Ellie Mertz Reed Hastings - Founder, Chairman, Chief Executive Officer, President and Member of Stock Option Committee David Wells - Chief Financial Officer and Chief Accounting Officer

Mark S. Mahaney - Citigroup Inc, Research Division Scott W. Devitt - Morgan Stanley, Research Division Youssef H. Squali - Cantor Fitzgerald & Co., Research Division Douglas Anmuth - JP Morgan Chase & Co, Research Division Richard Greenfield - BTIG, LLC, Research Division Andy Hargreaves - Pacific Crest Securities, Inc., Research Division Anthony J. DiClemente - Barclays Capital, Research Division Jason S. Helfstein - Oppenheimer & Co. Inc., Research Division Anthony Wible - Janney Montgomery Scott LLC, Research Division Heath P. Terry - Goldman Sachs Group Inc., Research Division

Good day, ladies and gentlemen, and welcome to the Netflix Third Quarter 2012 Earnings Q&A Session. [Operator Instructions] Later, we will conduct a question-and-answer session, and instructions will follow at that time. As a reminder, this conference call is being recorded. I would now like to introduce the host for today's conference, Ellie Mertz, Vice President of Finance and Investor Relations. Please go ahead.

Thank you, and good afternoon. Welcome to the Netflix Third Quarter 2012 Earnings Q&A Session. I'm joined here by Reed Hastings, CEO; and David Wells, CFO. We announced our financial results for the third quarter at approximately 1:00 p.m. Pacific Time today. The shareholder letter and the Q3 financial results and the webcast of this Q&A session are all available at the company's Investor Relations website at ir.netflix.com. As is our standard practice, we will begin the call with questions received via email. Please email your questions to ir@netflix.com. After email Q&A, we will also open up the phone lines for additional questions not covered by the email Q&A or the letter. The dial-in number is within our investor letter, but let me repeat it now. Please call (760) 666-3613 if you'd like to get in the queue. We may make forward-looking statements during this call regarding the company's future performance. Actual results may differ materially from these statements due to risks and uncertainties related to the business. A detailed discussion of such risks and uncertainties is contained in our filings with the Securities and Exchange Commission, including our annual report on Form 10-K filed with the Commission on February 10, 2012. A rebroadcast of this Q&A session will be available at the Netflix website after 6:00 p.m. Pacific Time today. Now, let's move directly to questions.

As is our standard practice, we've organized the email questions by topic after we received them this afternoon. So let's start with questions on domestic subscribers. Question one. Your original thesis was 3 years for full brand recovery. Is this still the case? Can you recover faster?

It's Reed here. I'd say the 3-year model is consistent with what we're seeing. We've made substantial progress. As you can see, in the engagement metrics, we feel really good about the originals, particularly Arrested Development, because of its broad existing reputation in terms of burnishing our reputation. But 3 years is still the right timeframe.

And this is David. I would add, like I've said before, that we are seeing recovery in our brand reputation metrics the likelihood to recommend, but it's not at the level that it was pre-price change in July of last year. So there's still some room for recovery, but we've seen it trend in the right direction.

What changed, as you move through this year, that has caused you to miss your earlier guidance of 7 million net domestic streaming guidance by so much?

Well, we were within the range of our guidance from 90 days ago, but not within the annual guidance that we gave at the end of Q1. And roughly, I think we're feeling our way along as the streaming market grows, and we miss-predicted, but I would call that more of a forecasting error than anything else. When we think about growing 5 million net additions in this year, if we had predicted 5 million, I think we'd all be feeling good, that that's great growth. So we own it in terms of a bad forecast. But in terms of actual performance of the business to grow 5 million net adds, domestic, is substantial, and we feel good about that and about the growth next year.

Follow-on question. What does the new subscriber guidance of 5 million compared to the prior guidance of 7 million suggest about the addressable market opportunity you have?

Well, I think it makes it all a little bit harder. We think harder about the 60 million to 90 million, and we'll approach it not as fast as we would have if we had done 7 million this year. But the long term of every household becoming an Internet video household, I mean, that's definitely happening and going to happen. People are using Internet video in huge numbers, and that's only spreading. And we continue to lead the market in a big way. So I'd look at the long-term picture as very strong and not changed.

You predicted that the London Olympics might dampen subscriber additions and/or engagement in the third quarter. Did that play out in your actual data?

Yes, it did. We had a strong July, a soft August, and then a slower bounce back in September.

How can you improve involuntary churn rates in more mainstream, lower income households? Why don't I take this one? Some of the color that we've seen in terms of trends over the last couple of years is an increased use of debit cards and prepaid cards over credit cards. And the unfortunate thing about these different payment methods is that they have lower approval rates. And so what we've been doing is looking at a host of different things to improve those approval rates, and, in particular, looking to recover as many sale transactions as we can. So we're hopeful that there's a little bit of room that we have there to improve the overall involved churn rates. Let's move to questions about content. You have the advantage of perfect information on what your consumers watch. You have licensed a wide range of content, some of which is not viewed all that much. With this information, how do you intend to handle negotiations when deals come up for renewal? Would you pay less for a certain content or pay the same or more but require programmers to provide you with newer, fresher or better content?

We don't get the luxury of requiring programmers to do anything. We certainly look at the viewing as an indicator, and that limits what we're willing to pay for certain content. And if it doesn't present a clearing price and there's someone else willing to pay that much for the content, then the programmer will go to someone else. So I would say it's an informed auction market. We do have a lot of data that's very helpful in our overall programming. But it's a pretty straightforward auction, highest bidder gets the contents kind of market.

Question on the streaming hours. You have not talked about monthly or quarterly streaming hours since your comment about June 2012. Could you update us on recent activity?

Yes, sure. We were over 3 billion hours in Q3, which we feel great about, and it's continuing to grow.

With 2/3 of viewing now television, are you limiting the size of the audience that you appeal to by not having the more in-demand or on-demand film content that help build your DVD and then streaming businesses? How would that impact your cost of content?

Well, I think we're expanding our market with our TV series. If you look on cable or satellite, what gets watched more -- I mean, TV shows get watched a lot more than movies. So I think there's a big expansion in addressable market, both in the sense for any consumer's time and in terms of people who are interested in the Netflix service. And we're trying to be very strong both in movies and in TV shows, and we're only getting stronger as we grow.

Just a reminder here, too, that even in our DVD business, about 1/3 of the DVD shipments were new releases and 2/3 were things that didn't fall into any release category or older content. So many subscribers just want something good to watch, and they're less engaged in terms of caring about the freshness of the data or the content.

Next, a question on fixed versus variable licensing. [indiscernible] some of its recent deals a ethic [ph] to have a variable component on pricing based on subscribers. Given the lack of visibility you now have on future subs, is this something you would be willing to do? You've always resisted and it has worked to your advantage as you grew faster than content providers anticipated.

Now we're very happy with knowing exactly what our commitments are, and we don't have any plans to change our licensing structure. And the people that we generally bid against other networks are in the same model, where they're bidding fixed fee.

In terms of the goal to focus on licensing, only the contents of viewers, when you negotiate with these videos to license content, their goal is to bundle together as much as they can to maximize the value of the deal. How do you balance your desire for more efficient spend with the studios' desire to bundle more content?

We really look at packages as a whole, so we'll estimate for a given package how much viewing it will generate, which includes the stuff that's most appropriate for our members and the stuff that's less appropriate, and then try to figure out a clearing price for that set of content. And then, there's a back-and-forth of how well if you added this or took out this, depending on are they revenue-motivated or packaged-motivated, and that's all part of the give-and-take across the hundreds of different agreements that we have.

Disney's current output deal with Starz ends in 2015 with the [indiscernible] through Starz until the end of 2016. Why should Disney choose Netflix versus renewing with Starz or starting their own streaming movie service?

Well, I think in each market and country, there's a different set of content at a different time. But in the U.S. context, the Disney deal comes up in 2015, so it's a little premature. But we would certainly be bidders for that or other pay one deals as we have been in the past.

Can you discuss the device mix of content consumption? How are tablets and smartphones faring in terms of consumer hour streams?

Tablets are growing. And the smartphones are growing, you probably would expect that. Tablets are a little bigger than smartphones for us. But the big viewing is really on the television, smart TVs, game consoles, Apple TV, Heroku, Blu-ray players, all the things that bring Netflix to the big screen.

Moving to questions about original content. How do you quantify the success of proprietary content in terms of consumption and financial investment?

Well, we look at originals like we do third-party license content in terms of how much it's going to be viewed or how much it is viewed relative to its cost. So that's the primary measure. And then outside of that, we do expect some ancillary PR value, public relations value, and subscriber excitement around the originals, but we don't rely on that in terms of the primary relative measure.

Yes, so I think that's right. The base case is that it's good content that we're producing, and the upside is that it transforms the consumers' relationship with Netflix because it's only available on Netflix, it's not also on cable and also on DVD and also on pay-per-view. It's really, fully, totally unique only on Netflix within our territories.

Investors are quite surprised that you're releasing all the episodes of House of Cards at once, ensuring they will watch all the episodes over a weekend or 2 and then churn off. We're excited that Netflix is the first TV producer to put the consumer first and give them the content the way they want to consume it. How do you look at the trade-off between churn and hooking people on Netflix's other content if you're willing to bomb through 13 hours of House of Cards?

Well, we do want to hook people, but we want to hook them for the long-term, and we hook them for the long-term by treating them right and making all those episodes available at once. And obviously, there's later Arrested Development and Hemlock and Orange is the New Black and then a season 2 of House of Cards. So if we do our job right, there's always a reason to be a Netflix member on the original side in addition to the license side. So I don't think that it will be material, they join only for House of Cards and then exit. And so that makes us comfortable with this very consumer first trade-off.

The original content efforts, to what extent you have ownership rights that could generate additional revenues, syndication, DVD, licensing, et cetera?

It's a great discussion item for us over time. But on this first round, we're a licensee in a certain window and so we're not essentially a producer that then manages and profits from those ancillary rights.

Moving to questions about international. Recently, you commented that the Nordics have the most content of any international launch. Can you quantify the level of content and/or magnitude of that spend versus other regional launches?

No. We've been very successful in the Nordics. We're very happy with that, but I don't have any specific quantification in terms of dollars or title count that would be helpful. We do feel like, market by market, we are learning and getting better at the right content mix, both to attract the subscribers and economically, and we look forward to having a fuller update for you on Nordics once we're a quarter in, in January.

And I'd say that the basis for making those statements is an amalgam of metrics, it's not any one particular title count or dollar spend. Our experience has been, when we look at box office or Nielsen ratings or other types of metrics, that it's a combination of all of those in terms of the subscriber perception of content quality.

In its early years, Netflix incurred significant losses in starting up its DVD-by-mail service. As you look at international streaming, do you see parallels to the startup of DVD, or are you encountering significantly more challenges?

Well, I think it parallels with DVD or parallels with any successful business, which is you establish a new market and there's losses for some time period. And then, the key thing is the durability of the profit stream after that. We never invested in having DVD expand around the world because, number one, of the mail system's difficulty; but two is the profit stream wasn't going to be that long. In contrast with streaming, if we're able to establish a significant market share and franchise in a market and we manage that well, that should be a multi-decade profit stream, and so the DCF on that is enormous, which is why it gives us the confidence to make these big and aggressive investments in establishing our leadership in these markets.

Have you considered taking on partners and structuring JVs in certain countries? It appears like it could have 3 benefits: gain traction in countries where you have no brand equity; minimize the level of investment; and could accelerate Netflix's global ambition.

Well, it's definitely a factor, for example, when we look at Asia and some of the more challenging markets where U.S. firms have had a tough time and had progress in that. In Europe, most of the consumer firms have really been able to execute much better, faster, by keeping those wholly-owned, and that's our model for now.

Assuming that you are able to maintain positive consolidated earnings in 2013, what would be the maximum number of countries or regions you would consider launching them to?

I'm not sure I would approach that in terms of a maximum number. I mean, we look at a territory launch, it could have 4 countries in the case of the Nordics, but it's sold as a content block. So I think what we've said is we're focused on global profitability, but also on a path to profitability for our large existing markets. And so those are the 2 gating conditions that we'll be looking for in terms of finding that next market.

With regards to Latin America, what steps are you taking to improve payments? Have you considered withdrawing from any markets within Latin America? Have you considered bundling your offerings other than that market to improve your ability to collect? Second answer, on the payment side: On the payment side, we've been looking across a host of options to expand our ability to collect payments from subscribers, so let me we give you some examples of things that we're doing. In Mexico, what we found is that not all debit cards are accepted for e-commerce, and so we've been working directly with the banks to allow them to, at the -- I should say, allow us to accept debit cards online. And in Brazil, what we're looking to do this current quarter is to expand the payment methods that we offer, including things like direct debit [indiscernible] expand the addressable market of folks who have the ability to pay for our service.

And then in terms of bundling, that's not something that we're actively engaged in. We are looking at payment with other subscription services and trying to think that through. And then in terms of withdrawing, we're not thinking about that. All of our content agreements cover the entire region and are long-term in nature. So we're making great progress on the revenue side. So we're continuing to invest on that basis.

If you return to profitability for the overall -- if the return to profitability for the overall business takes 3 quarters or more, is there a risk that you'll miss an opportunity to have first mover advantage in continental Europe? Or does first mover advantage not matter since [indiscernible] had first mover advantage in the U.K. and you came later and surpassed them?

It's a matter of debate for us. Certainly, first mover helps, but it really depends on how big a scale you get. So if a competitor gets to great scale in one of those markets before we do, we would tend probably to focus on other markets. So it's -- there's no easy answer to that one. It's something we try to balance, which is our rate of investments on the current business versus the risk of competitors getting ahead. So far, we'd like the set of trade-offs that we've made.

Moving to questions about the product. How do you think about improving the platform from here to stay ahead of where the competition is going?

We're making a great progress on the algorithms that drive the merchandising, and I know they're invisible to all of you just like the Google Search algorithms are invisible. The UI is not the main focus because it's pretty easy to copy, and our main focus is showing the right content to the right person. When you turn on Netflix, whether that's on an iPad or a PS3, you could see 15 or 20 block shots, and if those are the right 15 or 20 block shots for you, for your mood for the time of the day, we're very likely to get a play. If those aren't the right block shots, then we're much less likely to get a play. And then that builds our engagement, and the more people watch, the more they retain. And so that's really the importance of the algorithm development and testing that we do.

I would also say that, for some subscribers, we're constantly testing user interfaces. There's a lot of exciting user interfaces coming up enabled by devices that enable gesturings, voice and other things. You'll see us continue to experiment and play with those types of interfaces in tests. And if we find one, then we'll continue to innovate in those areas.

Amazon's Web Services had another significant outage this week. Did this impact Netflix in any way? If not, was that due to Netflix's planning for random outages or due to not using effective services in the first place? Do these events affect your confidence in using services like Amazon's AWS for critical infrastructure and customer-facing functions?

Okay. AWS has done a great job for us as they have for other customers by having an architecture that isolates faults, and so it's built upon the notion that occasionally data centers will go down, and they give you the components to build around that. That would be true if they were our own data centers or if they were AWS. And in recent outage that they had, our customers were not materially affected. The traffic switched over smoothly, as it's designed to, to other Amazon data centers. So we're extremely happy with the decision to expand within the Amazon Web Service's footprint.

Next, some questions on competition. Can you compare the value proposition of your streaming offering to that of Amazon today? Is the gap closing?

Well, Amazon as Prime is 2 different markets that we compete with Amazon in, and one is in the U.K. and there are -- there's almost no content that's on both services. So think of it as 2 really different services in terms of content. And we believe our viewing is higher. We think our content is better. We're growing faster, but it's a real head-to-head battle. We're both at the same price point, roughly speaking, within 20% of the -- in standalone services. In the U.S., Amazon is bundled with Prime and it's got a subset of our content. And then depending on how you want to think about the price of Prime as an annual $79 payment or free, with free shipping, you get 2 different views on it. But as we said on the letter, most of our subscribers want us to have more content. They're not particularly motivated by a service that has a significant subset of our content. So at this point, it's not in anything that we can measure affecting us directly.

Somewhat a related question. Reed Hastings recently stated in an interview that Amazon had thus far not had much of an impact on Netflix's growth trajectory. Yet given what we're seeing now, in terms of slowing subscriber growth and lower guidance, does this change Netflix's view at all in terms of how many customers may be choosing alternative providers?

Well, we ask ourselves the same question, obviously. But in every data source that we look at, unrelated. I think as Amazon builds out their original content, which they're very actively engaged in and they'll get exclusive deals against us, they'll more and more be a different service than ours. And then, many subscribers, potentially, will subscribe to both. So I think that's the way that it evolves over time, like it has in the U.K.

Now a few questions on the financials. Will the company be disclosing its obligations related to original content production costs?

Well, yes, we are in the sense that the obligations that we've signed up for on originals are included in our streaming obligation tables. So they are already in there. We don't break them out. We'll provide more color as they get more and more material.

You talked about your expectations for negative free cash flow for the next several quarters as you ramp spending on originals. But eventually, you expect to return positive -- return to positive free cash flow. What gives you the confidence that you'll be able to return to positive free cash flow? Is it a slowdown in content spend or an acceleration in subscriber growth, or both?

Well, I wouldn't characterize either of those that way. I would say that we're focused on 3 things from a cash flow perspective: One is our growth in profitability in the U.S. and what that looks like; the second is our international investments and the losses related to those and how they play out over time; and then, the third is our growth in originals, which takes more cash than ordinary types of content deals. So we put all 3 of those together, and we want to maintain a comfortable margin. And I think those statements are based on the fact that I think that we were in a position where we can grow next year under various scenarios and expand our content expense, not slow it down, but just slow down the acceleration of the content expense in addition to funding our international investment in originals.

Can you help us quantify the original content cash outlays in 2013, understanding that they will be higher than the P&L expense?

Well, what we said about those before is we do have some output style deals that have more cash than expense, and those are -- have been about 10% in terms of content cash relative to the expense. Originals only need that, that ratio goes up anywhere from 10% to 20%, but probably not more than 20%.

20% in excess of the P&L?

Correct.

How might we think of content spend growing year-over-year in 2013?

Hopefully, content spend will grow substantially as we continue to make our service better, and that's built into our plan, but we don't believe that we're overcommitted. We have some flexibility, though, depending on what the growth is because we're very conscious of the variability in our growth outcomes. And you can see that in the fact that while we're short on total subscriber growth this year, we are hitting our contribution margin targets. And so that's a demonstration of the flexibility that we have on the cost side.

On CapEx, what was the driver of the uptick in the acquisition of PCNEs [ph]?

It's cash boxes related to our Open Connect program.

And finally, there has been increasing talk that Netflix will eventually have to raise prices to increase ARPU. We've heard from management that this option is off the table, with the brand seemingly recovering from 2011's event. Is this something that is being discussed? Would it be unrealistic to expect an increase before 2014? Obviously, it cannot stay $7.99 forever.

Well, we're very excited about the $7.99 price point, and we have no plans to change that. Consumers valued the incredible bargain that, that is and that helps our growth. So we see increased monetization from increased growth rather than any change in price.

At this time, I'd like to turn the call over to the operator, and we'll begin taking live call-in questions.

[Operator Instructions] Our first question comes from Mark Mahaney from Citi. Mark S. Mahaney - Citigroup Inc, Research Division: As the usage model switches more and more towards TV viewing, could you talk about the impact that could have on the business model in terms of margins, visibility into revenue and into profits, and maybe any pricing options that, that would give you?

Mark, there's no fundamental difference in the way that we license TV shows from the way that we license movies. In other words, it's a fixed payment per year or per time period. The only difference, I suppose, is we'll do a contract that will have, if there's a Season 3, Season 4, Season 5 that's built in. So I suppose there's a little bit of difference there. But fundamentally, it's content over some time period. And -- but the great experience about the TV shows in discovering them is being able to go all the way back to the pilot and really get a lot of fantastic viewing out of it, which creates a differentially better experience than any other provider. So that's the big asset of it.

Our next question comes from Scott Devitt from Morgan Stanley. Scott W. Devitt - Morgan Stanley, Research Division: I had a couple. First on the domestic streaming subs. You mentioned the Olympics as a factor. You also mentioned the change in seasonality earlier in the years, so I was just wondering, given the way that you exited 3Q and the very back-end loaded nature of 4Q, how you approach the fourth quarter guide for streaming subs? And on that topic, as it relates to early 2013 as well as more of the originals in the U.S. market, House of Cards and Arrested Development start to flow through potentially. Do you anticipate that, that drives subs above seasonal trends or is content and original content not necessarily a direct driver of subs? And then secondly, Fox I think recently opted to renew with HBO with an exclusive long-term deal. I don't know, it's like 2 to 3 years ahead of the renewal. Just wondering why you think they would opt to do that rather than giving Netflix a chance to bid for the rights?

Scott, I can -- this is David. I can take sort of Part A and B, and I'll let Reed take the Part C. Part A, the seasonal pattern is still in place. That is that we still see greater sub additions in the fourth quarter and first quarter than in the second quarter and the third quarter. So I think even our midpoint of our guide implies that seasonal pattern will still be in place. The Olympics sort of affected the monthly progression of the acquisitions through Q3, but overall, seasonality of net additions is still in place. And then your Part B on originals and content, the originals we're very excited about we think that they will drive a lot of consumer excitement in it. But on a grand scheme, it's likely to be less than -- certainly less than 10% of hours viewed next year and maybe even less than 5% of viewed. So in terms of our planning, I don't expect that to drive a ton of additional adoption.

Scott, it's Reed here. You're very gracious because I imagine what you really feel is why do we make seasonality excuses and not Olympics excuses and aren't you getting tired of it. We are tired of making those excuses as opposed to getting back to our track record. On the seasonality -- or the originals, like David said, we may have upside from it, but we're not going to bake that into our spending plans. We're getting into unknown territory in terms of how many people come to us for House of Cards or Arrested or others. So we'll gain some experience in that, and then that will help us understand how much more investment we should do in originals after that. But at this point and in Q1, we will think of it just as upside. And then in Fox and HBO, I'm not sure. Maybe the deal from HBO is good enough, but I think you'd really have to ask them or Fox, and we'll take it from there. Scott W. Devitt - Morgan Stanley, Research Division: If I could squeeze one last one. The social integration update, and talked about historically there are restrictions in the U.S. Is there any update to that?

No, there's a bill in Congress now that is pending before the Congress, and it's anybody's guess if they're going to pass it in a lame duck session. And then that would enable all different types of consumer permissions for sharing, which would be useful in a Facebook context but also useful in any kind of meta-browse various types of scenarios where people want to allow a feed to go to a service to help them. So we're optimistic, but when we come to the Congress, it's hard to predict. And then in terms of outside the U.S., we're continuing to learn on our international markets. We've seen some nice take-up in Brazil, which is a very social place, obviously, as well as in the U.K. So we're continuing to work on it, but no one yet has correct formula, us or anyone else, in terms of real explosive generator. So it's in the category of a little bit positive, and we're continuing to work on it.

Our next question comes from Youssef Squali from Cantor Fitzgerald. Youssef H. Squali - Cantor Fitzgerald & Co., Research Division: A couple of questions. The -- I want to go back to the one of the first questions that was asked about the 3 years needed to restore your image. How did you get to that determination that it should take you guys about 3 years? And why isn't it taking shorter since you guys have basically, arguably after a year, I think people start forgetting, and to the extent that the quality of content keeps improving, it would kind of get restored faster than that. And then I have a follow-up.

Let's -- Youssef, let's hope you're right, but it's not something we wanted to count on. 3 years was something we felt comfortable we could deliver on in terms of our reputation. And some people do forget in 1 year, other people take a longer time. We're just going to have to work really hard on providing a great value and stick to our knitting, and then the brand will steadily recover. And if it's sooner than 3 years, that's great. Youssef H. Squali - Cantor Fitzgerald & Co., Research Division: But you don't think that's an indication rather of just maybe perceived lower quality of content?

No. The best indicator of our perceived quality of content is how much we're getting viewed, how often -- you go home, it's Wednesday night, and you want to pick something to watch and you have to decide whether you're going to pick up your cable or satellite remote or you're going to put on a DVD or you're going to watch Netflix. And the more of those moments of truth that we win, the more people are engaged with our service and the more they retain and the more they tell their friends about Netflix. And so our primary driver is winning those moments of truth, which, if you think about it, is based upon your expectation of finding something good. And the better that those first 20 titles are for you, the more likely you are to turn it on, Netflix first, as opposed to one of the other entertainment options first. And so because our engagement is higher than it's been in the past, we're really confident substantially up over 30%, we feel really good about that in terms of our content mix getting better and better. Youssef H. Squali - Cantor Fitzgerald & Co., Research Division: And, David, if I just may, can you maybe just talk about trends of usage and consumption of exclusive versus nonexclusive content, and maybe parlay that with the mix in your content cost between the 2?

We're seeing engagement across both exclusive and nonexclusive categories. So in some cases, the exclusive content may cost more, but it's viewed more because it's worth more. The consumer doesn't -- isn't able to get that somewhere else. So you can infer from that, and some exclusive categories have higher viewing. But we're seeing user hours across nonexclusive content as well. In fact, at this point in time, there's more nonexclusive content on Netflix than there is exclusive content, and our hours continue to go up. And then your second part of your question was around cost, is that right, Youssef? Youssef H. Squali - Cantor Fitzgerald & Co., Research Division: Right.

In terms of -- there's an expectation that we'll move to exclusive over time, and that means that the cost will be -- will pay more for exclusive content. Youssef H. Squali - Cantor Fitzgerald & Co., Research Division: But you feel that you're -- basically, if you look at ROI for exclusive versus nonexclusive, you don't think the ROI on the exclusive is lower just because you're paying more for it?

No. I mean, we look for engagement across that category of exclusive, so we look for deals and types of content that works well relative in that exclusive class. And we think that there is a point where we're not going to be 100% exclusive, but we'll move more and more towards exclusive content as a differentiator. And it is important for us.

And, Youssef, it's Reed. You really want to keep in mind how many different flavors of exclusive there are. There's exclusive for online, there's exclusive against DVD, there's exclusive against cable -- cable with TV Everywhere and without. So there's a lot of different options and degrees as opposed to a simple nonexclusive, exclusive.

Our next question comes from Doug Anmuth from JPMorgan. Douglas Anmuth - JP Morgan Chase & Co, Research Division: Just wanted to ask a couple of things. In your letter, you talked about the long-term domestic market opportunity potentially being 2x to 3x that of linear HBO. And I'm just curious how you get there. So if we think about broadband households, it would seem that the number is approaching the number of cable households, but at the same time, there's still a limited amount of time in the day and a limited amount of disposable income. So I'm trying to understand how you get comfortable with that market opportunity being 2x to 3x. And then secondly, can you just update us where you are in terms of your personalized plans, what you might do around dual streaming?

Sure. In terms of the 60 million to 90 million, we've got that online strategy deck, which outlines some of this. But I'll run through it quickly, which is if you think about us versus linear HBO, we're able to be consumed on multiple platforms: laptop, tablet, phone. We're purely on demand with an on-demand brand. We have much broader content range, including a lot of kid's content, and we have a lower price than HBO. So it's extrapolating those factors that make us feel comfortable about the 2x to 3x, or 60 million to 90 million. In terms of broadband households, we do have the assumption that all households become broadband households, and that, that will happen over the next 5 or 10 years. And so that doesn't become a practical limiter in the same way that cable is in every household. And so that we'll equal, essentially, HBO in terms of addressable market, but because of lower price and the other factors have a larger actual market. And then on the personal stream plans, that's still something that's in development. We've been prioritizing some other things because it's a fairly minor issue, so maybe sometime this quarter and next quarter.

I think you'll see some testing this quarter, and then depending on the actual performance of that, we'll role that out accordingly.

Our next question comes from Richard Greenfield from BTIG. Richard Greenfield - BTIG, LLC, Research Division: A couple of questions. One, just the explosion of tablets. Reed, you talked before about how TV is still the primary viewing place that you need to get to, to drive your business. But we just with such -- we're seeing so many different companies launch ever cheaper and smaller tablets to get them into more people's hands across the globe. I'm just wondering how you think that tablet rollout impacts Netflix's subscriber growth over the course of the next couple of years relative to the importance you've place historically on the TV? And then just, too, somewhat related, you resigned from the Microsoft board. I just guess it would be -- a lot of people are curious what should we read into that, if anything?

Sure. On the Microsoft board, I'm on 6 different boards, and I felt like this is a good time, because Microsoft is strong coming out with Windows 8, for me to trim that back and to focus more on Netflix. So that's the story there. In terms of the tablet question, it's definitely growing for us. The question is, is it a straight substitute for a laptop as it grows, in which case it doesn't particularly affect us, or is it a net addition because it's a better consumption device than a typical laptop? We think there's some truth to the latter. The other scenario where tablets are really interesting are choosing on the tablet and then consuming on a TV. So one-second screen scenario that people talk about is you watch on the TV and then all the supplemental information, the history, the actors, et cetera, is on your tablet. And that's nice. But what we see is a big opportunity in using the power of touch to really choose what you want to watch, and then to be able to select it and then it's automatically playing on your TV. We've got a demonstration of that now on iPad and Android with the PS3, and then that's something we're working with all the CE ecosystem to build in so that over the next several years, this becomes a general capability for Netflix's choosing, which is to be able to get used to the power of touch in intimacy but then be able to enjoy the large-screen audio and video experience.

The next question comes from Andy Hargreaves from Pacific Crest. Andy Hargreaves - Pacific Crest Securities, Inc., Research Division: Just wanted to ask on the international stuff. How long would you stay in the market before you decided the profit potential wasn't worth the ongoing investment if things aren't going the way you thought?

Andy, there's very little chance of that happening because we've got long-term content deals that we pay on, so it's not an active topic. As long as the revenue growth is good and there's no competitor ahead of us where we have to face a decision of doubling down or not, it's clearly the right economic course to push forward and to derive the benefit. And we look at Latin America will be an awesome profit source for us with a great service over time, like it is for DirecTV today. And then in the U.K., you know it's a more competitive dynamic, but we're really making great progress. And so both are really important to us. And of course, Canada, already, is very successful. Andy Hargreaves - Pacific Crest Securities, Inc., Research Division: And on the U.S. marketing expense, it looks that's come down through the year just on a run-rate basis from where you were in Q1. Is that accurate, and can you just walk us through how you're thinking about your marketing domestically?

Well, in markets that we've already got a very large share, there's less need to spend in marketing on the margin than in new markets where we're just getting established. So you're seeing a little bit of that.

Yes, we're just gaining in efficiency in terms of word of mouth being more and more effective. Andy Hargreaves - Pacific Crest Securities, Inc., Research Division: And then just last clarification. Was that 3 billion number you guys gave in terms of hours viewed, that was -- is that global or just U.S.?

All the numbers that we've been talking about are global, so the 1 billion for June, the over 3 billion for the quarter, Q3, are global.

Our next question comes from Anthony DiClemente from Barclays. Anthony J. DiClemente - Barclays Capital, Research Division: Question about your comments about Hulu and the release, talked about it being a wildcard. And you said in terms of U.S. viewing, Hulu is your closest competitor. And so I just wanted to ask a little more about that. Is there anything that you can add as to why Hulu's programming is more substitutable for Netflix than, let's say, Amazon or HBO? And what makes them a global wildcard?

What makes them the wildcard is the ownership owned by 3 of the largest content companies in the world. And in terms of their viewing, I think they're closest to us because they focus on television like we do now, so they've got a good match in that way. And then they have the -- in addition to some of the TV shows, they have some current seasons in sort of catch-up mode, where they have last 5 episodes and then it falls off the service. But -- so they kind of have all -- you can think of it as almost 2 different markets or segments, but they get a lot of viewing on that new TV. Anthony J. DiClemente - Barclays Capital, Research Division: And then one more. Just back to, I guess, one of the earlier questions about the studio output deals. It does just seem as though, with Epix and Starz and Fox and Warner, that many of the majors are spoken for. And so is there anything you can add to the comment that you plan to bid for further deals? Is it -- can you talk about other deals that could be coming up? Are they more so smaller studio output deals? Is it -- are we talking, I guess, Universal could be coming up? Can you just give us a little more in terms of elaboration on that comment that you wrote?

Sure. There's small ones potentially. Again, if we're -- I think you're probably asking U.S. because that's our biggest market. In the U.S., there's some smaller ones that we can work on. And then there's Universal, Disney, Sony, all of which come up over the next 5 years. But none of them are going to change, they wouldn't be live for us on the new movies in the next 2 years. So I don't think it's probably hugely material to your -- for this, for the long-term, good for us to be a bidder there. Anthony J. DiClemente - Barclays Capital, Research Division: And Dreamworks Animation movies go live soon, is that right? Could you just remind us on that one?

It's in 2013, Anthony, so we've got -- it comes up in 2013. So as soon as -- it depends on what your definition of soon...

We can find [indiscernible] what's the first one coming through on that deal. Anthony J. DiClemente - Barclays Capital, Research Division: Right. But it's for any movies released in 2013 as opposed to just starting in 2013. So, right, so it wouldn't actually be -- it would be any theatrical film released in 2013 be eligible for the subsequent Pay TV output window, is that correct?

That's correct.

Yes, that's right. Anthony, just one final comment. We do have existing deals that have plenty of film output coming through. But you guys focused on additional things that we could add. But there's deals with FilmDistrict, Open Road, Relativity and others that are flowing film to the site now.

Our next question comes from Jason Helfstein from Oppenheimer & Co. Jason S. Helfstein - Oppenheimer & Co. Inc., Research Division: Most of the questions have been answered. Just a quick one. So in the release, you do give some color about you don't see voluntary churn as an issue, and you continue to see gross acquisition up on a year-over-year basis. Do you have any regrets, just given the volatility in the stocks, for not putting out gross ads and churn anymore, just given that it would seem like additional metrics might help reduce the volatility in the stock and help people understand? And alternatively, how about providing churn kind of on a trailing 12-month basis or something, just to kind of get your thoughts about that.

I don't have any regrets in providing it, Jason. I think the volatility was introduced in the 7 million and not meeting the 7 million, so I don't think that would have helped.

Our next question comes from Tony Wible from Janney. Anthony Wible - Janney Montgomery Scott LLC, Research Division: In the past, you guys have given penetration of the service in the San Francisco Bay Area. Can you guys give what that adoption number is today for just the streaming component? And the second question is, have you guys seen any change in attitudes on the content producers in light of some of the recent ratings weakness? Have they been more eager to cut digital deals to kind of fiddle what may be a revenue shortfall, or are they becoming a little bit more weary? Any color would be helpful.

Tony, in terms of the regional penetration, we used to do that because on the DVD side, we had several years of overnight delivery in the San Francisco Bay Area but not the rest of the country, so it was several years ahead. There's not an equivalent market where we had streaming before another market. So we don't, internally nor externally, focus on a sort of market-level prediction. In terms of the content producers, you got a wide mix. You get -- some of you are concerned for all the reasons that you addressed, and some that are more interested in the digital revenue. So the great news for us is there's no single producer that's a material part of our content. There's many different producers for many different companies. And so that helps us, and they're all looking for the best dollar, and then we're a bidder for that content.

Our next question comes from Heath Terry from Goldman Sachs. Heath P. Terry - Goldman Sachs Group Inc., Research Division: Just wondering if you could give us a bit of an update in terms of what you're seeing in the device ecosystem going into the holiday season. What kind of device penetration you're expecting, both television and as well as -- as well as to the extent that there's a mobile metric that you could provide, as well, relative to what we saw last holiday season, particularly for the remote Netflix buttons?

Heath, it's Reed. I mean, we'll have more Netflix buttons on more Smart TVs than we've ever had. Smart TVs, as you know, are becoming a larger percentage of all TVs. It's becoming pretty standard and built-in, so that's a significant positive. There's the Wii U releasing shortly, that's a new game console. And so we'll see if that takes off and provides us a special lift. And then mobile is -- been very strong for several years and continues to be strong. I don't see a huge, like, CE-related hockey stick to that part of the business.

Okay, that's going to be the last question for today. Reed, would you like to provide any closing remark?

I just want to thank you all for the questions. We're continuing to work very hard and look forward to being in a better place relative to our guidance a quarter from now. We certainly try very hard every quarter, and it doesn't feel great to come in, in the lower half. But it is what it is, and we're moving forward.

Ladies and gentlemen, thanks for participating in today's program. This concludes the program. You may all disconnect.