Good morning, and welcome to TripAdvisor's second quarter 2018 earnings conference call. As a reminder, today's conference call is being recorded. At this time, I'd like to turn the conference call over to TripAdvisor's Vice President of Investor Relations, Mr. Will Lyons. Please go ahead.
Thanks, Dulann. Good morning, everyone, and welcome to our call. Joining me today are Steve Kaufer, our CEO, and Ernst Teunissen, our CFO. Last night, after market close, we distributed and filed our Q2 2018 earnings release, and we made available our prepared remarks on our investor relations website, located at ir.tripadvisor.com. In the release, you will find reconciliations of non-GAAP financial measures to the most comparable GAAP financial measures discussed on this call. You will also find supplemental financial information, which includes certain non-GAAP financial measures discussed on this call, as well as other performance metrics. Before we begin, I'd like to remind you that this call may contain estimates and other forward-looking statements that represent management's views as of today, August 2nd, 2018. TripAdvisor disclaims any obligation to update these statements to reflect future events or circumstances.
Please refer to our earnings release as well as our filings with the SEC for information concerning factors that could cause actual results to differ materially from these forward-looking statements. Now, here's Steve, who will share a few thoughts before we open up the call to questions.
Thanks, Will, and good morning, everyone. We're proud of how the business is doing. The hotel meta business has turned around, and we've got new media products that are gaining traction. On the non-hotel side, restaurants and attractions continue to grow, and we're on track to hit our targets for the year. We're now forecasting adjusted EBITDA growth in both segments while still investing in our long-term core growth initiatives. With that, let's open up the call for questions.
Thank you, sir. Ladies and gentlemen, if you have a question at this time, please press star and one. If your question has been answered or you wish to remove yourself from the queue, please press the pound key. We also ask that you limit yourself to one question and one follow-up. Our first question comes from Deepak Mathivanan of Barclays. Please go ahead.
Hey, guys. Thanks for taking the questions. Two questions. First, on non-hotel, margins came in a little lighter this quarter. You called out lapping of operating efficiency gains from last year, but maybe can you provide some additional color on areas that you're incrementally investing there? Is there any change in your view on the long-term margin profile of various business in this category? Second on hotel, nice to see the healthy gains in mobile monetization continue, but it seems like desktop monetization also improved nicely in 2Q. Can you talk a bit about the drivers of desktop monetization? Is that sustainable, or were there any sort of one-time gains this quarter? Thanks.
Hey, Deepak. Good morning. This is Ernst. That's a lot of questions in one question. We'll take them one at a time. First of all, non-hotel, I'll talk about the top line as well as the bottom line, and I'll ask Steve to talk a little bit about our investment initiatives as well. If you look at the non-hotel top line, we saw a deceleration from Q1 to Q2. There are always, from quarter-to-quarter, some lumpiness in year-over-year comparisons. If you look at the stacked two years growth, for instance, Q1 and Q2 were pretty much the same. We had a tough comp of growth in Q2. There's also always some timing of booking windows differences that cause the recognized revenues to fluctuate a little bit.
If you look at underlying bookings, for instance, in non-hotel, the deceleration between Q1 and Q2 was much less pronounced. We look at our non-hotel revenue path right now. It's pretty much where we expected it to be for the year, and feels pretty good internally, and we look at the rest of the year and are set to hit the growth targets that we set out for that business, which is similar to last year. We feel very good about the trajectory there. The headline print of a deceleration of Q1 to Q2 is definitely not how it feels like underlying in the business. That's on the top line. That flows somewhat through to the bottom line as well, obviously. We look at margin. We saw a big margin expansion last year.
We saw some efficiencies that we took last year, particularly on the marketing side, less so this year. First half margin up from last year's margin in non-hotel. There is potential for full-year margin to be higher, but we're also investing quite significantly in this business for long-term growth, which is not directly translated into quarter-to-quarter revenue acceleration. It's really setting us up nicely for the long term. Steve, do you want to talk a little bit about the investments we're making?
Yeah. Thanks, Ernst. We have actually quite a few different tracks of investment in our experiences business. Ton of faith that this is going to be a great driver for us in the future, which in turn drives short-term investment. Some of which you see pretty immediately in terms of things like 24-hour cancellation for most of the things that we sell, which is on the site live now, great for the consumer. Other things you see roll out over time, better user experience, a lot of research being done there on both the TripAdvisor and the Viator points of sale. Those benefits are seen a bit more over time.
More than a handful of very much under-the-cover investments, mostly on the tech side, that involve getting more supply, more accurate supply, more globalized supply, more internationalized, better language support, where, hey, in some cases, we're bringing on supply in anticipation of demand, but it's always a bit of a chicken and egg thing there. We finally have our internationalization of our platform moving the Viator tech base to be able to better support multiple languages, multiple points of sale. The whole global expansion playbook that TripAdvisor executed very well on over the past decade plus, a lot of that is still in front of us in terms of opportunity for Viator. You take that plus the Bokun acquisition, getting into the SaaS space, helping more and more suppliers get online.
You see great growth, or we think it's great growth in terms of online bookable supply, we're still overall at the tip of the iceberg. When we look at how this attraction business, this experiences business can compare to the trajectory of bringing hotel booking online 20 years ago, we still see remarkable parallels, similar margin structure, it starts with collecting demand and supply, bringing it together, that's exactly what we're doing.
Shifting to the hotel side was your second set of questions. We feel good about the trajectory of our hotel business right now, and it's performing well in Q1 and Q2 and ahead of the expectations we had earlier in the year. If you unpack that a little bit, one of the important initiatives has been rationalizing our marketing portfolio and particularly weeding out some marketing spend that we felt was not as efficient as we thought before. That has resulted in some shopper headwind. We saw negative 3% shopper growth in the quarter, which is actually pretty good considering the magnitude of what we cut back to the benefit of EBITDA there. It's a one-off impact this year, and sets us up nicely for next year.
We thought looked at the shopper growth and in light of the marketing efficiency that we took, was actually a pretty good result and setting us up well for the future. Then on the revenue per shopper side, yeah, we saw improvement there, sort of a lower drag on revenue per shopper year-over-year than we saw in Q1 and in Q4. That was a good result for us. If you unpack that a little bit, between desktop. Desktop was more or less flat year-over-year on revenue per shopper, mobile up. The fact that the aggregate was down is a continued mix shift, but both actually performed strong in light of the fact that we still haven't lapped the partner bid downs of last year.
The fact that, for instance, desktop was flat, shows you that the underlying performance of the shoppers is improving. The quality of the shoppers is improving, our monetization of those shoppers in terms of bookings generated for our partners is improving. That statistic is very encouraging to us and bodes well for the back half of the year where we see some easing of the year-over-year comparison on partner bidding levels.
Great. Thanks, Ernst.
Thank you.
Thank you. Our next question comes from Lloyd Walmsley from Deutsche Bank. Please go ahead.
Thanks. Wanted to follow up on some of the stuff, Ernst, you were talking about, getting more efficiency in direct marketing out of the hotel segment. It doesn't seem like it's having much of an impact on revenue. Wondering if you can just give us a sense for maybe where you're finding those efficiencies. Are there areas you're still growing marketing spend where there are more efficiencies? Kind of related to that, how much of the direct spend is related to non-hotel with an eye towards understanding how much room is left to keep cutting. A second one, just the deferred merchant payables growth slowed a bit.
Can you give us some color on how much of that is hotel instant book versus vacation rental versus attractions, and how that may show up in revenue in the second half and where it may show up? Thanks.
Sure. Good morning, Lloyd. Yeah, we don't break down our direct marketing spend between segments, but to give you some color, one of the statistics that you could read in the prepared remarks is that our total consolidated TAMG direct selling and marketing expense was negative 10% year-over-year. If you unpack that, we're growing our TV expenditure in hotel, which is there. We're also growing our expenditure in our non-hotel business. The negative, in spite of those growth areas, is driven by further rationalization in our online paid marketing in hotel. That's the driver of this. Logically, that percentage year-over-year was down more than 10%. We've talked about this before.
The source of this is an incremental, over the last sort of 12 to 18 months, incremental insight into our downstream booking behavior of the traffic that we buy, and particularly honing our internal evaluation models and buying models of external traffic to be consistent with the downstream conversion. That's opened up a much better insight into what investments we were making in the past were profitable and which were less profitable. It's that reorientation and that realignment that is driving the profitability. In our internal models for the year, we assumed a certain shopper loss associated with that and revenue loss associated with that. We're finding that the team is able to adjust their marketing campaigns in a way that doesn't make us lose as much revenue as we initially thought given the size of the expenditure reduction.
That feels very good to us. We've done quite a bit of it. The team is still identifying further opportunities, so in terms of which inning are you in in finding these efficiencies? We're well into the seventh inning, I would say. Of course, that benefits us every quarter on a year-over-year basis. There is some more efficiency that we think we can find, but that's the eighth and ninth inning rather than as large as what we had done before. On deferred merchant payables and the workings there, as I was saying also on the revenue side, there are sort of shifts in booking windows, and then you see an aggregate of impacts from rentals, as well as from our experiences business.
The trends are pretty much as we would expect them to be given our bookings growth, and there are some timing differences from quarter to quarter. You see this quarter, like you have seen in other years, significant cash inflow on deferred merchant payables, which is going to flow out again in the back half of the year. Those trends are pretty much in line where we expect them to be.
Thank you.
Thank you. Our next question comes from Michael Olson from Piper Jaffray. Please go ahead.
Hey, good morning. A couple questions, if I could. Within non-hotel, would you be able to provide a broad strokes breakdown between experiences and restaurants? I'm guessing you don't want to give precise %, but any high level mix you can share on that? Second, what are you seeing from a competitive standpoint in tours and attractions? It seems like this should be fairly unique inventory, but are you seeing any others coming in that are kind of building up much of that disaggregated experiences inventory? Thanks.
Thanks, Mike. I'll take the first part of this, and Steve will take the second part. On a quarterly basis, we don't break out our revenue between the different lines and non-hotel, but we did say at the end of last year that for 2017, the breakdown was roughly 50% of revenue in experiences, and then equally 25% in restaurants and in rentals. Obviously, that is shifting this year. The growth rate in restaurants and attractions is significantly ahead of rentals. That's going to be different for this year. We'll provide you with an updated view on the mix at some point in the future.
To your question on inventory, tours and attraction inventory, there's certainly several other players that have, and in some cases are still trying to gather the industry footprint. No one's particularly close to us at this point. To be clear, the inventory we have is not exclusive, and we don't see that that would be a key trend or an expectation going forward. There's so much that is already out there. Getting more of it to be bookable on our sites is the clear opportunity versus having to fight for market share from any of the other players. TripAdvisor sits in a pretty darn unique position in the industry by having such an impressive demand footprint of people looking for things to do. A lot of our brand is built around having that great vacation, people already coming to TripAdvisor for that, and we have the in-destination footprint.
When you wake up in your hotel room, and you're looking for what you're going to do that day or the next, we're still the natural site to go to. While, of course, we do pay attention to who the competitive players are there in the space, we're much more focused on growing the pie and helping the consumers who are on TripAdvisor and Viator, and the ones that we can attract with the best inventory, the best customer experience, the best customer service to make their stays delightful. If we keep our eye on that ball, I think we're in great shape for a long time.
Thank you.
Thank you. Our next question comes from Justin Patterson from Raymond James. Please go ahead.
Great. Thank you very much. I'll go back to the marketing theme. As you've learned more about downstream booking behavior and go into the seventh inning of these marketing efficiencies, what are the factors you're looking at as you think about reinvesting in TV or performance marketing? It seems like some of the benefit you've seen so far this year, beyond just channel efficiencies, is also competitors moderating some of their spend. Longer term, I'd argue that you need to start seeing site traffic growing again to really drive sustainable growth in both the hotel and the non-hotel business. Any color you can provide on what you're thinking about factor-wise to reinvest in marketing would be helpful. Thanks.
Well, you've certainly seen. Thanks, Justin. It's an excellent question. You've certainly seen our shift over time from some of the performance-based channels to the more branded channels like television. It's great to see those working for us. We already have such a global brand that, to our chagrin, was slightly misunderstood in some cases in terms of not being understood as a place that you could go to compare prices. Our TV campaign is successfully moving that perception over to more of what we want, which is the reviews, the great experience, and the ability to compare the price and find the best deal. That's what's driving some of our revenue per shopper improvements. That's what's driving some of our overall
Value from the traffic we already have. Of course, we would love to see more hotel shopper growth, but I urge you to remember we already have a phenomenal amount of hotel shoppers. Having these folks do a bit more of what we would like them to do in terms of being closer to the transaction, come back a few more times during their trip. You could paint, and we do internally paint, a wonderful picture that doesn't require us to buy a lot more new traffic through these other channels. TripAdvisor is a great site to be a member of. TripAdvisor is going to help you plan your whole trip.
If we can tap into those touch points a few more times for each visit we get, that just drops straight to the bottom line for either hotels or experiences or restaurants for us because we offer so many services across the trip. Of course, a lot of that starts with the hotel shopper, but we don't feel a need to get to double-digit growth in hotel shopper in order to drive business forward. We're fortunate to play in such a big category that there is that opportunity for double digit, but again, we don't need to feel it because our opportunity to improve the overall revenue per shopper for all those shoppers that we have is still there.
Great. Thank you, Steve.
Certainly.
Thank you. Our next question comes from Mark Mahaney from RBC Capital Markets. Please go ahead.
Hey, I think I'll just stick with that hotel shopper line of questions. Just to put it all in context, I just wanted to go back and check. This is the first quarter in which hotel shoppers have declined for you year-over-year, and I get the point that that's in part due to marketing efficiency measures on your part. Do you want to set some expectations as to what you think, what you'd like to have that hotel shopper growth be like in the future, even if it's not double-digits? Can you return it to growth? Just address the issue of, is it purely decelerating or declining year-over-year because of your marketing efficiencies, or could there be something else going on? Could there be a shift towards other hotel shopping channels just to throw out Google or anything else?
I would think long term, you wouldn't want a year-over-year decline in hotel shoppers, even if you get an improvement in revenue per shopper. Just spend a little bit more time talking about how we should expect that metric to look in the future and why you wouldn't want it to grow reasonably strongly going forwards.
Yeah. Thank you for giving me the opportunity to clarify. We don't want it to decline. We think we have all the reasons in the world to have it continue to grow. Does it need to get back up to double-digit growth for us to be successful? No, it really doesn't. We're just dealing with laws of very large numbers here. When we look at the decline this particular quarter, and perhaps for the next quarter, as we lap year-on-year, we do point very specifically to our paid marketing channels as having, since we're cutting back on those, that's having an effect on our hotel shopper growth. I'll note that we're obviously, and you can see it in the numbers, cutting back, we think in a pretty smart way, as in we're cutting back on the least profitable of these hotel shoppers for us.
Yes, the overall number may have declined, but it's the ones that were doing the least on our site. We have a bunch of other things, both in development and just as part of our overall TV strategy that counteracts the decline in hotel shoppers, and we expect it to continue to grow. I just don't want to set the expectation that we need the double-digit growth that we saw in years past in order to successfully grow in the hotel category and grow overall.
Thank you, Steve.
Thank you. Our next question comes from Eric Sheridan from UBS. Please go ahead.
Maybe a two-part question with respect to the non-hotel business. Is there any color you can give us on how much traffic or how much interaction there is between TripAdvisor in terms of feeding traffic into the non-hotel properties, just so we can get a better sense of how synergies might be able to be achieved between the various brands under the TripAdvisor corporate umbrella? Second, talk a little bit about marketing on the non-hotel side. Where do you find that you're getting efficiency in terms of driving traffic, new users, brand awareness? Just so we can get a little bit better sense on how that margin structure might evolve over time. Thanks so much.
Sure. We might tag team this a bit. This is Steve. On the non-hotel side, in general, if I understand the question correctly, the traffic that's coming to the TripAdvisor points of sale and interacting in the restaurant and attraction category, in general, are staying on the restaurant and attraction pages on Trip. You can book almost all, if not all of the attractions while staying on Trip. You don't need to be sent over to the Viator points of sale. On restaurants, again, you can make most of the reservations through La Fourchette while staying on TripAdvisor, and our entire restaurant media business is entirely on TripAdvisor. That TripAdvisor as the channel to sell and promote attractions is the fastest-growing. It's been that way for quite some time. It's our lead horse from a global perspective.
Not the only point of sale, obviously, but the one we're putting most of the muscle behind. The marketing on the non-hotel side, obviously, we try to employ many, if not all of the same paid performance channels. As I was talking about TV earlier, right now the message has been focused on educating folks that they can do price comparison, hotel shopping on TripAdvisor, not just reviews. It's likely going forward that we're going to step into the fact that our entire value proposition is all around everything you can do to have a fantastic trip. Hotel is part of it, restaurant is part of it, experience is part of it, I'd say it's likely our advertising campaign, our TV campaign also expands to cover, not just hotel price comparison, but more.
Great. Thank you.
Thank you. Our next question comes from Mark May from Citi. Please go ahead.
Hello?
Mr. May, your line is open.
Yeah, thank you. Sorry about that. Apologize if this has been covered already, but could you comment on the click-based RPS improvement and how much of that sequentially is due to maybe just a pickup, auction dynamics and a pickup in demand from advertisers from somewhat depressed levels in the second half of last year versus the result of maybe improvement in conversion rate? Thank you.
Hey, Mark, this is Ernst. The overall sort of bidding environment has been pretty stable, in Q1 and Q2 from those levels of Q4, obviously down year-over-year from levels we saw before those bid downs happened last year, but pretty stable environment. The relative improvement, we believe is more a function of improvements we're making to the quality of the traffic and the experience on the site.
Okay. Sorry if this has been covered already, but in terms of non-hotel revenue growth and the slowdown in the growth rate itself year-on-year, how much of that is driven just by law of large numbers? Are there one of your contributors to that line that you saw a slowdown that's kind of masking other dynamics in that segment?
Yeah, I did talk about that earlier on the call. What I would say is, it is the quarterly inconsistencies of how revenue comes in. If you look at a two-year stack basis, Q1 and Q2 was very similar in growth. We had a difficult comp in Q2 last year. If you look at underlying bookings growth quarter-to-quarter, much less pronounced difference Q1 to Q2. Some of this is the timing window of the revenue recognition as well. We're looking at Q1 and Q2, and we see pretty consistent performance of our non-hotel business. We look at the rest of the year, and our expectations are intact.
Thanks, Ernst.
Thank you. Our next question comes from Matthew Brooks from Macquarie. Please go ahead.
Good morning, guys. I want to get into that sort of non-hotel segment again. Maybe can you talk a little bit about the maturity profile when you add new listings to Viator? You basically doubled the listings in the last year, but does it take quarters or does it take years of those bookings per listing to start to mature?
This is Steve. Great question. Thank you, Matthew. Some of that is a bit new to us for the types of inventory that we're growing. Again, inventory is just one metric. It's a necessity if you're looking to grow globally, but as we bring on smaller tours and attractions in less traveled parts of the world, as that's what happens in the marketplace, they're not going to be performing to the speed or to the scope of your big, top-tier attractions in major tourist destinations, which for the most part, we already have. As we go forward, we thought it was helpful to show that we were investing a lot in supply acquisition on a very global basis, which we are. A lot of those attractions will become mature performers for us over time. Some of it is just make sure it's working properly on the system.
Other parts is us actually building more of the demand footprint in that particular part of the world as well. Again, it'll take time. I wish I had a few more specifics for you to your question, we're comfortable that we're certainly on the right path there.
Great. I guess as a follow-up on the sponsored ads, is there any sort of data you can point to highlight the traction you're seeing there in either restaurants or hotels? Are you looking at this product as something you want to use in experiences as well?
Yeah. We think it's kind of a great product that matches the fact that a lot of our hotel shoppers are high in the funnel, it's an opportunity for hoteliers to get themselves in the consideration set, right? You can only have 10 or 20 properties in the first page on New York or Paris, you have a lot of other terrific properties that really are great for consumers that are on page three or four in that sort list, here's the opportunity for them to be considered. Wide appeal amongst hoteliers. Similarly, in restaurants, we have four plus million restaurants on the site.
If you're already number one in your city, it may not be the right product for you, that leaves 3.9 plus million for whom this is a good opportunity for exposure to our couple hundred million monthly eyeballs for restaurants, for instance. Would we roll it out to experiences? Sure. Under consideration. Again, experiences is much more of a transaction model for us, so it's not as timely, if you will. To back it up, offering the sponsor placements for restaurants and hotels as a great media product, we see a ton of headroom going forward, and it's off to a very nice start. Okay, thank you very much.
Thank you. Our next question comes from Doug Anmuth from JP Morgan. Please go ahead.
Great. Thanks for taking the questions. First, just wanted to ask you about mobile revenue for mobile hotel shopper. Obviously, saw some strength there. Just hoping you could just quantify a little bit of what you're seeing in terms of the gap relative to desktop. Secondly, just given the time that you've spent in the travel landscape, Steve, and obviously how well you know the other players, how are you thinking about the sustainability of these stable auction dynamics going forward, given that we've seen different periods in time over the years, different behavior? Thanks.
This is Ernst. I'll start with the first part on the RPS, let Steve talk after. In terms of mobile RPS, we've seen continued improvement of mobile RPS, and it was up again double digit this quarter, year-over-year. We're very pleased with that. We're showing growth even while we're lapping some of the impressive improvements that we made last year. That is good. The monetization gap quarter to quarter didn't move very much. We're still at this approximately 40% monetization on mobile versus desktop. We also saw nice improvements on desktop, at the end of the day, the overall improvement is what really matters. We're pleased with that mobile RPS, and we see more headroom there of improvements.
This is Steve. On the stable auction dynamics, as Ernst mentioned, we've seen stability. We feel we have very strong relationships with all of our big clients, all of our small clients as well. We have a robust auction. There is the caveat that our partners at any point could make changes to what their bidding strategies are. The scale at which we operate and the position in the funnel of which most of our, or many of our customers are at, makes us a pretty unique traffic source for the major OTAs and the hotel brands. Simply because we're at the, I'm not sure where I want to go yet. I'm not sure where I want to stay, and that's a very up-funnel demand that is often ahead of where the big OTAs are. As such, it's traffic that's up for grabs. It's traffic that any of our clients can try to make loyal to their booking engine.
We're comfortable with that because of our up-funnel position, because of the fact that so many are loyal to TripAdvisor as the place to start their travel planning experience. Our overall value proposition is really quite different than obviously a hotel brand or a straight OTA. As everyone is looking to have travelers pick their site as, or their app as the place to start, we're comfortable with where we are in the ecosystem because of our hotel and restaurants and attractions and flights and guides and experiences and community as a way to differentiate from the other choices that travelers have, and our continued 10%, I believe it was 10% this past quarter, UU growth shows that, yeah, even at the large numbers we have, we're still attracting more and more travelers each quarter.
Just to follow up there, if you saw the environment heat up again at some point in time, there would obviously be some benefits to you, at the same time, I imagine you'd have to respond, in terms of how you're spending your marketing dollars ultimately. Is that something that you're prepared to do if things were to change?
We look at our different revenue streams and say we're investing in our media businesses because there's a ton that we feel we can offer to hoteliers and restaurateurs and eventually attraction suppliers all around the globe that isn't dependent upon auction, but is dependent on the fact that we have 400 plus million UUs a month. It's just the eyeballs that are on our site that don't go away, don't change depending on whether the auction is paying top dollar or lower dollar. Again, if we get paid more, then yes, by our return on investment marketing philosophy, we can spend more. If we get paid less, we choose to spend less as well. Again, we've been used to those dynamics for quite some time.
Okay. Thank you, guys.
Thanks.
Thank you. Our next question comes from Tom White from D.A. Davidson. Please go ahead.
Great. Thanks for taking my question. Was just hoping maybe you could give a bit more color on the early ramp of the premium media product for hotels, number of advertisers maybe, typical length of commitments, typical spend, anything like that. Then just on the auction, you talk about stability. I was just hoping you could maybe characterize that a bit. Is it still that the large advertisers have higher ROI targets across the board that haven't changed, or are you seeing signs that in particular markets or geographies some of these guys are maybe being a little bit more aggressive or selectively more aggressive to lean into the channel again? Thanks.
Sure. We're not at the point where we're able to disclose much by way of the media products for hotels or for restaurants, but I do enjoy being able to add the color that both are launched, are global. We're seeing meaningful product improvements in both categories on a quarterly basis. We continue to see sales grow. Since there's both the subscription elements as well as the pay-per-click elements, when we look at the growth of this category of product for us, the media as opposed to auction, it continues to show really nice promise with none of the potential threats of changing ROI behavior of our OTA clients. It's just a nice diversification, tapping into our overall traffic.
Both hotel and restaurants for sponsor placement, the pay-per-click are self-service, and we have, depending on the product, different levels of sales assist and business development activities going on against them. Obviously, once a client joins on the sponsor placement, it's very much on a performance basis. We only charge when the click happens, and it's their opportunity to get into the discovery set, that's what so many of these clients have been asking for so many years, because there isn't any other way. You still can't influence our overall ranking that's based purely on our best value algorithm. Sponsor placement lets you get visibility onto that page.
Yeah, in terms of the auction stability and any trends to call out, no trends really to call out. We've said we've seen stability since the mid Q4 into Q1, Q2. There's always little fluctuations around everything based on geo, et cetera, nothing to call out really here. Overall stability is what we'd like to emphasize.
Okay, thank you.
Thank you. Our next question comes from Jed Kelly from Oppenheimer. Please go ahead.
Great. Thanks for taking my questions. You said you were incrementally more confident for this year. I guess, does that mean you're incrementally more confident versus 1Q? Can you sort of unpack what you meant by that a little bit?
Yes. Thanks for the question, Jed. We did say in the prepared remarks, incrementally confident about our EBITDA growth. Look, we've had a good first half of the year. It's exceeded our expectations that we had in the beginning of the year. Q2 exceeded our expectations that we had at the beginning of Q2. We are looking at the rest of the year and compared to where we were three months ago, again, we have readjusted our expectations and have upped our own expectations internally. We look at that and we feel pretty good. We say we expect incremental growth. It's not what we expected in the beginning of the year. We did 9% year-over-year growth in the first half of the year on EBITDA. We've commented before that we see some of these trends that we have easing in the back half.
We said before that we expect sort of relative year-over-year performance of EBITDA to be back-end weighted. We still believe that. We look at a start off of the quarter, this quarter, July was strong, was solid. Things are looking really good for us from here. That was behind our statement that we're incrementally confident.
Just back on the marketing, can you give us a sense how much of your non-television marketing is actually being utilized to drive mobile shoppers to your apps and sites?
This is Steve Kaufer. Of course, all of our performance channels work on mobile. If you're buying traffic on a Google and you have our app installed, you're going to automatically be deep linked, so that'll end up in the app. If the question is, are we spending a ton of money on buying app installs, that's not currently part of our game plan. We have a lot of app installs because we have a great app that people want to use when they're traveling, and we rely more on that. Did that address your question or was it?
Yes. Thank you.
Okay.
Thank you. Our next question comes from Kevin Kopelman from Cowen and Company. Please go ahead.
Thank you. I have a question on TV ads. How far along are you as far as the global rollout? Do you see yourself bringing those TV ads to all of your key markets, and what timeframe? Thanks.
Well, we certainly are in the majority of our key markets already. As every quarter passes, we're more and more excited and pleased with what TV's been able to do for our brand and our revenue. The decision to increase our spend on TV, the decision to restart last year and increase our spend this year were clearly the right ones, and I think our financials show that, and we're educating and building the brand. I alluded to earlier the fact that, while the TV campaign may be just about hotel price comparison at the moment, it would be reasonable to expect that to extend to more of our community, more of our full value proposition that we can offer to travelers, including some of our newer businesses. We don't disclose, certainly this far in advance, whether we will be adding TV in additional key markets.
I can confidently say we view this as a great channel for us that we're happy that we got into.
Great. Thanks, Steve.
Thanks.
Thank you. Our next question comes from Brad Erickson from KeyBanc Capital Markets. Please go ahead.
Hi. Thanks. Just a follow-up. I guess two follow-ups. When you look at the hotel shoppers on the site you're seeing today, what portion of those would you say are coming to the site directly, and I guess how has that changed over the past year? Secondarily, related to your comments you made earlier, just seemed to imply that hotel shopper growth wasn't necessarily the most important thing necessary to grow overall. Is the intent there really just to monetize better on the most profitable part of your traffic? Just would like to understand what was meant by that a bit better. Thanks.
Yeah. I'd urge folks not to focus too much on that hotel shopper growth comment earlier, simply because I've said it multiple times in the past, in terms of we have so many hotel shoppers, getting them to stick around and return on their own accord within the same trip, just if you do the math, is a bigger opportunity than potentially attracting more. The law of big numbers says it will continue to be hard to grow that at meaningful double digits, so I didn't want to set expectations that we needed that to happen to grow the biz.
Our opportunity to take the traffic, the hotel shoppers, and other shoppers that are on our site that really are pretty far away from a transaction and to pull them by talking to them better with the better site experience, more functionality, and pull them down the shopping path so that when they are ready to in fact select a specific hotel, to select a specific booking site, that we're the site that they return to do that. At a couple of 100 million hotel shoppers a month, that remains our fundamental biggest opportunity, which was a completely true statement a year ago, and frankly, two years ago as well. We do hope to have hotel shopper numbers return to growth in subsequent years, of course, because the overall market of hotel shoppers still has plenty of headroom for us.
To the first part of your question, hotel shoppers coming to main [inaudible]-- W e don't split out that quite yet, but you can certainly understand that we do care about that as a metric or sort of a home page, first page view. TV, of course, helps drive some of that, as well as obviously the better site experience and the communication that we're able to do with the customers to pull them back, all of which it's sort of direct traffic as opposed to paid for or accidental traffic.
Got it. That's helpful color. Thank you.
Very good.
Thank you. Our next question comes from Peter Stabler from Wells Fargo Securities. Please go ahead.
Good morning. Thank you for taking the questions. A couple quick ones. In your letter, you talked about the non-hotel rental business you got into, some competitive pressure there. Can you give us a view on how you look at that strategically as you look out over the horizon? Is this an area where you're still investing significantly or investing at all in building supply? Could we expect some marketing support there in the future? Just to kind of some color on how you look at that opportunity in the meta channel. Then a quick one on the TV advertising.
In terms of the increase in the quarter, possible to give us any color on whether that was the addition of additional markets or whether in some of your big markets, whether you're adding weeks or weight, given the fact that some of the OTA competitors are increasing their brand television spending as well. Thank you very much.
Peter, this is Ernst. I'll take the first part of your question. The rentals business is a good business for us. It's an important business for us to be in for our users. It's part of our complete offering in accommodations, to have a hotel offering as well as alternative accommodation offering. We operate a number of brands in the rentals business, importantly, offer it on our TripAdvisor brand, which gives us a lot of volume in this category. It's a nice business for us. It's a profitable business for us. It is not a business where, to your question, where we're incrementally putting a lot of investment $ behind it compared to what we're doing in experiences and restaurants. The most important reason is that we see a lot of blue sky and green fields in the experiences space, and we have an early head start.
We're making most of our push there. Same on the restaurant side. On the rental side, it's a very competitive place, and we're playing it a little slower. As I said, it's a profitable business for us and we're happy to have the business.
This is Steve. On the TV, at the beginning of the year, we had launched a number of new markets expanding in Spain and doing some additional creatives as well. We had set expectations that the spend would be north of $100 million this year. The Q2 increase is really just the follow-through on that. It is to some degree, a combination of new markets, but just being on air for with we feel the proper seasonal weights, so that you shouldn't read anything into our spend being a reaction to what other folks are doing versus what we think is right to grow our brand.
As a reminder, last year, we started to spend in June, so only in one month of the quarter, last quarter, and obviously last year, and obviously this year, we were present on television in all three months of the quarter.
Helpful. Thank you.
Thank you. Our next question comes from Naved Khan from SunTrust. Please go ahead.
Thanks a lot. Just a couple. Steve, or maybe Ernst, you can answer this. How much headroom do you see to continue to drive improvement in mobile hotel shopper monetization? By my math, I guess you saw that metric grow 12% this last quarter, versus maybe somewhere in the 20s in the quarter before that. You probably anniversary-ed some of the changes from last year. How should we think about this going forward? Then I had a quick follow-up on the non-hotel side.
Sure. This is Steve. I certainly believe there's continued room for mobile monetization, improved mobile monetization. In part, our mobile site just continues to get better. All of the improvements that we do, all or almost all of them are rolled out mobile first. That's where our consumers are today. They're still migrating there in larger numbers. There's no doubt internally that continues to be a focus. We also just benefit from the fact that consumers in general are more happy to buy things on the phone. As an up-funnel consumer comes to us, clicks through to a partner, they're more likely to transact. That partner's more likely to give us credit for the transaction because we sent them the lead, the CPCs go up.
Some of it, we certainly believe is the great work that the team here is doing, there's just other secular tailwinds that we expect will improve our mobile monetization over time in all markets. There's no question in our mind.
Would you opine on which inning we are in? Is it middle innings or early innings?
Well, I think it's probably still early innings, in part just because consumer behavior, travel is still a big purchase, why is it our desktop CPCs are so much better than mobile CPCs, people wanting to stay somewhere? Our best answer is that people are still more comfortable with that type of purchase on their bigger screen, bigger device. I'm not sure they'll ever be equal. As everyone does more and more stuff on the phone, we're in a perfectly good position to benefit from that. We're in a better position than many others to benefit from it, in part because it's been such a headwind for us in the past, as consumers get used to buying it moves to being more of a tailwind, given so much of our traffic, over half, is already on the phone.
Okay, thanks. Just on the non-hotel side, what was the contribution from Bokun? Can you quantify it? How important is performance advertising for the growth in the channel?
The financial impact of Bokun was fairly limited in the second quarter, other than the investment that we made. We are going to make incremental investments throughout the year to drive the strategy behind Bokun. Bokun doesn't add a lot of revenue right now.
A great capability that we're going to invest behind, which is ultimately going to help us in the business. Not a big impact just yet in our financial model.
Okay. The performance or the importance of performance advertising to growth of non-hotel?
Yeah. Performance-based marketing as a % of revenue is the highest in our experiences business. It's, for instance, much lower on our restaurant business, so it's mixed. The intensity in the P&L is mixed. Particularly for our experiences business, performance-based marketing is an important channel that we continue to leverage.
Thank you.
Thank you. Our last question comes from James Lee of Mizuho Securities. Please go ahead.
Yeah, thanks for taking my questions. My first question is about the ROI requirement for marketing spend. In the past, you said one time for digital. It seems like, over the last few quarters, you performed better than expectation. Is that correct to say? Also, in the past, you said that if you perform better than one time digital, you would actually use the excess to invest in 2019. That doesn't seem like the case, given your commentary on hotel shopper growth. I just want to confirm that. Thanks.
Yes. We have turned our performance-based marketing channels more towards profitability as opposed to gross break even, with the profitability measured on a more accurate system that looks at the downstream booking revenue as opposed to just what we can measure instantly, the click base. It's a little complicated, but it's a better reflection of how valuable the traffic is that we're sending to our clients. We've certainly used some of the savings in our performance channels to help fund some of the brand building and the TV stuff that we're doing, and that's part of where that mix shift is headed. I don't believe we've said that we would be taking savings from performance-based marketing channels and doing anything specific with them next year.
Okay. Can I ask a follow-up question, Steve? You sounded very confident on your revenue per hotel shopper growth going forward. Help us understand how should we expect the growth for 2019, number one, and what are you doing specifically to drive better conversions there? Thanks.
We have multiple years history of driving improved revenue per shopper, per device on desktop and on the phone. The mix shift moving to the phone can make that overall number decline in spite of the kind of continued optimization on each device. What hit us was back half of last year with just the drop in the CPCs, which of course impacts revenue per shopper. I can't commit to improved revenue per shopper simply because I don't control what our big partners bid. We've always faced that issue, and all I can represent at this point is that given the public commentary of our bigger clients, given the dynamics as we see them in the meta space, and given the past two quarters, the auction looks stable at this point.
We're providing very real value, a better ROI for our clients, we have no reason to think that that's going to be changing, certainly not in the near term. With that as an admitted assumption, the improvements that we've been making to the product, both desktop and mobile web and app, we have a nice history of improving the revenue per shopper if everything else is constant. It's better pricing on the site. It's better supply. It's a better sort. It's the UX improvements, the preference improvements. I mean, sizable teams working against this every single day with a nice track record saying, "Yeah, we make improvements when all other things are constant." Again, history is the best indicator there.
Aside from any external bidding changes, I think we have the track record that backs up that statement, or that backs up my confidence in it.
Thank you. I show no further questions in queue. This concludes our Q&A session. At this time, I'd like to turn the call over to CEO, Steve Kaufer. Please go ahead.
Great. Hey, thanks everyone for joining the call. We had a strong first half of 2018. I really want to thank all the TripAdvisor employees that put in a ton of hard work to get us here. It's great to see the business making this rebound at this point. We've done a bunch of changes, strengthened the business in a whole bunch of different fronts. We're proud of our hotel segment, proud of the non-hotel, looking forward to an improved back half of the year as we shared, sets us on a nice trajectory for 2019. Thanks, everyone, and we'll update you again next quarter.
Thank you, ladies and gentlemen, for attending today's conference. This concludes the program.