Will Lyons - Former Senior Director of Investor Relations Stephen Kaufer - Co-Founder, Chief Executive Officer, President, Director and Member of Executive Committee Julie M. B. Bradley - Chief Financial Officer, Chief Accounting Officer, Senior Vice President and Treasurer.

Lloyd Walmsley - Deutsche Bank AG, Research Division Mark S. Mahaney - RBC Capital Markets, LLC, Research Division Eric James Sheridan - UBS Investment Bank, Research Division Kenneth Sena - Evercore ISI, Research Division Scott W. Devitt - Stifel, Nicolaus & Company, Incorporated, Research Division Nathaniel H.

Schindler - BofA Merrill Lynch, Research Division Michael J. Olson - Piper Jaffray Companies, Research Division Peter Stabler - Wells Fargo Securities, LLC, Research Division Douglas Anmuth - JP Morgan Chase & Co, Research Division Brian Nowak - Susquehanna Financial Group, LLLP, Research Division Heath P.

Terry - Goldman Sachs Group Inc., Research Division Ron Zember.

Good afternoon, and welcome to TripAdvisor's Third Quarter 2014 Earnings Conference Call. As a reminder, today's conference is being recorded. At this time, I'd like to turn the conference over to Mr. Will Lyons, TripAdvisor's Senior Director of Investor Relations. Please go ahead..

Thanks, Kate. Good afternoon, everyone, and welcome to TripAdvisor's Third Quarter 2014 Earnings Conference Call. Joining me today are Steve Kaufer, CEO; and Julie Bradley, CFO. We distributed our earnings release through our Investor Relations website located at ir.tripadvisor.com.

In it, you will find reconciliations of non-GAAP financial measures to the most comparable GAAP financial measures discussed on this call. On our Investor Relations website, we have also posted supplemental financial information including non-GAAP financial measures.

Before we begin, I'd like to remind you that estimates and other forward-looking statements included in this call represent the company's views as of today, November 4, 2014. TripAdvisor disclaims any obligation to update these statements to reflect future events or circumstances.

Please refer to today's earnings release and TripAdvisor's filings with the SEC for information concerning factors that could cause actual results to differ materially from those expressed or implied by such statements.

Finally, unless otherwise stated, all references to selling and marketing expense, general and administrative expense, technology and content expense and total expenses are non-GAAP measures as defined in our earnings release and all comparisons on this call will be against our results for the comparable period of 2013.

And now I'll turn the call over to Steve..

Thank you, Will, and welcome everyone. Total revenue growth accelerated to 39%, coming in at $354 million driven primarily by 31% growth in our click-based revenue as well as by our Viator acquisition. Adjusted EBITDA increased 14% to $119 million. These results, while strong, were not as strong as contemplated by our Q1 and Q2 guidance commentary.

Julie will provide more color in her remarks, but I would like to take this opportunity to address some of the factors that negatively affected our performance this quarter. In our click-based business, declines in clicks per hotel shopper on our site had been a headwind to our forecast since mid-August.

While we saw a similar decline in 2012, in years prior in our classic display, our results last year from our new metasearch display suggested a minimal, if any, seasonal decline. Current data suggest that we continue to be impacted by seasonality but to a lesser extent than pre-metasearch.

At this point, we believe the seasonal clicks per shopper trend may have been masked by optimization gains we were making in the metasearch experience throughout the back half of 2013.

While this continued seasonal impact makes us incrementally cautious, it's important to note that our click-based auction continues to be healthy, and our revenue per hotel shopper growth accelerated in Q3 to 14%, up from 11% growth in Q2.

Our EBITDA results reflect our revenue headwinds since August and also include acquisition-related expenses and the investments we're making to capture market share in restaurants and attractions.

Above all, our third quarter illustrates how we're executing on our strategy to deliver more value to users and partners throughout all phases of travel planning and trip taking. We're growing content and community. We're innovating our products and platform.

We're winning on mobile devices, and we're increasing the growth trajectory outside of hotels by welcoming Viator to the TripAdvisor family. In terms of content and community, we now have nearly 200 million reviews and opinions, growing at 115 contributions per minute in 28 different languages.

This fresh rich content helps travelers choose from more than 1.5 million places to stay including 890,000 hotels as well as 2.4 million restaurants and 480,000 attractions around the globe.

Traffic to our sites remains robust as Q3 monthly unique visitors was 315 million according to Google Analytics, up 20% year-over-year, and unique hotel shoppers grew 17% according to our log files. Nearly 50% of users visited TripAdvisor via tablet and phone, and we're continuing to see industry-leading uptake and engagement on mobile devices.

Traffic to non-hotel pages, including restaurants and attractions, continues to grow faster than our average. In a minute, I'll describe how we're leveraging our industry leading media assets and investing heavily to match more users with more advertising partners in these categories.

We're amplifying our consumer message through an offline ad campaign, and I'm pleased to add that we're seeing positive results. TV is increasingly cited in user surveys as the reason someone visited TripAdvisor. Google Trends data shows traffic uplifts in the U.S., France, Australia and Canada, the markets in which we have been on-air.

Traditional offline marketing is becoming a nice complement to our ongoing permanent branding efforts, the awards, certificates, stickers and badges that give consumers confidence at hotels, restaurants, attractions and on websites all around the globe. Based upon our experience this year, we expect to expand our TV ad campaign in 2015.

We're also in the early days of a number of product and platform initiatives that will make TripAdvisor more valuable to users and partners alike. In hotels, we have made nice progress on our Instant Booking initiatives since launching with 3 initial partners to a small percentage of U.S. smartphone users just 6 months ago.

On the user side, shoppers appear to understand and really like the Instant Booking experience on TripAdvisor. Users are clicking on the Book on TripAdvisor option more frequently than its metasearch equivalent and most importantly, users are converting to bookings at a higher rate. As expected, we started rolling out Instant Booking to U.S.

desktop and tablet users in Q3. Notably, Best Western was the first supplier to make all of their properties instantly bookable on all devices. For now, the user cohort that can experience this feature is relatively small as we focus on the U.S. first before rolling out more broadly.

A more deliberate rollout schedule allows our teams greater flexibility to onboard more partners to optimize the booking funnel and to ensure that every booking is delivered quickly, accurately and securely. Instant Booking is a really great feature, and we are eager to get it in the hands of more hotel shoppers around the world.

On the partner side, momentum is building globally. We have 6 partners live and another 16 signed and pending implementation. Specific to OTAs, in addition to getaroom and Tingo, AMOMA and Jetsetter are now live on the site, and we are working through implementations with another 9 OTAs, including Asiatravel, HotelsClick and otel.com.

On the supplier side, in addition to Best Western and Choice, we've signed contracts and are working through implementation with another 7 chains. We like our recent traction, especially with regional international players in India, APAC and LatAm who are looking to increase their share of voice on our platform.

We've also introduced Instant Booking to independent hoteliers through our TripConnect platform. While this is still in beta, we have already connected a handful of Internet booking engine partners, and we expect to connect many more over time.

As we onboard these booking partners, the focus shifts to educating independent hoteliers about how they can access our global audience and drive bookings through a commission model rather than cost per click.

We are offering independents a new way to advertise on TripAdvisor, and we see this as additive to any OTA relationship that independent hoteliers may have. Beyond Instant Booking, after almost a year of extensive testing and iterating, we've officially launched our personalization feature, Just for You.

This is a big change as it fundamentally transforms our one-size-fits-all sort order into a highly adaptive recommendation engine, matching travelers with their ideal hotels based on content, travel preferences, intent to purchase, site behavior and users like you.

Users can now see this feature globally in all cities that have 10 or more hotel listings on desktop and tablet web. Getting personalization right is extremely difficult in travel, a category in which consumers' needs are nuanced, contextual and episodic. We're excited about personalization.

We're applying TripAdvisor's scale and consumer-centric focus to deliver users a more helpful and differentiated experience. We have a unique understanding of travelers and with Just for You for hotels, we believe that we are just scratching the surface of what's possible.

Helping our users find, research and book the best hotel remains our top strategic priority. At the same time, we're engaged in the early stages of leveraging our significant media assets in restaurants and attractions to unlock even more value for TripAdvisor users, advertising partners and our business.

Restaurants are a natural extension for TripAdvisor given our fast-growing global audience of both travelers and locals, our rich content on 2.4 million restaurant listings and our growing lead in mobile.

In just a short time since our LaFourchette acquisition, we've streamlined the research and reservation booking experience and rolled out a new instant reservation feature. This has tripled the number of seated diners from TripAdvisor through the LaFourchette platform.

We're also working hard to expand LaFourchette's reach to more international markets. We're executing a similar playbook in attractions, our third largest source of demand after hotels and restaurants.

From a long-term perspective, our acquisition of Viator, a global tours and attractions leader with 20,000 listings in 1,500 destinations in 10 different languages, was one of the biggest developments this quarter. Similar to our strong position in accommodations and restaurants, we are already an established global leader in this category.

With Viator, we can now offer consumers an end-to-end research and direct booking experience. Since closing the deal in August, we have seen a significant uptick in attraction bookings through our platform. We see a lot of opportunity ahead in what PhoCusWright calls an $80 billion annualized market opportunity.

LaFourchette and Viator are exciting developments for our long-term growth. We've added 2 great teams, we've reinforced our user value proposition in these categories and I'm very pleased with the path we're on. In mobile, the ongoing shift plays to our competitive strengths.

Nearly 50% of our users are on mobile devices now, and we just crossed the 150 million cumulative download mark for our top ranked apps. This includes more than 135 million downloads of our best-in-class TripAdvisor app, which itself has been growing organically at triple digit rates for the past 5 straight quarters.

Looking specifically at smartphone. We've been executing well against our multiphase plan. We added metasearch in 2012, native apps last year and continued to reduce friction for more users on more screens in more ways throughout travel planning and trip taking.

As a result, we continue to see triple digit revenue growth on smartphone driven by increased traffic and better monetization on that device. Repeat usage on our native apps has been rising this year, another solid indicator of success.

On TripAdvisor, advertisers get the unique opportunity to deliver the right advertisement to the right user at just the right time in front of a massive and growing global audience.

As we look forward to 2015, we are carefully considering a number of media-related ad products, particularly on mobile that serve to reinforce TripAdvisor as the perfect travel companion and making it an even more attractive platform for advertisers. Switching gears to another adjacent growth opportunity, Vacation Rentals.

Our free-to-list transaction model continues to gain traction. We now offer users nearly 650,000 vacation rental listings, up 36%, providing consumers with greater choice. We're leveraging search and CRM in order to drive increasing demand through better user conversion.

Free-to-list bookings increased more than 200% in Q3, and we see this becoming an even larger part of our Vacation Rental mix over time. To conclude, it's been a very solid 2014 so far. The fundamentals of the business are strong, and I'm pleased with our progress in all of our key initiatives.

Continuing to move fast is imperative, and we believe investing today in our brand, product innovation and expanding our platform is the best way to drive long-term, broad-based and profitable growth. With that, I'll turn the call over to Julie..

Thank you, Steve, and good afternoon, everyone. Third quarter total revenue was $354 million as growth accelerated to 39%, up from 31% in Q2. This was driven by our PPC business as well as nice contribution from our new restaurants and attraction businesses, LaFourchette and Viator. FX provided a 1% tailwind.

Click-based revenue growth accelerated to 31%, up from 28% in Q2 and 13% in Q3 of last year due to stable, strong CPC pricing in the metasearch auction as well as from lapping our metasearch transition last year. As Steve mentioned, Q3 performance was not as robust as contemplated in Q1 and Q2 guidance commentary.

Since mid-August, we have seen greater seasonal decline in our clicks per shopper metric than in comparable periods of 2013, but which prominently existed in the years prior to the metasearch transition. Our most recent data suggest that seasonality is reduced but not eliminated by metasearch.

Given stable hotel shopper growth and stable CPC pricing, the clicks per hotel shopper trends were a tailwind -- excuse me, were a headwind to our Q3 click-based revenue growth, and our updated outlook factors in continued seasonality in Q4. This illustrates how impacts from transformative improvements can be extremely difficult to forecast.

Starting with our 2011 site redesign and then metasearch in late 2012 and 2013, impacts from these initiatives are amplified, yet are not always clearly visible until we fully lap the change.

Display revenue growth remains solid at 13%, despite a 14 point tougher comp as marketers look to gain access to our community of over 300 million monthly visitors.

We're seeing rising spend from the destination marketing organization sector worldwide, and hoteliers are increasingly shifting their budgets to an always on spend cadence rather than traditional, sporadic campaign investments. I will note that the 13% growth compares very favorably with declines posted by many other CPM-based businesses.

We have always maintained a premium position in the travel marketplace and continue to be considered a must-buy by many of the major travel brands. Subscription, transaction and other revenue grew 106% and included a full quarter contribution from LaFourchette as well as consolidated Viator results since August 8 closing date.

Steve noted the continued traction from free-to-list in Vacation Rentals, and we expect that our portfolio mix will continue to shift towards this transaction-based product. On the expense side, in line with our expectations, total expenses increased 56% driven by ongoing investments in online and offline marketing and headcount.

More specifically, we invested approximately $19 million in offline advertising in Q3, up for more than $10 million in Q2 and approximately $2 million in Q3 2013. We are on track to invest just over $30 million in TV this year, and we are actively laying the groundwork to expand the campaign next year.

We also had a full quarter impact from our 3 acquisitions in Q2, and we have consolidated Viator in our results since August 8. As a result, adjusted EBITDA of $119 million was up 14% and adjusted EBITDA margin was 34%.

We're focused on fueling business growth not only in the hotels, but also in restaurants, attractions, Vacation Rentals and China, inherently lower margin businesses that are also at an earlier stage of growth. We anticipate continuing to be a net investor in these non-hotel categories to rapidly scale and capture market share. Moving on to taxes.

Our Q3 GAAP effective tax rate of 28% decreased sequentially primarily due to changes in full year forecasted geographical mix. We continue to expect our GAAP effective tax rate for the full year to be in line with our 2013 rate of approximately 28%. As for share count, we had 146 million diluted shares outstanding during Q3.

In terms of cash flow, we generated approximately $58 million of cash from operations during the third quarter of 2014, primarily due to seasonality and timing in deferred merchant payables as well as timing of our offline ad campaign, customer receipts and vendor payments.

We also spent approximately $13 million on CapEx or 4% of revenue driven by capitalized engineering salary and leasehold improvements to support global office expansion.

From a liquidity standpoint, our cash, cash equivalents and short-term and long-term marketable securities decreased $107 million during the quarter to $614 million driven primarily by free cash flow of $45 million offset by the net cash outlay to complete our Viator acquisition.

We have just over $100 million remaining on our existing share repurchase plan, outstanding borrowings of $310 million as well as an undrawn credit facility of $200 million. As I turn to guidance, in click-based revenue, we expect that the clicks per hotel shopper seasonality headwind from Q3 will continue in Q4.

Given the high flow-through of the click-based business, its impact drops straight to the bottom line. In our display business, I'll remind everyone that Q4 is our toughest comp as we are still lapping the benefits from our delayed ad call innovation last year, and we also benefited from large end-of-year ad buys that may or may not reoccur.

In subscription, transaction and other revenue, our shift to free-to-list increases seasonality in Vacation Rentals, and our forecast for the balance of 2014 includes a full quarter's worth top line and bottom line impact from our Viator acquisition.

As for 2014 revenue expectations, we are reiterating our full year total revenue growth expectation of high 20s to low 30s. We now expect click-based revenue growth in the mid-20s.

We continue to expect display-based growth in the mid- to high teens, and we are raising expectations for subscription, transaction and other revenue growth to the low 80s, which includes a full quarter contribution from Viator. On the expense side, we are leveraging our very large and profitable hotel business to capture market share in attractions.

This is squarely in line with our stated philosophy, and we believe we are appropriately balancing our long-term and near-term growth objectives. Rolling it up, we now expect full year adjusted EBITDA growth in the low 20s.

Roughly half of our revised thoughts relate to the click-based seasonality headwinds, while the other half comes from acquisition-related expenses in ongoing investments to capture market share in attractions.

As a reminder, we do not manage the business to a margin target, in part due to our willingness to strategically invest in growth but also because our major product enhancements can impact top line.

While we measure countless metrics each day in order to run the business effectively and efficiently, we measure the success of our business over 2- to 5-year increments and firmly believe that we are making the correct strategic decisions to maximize value over that time frame.

I'll also note that these expectations are based on currently available data, including foreign exchange rates. As a reminder, our click-based revenue is highly sensitive to fluctuations in hotel shopper growth, clicks per shopper and CPC pricing, including those impacted by fluctuations in local currency.

Additionally, similar to last year's metasearch initiative, while Instant Booking is an exciting usability improvement for users, it may have financial impacts either positive or negative that we are unable to forecast. In closing, we're excited about what we've accomplished to date and believe we are well positioned for what lies ahead.

We will now open the call to your questions..

[Operator Instructions] Our first question comes from the line of Lloyd Walmsley with Deutsche Bank..

A couple, if I may. Just first, you mentioned in the prepared remarks, Steve, that when given the option, hotel shoppers click on the instant book option more often than the metasearch option.

Can you give us a sense for how revenue per hotel shopper is in these scenarios versus core metasearch? So is there enough conversion to make up for the reduction in metasearch clicks? And then secondly for Julie, it sounds like you guys are pretty upbeat on the TV spend, but it doesn't look like shopper growth is accelerating.

And then when you look at the direct marketing spend even excluding the TV, it looks like it's up about 6 percentage points year-over-year as a percent of click-based revenue.

Can you just kind of elaborate on what's driving that deleverage outside of TV and when we should start to see that impact shopper growth?.

How many clicks are coming in? What's the conversion rate within that funnel? What's the rate of credit card declines and all that? What's the commission rate that we're getting paid, which naturally varies by supplier, and multiply that out by the average daily rate and you'll get something that we calculate back to being an effective CPC and then we look and compare that against the other CPCs in the meta action.

Not surprisingly, early stage, we find sometimes the effective CPC is more than we're getting in the meta auction. Other times, it's less than we're getting in the meta auction, depending on who the provider is, depending on a bunch of other factors.

But frankly, we're still trying to understand in enough details so that when we go and roll this out more broadly, we're rolling it out in a smart way and effective way.

The last thing we want is for our consumers to be enticed with a Book on trip TripAdvisor click-through and then not get the experience that actually lets them finish the transaction, and we're pretty careful about not rolling that out sooner.

To jump in on kind of TV for a minute, it's always, and everyone will tell you, hard to measure the effectiveness of TV. We look at signs from our user survey. We look at Google Trends data. We look at our domain-direct. And it's pretty clear when we're on, we can see a lift and our consumers tell about that lift.

Why did you visit TripAdvisor today? "Well, I saw a TV ad." Always hard to know how many of those visitors would have come anyways because they would've stumbled across us through their normal research channels or come to domain-direct anyways.

So we've received enough signals to believe that it's working at a reasonable ROI for us, not a profitable ROI for us, as measured in a single week or a single month.

And we're keeping a sharp eye to understand how much of that return on investment takes time to build because after all, the whole point of that offline experience is to build up the brand awareness over time. We can't measure it in a single month or even a single campaign.

So tough for us to measure, but we've seen enough positive signs that we want to set expectations that it would be likely to be -- likely but not guaranteed to be growing for next year..

Great. Thanks, Lloyd. I'll take the second part of your question about the selling and marketing expenses.

When I look at the year-over-year compare as a percentage of revenue and exclude the net impact, which was about $16 million more TV spend in this past Q3 than a year ago quarter, we're only about 2% off -- 2 percentage points off for percentage of revenue.

I'll tell you, our LaFourchette acquisition and Viator to a smaller degree are more skewed towards sales and marketing expenses than technology expenses. So I think pretty much looking at as a percentage of revenue and also consistent in direct to indirect, which continues to be about a 70-30 split where we're right on the same path..

Our next question comes from the line of Mark Mahaney with RBC Capital Markets..

Two quick questions please. I think that your math says -- implies that you'll do very little TV advertising in the fourth quarter. Is that just for seasonal reasons? Because I think you talked about maybe amping that up a little bit more next year.

And then just briefly, are you seeing any signs of macro weakness in Europe? Priceline was talking about it this morning. It doesn't show up in your September quarter numbers.

Are they just seeing something different? Is there a reason they would see something different than what you're seeing?.

Sure, Mark. So you're right. We expect to be much quieter in Q4 in terms of the TV spend. It's just not an optimal season to be advertising in our current market.

So it's us sort of catching up to what always should be done, and then you'll see a sort of more spread as opposed to more concentrated spend next year according to our current thought process..

Sure, and I'll take the EMEA question. So our guidance does show that the trend line for EMEA is not much different than what we've seen in the past where we're always having a decrease in revenue -- percent of revenue from EMEA. However, we do include in our guidance the negative impact from FX.

So if I was just to look at October, for example, baked into that is about a 2% headwind and extrapolated over the balance of the quarter..

Our next question comes from the line of Eric Sheridan with UBS..

Steve, on Instant Booking, you made a comment that you're seeing a conversion of those bookings at a higher rate. Wanted to know if we can understand a little bit of quantification around that, and you also mentioned that there's another 7 chains you're working with.

Wanted to understand sort of the onboarding process and what the dialogue with the general hotel industry is about Instant Booking..

Sure. Happy to try to help, Eric. So we wouldn't ever go into specific conversion rate by partners, but what we have seen is that when it's particularly a supplier brand, we see a higher conversion rate than an intermediary like Tingo. That really wouldn't come as a surprise to anyone, but we've now seen it on desktop with Best Western.

I caution it's on a limited set of data. It's a limited number of OTAs and it's a limited number of chains. But the conversion rate, the presumed safety of booking with the supplier, we expect to continue. In terms of the chains, it's a bit of a new animal for a hotel chain to get their arms around.

What I mean is when we go talk with them, they say, "Another channel, sounds great. A commission-based channel, that's easier for me to understand than a click-based channel in terms of measuring an ROI for me. Sounds like you're close to an OTA." But wait a minute, we're not really an OTA.

We're facilitating a booking that's going to end up being supported service from the hotel itself, the hotel chain. So it's a little bit in between. It's an assisted book or an instant book, and that causes a little bit more confusion than we would like on the part of the hoteliers.

I think it's fair to generalize and say every supplier, every chain that we talk to would very much like to participate in the Instant Booking program. There's kind of no downside for them. The challenge is connectivity. There is some technical work that gets done on both sides in order to link the companies together.

And then there's a challenge of a contract in how they choose to allocate the commission dollars, whether we're treated like an OTA channel or we're treated like a direct booking, and each different chain has a different way of approaching that. Having said that, we now have a bunch more chains signed.

We're in legal review with quite a few others, and the pipeline continues to grow not as fast as we would like in terms of being live on the site, but it does continue to grow. And I have very little concern about not being able to sign up the chains because it's a fairly straightforward value proposition for them.

It's just the time frame that makes it very hard for me to predict..

Our next question comes from the line of Ken Sena with Evercore..

So you mentioned a rising usage rate or repeat rate on native apps. Maybe could you provide a little detail on that as well as some of the newer mobile marketing alternative channels that you're looking at exploring. And then also, just any color on maybe the use of app install ads and kind of what sort of efficiency you're seeing there..

Sure, Ken. So we are seeing our monthly active users growing on the app. I'd certainly expect that given the number of usability and other improvements that the team has put in to making the overall app just a much better experience.

When we refer to newer ad products, we look and say, as most companies of our size would say, there's a lot of traffic that's shifting over to the phone and a standard CPM ad doesn't work all that well there. We do have CPM ads available on the phone. They're not our preferred method of helping our advertisers reach the target audience.

So we look in 2015 and we have the challenge to say, "How are we going to monetize the eyeballs in a much more effective way than we have so far without obviously getting in the way of the user experience?" In terms of the growth of the actual app installs, we are particularly active in the download acquisition market, if that was the question.

So we get our app usage primarily from folks that find us in the App Store, have heard about us, recommended from friends or visit the mobile website and are teased to download the app, which is just a faster and better experience..

Our next question comes from the line of Scott Devitt with Stifel..

I had a couple. It does seem like the industry is becoming quite a bit more competitive, and TRIP -- the value in your system from a distribution standpoint resides in the traffic and the traffic growth, at least it seems.

And so I'm just wondering if the brand spend offline and otherwise is something that you think is going to continue to increase as a percentage of revenue over time as you secure your position of relevance with the OTAs and suppliers. And then secondly, Steve, you noted the click per hotel shopper being weaker than expected.

Expedia talked about the shift in -- and potentially toward their own platform in Trivago. And I'm wondering in terms of any success that maybe the OTAs are having with their metasearch product, whether that's shifting underlying budgets as well and your view on that..

Sure. So to take the second question first, I don't think budgets are shifting. All of our OTAs continue to want to spend as much as they can effectively on our platform, and when other clients talk about TripAdvisor's not being the most ROI-effective channel for them, that's really not us setting pricing.

That's the auction dynamic saying this is what the competitive marketplace will pay for the lead. So it's kind of a sign that pricing is pretty effective from our perspective. I haven't -- I don't think I've ever spoken with an OTA who was cutting a budget with a particular source so long as that source was profitable.

Every -- all of our clients want more and more traffic so long as the unit economics work for them. So less of a fixed budget mentality than anything that's profitable. In terms of the offline brand spend, it's absolutely a very competitive market out there trying to get first crack at a traveler who's planning a trip.

We have some inherent advantages in the sense that our content and our brand is plastered all over the place, offline, in the restaurants, in certificates in hotels. It might be the guest card after you leave a hotel. It might be an email, a solicitation to collect a review sent on the part of a hotel.

It might be the logo or the ratings on a destination marketing site or an OTA site globally, or just the individual hotelier's Facebook page that says, "Hey, you can come see the TripAdvisor reviews that are posted there through our widget." So we get a ton of wonderful branding that way that we have complemented with our offline campaign.

So when we try to estimate the amount that some of our competition is spending offline, we don't feel like we have to match that in order to achieve an even greater penetration in the consumers' mindset because of all the other activities that have been going on and are representing our content and our brand value proposition in a way that we really like.

It's decision support. It's helping the traveler plan that right research experience, plan that right trip and it all points back to TripAdvisor. So while we do currently expect an expansion of the ad campaign compared to our partners at least, it's off of quite a small base..

Our next question comes from the line of Nat Schindler with Bank of America Merrill Lynch..

You've said in the past that monetization of the smartphone was less than 5% of desktop. Then you've also said here that your monetization on -- your revenue on smartphones is triple-digit growth. So I'm assuming that distance has shrunk a bit.

Can you give us an update on that? And also, a little bit of discussion what has been the biggest driver of increased monetization on smartphones..

Well, I'd say that the biggest driver of revenue from smartphones has certainly been the growth in the installed base or the overall traffic growth split between mobile web and app installs. You take that, plus it's getting to scale.

So at scale, it's starting to fetch maybe a more active CPC bidding environment as more people are interested in the traffic. And then we've done a lot of hard work to make the experience and the frictionless part of the experience better. And so we're not at the point of kind of disclosing overall smartphone revenue.

But when we look forward and we see that traffic trend increasing, we feel like we're in a much better place than we were a couple years ago in terms of us being able to monetize that traffic shift..

Okay. On a related question, you can kind of tease out from some of the information you've given over the last several quarters that it looks like most, if not all, or potentially even more than all of your traffic growth has come from mobile.

Are you seeing growth in your desktop users still?.

So the combination of desktop and tablet, I believe, continues to grow, not nearly what it was several years ago as consumers shift over to using the phone for lots of browsing.

When we look at, and it's one of the things that's a challenge for us and probably every other Internet company that doesn't have signed-in users, how do we know how much that phone traffic is doing the research and then coming back to their familiar or big-screen desktop tablet to finish the research and do the purchase? Anecdotally, we know it's a chunk.

Metric-wise, it's a bit hard to actually measure..

Our next question comes from the line of Mike Olson with Piper Jaffray..

So you've gotten into restaurants and attractions.

Are there other categories that make sense to expand TripAdvisor as a booking channel and monetize some of the content that you have? I guess in other words, are there other categories where you have strong content that don't currently have any accompanying booking channel through which to monetize that content?.

Well, I guess sort of restaurants and attraction kind of flushes out the list for us. So it's the hotels, the airfare and the airlines, vacation rentals and now restaurants and attractions. And if you look at TripAdvisor, that's kind of most of the transaction-related things that we do.

There's -- I guess there's car rental we don't really touch and don't have any immediate plans, and we have a subsidiary, Cruise Critic, that takes care of the cruise area for us. So long-winded answer to kind of saying, no, I think we're pretty much set in terms of looking at the categories in which we can monetize our traffic..

Okay. And then separately, you mentioned China as one of the areas of continued investment. There's been signs of incrementally more competition in that market.

Are you guys seeing or anticipating China will become more competitive in the coming quarters?.

While we do view it as phenomenally competitive already with meaningfully different dynamics than any other country that we're in, you're right, we continue our investment because we look at the size of the outbound travel growth over the past couple of years and the next several and say getting a decent share of that is a pretty big win.

And we look at the outbound travel segment as an area where we have some pretty meaningful content differentiation from most of the other players.

And it's not that we plan to give up on any of the domestic travel, but focus on where our strengths are and our strengths are clearly leveraging the database of points of interest of hotels, attractions, restaurants and reviews, photos and the rest for any outbound destination that a Chinese resident would look to travel to..

Our next question comes from the line of Tom White with Macquarie. Our next question comes from the line of Peter Stabler with Wells Fargo Securities..

weight, weeks or countries. Could you comment on your expansion? And should we expect this to be adding geography or more weighted to the countries you're in? And then secondly, wondering, Steve, if you could elaborate a little bit on the Just for You personalization.

What kind of data signals are you getting? And wondering if you could share any color on what percent of your users routinely log in..

Sure. 2 excellent questions, Peter, thank you. In terms of the TV advertising, yes, we would expect to add new geographies to advertise in, and we're not sure whether we would have a meaningful up weight to our current markets. We'll figure that out as we go, can't give you much more color there.

In terms of the Just for You, we specifically built it so that you did not need to log in to get a benefit from that personalization. So right now if you're living outside of the U.S.

and you are looking at a major city, you'll see a Just for You list that's different in terms of the hotels that might be recommended for you than if you were looking at that same list in the U.S. In part, that's a simple filter on -- as so many of our reviews are written by Americans.

We're helping folks from other countries perhaps see a better recommendation not as tilted towards Americans. When you browse around the site, we're also going to start to influence the results, and browsing takes a number of different forms. You can browse, you can show much deeper purchase intent by looking at or even booking a particular property.

And so we want the experience to be valuable not just to our logged in users, but to those that browse.

If you're logged in and you've written a review in the past and we've seen what you've done, wow, we can go to a whole new level in terms of looking at the things that you've liked as in the positive reviews that you've given as well as what other people like you have also enjoyed and make more specific recommendations.

I take a moment to point out, as I did in my prepared remarks, that we do think it is just the beginning of what we're going to be able to do here. We have a tremendous amount of data.

Only a few other companies on the planet have the amount of data that we have on traveler preferences covering any type of lodging, all of the restaurants, all of the attractions that we have and how we can use that to personalize the experience that, at the end of the day, give you a better recommendation.

How do we measure that? Well, one of the things when we look -- how do we measure whether a recommendation is better is that as we give you a new sort on a hotel list, we're seeing people click on the hotels that are listed higher in the new sort order versus the old one.

Again, a simple indication that says the way we're presenting information is now better than it was before, and we're getting people to take the next step to read the reviews or go and book the properties.

Amusingly, they have some unintended short-term consequences, right, because if we get people to the exact property super quickly, they might only generate 1 or 2 clicks in that session versus 3 or 4 or 5. And since we charge on the meta land on a per-click basis, that might actually generate a little bit less revenue.

But wait a minute, since we're sending the user downstream to a client, and those clients fundamentally pay us for bookings not just clicks, because we're sending a better recommendation, they're much more likely to actually book, consummate the transaction so that the quality of our leads improves.

Quality of leads improves means the clients can pay us more, and we see that time and time again. So in turn, it might be a few -- it might be a little bit fewer clicks to, in fact, generate higher CPCs down the road as the way we actually make a bit more money by giving a better experience to the user.

We think to that specific example, it's far too much in the noise to call that out as a kind of revenue risk headwind, whatever. We think the personalized experience in getting the user to come back to TripAdvisor, because we made a better recommendation, is far and away the win as opposed to a specific monetary gain short term..

Our next question comes from the line of Douglas Anmuth with JPMorgan..

First, I just want to ask about seasonality, the comments that you had made earlier, Steve. I just wanted to understand what was so different here, and my understanding over the years is, in the travel sector, that seasonality is kind of generally more predictable.

Is it just that last year, you didn't see it as much because of the meta moves, but you had actually seen it in the years prior to that? And then just secondly, I know it's early here in terms of '15, but can you give us some thoughts on your kind of puts and takes around hotel shopper growth and revenue per hotel shopper as you think about it into '15?.

Doug, it's Julie. I'll take the first part of it. The seasonality that we are talking about related directly to the clicks per hotel shopper, not overall seasonality. We still do have Q3, revenue is much stronger seasonally in the third quarter than the other quarters.

But when we were looking at 2013 trend lines for clicks per hotel shopper, that metric, it stabilized last year once we had fully rolled out the metasearch. So we took that to mean that the metasearch significantly reduced the seasonality and -- because previous years, we had seen considerable declines in clicks per hotel shopper.

And -- but then based -- once we got into this year, mid-August in particularly, what we were seeing made us believe that historical seasonality does still exist, but it was just being muted by all of the great revenue optimizations that we were making last year.

So we believe it's less going forward with metasearch, but it is not eliminated and there's been no other -- given that everything else is stable, there's no other structural changes. It was just a seasonality that we didn't see in previous data..

Our next question comes from the line of Brian Nowak with SIG..

I have 2, please. Just to go back to the greater-than-expected seasonality in the clicks per hotel shopper.

As you study all the data from your site, is there anything that changed throughout this quarter that's kind of causing you to have to guide down the revenue and the decline in clicks per user? And what gives you confidence that this isn't a structural change or a change in the quality of the traffic so that 2015 clicks per shopper will be better? And then secondly, is there any way just so we can better understand -- can you quantify that metric a little bit more? Roughly, how many clicks are there per hotel shopper right now?.

Sure. When looking at the data, especially it's helpful to look back several years, and you can see the seasonality patterns in the year kind of increased in clicks per hotel shopper going into Q3 and then it declines coming out of Q3.

Now with meta, when we looked at that as we were rolling it out last year, there was -- once it was fully rolled out in the June time frame, we saw the stabilization in the clicks per hotel shopper and -- which would lead us to believe that this historical seasonality didn't exist because the consumers were able to see price and availability right online and they were further down the purchase path when they eventually clicked off.

So all that activity happening in Q3, they're on TripAdvisor but they're -- they didn't need to click in order to keep researching, so it made sense. What we saw in 2014 was some of that seasonality did exist, so we saw an uptick in clicks per shopper. And then starting in Midsummer, we saw a slight decline in clicks per hotel shopper.

So it's definitely being muted. It does not -- it was not eliminated. So we were able to go back to previous year's trend lines prior to the meta rollout and understand what a muted seasonality would look like versus an elimination, and that's what we've baked into our guidance for the back half of the year..

Great.

And any help at all on the rough number of clicks right now per hotel shopper?.

So we don't break out that number..

Our next question comes from the line of Heath Terry with Goldman Sachs..

I guess just if you could give us a sense, Steve. You've talked in the past about sort of the holdups to signing more hotel partners into instant book. You mentioned having sort of 7 more that you've been able to sign or begin the process of signing or bringing on board.

Can you give us a sense of what it was that was different about this 7 in terms of their willingness to join and maybe a bit of an update on what the existing roadblocks are for other hotel suppliers coming on to the platform?.

Sure. Thanks, Heath. I won't call out or I can't think of a specific roadblock. There's a standard commission negotiation, but what we've got is something different than what they're used to dealing with.

They're -- in general, a chain is used to dealing with a channel at a 12% commission, a 15% commission, a 20% commission and that's what every room is sold at. And in our case, we want partners to come on board, but it's still a sort of a quality score, if you will, on the Instant Booking opportunity.

So what I mean by that is we're looking to present the booking option to our consumers that kind of -- that does the best, the combination of has a good or the best price and has the brand that lets the -- that encourages the consumer to actually finish the transaction and that has a good level of service so that it can keep the TripAdvisor name behind it.

And when you take all those into account, it's just different than what a chain normally deals with when they sign up a channel. So with these 7, we got them through the process. We got them to understand what it is. They said, "Yes, okay.

I'll try it, sounds good," and they're in the implementation stages, which in turn varies quite a bit between the different chains depending on the level of APIs that we already have with other providers.

I go back to there isn't a particular reason why a chain doesn't want to participate, but there is a commission negotiation and there are some other terms that slow things down, for lack of a better way to phrase it..

Our last question will come from the line of Ron Zember with Nomura..

Quick question on Instant Booking on desktop and tablets versus mobile, and I recognize that it's still early innings here in terms of the rollout.

But if we look at the comparable time frame, is there a similar impact on desktop and tablets versus the impact that you saw on mobile in terms of pricing and in terms of conversions? And then I know you guys aren't quantifying clicks per shopper, but was wondering if you might be able to give us a qualitative understanding of how that's -- what the trajectories look like for the last few quarters..

Well, I can try on the Instant Booking. I'm not sure I can shed too much light on it.

Our rationale for kind of the more complete rollout on the phone was made as much on the user experience is just so much better on the phone with Instant Booking that the click-off model, regardless of whether it was going to be revenue positive or negative for us, we just needed to get there, learn our way, optimize the conversion path, which is what we did.

And part of the gains that we've seen are -- on the phone in terms of monetization and active usage is probably related to the Instant Booking, but we never kept a control group out there to really know exactly how much the Instant Booking drove our monetization improvement on the phone.

So for desktop tablet, we are being a lot more careful about it because monetization is one way that we look at measuring the success of it.

At the moment, I'd say the deliberate rollout is less on the monetization factor and more on the -- we want a bunch of partners in the store and with the ones that we have signed, that's a set of meaningful additions so that our consumers will be able to find good pricing, great availability as they go through the Instant Booking process.

We're still limited to the U.S., as I'm sure you understand, and so we'll make sure that we'll do our best to get that going before we look at a much broader rollout..

On the clicks per shopper side, last year, as we were rolling out metasearch, clicks per shopper decreased significantly throughout the rollout and then stabilized once we were at 100%.

And if I look at our exit rates for 2013 and compare it to the averages in 2014, kind of neutralizing seasonality, up a little in the first half of the year, down a little in the second half of the year, we're pretty close to our exit rates coming out of 2013..

I'd like to turn the call back over to Steve Kaufer for closing remarks..

Well, thank you. Everyone, we feel we have a tremendous number of exciting opportunities ahead of us. We've talked about many of them in the call, but it's very exciting to see the entire TripAdvisor story come together in terms of planning all the different aspects of a trip with our acquisitions in the restaurants and attractions category.

I really want to welcome the new teams to the family, and thank all of the TripAdvisor media group employees for the remarkable efforts that it takes everyone on the team to make this stuff happen. Thanks, again, everyone. I look forward to updating you all in 2015..

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