Greetings, and welcome to The Trade Desk's third quarter 2019 earnings conference call. At this time, all participants are in a listen-only mode. A brief question and answer session will follow the formal presentation. If anyone should require operator assistance during the conference, please press star zero on your telephone keypad. As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Mr. Chris Toth, Head of Investor Relations. Thank you, Chris. You may begin.
Thank you, Operator. Hello, welcome to The Trade Desk third quarter 2019 earnings conference call. Our call today is taking place from our Ventura headquarters. On the line today is our Founder and CEO, Jeff Green, and Chief Financial Officer, Paul Ross. A copy of our earnings press release can be found on our website at thetradedesk.com in the Investor Relations section. Before we begin, I would like to remind you that except for historical information, the matters that we will be describing will be forward-looking statements which are dependent upon certain risks and uncertainties.
I encourage you to refer to the risk factors referenced in our press release and included in our most recent SEC filings. In addition to reporting our GAAP financial results, we present supplemental non-GAAP financial data. A reconciliation of the GAAP to non-GAAP measures can be found in our earnings press release. We believe that providing non-GAAP measures combined with our GAAP results provides a more meaningful representation of the company's operational performance. I will now turn the call over to Founder and CEO, Jeff Green. Jeff?
Thanks, Chris. Thank you all for joining us. I'm going to quickly cover the highlights of the third quarter. Then I wanna dig a little deeper into a couple key areas that are rapidly coming into view in our industry. It will provide context for our performance this quarter, just as importantly, for our confidence in the fourth quarter and moving forward into next year. Once again, in Q3, we executed well and exceeded our expectations. We saw revenue grow 38% year-over-year. Revenue was $164.2 million, surpassing our guide of $163 million. Just so you have a frame of reference for that growth, remember that Magna Global estimates the programmatic advertising market is growing at around 20% this year. The overall advertising industry is growing at 4% according to IDC.
In the fastest-growing segment of an industry that is expected to reach a TAM of $1 trillion in the next decade, we are significantly outperforming. In fact, we are once again growing at about double the pace of the industry. Our outpaced growth and market share gain are the result of investments we've made in key channels and markets. It's also because of our commitment to objectivity and independence, and it's the work we do across the industry with partners and standards bodies to make this an industry that advertisers and content providers trust. As you know, one of those major investment areas is CTV. Spend on our platform in Connected TV was up 145% from Q3 of last year. We've seen strong growth in available CTV inventory, especially in live events.
We're especially excited about the upside potential for live events in 2020, including major sporting events and the U.S. elections. After multiple years of triple and even quadruple-digit growth in CTV, Q3 year-over-year growth of 145% is definitely one of the biggest bright spots in our business. CTV is the most strategically important focus of our business going into 2020. We invested early in our CTV infrastructure and in the supplier ecosystem, and we're seeing the payoff of those investments accelerate. I'll say more about this a bit later. The growth is not just in CTV. Audio spend was up a stunning 162% year-over-year. Like CTV, audio is a large and growing market, about $3 billion according to PwC digital radio estimates. In terms of pure percentage growth, it continues to be one of our fastest-growing channels.
Audio is highly attractive to our customers because it regularly delivers high-performance metrics such as completion rates. We continue to integrate new sources of audio inventory worldwide and expect to see broader adoption of this channel by advertisers in the years to come. Music dominates audio today, but I'm just as bullish on other nascent audio opportunities such as podcasting in our future. If you believe as I do that CTV and audio are two of the most effective forms of advertising because of high audience engagement, this means that TTD is growing fastest in the forms of advertising that have the most staying power, the most efficacy, not just driving clicks. More and more, the way the world accesses the Internet is through mobile devices. This is especially true in the high-growth markets of Asia, where the largest middle class in history is emerging.
In Q3, almost half our revenue came from ads on mobile devices, which includes mobile video. Mobile in-app spend increased about 58% from Q3 of last year. In Q3, the move to mobile helped fuel our Asia growth. Spend in our offices in Seoul and in Tokyo grew 166% and 65% respectively. Our Seoul office achieved record spend. We added more major advertisers and more accounts grew their spend with us. We are seeing similar trends in Europe, where spend growth in our offices in Madrid and Paris grew 75% and 57%. In China, we continue to see very promising signs. We hired a new Country General Manager, Calvin Chan, who was previously the CEO of AdMaster, one of the leading ad tech companies in China. You can think about them as the local equivalent of DoubleClick.
Calvin has already forged a new partnership with BlueMedia, one of the largest Chinese media agencies. They see the value in working with us to help their advertisers tap into our global footprint. As Chinese companies look to leverage data-driven advertising to market directly to consumers, both within and outside of China, there are a few major agencies that they turn to, and BlueMedia is one of the largest. BlueMedia is partnering with us not only because of the benefits of our platform, but also because of all the work we've done to build the right local ecosystem, including partnerships with Baidu, Alibaba, iQIYI, and Tencent. This deal is important because BlueMedia is one of the local programmatic pioneers in China. As in many emerging markets, much of our go-to-market strategy focuses on educating brands and agencies about the benefits of programmatic.
We are taking this to the next level in China with the launch of a new iteration of Trading Academy called The Trade Desk Edge. I'm going to be in Hong Kong and China in the next couple of weeks to record new courses that are specific to the China market. Participating in these sessions will be many of China's marketing leaders from brands, agencies, and standards groups. Calvin has been instrumental in getting all of this work in motion. Through this work, we are laying the foundation for long-term success in China. 2019 has already been a year of significant progress. We are still in the early stages, but I am very optimistic about our ability to drive significant growth in what will be the world's largest middle class in the years ahead.
We continue to invest and place bets on China, we're convinced that we can help Chinese companies grow. Regardless of channel or region, one underlying constant is the importance of data. As you'll recall, last year, we revamped our entire platform to put data front and center in the user experience. In Q3, data was up 63% year-over-year. In the last four quarters since that launch, data spend increased about 80% on a year-over-year basis. We've seen similar gains in cross-device spend. With cross-device, advertisers can target across the range of devices a consumer uses in a way that drives a more pleasant, integrated advertising experience. Advertisers have embraced our cross-device solutions, spend is up 132% year-over-year.
As advertisers become more comfortable with programmatic advertising and the power of data in driving greater precision, they are expanding the types of data they use and becoming more sophisticated in how they apply it. Procter & Gamble, one of our bigger customers, summed it up well when their Chief Brand Officer was asked about the power of data-driven advertising. He said, and I quote, "Moving from mass blasting to mass reach, one-to-one precision has helped us get substantial efficiencies in media, which then allows us to grow. That's helped us in the cycle. You raise the bar on superiority, you find ways to get more productive, then you reinvest back in superiority, reach more people, and the cycle continues." It's that kind of growing awareness of the power of programmatic that helped us sign many new MSAs with some of the world's top global advertisers in the quarter.
Among them were a large entertainment company, a global consulting business, multinational beverage company, a major insurance firm, and a global travel marketing company. As they always do, these new customers have started relatively small as they test the platform, and then they grow exponentially into 2020. When we signed many large brands on our platform in second half of 2017 and 2018, they too began with small campaigns through their agencies. When they saw measurable results, they increased their spend. In some cases, within just two quarters, these brands have increased their spend on our platform nearly 800%. These large advertisers were a key factor in driving our 40% year-over-year revenue growth through the first three quarters of 2019. We expect similar results from our new advertisers.
Among those existing advertisers who are increasing spend with us, a major auto manufacturer reallocated significant spend to programmatic on our platform, and we won multi-million-dollar incremental spend from a global restaurant brand that had previously relied mostly on one of the walled gardens for some of their spend. These dynamics position us very well for continued growth, not only in Q4, but also in 2020. We are more confident than ever in our ability to aggressively grow our business more rapidly and profitably than our peers. As I said, I wanna dig a little deeper to provide you more context for that confidence. I believe we are well-positioned to take advantage of significant shifts in the advertising industry, which makes us as competitive as anyone for the next advertising dollar. Nowhere is this more apparent than Connected TV.
As we have discussed in the past, the nature of TV viewing and TV advertising is changing right before our eyes. This is significant because in many ways, TV is the most important frontier in digital advertising. For many of our customers, such as major CPGs, retailers, and automakers, TV represents the largest piece of their massive ad campaigns. Until now, they have only been able to run those TV campaigns based on the very broad demos on linear TV. They've had very little detailed feedback on how these campaigns perform. Connected TV is changing all of that. As more viewers access TV content via connected devices and smart TVs, and as more content providers build and launch new streaming platforms, advertisers can apply data to their TV campaigns for the first time.
It's a game changer. As I said two years ago, companies like Hulu, AT&T, and Spotify were pioneers of this ad-funded streaming revolution. I said that they were what I call tea leaf companies. If you watch what they do, you can predict and know what others are going to do. They developed new TV and audio revenue models. They took strong positions on ad-supported options. In some cases, they offered premium offerings with no ads, and in others, they offered a discounted offering with targeted ads. These tea leaf companies proved the model, and most consumers chose the ad-funded models. As a result, most of those companies ultimately laid out strategies that went all in on programmatic ads. Whether any individual tea leaf company executes well is less important. Some will execute, and some will not.
What's undeniable is that these companies have changed the game in streaming content and programmatic advertising. They were the pioneers, but now it's a movement. AT&T's HBO Max is planning an ad-funded tier. When NBC launches its Peacock service next year, there are reports that it will be entirely ad-funded. When Disney launches its Disney+ subscription service soon, it will sit alongside their ad-funded options for properties such as ESPN, ABC, National Geographic, and FX. Disney also owns Hulu, which of course has an ad-funded tier and has been leading the fusion of programmatic ads and CTV. These companies and others are giving consumers various ways to pay for the access to their content, and in doing so, are maximizing their revenue potential. I believe 2019 is the year that CTV proved that its future will mostly be ad-funded.
Given the current economics and the current state of competition in the TV industry, all providers will have to explore ad-funded CTV models. On the linear side, figures from Leichtman Research Group show that the entire traditional linear TV industry lost about 1.53 million subscribers in Q2 2019. Linear broadcasters are fighting for fewer viewers while content costs are going up. That's a ticking time bomb. For advertisers, that means their ad-to-viewer ratio is worse than ever. Until recently, there's been a sense that they have nowhere else to go. Now they do. As more providers shift content to streaming platforms, competition among them is becoming more intense. As we have seen, platforms such as Netflix are fighting tooth and nail for subscriber growth while having to issue new debt just to keep pace in the war for premium content.
There's only so much subscription demand, and there are new competitors every month with massive established content libraries. It all points to growth in ad-funded models, and I firmly believe that even Netflix will start to experiment with ad-supported services in the future. I believe they'll eventually adopt what others have, giving consumers the choice to pay more and avoid all ads or pay less and see a few highly relevant ads. As content providers create and launch their streaming services, almost all of them are working with us to figure out how to optimize for programmatic demand. Since we last reported earnings, we announced the details of a private marketplace agreement with Amazon. This deal is a game-changer for the industry and represents another tea leaf offering. We are now one of the preferred DSPs for third-party premium content in their Fire TV marketplace.
We have access to all their premium content inventory, not the remnants. As Amazon Publisher Services, APS, stated when we announced this deal, they are coming to us in order to maximize demand, to ensure fee transparency for the advertisers they are trying to attract, and to improve ad frequency for consumers. As you may have heard, we have also started to work with Disney as a preferred partner as they start to stream more of their premium content. Let me just pause a moment and reiterate, these content providers are not our direct customers. They are our partners. More and more, they are asking to work with us. It's not just Amazon and Disney, as it's other major global providers worldwide, including Channel 4 in England, ProSieben in Germany, TF1 in France, and pretty much every other significant network and content provider.
They are coming to us because they want to make sure they do everything right so that advertisers can apply data-driven strategies to their content. On a more micro level, what did all this change mean for us in this quarter? Three things. First, as the quarter progressed, more and more premium TV inventory became available on our platform, and more of our customers started to access it. This helped drive revenue growth acceleration for us as we moved through the quarter. As a result, our last month was especially strong in CTV growth. That's in part the result of inventory coming online from partners such as Amazon and Disney. With Amazon, the number of impressions on our platform increased 21x during the quarter. We're also seeing game-changing progress from other partners.
For example, FreeWheel, the largest ad server in the CTV space, launched their version of header bidding, Unified Yield. This is the best thing for them to offer content owners because it gives them access to the most demand. It also means, in a nutshell, that we can compete on every impression. As always, if we can level the playing field, this becomes a fight between data-driven ad selection and guessing. Whenever this is the challenge, we're confident we'll win way more often than we lose. As a result of FreeWheel's leveling the playing field, we saw available CTV spots increase significantly through the quarter, up as much as 300%, with some premium content providers using this new feature on FreeWheel.
Second, upfront commitments by major brands to programmatic TV advertising campaigns, which were signed back in the spring, started to go live in late August and September as broadcasters launched new content. Last, but perhaps most interesting in terms of the fundamental difference between linear and Connected TV advertising, is live sports. Let me spend a minute here. The unpredictability of live events is a huge opportunity for both content providers and advertisers. If you look at what Disney is working on with us as they build their streaming platform, it's all about optimizing the value of unpredictability. Think about it. A Major League Baseball game goes into extra innings, or an NFL game goes into overtime. This is when viewership is at its highest and most engaged level. Everyone's watching. This should be the broadcaster's most valuable ad inventory and the most desirable for the advertiser.
In a linear environment, these ad spots are often wasted. The advertiser can't plan for them, and the broadcaster often ends up giving them away at a discount for repeat ads. In a data-driven environment, the advertiser can optimize for those opportunities, automatically recognizing value and shifting investments in real time. The broadcaster can then realize the full value of those spots. As the third quarter progressed, we saw the NFL kick into gear, college football start, baseball playoff races heated up, and the start of the European soccer season. All of these are huge opportunities for our advertising clients, and this momentum will stay with us through the quarter and into next year, where we'll have not only major sporting events, which will continue to benefit from programmatic TV advertising, but also highly charged political events, such as the presidential election, which will also drive unpredictability in content.
All of that said, not only did CTV help drive revenue growth acceleration for us through the quarter, I firmly believe that CTV advertising is coming of age right now. It will fundamentally alter the economics of television moving forward in a way that will benefit The Trade Desk. This shift is important in another context too. The progress we've seen in CTV is a direct result of the investments we've made in our platform and in our partner ecosystem. This commitment to investment is a major differentiator for us. Advertisers recognize that we are always investing to help them be as successful as possible. As I have said almost every quarter, we retain the financial flexibility to invest wherever necessary to stay at the bleeding edge of our industry, and that's not just in Connected TV.
We're building on our Adbrain acquisition and investing heavily in identity graph solutions that enable our customers to drive data-driven precision in their campaigns, whether they are operating in a cookie-based or a cookie-less environment. Our investment also includes areas such as measurement as advertisers look for the most precise information on campaign performance. Our measurement tools allow an agency or advertiser to use third parties for verification. This helps the ecosystem become more transparent and avoid the grading your own homework syndrome that advertisers experience with walled gardens, and which often obfuscates ROI and data. We also continue to invest in our infrastructure to support business and data processing growth worldwide. You already see some of these infrastructure investments flow through to our gross margin or platform development expense line. It is these types of investments that help drive leverage in our model.
We do not have to choose between growth and profitability. We do both. This is highly appealing to our customers and a major factor in our 95% + retention rates. They know we will continue to invest for their success. This puts us in a position of strength. For companies of similar size and profile, we are growing faster, and we are significantly more profitable. In our view, profitability is good business discipline. It forces us to focus on how we're delivering premium value, and we constantly ask the question: Are we providing value that exceeds the fee we charge for in our platform? That focus is driving more major brands and agencies onto our platform. It's what's prompting a major beverage brand to work more closely with us to apply advanced tools to improve frequency management across all digital advertising channels.
It's what's driving all the major broadcast providers to work with us as they strategize on their new streaming platforms. I've covered a lot of ground, but I think it's all important to understand what's going on in our industry and our position within it. Given everything I've discussed, I feel very confident that we are well-positioned to continue to lead our industry and drive momentum for our business. As we exit 2019 and look forward to 2020, there are six specific areas that make me extremely positive about our future prospects. First, brands and agencies are driving more data-driven strategies across their campaigns to stay competitive and to protect their brand value. This is good for them, and it is good for us.
In the long run, moving to data-driven choices leads advertisers to global omni-channel platforms like ours. Over time, we expect to see the amount of data per impression increase. Second, we have more large advertisers on our platform than ever before. More than 70% of the Ad Age top 200 advertisers have spent on our platform in the last 12 months. Despite this momentum, we still see significant growth potential as data-driven advertising becomes an even larger element of their campaigns. As a result, their current spend with us is still small relative to what we expect them to do in 2020 and beyond. The third area is Connected TV.
As I discussed, the progress we have made in adding premium inventory supply from the likes of Amazon, Disney, and others is driving more and more advertisers to apply data to their huge TV ad campaigns for the first time. This has helped our CTV spend increase materially. We expect CTV growth of over 100% again in 2020. Fourth, global expansion. We've made significant investments outside of the U.S. over the last two years. We believe we're in a position to accelerate our international growth in 2020. Fifth, we continue to invest heavily to stay at the forefront of innovation in our industry. We have seen major success with products such as cross-device data and identity solutions, and expect to continue that trend with new products that we launch in 2020.
Finally, we have extremely strong customer retention rates and expect to maintain those rates even as we expand our customer base. This model of excellence in customer service has fueled this business since its inception. Our goal is to continue to offer the best scaled, objective media buying platform in the world. All of that said, and with all that great momentum, we're still just getting started. Our numbers quarter after quarter show, The Trade Desk is growing much faster than the market. We anticipate these trends will continue for the rest of this year and into 2020. I'm going to turn the call over to Paul to discuss our financials.
Thanks, Jeff. Good afternoon, everyone. Q3 was another record quarter for The Trade Desk, and we were really pleased with our Q3 financial performance and execution. Revenue increased 38% year-over-year. Adjusted EBITDA increased to $47.8 million, a Q3 record, and net income was $19.4 million. This marks our 14th consecutive quarter in a row of GAAP net income, all while we continue to invest aggressively for future growth. Revenue for the third quarter was a record $164.2 million, which was above our expectations and reflected increased spend by our existing customers and the addition of new customers and advertisers. For the quarter, approximately 89% of our third quarter gross spend came from existing customers who have been on our platform for longer than a year.
Q3 marks the 23rd quarter in a row where customer retention was over 95%. With the growth of our business, our operating expenses grew to $142 million in Q3. This increase was primarily due to sales and marketing and engineering as we continue to scale for growth. The year-over-year increase also reflected higher G&A expenses, which take into account stock-based compensation. GAAP net income was $19.4 million for Q3 or $0.40 / fully diluted share. Our adjusted net income was $36.1 million or $0.75 / fully diluted share, compared with adjusted net income of $30.2 million or $0.65 / share in the comparable period last year. Adjusted EBITDA was $47.8 million with a corresponding margin of 29.1% of revenue during Q3 2019.
The increase in adjusted EBITDA dollars reflects the strong growth of our top line, partially offset by our increasing investments across our operating expense lines. Net cash provided by operating activities was about $67 .5 million for Q3, and our trailing 12 months of operating cash flow and free cash flow were $136 million and $100 million respectively. As a reminder, the timing of our payments and receivables in any given quarter can swing our cash from operations significantly. We continue to have zero debt on our balance sheet, and our cash and short-term investments position continues to climb, exiting the quarter at $297 million. Our DSOs at the end of Q3 were 96 days, a decrease from four days from the same period a year ago.
DPOs for Q3 were 77 days, also a decrease of four days from the same period a year ago. For Q4 of 2019, we are expecting revenue of $213 million and adjusted EBITDA of $78.5 million. For the full year 2019, inclusive of our guidance for Q4, we now expect revenue for the year to be at least $658 million, which approximates to 38% growth year-over-year, and corresponding adjusted EBITDA to be $209 million or 31.8% of revenue. With that, I will hand the call back over to Jeff for any final comments, and of course, Q&A. Jeff?
Thanks, Paul. Q3 was another very encouraging quarter for The Trade Desk as we continue to see our strategy and execution pay off as more advertisers commit their budgets to us. We exceeded our expectations for the quarter and are raising them for the year. The fundamentals of our business are solid, and we continue to scale across markets and channels. Looking at the rest of 2019, we anticipate a strong Q4. The advertising commitments that were made in the television upfront earlier in the year are now kicking in. Of course, the holiday advertising season traditionally makes Q4 the strongest quarter of the year. We see similar opportunities in 2020. We expect another strong growth year in CTV, fueled by live events such as the U.S. elections and the Summer Olympics, and more quality inventory than ever on our platform.
As the worldwide advertising market moves towards $1 trillion, The Trade Desk is perhaps the best-positioned company to win the largest share of the programmatic portion of that market, which is the fastest-growing segment. We feel confident in our strategy, our investments, and the momentum we are seeing from both advertisers and content providers. We see nothing but upside ahead. That concludes our prepared remarks. Operator, let's open it up for questions.
Thank you. We will now be conducting a question and answer session. If you would like to ask a question, please press star one on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star two if you would like to remove your question from the queue. For participants using speaker equipment, it may be necessary to pick up your handset before pressing the star keys. The first question comes from Michael Levine with Pivotal Research Group. Please go ahead.
Two questions, guys. You know, first, I'd love to understand, you know, how we should be thinking about the linearity of 2020 with regards to Connected TV and, you know, both what gives you as much confidence as you do in terms of that very strong outlook, as well as trying to better understand the move by Roku in terms of acquiring DataXu and, you know, how you think they end up being positioned as the media companies get a little bit more sophisticated about what their inventory looks like online.
Awesome. Thanks. I appreciate the question. First, to talk about Connected TV. This year has been something of a dream as it relates to the way that we've set up 2020. If you go back in time and ask me to write down on a sheet of paper what are the best partnerships that we could form this year, I don't know that I could have dreamed up a better year than what has happened between the partnerships with Amazon and with Comcast and what we're doing with Disney and even what we're doing with Roku. I'll come back to the second part of your question in just a second, but it's just been an unbelievable year. Q3, of course, we just highlighted that our Connected TV spend increased by 150% year-over-year.
We talked about how Amazon inventory went up 21x throughout the quarter. You know, we've used that 21x before in our Connected TV numbers. To be there again with a partnership that is as new as it is with Amazon, but to go up that much is pretty exciting. Equally exciting is, you know, FreeWheel launching a product that is the equivalent of header bidding so that programmatic demand can compete with sort of the ordinary sales force in TV. Seeing that go up by 300%, also amazing. It's been amazing to see our partnership with them expand.
You know, to sum it all up, I heard a number on CNBC recently, which was 20% of television content has moved to streaming, but only 3% of ads have. If you look at that as sort of the macro direction that we're all heading in, and if you believe that ultimately all the content's gonna be streaming because it's better to be on demand, then, you know, we're 3% done, and this year just teed things up. If that wasn't enough of a dream year, you also have Disney going all in on what I think is a pretty amazing strategy.
Where, you know, they've recently put together the pieces from Pixar, Marvel, Fox, Lucasfilm, put all this amazing content together, and it's no wonder that my 14-year-old, when I was dropping him off on to school today, we passed an ad for Disney+. It wasn't my eight-year-old girl who said, "Can we get Disney+?" It was my 14-year-old son saying, "Hey, there's gonna be Star Wars content that comes out on November 12th when Disney+ launches. Can we get that?" I think it's, I think it's teeing up what's gonna be an amazing 2020. All of those to me represent tailwinds that are just helping. As it relates to the Roku move, let me just give a little bit of context to it.
First, Roku is a great partner for us. We buy inventory on their platform. They obviously have an amazing run to date. I think they're gonna keep going. They're gonna benefit from the move that we just described. It's an interesting move that they buy DataXu. To just give a little bit of context in terms of size, this year, just round numbers, we'll control a little over $3 billion in spend. They'll do a little over $100 million in spend at DataXu. Of course, in comparison, it's extremely small. I think the more impressive thing that they bought is technology talent that can help them build more of a self-service platform and help expand the Roku platform itself.
The thing that I find especially exciting about the move is actually all the discussion and all the things that Roku itself said about the acquisition. Most notably, that they're more committed than ever to, one, our partnership, and two, to being a part of the open internet. That language open internet, I think is really important to underline, which is being a part of the open internet means that you're willing to make your content available for purchase from all the major buyers of ads, of content. By having an open platform, it means you're willing to let the market compete.
That also means that you're willing to share the ID that makes it possible for you to compare what happens on the Roku to what happens on Amazon Fire, to what happens on any other device or any other content. Roku, of course, is dependent on other people's content primarily for them to produce the ad revenue that they have. This commitment to the open internet, to do symmetrically what Amazon does, which is share the ID with DataXu. I think that is the most exciting part about what they just announced. It's less about them having a DSP. It's more about them expanding their tech. Especially, it's exciting about them joining the open internet, which I think is really promising for TV content in general because no one will have a monopoly in television the way that there are in social or search.
Thanks, Mike. Next question, Operator.
Thank you. The next question comes from Youssef Squali with SunTrust Robinson Humphrey. Please go ahead.
Great. Thank you very much. Hi guys. Jeff , you've been very keen on investing aggressively against all these future opportunities while still showing margin leverage. I think this quarter shows that as well. What are the key areas of investment as we go into 2020? How should we think about the margin impact for the year? Paul , trying to just reconcile, you know, all the excitement that is palpable in Jeff 's presentation with your Q4 revenue guide, which basically implies growth dropping from 38% to 32%. Just how should we be thinking about growth as we move into 2020 as well? Thank you.
Awesome. I'll take a stab at both of those, and I'll let Paul add on to the second. In terms of the things that we're especially investing in as well as the things that we're especially excited about as we head into 2020. I know I just covered it a second ago, but I just need to underline there's nothing we're more excited about. There's nothing more game-changing than what's happening in Connected TV and that 150% year-over-year spend increase that we experienced in Q3 is just indicative of that. The second thing that we've been investing in quite heavily is international. This has definitely been an investment year where we need to help take all of our international markets into the next level in broad strokes.
You know, when we started this business. Actually, we're about to celebrate our 10-year anniversary. When we started this business, global advertising was roughly a $500 billion industry. It'll be a $1 trillion industry in six or seven years. At about that time, we expect a much greater percentage of our revenue to come from international, given the fact that right now the U.S. only comprises about 40% of the global advertising pie. We feel like we have a lot of catching up to do, and that means that we need to make investments.
This year has been a year of investing, where we expect to see in many of those markets, dividends start to pay off next year, and of course, they'll continue to pay off in years to come in order for that balance to come, to come to pass. We also have just been investing a lot in tools related to measurement. This somewhat relates back to Connected TV, which is that we need a better measurement in television so that we can show people what they're getting incremental, what is better about Connected TV than in traditional television. Because traditional television hasn't had very good measurement, it's just a lot more work to show people why they're getting better reach and frequency management in Connected TV than they are in linear television.
Additionally, in 2020, I think partly because of the political environment, partly because of just media growing up, and partly because of Connected TV, the role that objectivity plays is going to play a much bigger role. As a result, we've invested a lot in products where we can onboard first-party data on behalf of clients. We're continuing to do a more and more technological investment so that we can be the best partner in the world for companies like Comcast and like Disney and like AT&T and of course like ITV and ProSieben and RTL and Channel 4 and so many others around the world, TVB in Hong Kong.
The last thing I'll just touch on before I talk about the second part of your question is another huge opportunity in 2020 will of course be political spend. We view that as a humongous opportunity. I know that a number of tech companies, ad-funded tech companies, have decided to sit out this election. I think that's a mistake, not just because of the opportunity that political ads represent, but also I just think there's a civic duty, a role that we play in helping that process get better, and it can get better and will get better. That's something that I think that we can help in, so we're really excited at the opportunity that that represents.
As it relates to your question around slowdown, I'll just give a little bit of macro context before Paul weighs in. If you were to look at our 2017, you'd see that at the beginning of the year, we started at 70%, we ended the year at 40%. If you'd spent too much time looking or expecting a straight line that is simply straight up and to the right, even though the overall trend for us has been pretty steadily up and to the right, you would have missed out on a 2018 where our growth rate was 55% being misled by that Q4 of 2017 that was 40%.
It's easy for you to look at this year or this moment and make the same mistake. When you look at the untapped opportunity, especially in Connected TV, I think you just have to look at the general trend and not get too caught up in the short term, which is that overall, our biggest opportunities are ahead of us. I'll let Paul weigh in on any other parts of the financials if you'd like to.
Yeah. Hey, Youssef. Just a few general comments. We executed really well in Q3, we're raising our expectations again for Q4. We're comfortable moving expectations up as we go along, we're continuing to win share. We expect that to continue to drive growth. In the current environment, we think it's prudent to be measured. You know, I'll just remind you that surprises tend to be to the upside, we do what we say we're gonna do.
Thanks, Youssef. Next question, Operator.
Thank you. The next question comes from Vasily Karasyov with Cannonball Research. Please go ahead.
Thank you very much. Jeff, I wanted to talk about international for a little bit. Can you tell us what, maybe in more detail, what kind of trends you see in key channels, key territories? I know you touched on China. You did a lot of work building out overseas in the past couple of years. Can you share maybe more detailed thoughts on what needs to happen next for us to see an acceleration in share gains overseas? Do you think there is room? I think you just implied that you think that gross billings growth can accelerate in 2020. Just wanted to make sure I understood that correctly. Thank you.
Yeah. I mean, to be very direct, yes, I do expect revenue acceleration from our international teams in 2020. You know, there is a huge amount of opportunity in Connected TV in particular. You know, having met with some of the leadership of the biggest media companies in Europe in particular, in the U.K. and in Germany, I'm especially excited about the opportunity there. As you know, as anybody who's been following us for a while knows, we've also just been making big investments in Asia, especially in China. Those investments are showing green shoots finally.
It's been a long time of simply investing, and it's largely because of the fact that we've now put together a client-facing team inside of China, and because of the strong partnerships that we have with Baidu and Alibaba and iQIYI, and of course, Tencent, we feel more comfortable than ever to put more feet on the street. You know, the partnership that we announced for the first time on this earnings call with BlueMedia represents the first of many, but also, I mean, they're a massive partner for some other big American tech companies, and this is a huge opportunity. I think we've made a lot of great investments there.
I do expect revenue acceleration internationally. We're gonna keep doing that so that we continue to make up the ground, because that's the way I look at it. The international markets need to grow faster than the U.S. in order for us to rightsize that pie as we get to the $1 trillion. I just keep looking at that long-term view. There are some markets that are nascent. There are some where there's opportunity for us, where we need to go catch up. In most cases, nearly all cases, we are investing early rather than late so that we don't miss the opportunity, and we won't.
Thank you.
Thanks, Vasily.
Thank you. The next question comes from Brian Fitzgerald with Wells Fargo. Please go ahead.
Thanks, guys. Jeff, how has the rollout of first price auctions on ad exchange and other supply sources played out in the quarter? And what kind of opportunity do you see in predictive clearing related to that? It seems like when we initially talked about first price machinations back with header bidding back in the day, it took kind of quarters to normalize. As it moved from the publisher pages to the SSPs to the exchanges, the adoption and the adjustment period meters through quicker. Thanks.
Yeah. Just to give everybody a little bit of context, just because I imagine this is a pretty detailed question for a number of people on the call. For the last few years, there's been a debate, if you will, or a disparity in practices among sell-side platforms and ad exchanges about how the mechanics of the auction should work. Should it be a first price auction? Should it be a second price auction? At bad times, it's been what some have called a modified second price auction, which I would say at bad times has meant people haven't always been consistent in the way they've run their auctions.
What's happened over the last few years is that in part because of the competition that header bidding created, people have to run an auction with integrity or they don't have a place in the market. Whether you call it a first price auction or a second price auction doesn't really matter as long as you run an auction with integrity. If they do run a first price auction, I'm sorry, if they run a first price auction, there's an opportunity for us to add a little bit more value in predictive clearing because, you know, if you run a second price auction, then the auction itself is going to lower the price on behalf of the buyer. In a first price auction, we have to do that for them.
We have to predict where it's actually going to clear so that we can save them as much as possible, and we have to be careful about overbidding. That puts more opportunity for us to perform and to add value. In some big cases, that represents an opportunity for us to make more money. Overall, what we really just want is an ecosystem where there's integrity and consistency. We're happy that Google's Ad Exchange moved to first price auction. It helps us on both fronts, and just delighted to have some consistency.
On the big picture, it really doesn't have that much impact. It's slightly positive for us. Overall, it's a really a win for the ecosystem because people know how the auction is running, how it's working. In my view, the changes that they made to the auction made the ecosystem just a little bit more of a level playing field, which is good for everybody.
Thanks, Jeff.
Thank you. Next question, Operator.
Thank you. The next question comes from Mark Zgutowicz with Rosenblatt Securities. Please go ahead.
Thank you. Jeff, you commented on your revenue outperformance, which, you know, relative to worldwide programmatic, which obviously continues to be impressive. You might even argue that it's even more impressive given, you know, majority of your growth has been coming from the U.S. while it's really, you know, international and particularly China that's been really pressing that, you know, worldwide growth number.
The question is, what, you know, at what level do you need to see international mix really begin to grow? I mean, at what pace does it need to grow, you know, relative to the market? It seems to be obviously trailing, which you've acknowledged. The question is sort of how quickly that needs to grow to prolong, you know, what is, you know, eventual mean reversion at some point. Just more specific to that, what markets do you see really driving that acceleration in 2020 and then beyond, 2020? Thanks.
You bet. To answer your question about what pace we expect it to grow, I'll just say in broad strokes, I expect it to grow at a faster pace than the U.S. That's it. It's not gonna be straight line. Lots of these markets are very nascent as it relates to programmatic. You know, we've had a couple years, for instance, in Germany, where they've been up hundreds of percentage points every year. That's not gonna happen every year, especially when we're at the scale that we're at with them. We've had the same thing happen in the U.K. We've had similar things happen in Australia. We've had similar things happen in Singapore. Of course, we've had some amazing years and quarters in Hong Kong.
It's not gonna be linear, but we do expect the international markets as a whole to outpace the U.S., for as far as we can see into the future. In terms of what markets we expect especially to contribute in 2020, you know, it depends on whether you're talking about actual dollars or whether you're talking about growth rate. We have a much stronger presence in Europe just because we've been in the market longer than we do in some of the markets in Asia.
Asia has so much growth potential. We're gonna continue to invest, but it's possible to see some serious pops or inflections in some of the markets that we're in in Asia. Without going through each market by market, it's really hard to identify which ones we're expecting the most from in terms of growth rate or absolute dollars. We expect all of our international markets to continue to lead the way.
Thanks, Mark. Operator, next question.
Thank you. The next question comes from Mark Mahaney with RBC Capital Markets. Please go ahead.
Great. It's Zach Schwartzman on for Mark. Thanks for taking our question. At the end of your prepared remarks, you spoke about products such as cross-device data and identity solutions, and that you expect that to continue that trend with new products that'll be launched in 2020. What potential synergies do you see here with these trends for live TV, sports, and really just Connected TV as a whole? Can you elaborate how you think this could evolve and how you can capitalize on this trend in 2020? Thank you.
You bet. I'm so glad to get this question because it gives us a chance to talk about part of the heart of our value proposition, which is, you know, media is getting more and more fragmented, especially in CTV. Because of that fragmentation, if you're a buyer of ads, it's really hard to control what we call reach and frequency, which means how many people are you reaching and how often are you reaching them? If you're trying to launch a new movie and you want everyone in America to see the ad three times, doing that is harder than ever.
Not because there aren't people watching TV all the time, in fact, we're consuming more TV content apparently than ever, but because it's just harder to know where they are and how to measure where they are. Because there's 20 ways to watch ESPN on your TV. You could do it through your cable company, you could do it through Sling TV or Tubi TV, or you could do it through the ESPN app. There's a lot of different ways to get access to it. If those aren't all coordinated, if the buy isn't coordinated so that you're showing the ad three times to one person and three times to another person, you're gonna waste a bunch of money, even worse, you might make the consumer hate you because you showed them the ad 57 times instead of three.
The only way to fix that is to have a platform that is objective, sitting in front of all the different media options. The only way to partner with all those media options is to be objective, to not own any media, so that when you partner with them, you can then say, "Oh, we bought one here, and we shouldn't buy one over here." In order to do that, not only do you have to be objective and you have to partner with everybody, but you also have to have technology that helps you identify, oh, wow, this phone is the same user as this TV. It's gonna change the way all of TV is bought so that we're actually respecting the consumer more than ever.
You're not seeing ads that are irrelevant on TV, the way that most of the ads you see on TV are irrelevant to you today. That will all be fixed because of things like that cross-device technology that is growing at, I don't know, triple the rate of our business. We expect that to continue just because it's at the core of what we do, which is help buyers make very deliberate and data-driven decisions so they're talking to the right people the right amount. And as it relates to TV, or the second part of your question, I think just one other thing to mention, and that is that a lot of the data that gets used doesn't originate from TV.
The best way to learn something about the consumer is to listen to them and on in all of their environments where they're sharing their preferences in consumer safe ways so that we then can tailor the ad to them in a respectable way that if we were just in TV, if we were just in Connected TV, we wouldn't have the data or insight to do that.
Cool. Thanks, Zach. Really appreciate it. Operator, next question.
Thank you. The next question comes from Shyam Patil with Susquehanna. Please go ahead.
Thanks, guys. I had a question on the Fire TV partnership. Surprising to see the 21x growth rate occur so fast. Jeff, you mentioned you have access to all the premium inventory, not just remnant. I was wondering if you could just talk, you know, more about this opportunity with Fire TV for Trade Desk. Also just, you know, how economics here compare to other CTV partnerships you guys have in place.
You bet. Shyam, thanks for the question. I'm if I had to pick a partnership that we have just under discussed and is a very big deal that we've yet still failed to express what a significant deal it is, it's this one. Part of the reason why it's such a big deal is because of the way that Amazon is approaching it. I know we've taken a stab at explaining this before, but I wanna do it again because I think we're getting better at it. First, you know, Amazon Publishing Services, APS, is started with all the Fire devices.
They're showing ads on the Stick and the Puck that you buy from Amazon so that you can access all of your content, basically their competitive product to Roku. They're starting with just the third party content on it. We're not talking about Amazon apps or Amazon Prime Video. We're talking about, you know, things like Discovery Channel or the Turner products or any of those sorts of things that are on the app that you or on the apps that you would have on your Amazon Fire device.
Any of those that have partnerships with Amazon that are on their device, we're able to show ads, assuming that the content owners have opted in to showing ads as their means, or at least part of their means, to monetize their content. That's exciting by itself just because Amazon Fire is growing and they're competing. What's especially exciting is that many of the MVPDs, the distributors for linear television, charge a pretty aggressive rate. In some cases, in Connected TV, because rates have gone higher in Connected TV because of the data component that makes them more valuable, they're charging an even higher percentage. If you're a content owner and you didn't like the economics of linear TV, in some cases, digital is worse.
You're not racing to put your content on unless the consumers drive you there, which a lot of consumers are driving people there, so they're moving over because they're being forced to by the consumer. What Amazon did in this deal is they made very transparent that they're charging a much lower rate at 10% and 15%, which is much lower than any of the MVPDs did in television, in linear television. What that means is that now not only are consumers dragging content owners to streaming, but the economics are now too. I think that means very good things for our industry in addition to our partnership.
The fact that there's so much more content and so many more ads, that 21x, I think is largely driven by the economics that made it obvious for content owners to go all in on streaming. Because they're doing that, it puts pressure on other players to make the economics more competitive, and that too is going to fuel more people coming into streaming. I don't know that there's a more macro move that I can point to that has more bullish signs for us and for all of Connected TV in 2020.
Thank you very much.
Thank you.
Operator, next question.
Thank you. The next question comes from Mark Kelley with Nomura. Please go ahead.
Jeff, can you just remind us how much of your spend was political during the last election cycle in the U.S.? You know, it'll be a pretty big opportunity next year. It'd be good to have kind of a baseline to think about. Second, just bringing it back to CTV, I wanna make sure I understand your prepared remarks a bit more. You know, you called out a lot of the positives from gaining access to Amazon and some of the other growth metrics. You noted, I think, that in particular helped re-accelerate growth. It looks like CTV growth was basically flat versus the second quarter. You know, corporate level revenue growth was down a few hundred basis points. I guess, are you talking about spend, or did I mishear something? Thank you.
Let me first talk about the percentage that came from political. In the last election cycle, it was low single digits. Of course, when you've grown as much as we have as a company since the last election cycle, if we add low single digits to 2020 in political that wouldn't be here otherwise, that's a big number. That's a material number. I'm really excited about what we proved in the last election and excited to be a part of the process in this upcoming election. As it relates to the prepared remarks and slowing down in Q3, I actually I'm not sure what you're referring to.
I don't know if you're referring to just a 150% year-on-year growth and you're looking at from that perspective, Mark?
Yeah. You guys said 145%, and then, you know, in Q2, you guys said it was 2x year-on-year.
Yeah.
Yes, you get to basically the same number.
That doesn't mean that it's necessarily slowing down. You're just looking at the growth rate.
No. You guys said re-accelerated-
It's year-over-year.
Yeah. You guys said that it helped to re-accelerate the business. I just wanna know what re-accelerated. That's all.
Throughout the quarter?
Yeah. I see what you're asking for.
Oh, got it.
It was really exciting to see. I was just trying to give a little bit of a glimpse into what happened inside of the quarter. It wasn't that there was a slowdown or anything like that. 150% year-over-year growth is incredibly exciting, especially after a quarter where you put up 200% growth year-over-year. However, what I was trying to point to in that comment of reacceleration was simply that the last month of the quarter was especially strong. I was just trying to give some indication that that's part of our reason for bullishness on CTV as we're looking at Q4, but also as we're looking to 2020. That 21x is super exciting, but a really strong end to our Q3 is something I was just trying to share a glimpse of with shareholders.
Got it. Okay. Thank you, Jeff.
Thanks, Mark. Last question, Operator.
Thank you. The last question comes from Brian Schwartz with Oppenheimer. Please go ahead.
Yeah, thank you for taking my question this afternoon. Jeff, just one question. Just an operational question on the management team. You had talked previously about looking to help scale the leadership team to help you scale the business. Just wondering if you have any updates in terms of that initiative. Thanks.
Yeah, thanks, Brian. It's been a very busy year. I would summarize 2019 thus far as a year where we've been adding a lot of leadership from the highest levels of the organization, which we've added a few people to. Also that we've added a lot of management and leaders in the middle, if you will. We've never had a year where we've added so many leaders from outside the company at all levels of the organization. It's been an adjustment for us.
The adjustment has been extremely positive because, you know, we had a company with an really amazing culture going into the year, and we leave having added to it and added this amazing team with fresh blood that can actually influence lots of other people. By keeping our bar really high on the people that we'll add to the team, at times we've wanted to grow faster than we have. I still wanna grow faster than we have in terms of adding to the team and growing our business because I think it is land grab time. I'm really happy at what we've done this year in terms of adding to our leadership.
We still have more work to do, but I think this year marks the most significant year in the history of our company in terms of the number of leaders that we've added to the company. I'm really pleased with what is often a very tough sort of stairstep for growing and scaled companies when you get to our size. It's an exciting time.
Thank you so much, Brian. Operator?
Thank you. This concludes today's teleconference. You may disconnect your lines at this time. Thank you for your participation.