Tripadvisor, Inc. (TRIP) Q3 2012 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the TripAdvisor Third Quarter 2012 Earnings Conference Call. [Operator Instructions] As a reminder, this call may be recorded. I would now like to introduce your host for today's conference, Senior Director of Investor Relations, Will Lyons. You may begin.

Thanks, Ashley. Good afternoon, everyone, and welcome to TripAdvisor's Third Quarter 2012 Earnings Conference Call. We appreciate your joining us today. I'm Will Lyons, Senior Director of Investor Relations for TripAdvisor. And joining me on the call today are our CEO, Steve Kaufer; and our CFO, Julie Bradley. Before we begin, with many of our analysts and investors located in the areas impacted by the recent storm, we thought it best to delay our earnings call until today. Our thoughts are with those affected, and we appreciate everyone's patience. I'd like to remind you that the estimates and other forward-looking statements included in this call represent the company's view as of today, November 1, 2012. TripAdvisor disclaims any obligation to update these statements to reflect future events or circumstances. Please refer to our -- today's earnings release and TripAdvisor's filing with the SEC for information concerning factors that could cause actual results to differ materially from those expressed or implied by such statements. You'll also find reconciliations of non-GAAP financial measures to the most comparable GAAP financial measures discussed on this call in our Q3 earnings release, which is available on our IR site, ir.tripadvisor.com. Finally, unless otherwise stated, all references to selling and marketing expense, general and administrative expense and technology and content expense excludes stock-based compensation, and all comparisons on this call will be against our results for the comparable period of 2011. With that, I will now turn the call over to Steve.

Thank you, Will, and welcome, everyone, to today's conference call. Before starting, on behalf of the entire TripAdvisor Media Group family, our thoughts are with everyone that has been impacted by Hurricane Sandy. Special thanks goes out to those of you in the Northeast region who are joining us today. I'm happy to report that TripAdvisor had a strong third quarter. We generated record quarterly revenue of $212.7 million and adjusted EBITDA of $107.1 million. Across the globe, traffic to our site continues to be strong, as total traffic to TripAdvisor sites grew in the mid-30s and, even more importantly, our hotel shoppers have continued to grow at approximately 30% based on our log files. I'm also pleased that we made great progress this quarter rapidly innovating the TripAdvisor experience for both our users and our advertising partners alike. Our core click-based revenue accelerated, growing 15% year-over-year based on continued strong hotel shopper growth. This revenue growth can also be attributed to improvements in our on-site conversion and personalization efforts, offset by a still weak, yet stabilizing, Southern European market. Unique to the third quarter, we estimate that the Summer Olympics had a 2% negative impact on hotel shopper traffic in click-based revenue. We were also pleased to see a 5% to 6% uplift in click-based revenue very late in the quarter, when we lapped last year's site redesign. We expect this uplift to continue in the coming quarters. One of the most exciting things to see is the continued strong growth and deep engagement of our growing traveler community. Our marketable members grew over 70% year-over-year to more than 36 million, fueling our robust and growing email campaigns. Our members love sharing their travel experiences with other travelers and friends, and they continue to share content in 21 different languages at a remarkable rate of more than 60 contributions per minute, which equates to over 30 million contributions in a single year. During the third quarter, we were thrilled to announce content syndication agreements with Best Western and with Wyndham Hotels Group. The Wyndham partnership, which has been expanded to 15 of their hotel brands, also includes a review collection agreement, where Wyndham will encourage their guests to share their experiences through a post-stay email and survey powered by TripAdvisor. Including these latest wins, we now syndicate our traveler reviews to more than 500 major travel partners, including more than 50 hotel chains, reinforcing and enhancing the visibility of the TripAdvisor brand as the de facto standard for travel research. We also appreciate that tens of thousands of individual business owners around the world proudly display their "Reviewed on TripAdvisor" stickers, their certificates of excellence and their Travelers' Choice plaques on-site at their hotels, restaurants and attractions. The brand flywheel is humming, but with TripAdvisor sites accounting for only 10% of online travel uniques, we have a lot of runway left for continued growth and adoption. At TripAdvisor, we are always innovating our products and extensively testing and discovering new ways to improve site usability and experience for our users. We strive to create the very best experience for every user, on every device, in every geography. You can see this in our social and personalization initiatives, great vehicles through which we learn about our users, their friends and their preferences, all of which enable us to deliver more satisfying, inspiring and customized travel planning experience for everyone. Along these lines, we continue to leverage the Facebook platform and, subsequent to the end of the quarter, we were pleased to welcome aboard via acquisition the talented team from Wanderfly, whose focus on blending inspiration, personalization and social, nicely complement our ongoing efforts. In addition to social and personalization, mobile is a key area of focus and investment as we innovate to deliver the best user experience on every device. Year-over-year, we've seen downloads of our smartphone and tablet apps more than double, reaching more than 26 million total downloads. Over the same timeframe, our average monthly mobile uniques are up over 130%, more than 35 million, according to our own log files. We rolled out an improved tablet interface during the quarter and have seen tablet monetization improve nicely since the -- since earlier this year. It's now more closely resembling desktop, an important trend given that tablet comprises about 3/4 of our mobile revenue. On the other hand, smartphone monetization isn't growing as quickly as tablet monetization, which is not surprising to us given the use case and form factor of that device. During Q3, we began testing hotel meta display on smartphones, an exciting step towards optimizing the user experience. Similar to our Flights product, user interface shows partner prices and availability in line, allowing a hotel shopper to more easily compare booking options. The tests are progressing well, and barring any unforeseen challenges, we expect to roll out hotel meta display to all phone traffic later this quarter. For us, the phone is all about optimizing the user experience. And we expect over time, this innovation will drive better user engagement, better conversion and therefore, better monetization on the phone's smaller real estate. Now the logical question to ask is, are we planning to roll out a meta display to all users on all platforms in all geographies. And the answer is we don't know yet. We think the meta experience on phone makes a lot of sense and is worth testing on desktop, but we aren't making any commitments beyond testing at this point. We've contemplated a meta display for years. And as with all of our product releases, there are 3 main things that we consider: Is it good for the user experience? Is it good for our clients? And what does it do for TripAdvisor economics? We always weigh these factors very carefully, and our objective is to find the solution that works best for all 3. As it relates to our OTA and hotelier partners, looking at the global landscape, we have seen improvements throughout Q3, especially in our newer international markets. In the EMEA, we added Geremia hotels to our growing list of partners in the Middle East. In APAC, we added 3 new regional Asian OTAs, including asiarooms. On the bidding side, we started testing property rank reporting with a select set of major OTA clients in 3 test markets in early October. And while that testing has begun, we have not been surprised by any of the results so far. The goal of providing these reports is to allow partners the chance to better understand their ROI on TripAdvisor, and help them rationalize higher per-unit or per-click spend. We expect the test to be successful and expect to have minimum CPC per property thresholds in place in early 2013. At that point, we would consider moving to a worldwide rollout. During our relatively short history as a public company, you've heard us describe a few initiatives we've undertaken to strategically position TripAdvisor for the long term. From our site redesign to our traffic quality improvement measures, the data sharing with large partners and our new meta interface on smartphones, we remain focused on optimizing user experience, enhancing client satisfaction and maximizing TripAdvisor economics over the long term. This has been TripAdvisor's DNA since inception. In summary, we had a great quarter. We're squarely on track to deliver, to achieve, our 2012 business objectives and are excited about how we position the business for continued growth in 2013 and beyond. I'll now turn the call over to Julie, who will provide some color on the Q3 results and our financial outlook.

Thanks, Steve, and good afternoon, everyone. Third quarter revenue increased 18% year-over-year to $212.7 million, in line with our expectations. Assuming constant currency, revenue would have been 4% higher. Click-based revenue was up 15% year-over-year to $168 million. When comparing year-over-year hotel shopper growth, approximately 30%, to click-based revenue growth, please note that revenue growth this past quarter was negatively impacted by approximately 5% from foreign exchange, 5% to 6% from last year's site redesign and 3% to 5% from the spin-related Expedia pricing changes. Additionally, as Steve mentioned, we believe that our hotel shopper traffic and revenue would have been 2% higher were it not for the lower traffic during the Summer Olympics. Display-based advertising revenue grew 12% year-over-year to $23.6 million, driven by solid growth in the Americas, offset by weakness in EMEA due to persistent macroeconomic softness in Southern Europe. Assuming constant currency, display-based revenue would have been 2% higher. Subscription, transaction and other revenue, which includes business listings, vacation rentals and our transaction businesses, had a solid Q3, growing 53% to $21.1 million. In business listings, new sales reps are ramping globally, and we expect to see continued productivity improvements as we enter 2013. As it relates to geographic mix, despite macroeconomic challenges and unfavorable foreign exchange movement, non-U.S. revenue was 50% of total revenue during the third quarter, compared to 49% in Q2 and 47% in the third quarter of 2011. In general, we saw click growth accelerate in our core markets. And on the pricing side, we saw stabilizing to improving pricing abroad. Specifically in EMEA, we saw accelerating growth in the U.K. despite a persistently tough pricing environment. Also, Southern Europe, although still weak, showed encouraging signs of pricing stabilization. Our newer Russian market remains in hyper-growth mode, albeit growing off a smaller base. In our fast-growing Latin America and APAC markets, we saw continued strong hotel shopper growth and solid pricing trends, driven by stronger coverage, traffic acquisition, brand building and content collection initiatives. Notably, Brazil remains in a hyper-growth mode as well. And to put it into perspective, it's fast approaching Spain and Italy in terms of click-based revenue. On the expense side, non-GAAP sales and marketing expenses were $66.5 million or 31% of revenue for the quarter, consistent sequentially on a percentage of revenue basis. Given that Q4 is our seasonally lowest revenue quarter, we expect to see deleveraging in our non-GAAP sales and marketing spend. This translates into sales and marketing spend in the high-30s to low-40s as a percent of revenue for the fourth quarter and mid-30s as a percentage of revenue for full year 2012. Moving forward, we continue to prioritize traffic diversification, international expansion and a hotel shopper growth as we identify underleveraged marketing channels to meet these growth objectives. Non-GAAP tech and content expenses were $20.3 million or 10% of revenue for the quarter, and non-GAAP G&A expenses were $16 million or 8% of revenue. Both were in line with our expectations, and we expect these expense lines to stay relatively constant on a dollar basis in Q4. Our Q3 effective tax rate of 34.4% was higher than we expected due to an income mix shift towards jurisdictions with higher statutory rates, as well as an increase in valuation allowance for one of our U.S. subsidiaries. For the balance of 2012, we expect our effective tax rate to remain relatively consistent with our year-to-date rate. Cash flow from operations was strong at $76.5 million or 36% of revenue for the quarter. We continue to expect full year 2012 cash flow from operations to track with adjusted EBITDA after providing for interest and taxes, or low-30s as a percentage of revenue. In terms of headcount, we ended Q3 with 1,483 employees, up from 1,444 employees at June 30 and 1,196 at the end of Q3 2011. With that, let me turn to our outlook for the balance of the year. Given our year-to-date performance, we continue to expect 2012 total revenue growth in the high teens. Driven mainly by continued strong hotel shopper growth and the tailwinds created from lapping last year's site redesign, we now expect click-based revenue growth to be in the high teens for the full year. While fourth quarter-to-date volume and conversion remain strong, pricing continues to be a headwind given the persistent macroeconomic softness overseas, as well as the fact that we are still comping higher pre-spin Expedia pricing. We continue to expect low double-digit annual revenue growth for display-based revenue, as it faces a tough Q4 comp based on unusually strong demand in Q4 2011, which may not reoccur. We also expect full year subscription, transaction and other revenue growth to be consistent with our year-to-date results. Based on our continued profitability outperformance, we now expect full year 2012 adjusted EBITDA growth in the high single digits. We are in the midst of our 2013 planning process and will provide our expectations for next year on our Q4 call. In closing, our year-to-date performance keeps us well on track to achieve our 2012 goals. We will now open the call up to your questions.

[Operator Instructions] Our first question is from Mike Olson of Piper Jaffray.

Could you just talk a little bit more about kind of the traffic acquisition strategy and how anything may have changed versus what you talked about last quarter and, ultimately, kind of how you're thinking about it going forward?

Sure, Mike. Thanks for the question. I'd say we had put the plan in place when we spoke last time in terms of reducing some of the SEM spend on traffic that wasn't profitable when we measured it through to conversions, and that has remained the plan. It was a relatively easy change for us to make back then, and so the search engine team has been focused on continuing to grow that channel and they've had some success. That's still just one piece of the overall picture. We have essentially all of the different traffic acquisition vehicles and customer acquisition vehicles in terms of email to marketable members ongoing with respect to Facebook, affiliates and all the rest of the mechanisms by which we can attract traffic to our site. Shorter answer is not much of a change from prior quarter.

Our next question is from Laura Martin of Needham & Company. Laura A. Martin - Needham & Company, LLC, Research Division: I just wanted to check in, Steve, it sounds like maybe we've had some movement on mobile. You said that desktop is 3/4 of total mobile usage. My recollection, my estimate there, was about 70% pricing as a percent of desktop and smartphones at about 10% of pricing. It sounds like maybe both of those are getting better. Is that a fair readthrough from your comments?

Well, let me try to paraphrase. We haven't actually shared the specific breakout between desktop and mobile in terms of traffic. What we're saying is the -- when the phrase mobile is used, we often refer to both tablet and phone and that 3/4 of the revenue generated from tablet and phone remains on the tablet. We have seen some improvements in our monetization rate on the tablet. That's all good. And it's kind of what I had told you we expected to see over time as the tablet usage much more closely resembles desktop. On the phone, we haven't seen a meaningful change in the monetization rate, i.e., it's still a very small percentage of desktop. And we think it's form factor. We think it's in market activity. We think it's folks who are thinking about the trip, checking out some hotels and some whatever, who just aren't ready to actually spend the money right then and there because our typical usage pattern is not, "Last minute, landed in the city, need a hotel room," that some other apps may have. So we've introduced, from a testing perspective, meta display to the smartphones because we think that's just a much better experience when someone wants to check the prices and do that base level of research. Because we're showing the prices in line, we think the conversion rate of the leads will be higher downstream. So the overall monetization rate of that display will be better for us. And I'm just cautioning that in that mobile environment, a, it's a small part of our traffic; b, it's a poorly monetizing, small part of our traffic. So the shift from the current monetization model to a meta display on phone isn't a particularly material financial event for us.

Our next question is from Douglas Anmuth of JPMorgan. Kaizad Gotla - JP Morgan Chase & Co, Research Division: It's Kaizad Gotla in for Doug. I was wondering if you could just give us some color on the feedback you're getting around the traffic quality changes and the new bidding system. Any feedback from OTAs that you're getting would be helpful.

Sure. So it's always kind of hard to parse out what they're saying and what they're doing. We've seen some cases where partners have lifted their bids. Is it seasonal? Is it more hungry for reach? Or is it they can afford more because our traffic is better? So I don't, unfortunately, have a whole lot to offer by way of direct feedback. In terms of the bid piece, we haven't really changed anything in terms of how partners can bid on our traffic. What we've really done -- started to do in these tests markets is provide more transparency about what their bid currently gets them in terms of the rank reporting. And as I said, we -- some partners have started to use that information. There hasn't been any meaningful changes that we've been able to detect, either up or down, with respect to that having a material impact on us. We can say that the partners that have some of this information are grateful for it. They appreciate us giving it to them, and it's working their way through their systems as to how they might be able to take advantage of it.

Our next question is from Nat Schindler of Bank America Merrill Lynch.

One, your tax rate of 34.5% and -- or even the 30% that you've had historically, is still quite a bit higher than other Internet companies that are running about 50% domestic and 50% international. Is there a chance that you can get that lower, much closer to the Google or/eBay rates that they see in the, near the 20% by different Irish, Dutch Sandwich things? And on a second question, given the last 2 quarters at 49% and then 50% this quarter EBITDA margins, have you pretty -- are you willing to kind of throw out your 42% to 47% EBITDA margin guidance that you have -- you gave when you did the spin?

It's Julie. I'll take those. So first, on the tax rate, and one thing to keep in mind, up until December of -- end of December of last year, we were part of Expedia and part of their tax planning and mobile IP strategy. And now that we are a stand-alone company and have our own tax department and are looking at where our IP is located and what makes the best business sense, we're definitely looking into that. One thing that also plays into TripAdvisor's negative impact on the rates is that we're headquartered in Massachusetts, which has a higher tax rate than where some of the other players may be located. So we're working on it, I'd love a lower tax rate, and it is definitely part of our strategy going forward. Regarding the EBITDA margins, I want to remind you that we did not give that. That 42% to 47% was not guidance and, or a long-term target model. It was just to illustrate the effect of the spin on our 2010 numbers. So we've had things like the additional G&A costs, the decrease in Expedia pricing. Those -- that impact was applied against 2010 just to give an illustrative model. We're in the process of -- we've had a lot of movements in 2013 -- excuse me, in 2012. We are actively looking for investment opportunities for sales and marketing, for traffic diversification and so forth and -- even from quarter to quarter, we are able to spend less or more in sales and marketing. And we're looking at what 2000 -- we're in the planning side for 2013, and we'll come back to you on that in -- on our Q4 call.

Our next question is from Lloyd Walmsley of Deutsche Bank.

Just wondering if you could comment a bit. It looked like we saw big ramp in Expedia's general direct marketing expense in terms of year-over-year growth in the third quarter. And obviously, they've made it clear that they're adjusting spend levels vis-à-vis, you guys. But I was just wondering, at what point do you think you can be participating in some of that growth again from Expedia? And where do you think they're spending that incremental share as they accelerate spend?

Sure. Perfectly fair, interesting question. Unfortunately, I really don't have any insight into where Expedia is spending their funds on the marketing side. I would expect as our -- presuming our traffic continues to grow at the rate we have been growing, one would naturally assume that Expedia would be spending more with us in coming years. We've gone through the step function. We've brought things down to what we believe are much closer to kind of market rates. And now it's a bidding environment. And to the degree to which we can get everyone to be bidding up for the traffic we have, we'll see that increased spend. I think Expedia, like our other clients, are very smart on the traffic acquisition side. So we remain, to the best of our knowledge, an incredibly important, valuable partner of theirs in terms of buying direct leads. Their sales and marketing line, I believe, encompasses quite a bit more than just the direct marketing. So in that -- in their marketing spend, it could well be in other areas that we don't participate in: TV, radio, whatever it might be. You'd have to ask them about that part.

Our next question is from Anthony DiClemente of Barclays.

Just, as we look into 2013, I just wonder if you guys can bridge through for us the lapping or the anniversary effect of a number of the changes that you've already alluded to on this call and how we should think about maybe even just modeling through things like lapping through the CPC step-down, the site de-cluttering. Can you call any of those out as we model through for even just the first half of 2013, on revenue at least? That will be helpful for us.

So we tried to provide you some of that bridge in my prepared remarks by calling out things that were either onetime or not normal course of business. So FX. And we forecast in constant currency. So anything that we talk about on a forward-looking basis is using exchange rates as of today. So we're not trying to predict what's going to happen there. So you can make your own decision. So we've had some, just looking at Q3, some onetime, such as the Olympics. So that's 2% on overall -- on CPC tax-linked revenue. And when you extrapolate that out, it's probably about 1.5% off of our total year-over-year growth rate. The site redesign is probably the largest one. We -- as Steve mentioned, we lapped that at the end of Q3. At that time, we were able to clearly see that it had negatively impacted our year-over-year growth rate by about 5% to 6%. So like we would expect that trend to continue forward into Q4 and beyond. And then the other mention that we made was the Expedia pricing. Now that contract anniversaries the date of the spin. And we had said that originally, it's probably a 3% to 5% negative impact on revenue. As of the last couple of quarters, we've been saying it's at the high end of that range. So that -- we'll anniversary that at the end of this quarter.

And Julie, that's all helpful. One follow-up on that last one. Did you have a -- was there a spending floor that was in place? And is there -- should we be concerned like about the possibility of another step-down after lapping the expiry of the post-spin contract? Like can that -- can Expedia spend step down again? Or is that likely or unlikely?

I think we put the agreement in place to make for a smooth transition from where both sides didn't quite know what would happen when the companies were unbundled, if you will. We think the prices have gotten much closer to market rates, so we're not forecasting a big step-down function. Regardless of what we forecast, truly, it is a bid environment. So there are meaningful questions one gets to ask Expedia and all the other bidders, in terms of how much are they interested in growing their share, what are the competitive forces that are going to get the bids to go up. So yes, as Julie said, we anniversary the agreement. But obviously, we've known it's been coming for quite some time.

Our next question is from Peter Stabler of Wells Fargo Securities.

Now going back to the property rank reporting data that -- where you have the test -- 3 test markets I think you said, could you envision a situation where you would not elect to roll that out widely? It would seem like very valuable data, irrespective of whether an OTA partner has explicitly asked for it or not. And then secondly, wondering if you could comment on kind of the lifetime value of your reviews? You continue to add review content kind of hand-over-fist. But at what point does the value kind of expire a little bit? I mean, the 6-year-old review is clearly not as valuable as the 6-month-old review. I'm wondering if you could just kind of comment on those trends.

Sure. Two good questions. So on the rank reporting, I haven't heard any feedback from any of our test partners that say, "Thanks, but no thanks, we weren't really interested in this information." I think we more hear, "Wow, it's a lot of information. It's a lot to digest." And to put it in perspective, if we have a client that's bidding on 100,000 properties times, call it, 20 different markets, that's just a lot of data to digest and make sense of. So we're having to be measured in our rollout per our client request, basically. So I don't -- it's a little hard for me to imagine a situation where over time, we don't have this information available to everyone. But I'm a lot more cautious about trying to predict in which quarter it will be both available and consumed for everyone. To your second part of your question, lifetime value of the reviews, we heartily agree that the 6-year-old review is perhaps nice to have on the site because it mentions something that's unique to that review, a tip in the neighborhood or something like that, climbed on top of the hill to watch a beautiful sunset. The 6-year-old review really isn't part of our popularity index calculation anymore either, because it's so old and so much can change in the 6 years. Similarly, we have some hotels north of 8,000 reviews. And they may be all current, but you still don't need 8,000 to have a lot of liquidity of opinion for that property. The beauty of what we're able to do with generating so many more reviews is sort of provide fantastic coverage on not only the top properties in the city, but kind of all the properties in the city, in the country, all the restaurants. So it's getting to the point where it's tough to find a hotel with less than 50 reviews. And that's what we're getting pretty excited about because everyone can look at 5 reviews on a property and say, "What's the big deal?" With 50, you start to really understand what you're going to get when you stay there. And the same applies to restaurants and the same applies to tours and attractions and museums and kind of everything else that goes into what folks are looking at. So sure, if you took the 6-year-old reviews off the site, no one would notice. The volume was so small anyways, it practically doesn't matter. We keep them on there just as a convenience function for our current customers.

Our next question is from Heath Terry of Goldman Sachs.

I was just wondering if you could give us a sense of what you're seeing from a usage perspective within mobile, whether or not as the medium matures for you, whether you're seeing more transactional activity within mobile versus research and whether or not the -- how different the segments, hotel versus activities, are within your mobile channels versus desktop?

Well, I'm pretty sure that the activity is distinctly different. There's more non-hotel activity on the phone versus the desktop or tablet. Monetization is certainly -- a lot fewer people are clicking off to see the price, perhaps because they know when they do, that function on our mobile device right now, on the phone, it's a pretty painful experience in order to get back to the app and all that. So we'll let you know on the next call how the meta experience is being accepted by the user on the smartphone device and whether that's in fact encouraging more commerce clicks and in turn, more conversions. I still kind of stand by my original comments in terms of the use case of the mobile phone. While some other apps or the OTA apps, if you find yourself in a city and you need a hotel room and you haven't done any advance planning, you're just probably more in the, "Hey, find me something close to where I am and I'm going to book it," and an OTA web app can -- an OTA app can do that for you quite well. We're in the more expensive, the more time-consuming, the more enjoyable research experience, which generally happens long before the day of booking. So we believe our usage will continue to grow dramatically in mobile. We believe we'll be able -- the phone specifically. We believe we'll be able to grow our monetization rate. But I am personally skeptical that we'll ever achieve the monetization rate that we have on desktop on the phones purely because of that usage case.

Our next question is from Herman Leung of Susquehanna International Group.

Great. This is Deepak sitting in for Herman. So just a question on Facebook as a channel. So could you discuss how the ROI from Facebook as a channel was during the quarter? Maybe give some qualitative color on that. And then what other ways, maybe some social ways, in which you can acquire customers using the Facebook channel?

Sure. So Facebook remains a meaningful channel to grow our brand awareness and to grow our membership. I've tried to be reasonably clear that we don't view it as a very large traffic acquisition channel, certainly not compared to a channel like search. So you have folks on Facebook that are engaged in sharing and learning about what their friends are into, posting photos and the rest. And at many times, they're interested in interacting with our brand, with our Cities I've Visited or seeing a status update from their friends about the trips that they've taken. The Cities I've Visited app, to use it, you're joining TripAdvisor, and you're telling us something about yourself, namely all the different places that you've been. You're also giving us your friend network so that we can share all those places that you've been as data when your friends come and browse on TripAdvisor. To recap, Facebook as a member acquisition channel is quite strong for us because we spend a certain amount to advertise our map and TripAdvisor applications, we get a certain uptake, that's at a somewhat expensive rate. That's sort of qualitative because it's just a member, it's not a travel booking yet. But because of what those folks are doing with the app, it's generating invitations to other users, it's generating status feeds, it's generating Open Graph items and I can go through the laundry list of Facebook viral channels, which in turn are generating a lot more members for us for free. We're not having to pay to acquire. So we continue our spend on that channel because it helps us grow our membership and it helps us grow our social content, namely knowing where your friends have been to in order to make progress on our social and personalization efforts, which I've touched upon. So think of it as building a huge database of travelers, travelers who happen to be Facebook connected. We're learning more and more about them, whether or not they've been to TripAdvisor yet, we're teaching them about TripAdvisor, and we're getting the click-through rates and the emails that say, "Hey, they're interested in this. So they're coming back." And then the longer-term question is and through that mechanism, will they be converted to the more domain-direct, the more branded association. We see positive signs, but the jury is still out by several quarters, in our opinion, as to whether that's going to move the needle as much as we hope going forward.

Our next question is from Brian Nowak of Nomura.

The first one is with Expedia kind of looking to grow the agency business in the U.S., I was wondering if you could talk about any differences in cost per click or monetization that you see in agency-dominated markets versus merchant hotel-dominated markets. And then the second one, I'm curious to hear about -- I know there's been a lot of focus in the kind of the conversion of your OTA partners as a focal point in the last quarter. So I was kind of curious about any commentary about the direction of the ROI and your partner conversion this quarter versus last quarter?

Sure. So on the Expedia agency business, we're certainly aware of their offering choice to their consumers. We don't have inside information, but it stands to reason one of the opportunities for them is to increase their conversion rate even if the margin or the finances behind it might be different. And I say might because again, I'm not part of it. Our respective interests are perfectly aligned because if offering the pay at checkout feature improves conversion, then they're making more money per -- not necessarily per booking but per lead I send them because of the higher conversion rate. Therefore, they can afford to pay me as much as they were before or even more if they dip into some of their efficiency numbers. So we don't view the percentage uptake of their agency business as bad in any way should it be strong in certain markets. If it helps -- another way to look at it is if it helps them be competitive in certain markets, where they didn't have the product offering before, fantastic, because then they'll be able to bid more for traffic in those markets. Again, the conversion rate is expected to make up for any loss in merchant revenues. It's actually a little bit related to the OTA conversion, the second part of your question. So when we look at the leads that we send to our clients, I don't really ask them, "Are my leads better converting than a Google or somebody else?" I look to see, on a per-dollar spend basis, "Do we generate more profit for you folks than your other channels?" And we continue to hear that our clients are bidding independently, i.e., they're not favoring TripAdvisor over Google or vice versa. They'll spend wherever they get the next best ROI. So if a TripAdvisor lead converts at twice the rate of a Google lead, then they're probably paying twice as much for it. Then the converse is true, too. Will comparing the actual CPC or the actual conversion independently, not really the question you want to ask, it's the "Hey, how's the efficiency of TripAdvisor as a traffic source to these clients?" And while every one of our clients, no doubt, would like an even more efficient source, they're all free to lower their bids at any point, which will get them less traffic but presumably a little bit more efficient. And they're not doing that. So they've -- it's always changing, but it seems like they've found a happy balance in terms of the efficiency that they're willing to spend on TripAdvisor and the conversion that, that drives.

And I'm not showing any further questions in the queue.

All right then. Well, as we closed with before, we're very proud of the quarter, and we see trends going forward that we very much like. I want to thank everyone for joining our call today. And I really want to thank all the TripAdvisor employees around the globe for their hard work throughout the quarter, and we know we can count on it going forward. So we've made some great progress on our growth initiatives, very excited about it. And we look forward to updating all of you on our progress in early 2013. Thanks again.

Ladies and gentlemen, thank you for participating in today's conference. This concludes today's program. You may all disconnect. Everyone have a great day.