to Q3 and Q2, and January is so far, so good.

Paul Bieber, Credit Suisse.

Thanks for taking my questions. There are two quick questions. How should we think about the conversion-rate opportunity on Expedia.com in 2017, given your plans for allocating engineering resources to different projects? And then, any early takeaways from your tests with TripAdvisor?

Sure. As far as the conversion opportunity, it's pretty similar to what it's been in the past as it relates to Expedia, Hotels.com, all of our significant brands, and for us, the overall growth opportunity is -- it's conversion, it's higher conversion allowing us to reach into additional marketing channels, so it's conversion and audience growth, combined with supply growth and repeat growth, makes for a pretty good formula where if you modestly increase each part of that formula, you can get into pretty good room night volumes.

So, we're constructive on conversion. Obviously, the shift to mobile continues pretty quickly on a global basis, so that's a headwind, and also the shift internationally for us in general, our domestic brands convert at higher rates because we have a higher percentage of repeat customers than our international brand, so the shift to mobile and the shift to international points of sale, especially APAC, is a conversion headwind. But when you reverse those trends out or you look on a same-store basis, conversion trends in general are looking good and we think that we can kind of continue on the test-and-learn path that we've had for many years and kind of get going on the same formula that carried us here over the last four to five years.

As far as TripAdvisor goes, it's very, very early. We've ramped up on TripAdvisor. I'd say the volumes and the impressions are relatively limited at this point, for various reasons, so we're not going to have exposure to, call it, 100% of the TripAdvisor instant book audience. It will be substantially less than that. So at this point, it's not a significant factor for us, but it's a positive factor in our volumes early on and we hope to build on those volumes as the years progress.

And as a reminder, as TripAdvisor was rolling out instant book and we were not a participant, the headwind was between 100 and 300 basis points on room-night growth, so you can think about that as the possible maximum opportunity on a go-forward basis should we have full participation, which we're not doing right now.

Yes, we're not going to have full participation this year or next year.

Brian Fitzgerald, Jefferies.

Hey, guys. This is Alex Giaimo on for Brian. Thanks for taking my question. Can you just give us a little color on your cloud migration efforts? Maybe just how much of your operations are moving to the cloud and how far into the process are we? Then maybe, if you could, some color on the specific cloud providers you're using. Thanks.

Sure. As far as the cloud migration efforts go, we're, I'd say, early to midstream. There are parts of your stack that you can easily take to the cloud and we've certainly done that this year. The teams have worked very, very hard to refactor the code and architecture to make that possible.

One of the issues as far as cloud migration goes is to make sure that the applications that you take to the cloud aren't particularly chatty with your mainline data centers. The big cost becomes the transfer cost of data between your cloud centers and your mainline data centers, so we are migrating keeping that in mind. The teams are really doing a good job of taking parts of the code base to the cloud, then optimizing them from a cost basis, and then continuing to take other parts of the code to the cloud.

The significant increase in movement to the cloud reflects at least the plans for us to lift a significant part of our lodging search stack onto the cloud. Our significant cloud provider is AWS at this point, although we are looking at making sure that we have not an overall dependence on one player. So, as you can imagine with the volumes that we are running at right now, we have quite a few people calling us and offering their services and we'll determine whether we take them up or not, but right now the AWS team has been really terrific as far as their partnership goes. The engineering teams are working really well together, and I think that over a period of time this year, we will shift pretty significant parts of our throughput to the cloud.

Mark indicated the estimate for how much we're going to take to the cloud. These are estimates at this point, and we may wind up being successful in taking a higher portion of our code base onto the cloud or we may not be successful, and at this point, this is a pretty speculative effort, but we just wanted to give you as much -- kind of as much of a look ahead as we can and we'll update you on a quarterly basis as to how we're doing.

But this is unquestionably really, really good efforts by our engineers, and we think it's going to result in a consumer experience that's going to be substantially better.

Douglas Anmuth, JPMorgan.

Two things. First, Mark, I was hoping you could talk a little bit more about the margin trajectory for HomeAway. And in particular, you talked about the substantial investments in 2017. Can you just give us some more color on where those dollars are going to go, and then, obviously, kind of how you come out of that into 2018 if you work towards that $350 million?

And then, Dara, just to go back to your comment on the Trip Instant Book, just curious when you talked about not being at full participation this year or, it sounds like, in 2018, and I guess it's maybe a little bit surprising to me, but just wondering why that's the case. Thanks.

Hi Doug, so in terms of the margin trajectory for HomeAway, I would expect margins to essentially start to contract in 2017 and then rebound in 2018. In other words, I wouldn't draw a straight line between 2016 adjusted EBITDA and the $350 million that we laid out. We will be making some significant investments in 2017, which will ultimately then annualize in 2018.

The investments are predominantly going into technology and product, as well as sales and marketing, and you started to see a little bit of that impact in Q4 and that's just going to get even heavier as we move through 2017.

I think on HomeAway, too, when you look at our seasonality, as you know a significant amount of our stays happen in Q3, over the summer period, and I think for HomeAway you're going to see it be even more concentrated in Q3.

So the expenses and the negatives on subscription revenue kind of hit us consistently on a quarterly basis, and then the significant offset -- the positives on the stays are really going to be focused on Q3. So I think Q1 and Q2 are going to see some margin pressure. I think Q3, we will see a very significant payoff on the investments that we're making there.

You know, as far as our Instant Book participation goes, we're testing and learning. Its early results are good. As far as our ability to participate and the volumes there, I think you'll have to ask that question of the TripAdvisor team.

Jed Kelly, Oppenheimer.

Can you provide a little more color on HomeAway's performance marketing? Like what stages are we in terms of being a more aggressive advertiser in the US? And how does that domestic progress currently compare to initiatives in other regions?

I'd say that we're pretty early. We've actually brought in some excellent talent to really build up that team, some talent from within the Company.

And I think that HomeAway as it's transitioning from a business that -- whose economics were based on subscriptions and subscription renewals, and there's a certain marketing strategy to drive subscription renewals, to a business whose strategy is based on driving transactions, that's our bread and butter as it relates to our OTA brands and we have some good talent there. We're making some investments in data infrastructure and kind of the tooling necessary to be able to do this at scale. And I'd say we're pretty early in the process and we think that there is plenty of improvement ahead of us.

The other factor, obviously, as it relates to HomeAway is that as we move the HomeAway inventory onto Hotels.com/Expedia, we think that Hotels.com and Expedia and Travelocity's ability to bid in these variable channels, especially in the HomeAway's strength, inventory markets improves as well. That should happen late this year towards 2018.

Ron Josey, JMP Securities.

Dara, can you talk a little bit more about your international booking plans for this year? I think you mentioned in the past that was an investment focus for you all and wondering how you're thinking about jumpstarting growth there.

And then, Mark, I think you mentioned room nights on Hotels.com, Expedia, EAN, Egencia all grew faster than the 16% organic rate. Any additional color you can provide there? Like maybe high teens or 20%, that would be helpful. Thank you.

Sure. As far as our international booking plans, as you can imagine with the integration of Travelocity and then Orbitz, there was just a significant amount of work done as it relates to our brands and it was work that was focused domestically.

So I think in general when we look at our marketing investment plans and capital allocation plans, we're certainly going to allocate some incremental capital that's international capital, but the fundamental, call it, operating formula that we have in place isn't going to change. It's just that we have an opportunity now to focus in international markets, in growth markets, and read Europe and especially the Asia-Pacific regions as being the top two that we're going to be focused on. We saw decent trends in Q4 and we expect those trends to improve as we go into 2017, based on our expectation of the teams really focusing on these markets and really executing.

So, it's not anything substantially different, other than the opportunity for the teams to really focus and work through these opportunities.

And then, Ron, on the room night question, collectively the global brands grew a couple hundred basis points faster than the rest of the portfolio. I would also just call out the room night disclosure that we gave on HomeAway, and if we had included the HomeAway numbers in our 2016 full-year results, that would add another couple hundred basis points to the overall room night growth as well.

Got it. Thank you.

Mike Olson, Piper Jaffray.

A couple of questions, if I could, just to clarify on the increased spend in 2017. Outside of the cloud migration, does it essentially come down to marketing and product development spend on Expedia, Hotels.com, and HomeAway? And if that's the case, how should we think about the investments you're making in 2017 relative to the impact on night growth -- organic room night growth to be able to accelerate in the year based on that investment? Thanks.

Well, to answer your first question, those are the primary areas of investment. It is cloud and then it is continued spend on sales and marketing, both in terms of direct sales and marketing for our core OTA brands, as well as trivago and HomeAway, and then also with respect to our hotel market management team to continue to scale that up and be able to add new properties.

In terms of the impact to room night growth, listen, this is part of the formula. We're happy with the momentum that we've got right now, and our goal is to -- if we can do it, do better, but we're not going to guide on the room-night growth trajectory for 2017.

And I do want to be clear on the investments that while we continue to invest in our businesses and drive growth aggressively, we do think that we're going to leverage nicely on a fixed-cost base -- up sales, G&A, even R&D if you don't include cloud spend should leverage pretty nicely.

We're aware that we've invested in these categories pretty aggressively, and I think that now as a business we are in a nice position to leverage across our fixed-cost base, and on sales and marketing, whether it leverages or deleverages is really a factor of the channels that we're able to reach into and whether they are domestic or international or not. And if anything, we want our sales and marketing to some extent to deleverage because it means we're able to reach into some interesting new channels and usually interesting new channels tend to be a little less efficient than direct channels, but is a key factor in our growing our repeat base, and we think that if we can drive sales and marketing efficiently across these channels and then leverage the fixed-cost base of the business, we have a pretty good P&L ahead of us.

Justin Post, Merrill Lynch.

First, it looks like conversion rates kind of stalled out, and maybe that was due to Orbitz integration. Can you talk about your pace of conversion-rate improvements and could that pick up this year as you enter the important summer travel season?

And the second thing is on the cloud migration. I haven't heard other companies really have to see a big expense ramp as they migrate to the cloud. What's kind of unique about your ramp? And will we see leverage in 2018? Thank you.

Sure. As far as conversion, we don't talk about our conversion rates too specifically, but I do think that we were pretty open about our product conversion increases that had been a part of our significant growth that we've seen for the past four to five years not being where we want to be in the middle of the year.

So as we have focused on the core product, as we put the integrations behind us, we are in general more pleased with the trends that we are seeing. They're early trends, but you can certainly see it in the room-night trends in Q4 and our at least discussion or early discussion about January rates.

And as far as our cloud migration efforts versus other companies, I can't really speak to some of the other companies' efforts. I think that many of, call it, the new-generation companies kind of start up in cloud from the very beginning. I think that we are one of the, I'd say, few large-scale technology companies that is making this transition very aggressively. That's certainly what we hear from some of our cloud vendors.

So I think we're a little bit ahead of the pack here, and while it hurts our EBITDA as a print, so to speak, when we look at our net cash flows and when we look at kind of the long-term CapEx of the business and the operating expenses of the business, we think that cloud migration is going to be very, very important for us going forward and definitely a net positive.

And maybe one follow-up, just thinking about room-night growth. You talked about a continued growth on Hotels.com and Expedia and maybe slower growth on Orbitz and Travelocity and a few of the other brands. Can you give us any sense of the fast growth mix versus the slower growth mix as a percent of room nights?

Not specifically quantitative, but the big global brands are significantly larger, and you can back into a rough number based upon my commentary around 200 basis points of incremental growth on those global brands versus the overall portfolio.

And also keep in mind that the global brands are able to participate in Asia-Pacific markets, European markets that are faster growth by nature, so the regional brands, the expectations are for them to grow to the extent that they do closer to the domestic market, which is a little more mature than the international markets.

(Operator Instructions). Heath Terry, Goldman Sachs.

Can you give us a sense of the traffic dynamics at HomeAway? Some of the third-party reports are suggesting that traffic growth was negative in the fourth quarter, at least in the US part of the business. I guess, first, is that accurate? And second, is there anything structurally or from a market-focus perspective that explains that? And, I guess, just how do you think about the importance of traffic growth as you are trying to increase monetization on the platform and more of the inventories particularly -- potentially moving onto the Expedia platform? And then, just any update on what you're seeing in the direct booking pricing efforts by the hotels?

Sure. As far as the traffic dynamics of HomeAway, we have been obviously measured by third-party reports for many, many years. We have attempted on numerous times to compare those third-party reports with our internal metrics, and I think we gave up about 5-1/2 years ago.

So we honestly don't pay too much attention to those third-party reports. Certainly within -- on a quarterly basis, we look at them over a long-term period, but they often disagree with our internal metrics and I have no idea why and we're not going to spend a bunch of time trying to figure out exactly why.

What I can tell you is that our internal metrics indicate with HomeAway that traffic is up year on year. We have talked about natural search, Google natural search, being a headwind in general and I think a lot of people have talked about that as Google kind of takes more and more of their screen space and monetizes that screen space, and that's certainly their right. But players like ourselves don't have much of a choice because of Google's market power. So, that has been a negative for HomeAway, but we are able to offset that through very effective marketing on the brand side and in general just a better service that's attracting more people into a lodging category that in general is growing.

So, we're pleased with the progress of HomeAway. We always want traffic growth to accelerate and we'll certainly try to make it accelerate over the next couple years. But at this point, it is positive and we're satisfied with the results there.

And as far as the direct booking efforts by some of the chain hotels, no big news. I think not much significant has happened. You can certainly see from our room-night volumes that our results have not been significantly affected by those efforts one way or the other.

We do continue to have independent hoteliers be a higher and higher percentage of our overall booking. We think that is healthy, and we continue to really work constructively with a lot of these hotel companies to really look at value-added ways in which that we can work with them, whether it is trying to drive more direct traffic through them as a marketing channel; whether it's just to drive bookings, bookings, bookings; or it's actually to work with them on a more core technology basis to help them increase their effectiveness at far, far lower costs, based on the investments that we're making in our global distribution stack that they can take advantage of.

I think the conversations are rich. I think things have settled down a bit, and we think that there are certainly going to be many opportunities for us going forward.

Mark May, Citi.

Chris Merwin, Barclays.

Thank you. So I just wanted to ask about margins a bit. On a consolidated basis, it looks like you've guided to relatively flat margins this year at midpoint as you reinvest in the business. But when we think about the core OTA business, margins had been coming down slightly in the last few years, but finished up in 2016, so how should we be thinking about the normalized margin trajectory of the core going forward?

And then just quickly, for HomeAway, you reported online gross bookings of $6 billion. If we assume total are still in the $15 billion range, I guess that's about 40% of total, where do you see that percentage going in 2017? And when, maybe, can we expect to see HomeAway included in room nights? Thanks.

Thanks, Chris. So, listen, on margins, again 2016 was a bit of a noisy year, and, of course, with all of the integration efforts that happened with Orbitz, you had double cost at the beginning of the year that then came out and a lot of the synergies that we ultimately got were cost synergies, so it's hard to look at 2016 and draw much in terms of conclusions.

What I will tell you, though, is that, and we've said this before, is we are not solving for adjusted EBITDA margins; we're solving for adjusted EBITDA growth, adjusted EPS growth, and free cash growth. And as Dara mentioned, the general theory here is that we're going to keep very, very disciplined on our fixed-cost base, and then drive sales and marketing to really drive to that almost last marginal dollar of variable profit, and that is the growth-maximizing formula, but it's not necessarily the margin-maximizing formula for us and we are really going for growth here.

In terms of HomeAway room-night growth trajectory and sort of the overall opportunity, I do think that that $15 billion opportunity is -- it's a good proxy. We're not really thinking about it that way; what we're thinking about is how fast can we grow the $6 billion, and we think the opportunity is pretty rich.

And as Dara mentioned, we're still in the early stages of getting variable marketing ramped up, not only just getting the data infrastructure and the capabilities embedded in that organization, but then also HomeAway itself is quite early in the product innovation and test-and-learn velocity cycle that has become such a core part of our operational formula in the core OTA business. And as they ramp up that, that should open up even more opportunities for very strong growth for them going forward. So we're very optimistic of the opportunity and, again, the team is executing very well, but it's quite early.

In terms of room nights, we plan to, beginning on our next call, disclose the quarterly room nights for HomeAway, and before that call, we will disclose the quarterly room nights for 2016 as well, just so you've got the comparable periods.

Lloyd Walmsley, Deutsche Bank.

Thanks. Following up on that last question on the $6 billion, maybe you can give us a sense of what the growth rate was from HomeAway bookings exiting the year and what percent are carrying a traveler fee, I guess last year, of that $6 billion?

And then, another one if I can. Can you just, I guess, give us a sense of the overall health of the travel market? It sounds like you are seeing a pretty good growth into January, but looking back to last year, you had a pretty tough comp in the first quarter and had a very strong quarter despite that. So, wondering if there's just anything we should keep in mind as we head into February or March from last year that was particularly strong, or if you think the coast is clear through the rest of the first quarter to keep these kind of January growth rates up?

So the $6 billion represented 46% year-over-year growth, and we're not going to disclose the precise percentages, but generally everything that's online has the online booking fee or the traveler fee attached to it, so it's -- really, the question around monetization is not around the traveler fee, but rather whether the supplier or the vacation property owner or property manager is going by the subscription model or whether they are going by a commission model.

In terms of comps for Q1 and how to think about it, the big thing I would call out is just the Easter comp, which it moves into Q2 this year I would just ask you to take a look at our disclosure on the last call when it went the other way and you would be able to get a rough idea in terms of what the sizing of that might be.

Justin Patterson, Raymond James.

Kevin Kopelman, Cowen and Company.

Just following up on your comment about the subscription model versus the commission model on HomeAway, where are you there and how you see that trending? And then, also, on HomeAway ad spend, I think you had planned to grow that 80% in the first half of 2016. Can you tell us where you ended up on that for 2016 and how you are thinking about the ad budget in 2017? Thanks.

So Kevin, I think I'm going to pass on both of those questions, with apologies, but I'll give you a chance to ask another one. And the reason is that we did want to give you this snapshot of a lot of extra data in my prepared remarks to give you sort of a jumping-off point to help model, but on a go-forward basis, we continue to want to make sure that the HomeAway team has the latitude to do what's right for the business and make this transition and not be held to a bunch of metrics that we share with you all.

I think the classic media training will tell you to answer the question that you want to answer. So the weather here is not very good, Kevin. It's raining.

No worries on that one, but maybe a different direction, then. Orbitz was in kind of a transition year as you made some changes there at the beginning of 2016. Should we think about it as having stabilized and being more in line with your other kind of regional brands in 2017, or what's the update there? Thanks.

So Orbitz, they are going to continue to have tough comps for particularly the first half of 2017. As a reminder, we essentially took the first half of 2016 to migrate the vast majority of that business to the Brand Expedia and Egencia platforms.

When we did that, a number of things happened. One was that we did not build all of the functionality that the Orbitz team had on the Orbitz platform, which was phenomenal functionality, into the Brand Expedia platform; there was pricing matrix and a few other features. We are building that in, but we don't have it right now, and as a result, our air volume did suffer a bit and there was -- growth rates there were a bit depressed.

Secondly, for both Orbitz for Business and Orbitz Partner Network, when we made the migrations, there were a number of partners or clients that either we decided not to migrate because either they wanted too much customization or they weren't particularly profitable or they decided not to come with us, so that's a headwind.

And then, thirdly, there were some brands, such as HotelClub, and some points of sale for other brands, such as -- a number of points of sale for ebookers, where we simply just did not transition them. We essentially shut them down.

And so, those factors mean that there are topline headwinds for Orbitz until we lap those migrations and that's probably -- think about that as a Q3 clean comp. The good news is that underlying all of that, hotel conversion rates look solid, the teams are working hard to get the air functionality up to parity, and we continue to like the core parts of what we see.

I think, just adding to that, the technical capabilities of the Orbitz team, of the product folks, the engineers, the marketing teams, and the data teams there have been really, really impressive. So one of the adds that, frankly, we weren't counting on going into the deal was that we are getting some seriously smart people who now are contributing to the overall ecosystem within the Brand Expedia group. They are some of our top engineers, and especially as it relates to the air product, Orbitz, because of its historical focus on air, had been doing some things, technical things, on the air side that we are now picking up and it will certainly help Orbitz air volumes, but it's going to help all of the air volumes across our various brands.

Another area where the Orbitz teams were very, very advanced was on the private-label side. The private-label capabilities that they had built for some of the partners were substantial and included the capability for our partners to build out functionality that the Brand Expedia platform didn't have. We're essentially taking -- we are taking a significant portion of those capabilities, building them on top of the Brand Expedia platform, and these are feature sets and benefits that are going to roll out really later this year. It's going to take some time, but we're pretty optimistic about those private-label capabilities and the ability for us to then go and market a full-service travel stack to players on a global basis, whether it's banks who want to burn points or whether it's supply partners who want to sell packaged product. That is yet to be seen by the public, but when it comes out, it's going to be pretty dynamite.

Brad Erickson, Pacific Crest Securities.

So I guess with the ramp in sales and marketing spend we're seeing here with these big funnels, like trivago, what's a reasonable time frame by which those newly acquired customers can start to really affect the EBITDA line as repeat customers? And then, given the step up in the quarter in sales and marketing, what are you finding thus far in terms of repeat rates on that new traffic you've acquired relative to your historical averages? Thanks.

You know, trivago and the OTA brands are different animals. trivago is a marketing company, and often you can't exactly measure whether or not a customer is a repeat customer or not. You can put cookies on someone's computer, but sometimes cookies are erased, and consumers now are moving across devices much more significantly now than they ever have.

So when we look at a trivago business, we look at our business -- our marketing efficiency in aggregate on a country-by-country basis, and this team has been doing what they're doing for 11 years, and on a consistent basis we see the return on advertising spend on a country-by-country basis improve over a period of time. The team keeps pushing the marketing envelope to see whether or not incremental marketing comes back with incremental revenue. To the extent that we observe that it does, they will keep pushing marketing, and that results in the revenue growth that you have seen, which is market leading and truly impressive at the scale that these folks are playing at, and it's a formula that really no one else is playing.

And you essentially have the global television audience to work with, and at this point, at least, the global television audience is pretty big, and as the trivago brand improves on a local basis consistently, we see the online marketing return on investment also improve as well.

So, it first starts with brand, but it certainly has an after-effect in the other channels as well. So this is a formula that's working very well. I wouldn't put it in terms of return customers, but when you look at the return on advertising spend, it's unmistakable that the brand is growing and it is clearly getting lots of loyal customers.

We also see it on the other side as far as our being a customer of trivago's, Expedia, for example, bidding on trivago. When we bid on a variable channel, we measure what percentage of customers of that variable channel we can turn into, let's say, a Hotels.com repeat customer or an Expedia repeat customer. From that basis, trivago is actually very difficult -- a trivago customer is very difficult to turn into a Hotels.com or Expedia customer, which is great for trivago and it's great for us as well.

So any way we look at it, this is a brand that's pretty sticky and this is a brand that is clearly resonating on a global basis.

And then, Brad, just in terms of the spend that we spoke about in Q4 and you saw the acceleration in spend there, I'd say very consistent with what we've seen historically. Again, as conversion rates improve and we're back to business as usual, we're seeing very similar dynamics to what we would normally see.

Eric Sheridan, UBS.

Maybe one bigger-picture question for you, Dara. We've seen -- we continue to see a lot of strategic moves in this sector, a lot of people shoring up assets, continuing to think about their asset mix on a global basis. Maybe if we can get your updated thoughts on how you think about the asset mix at Expedia, how you see the strategic landscape, and then maybe dovetail it into a question for Mark of how it fits into the broader capital return strategy for the Company. Thanks, guys.

Sure. Listen, we're very happy about our asset mix at this point. We've got a core OTA business that has really first-rate brands that are now able to grow on a global basis organically. We have some regional brands that are driving pretty strong profitability, combination of profitability and growth as well.

We have Egencia that's getting big scale in the corporate sector, and really we don't see anyone else who is really a scaled technology player on the corporate sector and I think the Egencia team is just that, and I think you're going to see improved organic growth with that business over the next three to four years. And then, we have just two very big growth opportunities ahead of ourselves as it relates to HomeAway in the alternative lodging category and trivago, which we just spoke about.

So, we like our portfolio a lot. It's a combination of kind of highly profitable, consistent growth businesses and then very, very big growth businesses, if we execute. We will be opportunistic. We're constantly looking for opportunity. We will certainly have deals ahead of us, but at this point, the deals are going to be driven more by opportunity, rather than necessity.

And Eric, as a result of that approach, I would say that on a go-forward basis, our capital allocation strategy is broadly very consistent with what you've seen from us over the course of the last five years, which is we will opportunistically do M&A when we see attractive opportunities, but we remain over the long term net absolute believers in this Company and we have a willingness and a desire to shrink our share count over a long period of time.

Mark Mahaney, RBC Capital Markets.

Okay. I was going to limit myself to one question. Dara, the publicity over the immigration ban, whatever you want to call it, have you seen that have any impact on travel demand, inbound/outbound into the US? And then, Mark, in terms of cash overseas and the ability for some sort of loosening of repatriation regulations -- I'm sorry, taxes related to that cash, could that be material to Expedia in terms of how you think about dividend policy and share repurchases in the future? Thank you.

Mark, as far as the current events and the volatility there, we have seen an effect on trading on a short-term basis. The weekend of the executive order, certainly we saw a negative effect on trading. Election, big events usually have a negative effect on trading.

We haven't observed anything meaningful on a trend basis so far, fortunately. We were frankly worried about kind of the chaos and all of the volatility and uncertainty and the effect that it would have on general business trends and especially travel. We haven't seen any meaningful effect at this point, which is good news. We'll be watching it closely.

And then, Mark, on cash, so we ended the year with about $1.9 billion of cash. About $1.2 billion of that was offshore and about -- just over $800 million of that we would think of as being, call it, trapped cash.

In terms of the impact of a repatriation holiday for us, I think generally we'd look at that as a net positive. I don't think it's necessarily a game changer for us, however. We've been pretty effective at putting that cash to work in the form of acquisitions internationally, and when we look at the opportunities we have ahead of us and sort of the opportunistic M&A that Dara mentioned, it's more likely that that's going to be outside of the US than inside of the US. So, again, I don't think it's going to be particularly material, but it could be an opportunity.

Peter Stabler, Wells Fargo Securities.

Good afternoon. This is Rob on the phone for Peter. Two questions. Now that you are a couple of quarters in with best match at HomeAway, I'm just wondering if you could maybe talk about runway in terms of what inning you are in in terms of the improving conversion on that basis, what the response has been from consumers, as well as homeowners and managers?

And then, with respect to your comments just a few questions ago on Orbitz and the matrix, just wondering what kind of headwind was losing matrix to their business? And if you were to add that into PFS and bring it across the business more generally, what kind of tailwind or benefit could that be to Expedia's broader air business going forward? Thanks.

Sure, Rob. As far as HomeAway goes, we think that there's a significant amount of opportunity ahead of us, both in terms of best match sort, but also in terms of the site design and introducing the same pace of experimentation that you see on Hotels.com or in Expedia.

As you know, John Kim, who ran product at Brand Expedia, is now CEO of HomeAway, so we have a lot of experience there. We have ramped up very significantly our investment in product and technical capabilities there, and so we think that we're going to have much, much more experimentation both around sort, but in terms of the design of the site and the many feature sets that we are going to try on a go-forward basis both on desktop and on mobile as well.

So I would say that we are in the early innings. We need to drive conversion in order to hit our targets in 2018, and at this point we believe that we are capable of driving significant conversion increases both in 2017 and 2018.

As it relates to our homeowners, I do think that one of the clear areas of opportunity for us are to communicate more clearly with our homeowners as to why the sort is what it is, and I think that as we have moved from a subscription-based sort to a conversion-based sort, I think that we can more clearly communicate to our homeowners what that means to them.

We are working on radically improving our homeowner tools in general, their dashboards, so that our homeowners can understand what it is that they have to do in order to improve their sort position. Certainly getting their inventory instant bookable is going to help. Increasing their acceptance rates is going to help. Improving their pictures, descriptions, et cetera, is going to help as well, but we are going to lay it out for them so that a homeowner who is making $20,000 a year knows exactly what they've got to do in order to make $70,000 or $80,000 a year. Those are tools that we are investing in pretty aggressively as well and we think there's a significant amount of opportunity there.

So, we're certainly going to drive, I think, a lot more homeowner satisfaction in 2017 than what they saw in 2016.

And then, on Orbitz, we haven't broken out the specific impact of the air business. I will tell you, though, on a standalone basis looking at Orbitz in 2015 versus 2016, gross bookings in total was down, call it, mid-teens on a year-on-year basis.

And I would think about the air impact as being broadly consistent, but within that air impact, it's not only actually just losing that functionality, it's losing the clients and customers and shutting down the points of sale as well. So I do think it is an opportunity for us in terms of when we build that functionality in, both at Orbitz and Expedia potentially seeing an uplift, but it's tough to quantify at this point.

Dan Wasiolek, Morningstar.

Thanks for taking the question, guys. Just looking at the core OTA segment, it looks like the take rate there, the last couple of years it had been in kind of somewhat of a gradual decline, and then the last two quarters, there appears to be some stabilization, even some nice pick-up this last quarter. Just wondering how I should think about that take-rate metric going forward, whether or not maybe it's bottomed or if there's something to call out there, some type of positive mix shift either toward international boutique hotels or hotel mix? Thanks.

Sure. There's a few things going on there. One is there is some mix shift, and it's really mix shift from air to hotel. Generally speaking, our hotel business has been and we expect will likely to continue to grow a little bit faster than our air business, so that's a net positive.

I would also call out that over the course of the last good three years, we've been in a process of resetting our hotel commission rates from a sort of higher premium price to a low base rate, with then upsell opportunities and bid for placement type opportunities. We're in a spot now at the end of 2016 and going into 2017 where that work is largely complete, and you started to see that -- the impact of that actually show up in our margins in Q3 and Q4, and the other metric where it shows up, obviously, is revenue per room night and the gap between that and average daily rate.

So in terms of what to expect for 2017, I think we are in a spot where we are in a much more stable environment and I think there is opportunity absolutely for stability in revenue margins, and when you look at revenue per room night, again this quarter -- this past quarter, it was 110 basis points of gap between revenue per room night and ADR. We, as we look out into 2017, expect it could be a little bit higher than that as we annualize the margin reductions also from the impact of our rewards, loyalty programs, et cetera, but certainly much, much better than we've seen over the course of the last couple of years.

That does conclude our question-and-answer session. I'll turn the call back over to our speakers for any closing remarks.

Thanks, everybody, for joining the call today. As usual, there will be a replay available shortly. Dara, any closing thoughts?

Just a big thank you to our global employee base for an improved 2016 and certainly an improved end to the year, and hopefully we will all be alive to see the end of next year. Thank you.

Once again, this does conclude today's call. We appreciate your participation.