The Trade Desk, Inc. (TTD) Q3 2026 Earnings Report, Transcript and Summary

Good day, everyone, and welcome to The Trade Desk, Inc. Third Quarter 2026 Earnings Conference Call. [Operator Instructions] It is now my pleasure to hand the floor over to your host, Chris Toth, Vice President of Investor Relations. Sir, the floor is yours.

Thank you, operator. Hello, and good afternoon to everyone. Welcome to The Trade Desk Third Quarter 2026 Earnings Conference Call. On the call today are CEO and Co-Founder, Jeff Green; and our Chief Financial Officer, Nate Olmstead. A copy of our earnings press release is available on our website in the Investor Relations section at thetradedesk.com. Please note that aside from historical information, today's discussion and our responses during the Q&A may include forward-looking statements. These statements are subject to risks and uncertainties and reflect our views and assumptions as of the date such statements are made. Actual results may vary significantly, and we expressly disclaim any obligations to update the forward-looking statements made today. If any of our beliefs or assumptions prove incorrect, actual financial results could differ materially from our projections or those implied by these forward-looking statements. For a detailed discussion of risks, please refer to the risk factors mentioned in our press release and our most recent SEC filings. In addition to our GAAP financial results, we present supplemental non-GAAP financial data. A reconciliation of the GAAP to non-GAAP measures is available in our earnings press release and investor presentation. We believe that presenting these non-GAAP measures alongside our GAAP results offers a more comprehensive view of the company's operational performance. With that, I'll now turn the call over to CEO and Co-Founder, Jeff Green. Jeff?

Thanks, Chris, and good afternoon, everyone. Thank you for joining us. Our third quarter revenue was approximately $736 million, meaningfully ahead of the outlook we provided in August. Those results were much better than we expected 3 months ago, and it's approximately flat on a year-over-year basis. While these results are above what we expected for this quarter, this level of growth is, of course, well below the level of growth that we expect from ourselves in the future. Last quarter, I talked about the opportunities for improving our execution and the things we can directly influence to improve the trajectory of our business. Since then, we have been improving our operations at an increased pace across the company. You can see that reflected in the way we are working with our customers, how quickly we are improving our platform, how we are allocating resources and how we are operating as a company. I'm encouraged by our progress, and I am proud of our team. But I want to be very clear, we are not satisfied. The questions that we are focused on every day is how can we build on that progress? How can we improve the underlying trajectory of the business and how do we create more value for our clients. We also believe that by helping the world's largest advertisers drive better decisions, use their data more effectively and more importantly, drive better business outcomes that we can continue to help move the industry and the open Internet forward. We marked our 10-year anniversary as a publicly traded company in Q3. We've had tremendous growth since our founding and since our IPO. The complexity of the ecosystem has also grown and the way we operate must evolve alongside it. Given it was our 10-year anniversary, it is fitting to express my personal excitement to have my co-founder reengage more deeply at The Trade Desk once again. Dave Pickles and I founded the company 17 years ago. And Dave has continued to support The Trade Desk as a long-term adviser since stepping down from his role as CTO in September of 2023. But as we enter the next stage of our product innovation, I am very pleased to share that Dave, the architect of the platform and the key driver of The Trade Desk's revenue, has agreed to substantially expand his role once again. Together, in many cases, he and I led the growth of the programmatic space and the ad ecosystem that funds the open Internet. I'm excited to work with him again and to tackle new challenges. This is an area where we believe The Trade Desk is stronger than ever to make an impact at greater scale than we ever have. Bringing Dave's experience, his technical judgment and deep understanding of our platform more directly into that work will help us shape what we believe is the next generation of advertising technology. I'm extremely excited to have him more deeply engaged with the team as we move into this next chapter. We spent the last year changing our company because the industry has changed and is continuing to change. However, our vision and our values remain the same as what Dave and I established at our company founding. For years, we have primarily served the largest agencies and brands in the world. We have grown, but so is our ecosystem. Programmatic advertising was once a line item on a media plan. Now it's the heart of the plan. With the growth of the ecosystem of programmatic and digital advertising, we've also seen decision-making move inside of brands and agencies to different parts of the organization. Often, the decision-making has moved up, sometimes all the way to the C-suite of Fortune 500 companies, but in other cases, it's often also been distributed. But in all cases, there are more seats at the table. Decisions about what advertising to buy that were once made primarily by hands-on-keyboard traders are increasingly being made by senior agency leaders, procurement groups at large brands and senior executives. In many cases now, the CFOs, CEOs and even CMOs are quickly learning the esoteric details of programmatic advertising and the marketplace of buying media. Put simply, we must meet our customers where those decisions are being made and decisions are being made at a variety of levels of those organizations today. But that said, data-driven advertising is more important than ever because most of the Fortune 500 are searching for new growth. In most cases, growth isn't possible without branding and advertising. And as linear TV goes away and becomes less and less relevant relative to streaming, brands will win or lose based on their ability to understand and use data-driven advertising. This is an important moment for the advertising industry, but it is also an important moment for our company because we have spent the last year fine-tuning our company. There is a massive TAM and a huge opportunity for growth ahead. To capture that, we've had to refine our broader organization into one capable of greater scale than ever before. We have brought together perhaps the best team assembled in ad tech, which includes those who know the esoteric world of programmatic advertising and others who drive and operate at unprecedented scale. The Trade Desk has always differentiated with greater operating discipline in how we build our products, how we go to market and how we align our interest with our clients and how we execute. We are setting clear priorities in product development, developing more consistent sales and account planning processes, enforcing stronger accountability through our organization and allocating resources more deliberately against our highest return opportunities, which include agentic AI, CTV and measurement. AI has produced a tremendous opportunity for us and our clients. Because of AI, TTD is able to refine its product faster than ever. AI is helping us operate. AI is helping us build product and AI is in the product to help our clients manage their advertising. We've been transforming our business and are starting to see the fruits of our labors. We have strengthened our key leadership team and now have most of the key players on board. And as a result, we have organized our go-to-market teams differently. We've added experienced leaders throughout our commercial organization so that we are better strategically serving our largest customers, bringing together product and commercial teams to address our clients' need to find new growth. That includes better measurement, more effective use of data and AI and more strategic approaches to how we plan and grow together. One of the clearest indicators of that progress is our joint business plans or JBPs. Through the end of Q3, the number of JBPs was up more than 30% year-over-year, including growth of more than 20% in the third quarter, representing hundreds of agencies and brands. Most importantly, customers operating under JBPs continue to grow meaningfully faster than those under traditional MSAs. JBPs are much more than commercial agreements. They create a structured framework for brands, their agencies and The Trade Desk to plan, innovate and measure success together. These JBPs and client relationships that have JBPs tend to grow faster because they are built around long-term alignment rather than individual campaigns. JBPs provide the framework to plan together, to innovate together and to measure success against shared business objectives, and as our engagement with our clients deepens, our JBPs continue to expand. We currently have more than 100 JBPs in the pipeline, which gives us additional confidence in the opportunity ahead. One great example of where a clear JBP with bigger common goals than ever is driving growth is with one of our large financial services customers. Our team developed a deliberate strategy to engage more deeply with the leaders of individual business lines across the organization where we had not previously built relationships. Over several months, the team held dozens of conversations with business leaders focused on demonstrating the value that The Trade Desk could create against their specific business objectives. That engagement recently resulted in millions of dollars of incremental spend from one line of business alone. It is a good example of the opportunity we see when we move beyond simply being a platform that clients use to buy media and instead become a strategic partner they build with. We're seeing encouraging indicators elsewhere in the business as well. Outside the United States, we are seeing strong growth from our APAC business, which is up approximately 30% year-over-year in the first 9 months, led by China, which grew 45% year-over-year in Q3. We are also seeing strong momentum in EMEA, where growth is up more than 20% year-over-year in the first 9 months. We are also seeing encouraging progress from our growth team, which is focused on winning new brand clients, including large and middle-market clients and reengaging advertisers where we know we can create more value. Through the first 3 quarters of this year, the team has increased its client count by over 200% year-over-year, including 23 new clients in Q3 alone, and this team is a meaningful contributor to our overall revenue. Our success is led by people. However, we are a product-led company. Every product we ship aims to create greater returns for our customers, more value for The Trade Desk and our shareholders and represent an improvement to the ecosystem overall. Today, I'll highlight some of the big products that we've been working on for the last 1.5 years. We'll highlight some of the platform engagement we've improved with our system upgrade called Zuma, and we'll share more about our most exciting product, which is a measurement product and a marketplace that will transform The Trade Desk, the open Internet and premium media. And lastly, we'll provide an update on the progress on Audience Unlimited, which represents another way for TTD and advertisers to unlock the benefits of data and AI. Let's start with measurement. The advertising industry has a fundamental measurement problem. Last click and last view measurement naturally favor advertising that occurs right before a transaction takes place. What last click and last view attribution do a poor job of is recognizing all the advertising that created awareness, preference and the brand in the first place. Branding will always suffer and be underfunded when last touch attribution is the default because the heavy lifting of branding is done in the premium media. And the premium open Internet is where The Trade Desk leads in demand aggregation, and now where most premium media is consumed. Today, the heavy lifting of branding is done in CTV. It's done in sports, in movies, in news and in music. Great branding is almost always the process of associating a great brand with great media. As one client shared with me earlier this year, often the best and most effective ads for our brand are the most expensive. Our open measurement framework is designed to more accurately assign value across the customer journey and give marketers greater confidence in where advertising is creating incremental business results, whether that influence happens near the top of the funnel or at the bottom of it. Unlike the walled gardens of Amazon and Google and other user-generated content platforms, we are not grading our own homework. We are not simply measuring our own success. We are providing real sales data through real transactions. We are in the pole position here because we operate objectively on the buy side. We don't own media that we are trying to sell. Our aim is to help advertisers understand what decisions are actually working and actually driving growth and allocate their next dollar accordingly. We are trying to be their partner and actually help them grow and find incremental growth, not just take credit for the growth that would have happened anyway. Open measurement is designed to bring more sophisticated measurement directly into the platform. For advertisers with first-party data, that means more granular attribution. For advertisers that do not have access to sophisticated measurement today, we intend to make attribution available more broadly. Our goal is to evolve measurement beyond last click and last view toward a more complete understanding of the customer journey. In our product design, incrementality can help marketers better understand which investments are creating additional business results, including the value being generated across the open Internet. I'm excited to share more, and we will when we formally launch this open measurement product later this quarter. The second product area I want to discuss is Zuma. We launched Zuma during the third quarter as an important evolution of our product. Zuma upgraded Kokai. Zuma improves navigation, workflows and troubleshooting. It brings more AI directly into the user experience and improves the dialogue between human and machine. Increasingly, we want Koa and our agentic workflows to handle more of the repetitive and operational work, freeing clients to focus on strategy, creativity and business outcomes. We are already seeing early indicators of progress towards this vision. Clients are moving faster. Median active time to complete core setup workflows has declined, including 9% for campaign creation, 15% for ad group creation, and 9% for uploading creatives. Koa is doing more of the work. Koa optimizations are now enabled on 22% more strategies than before. Results are improving. In A/B tests across 62 campaigns, the upgraded Koa CPA model improved CPA by a median of 21% versus the prior model with no additional setup changes required. And we are only beginning to see what agentic workflows can make possible. Take campaign creation. In early testing, our campaign creation agent read a client's media plan and set up the campaign structure, ad groups, budgets and flights in a matter of minutes, automating work that was traditionally done manually and can take hours. Early adopters highlight meaningful time savings and more and more clients are asking to use it. We are still in beta, and there's a lot more work to do, but it is a good example of where the platform is heading. We are applying the same agentic approach to audiences and extending it across our partner ecosystem. Using MCP, an open standard adapter that allows AI systems to communicate with one another, Koa can share an advertiser's audience objectives with participating data providers. Then their AI systems can respond with custom segments designed for that campaign. Koa can then evaluate those options and use a segment only when it believes it is likely to improve performance. That creates an open agentic ecosystem in which our agents and our partners' agents can work together on behalf of advertisers. We are also beginning to work more closely with some of the world's leading agencies to co-develop agentic workflows. Previously, we highlighted our partnership with Stagwell. Today, we have extended that collaboration model, incorporating agentic workflows into the planning, activation and optimization processes that power campaigns for many of the world's largest brands and their agencies. We are still in the early stages, but these collaborations are important because our goal is to build agentic capabilities alongside the agencies and advertisers that will use them. Our partners operate the most sophisticated marketing organizations in the world. So we aim to create technology that solves real workflow problems, creating more value for brands. Over time, we believe agentic workflows can materially change the productivity of advertising teams. Work that has historically required many people and significant manual effort can increasingly be assisted or executed by agents, allowing leaner teams to operate even more sophisticated programs while keeping people focused on the decisions where judgment, creativity and strategy matter most. Put simply, Zuma is making the platform easier to use, accelerating the adoption of agentic AI and helping drive better outcomes for advertisers. We have a lot more work ahead of us, but the initial response reinforces our belief that we can preserve the sophistication that makes our platform valuable while making it more intuitive, more efficient and increasingly powered by agentic AI. And Zuma is not a onetime release. It represents the pace at which we intend to improve the platform and expand the capabilities of our agentic offering for advertisers and agencies. We are also seeing encouraging early traction with Audience Unlimited. More than 100 advertisers have now tested the product and several hundred campaigns are live. The idea behind Audience Unlimited is simple: make high-quality third-party data easier and more predictable to use while leveraging AI to help advertisers identify the audiences most likely to drive performance. And the early results are promising. In recent campaigns, advertisers have improved CPA by more than 30% and simultaneously reduced CPMs by approximately 14%. Multiple advertisers have also used Audience Unlimited successfully in direct competitive tests. These early results increase our confidence that Audience Unlimited can drive better advertiser outcomes while significantly simplifying the way data is used on our platform. As we continue to innovate, I expect Audience Unlimited to be a huge beneficiary of agentic AI as well. Measurement, Zuma, Audience Unlimited, agentic workflows and the work we're doing across our commercial organization all connect to the same objective, helping the world's largest advertisers make better decisions and generate better business outcomes. When I look towards 2027, we have more opportunity in front of us than ever. The TAM is growing. AI is fueling change and disruption, but our AI, joined with the trust and partnership we've built over many years, makes it especially valuable to The Trade Desk. Over the last couple of years, we've been upgrading our business. And as I previously mentioned, we have strengthened our leadership team and continue to solidify that team. We've evolved our commercial model, and we've been more deliberate about where we allocate resources. We've continued to invest in the platform, AI, measurement and data, and we're creating systems and processes that allow us to operate at scale. We've learned a lot along the way. We've made mistakes. And when we do, we try to correct them quickly. The organization we have today is the organization we need for the next chapter of The Trade Desk and the opportunity in front of us remains enormous. Our market share represents only a small fraction of roughly a $1 trillion global advertising market. Digital advertising continues to gain share and Connected TV continues to reshape the largest media market in the world. More of the world's premium content is becoming digitally addressable. Retail and commerce data are becoming increasingly valuable in an AI world, and AI is creating more data, more choices and more complexity. All of those trends increase the importance of objective decisioning and the role of The Trade Desk. The supply of the advertising opportunities continues to expand dramatically, faster than the demand. In a world where supply significantly exceeds demand, advertisers have extraordinary choice. But with a finite amount of ad dollars, more choices make decisioning even more valuable and more important. And objectivity also becomes more important. Access to inventory is increasingly commoditized. The value of decisioning comes from the use of data and AI to evaluate those opportunities and determine which ones are actually the right opportunity for a specific advertiser. The ability to evaluate millions of opportunities and determine which are right for a particular advertiser is what matters. And increasingly, the data required to make those decisions belongs to our customers. In an AI-driven world, proprietary data becomes even more valuable. Partners that help you protect your data become better partners. The world's largest brands need partners they can trust with that data, partners whose incentives are aligned with theirs and partners who are not using their data to their competitors' advantage. Our independence and objectivity have always mattered, and I believe that they matter more today than ever before. As most of you know, the U.S. just wrapped up its midterm election. We are proud of the role that we, The Trade Desk, played in the election. We aim to provide all political parties equal access to our platform so that they can enable their ideas to compete fairly in the marketplace of ideas. I consider it a great achievement of our company to have 2 parties with competing ideas, both use our platform at the same time in hopes of swaying voters and winning elections. This election was our biggest yet. Given this was a midterm and not a presidential election, we see this as a strong indicator of the role that we can play in the future as we continue to focus on building and maintaining trust of all of our clients, including the political clients. So let me wrap up with a few thoughts about the future. As we head toward 2027, our priorities are clear: one, to continue improving how we serve the world's largest brands and largest agencies; two, to make the power of our platform easier to access while preserving the sophistication that differentiates it; three, to change how advertising is measured so that marketers can better understand what actually drives growth; four, to continue putting AI to work in ways that make marketers more productive and their advertising spend more effective; and five, to operate The Trade Desk with greater agility, greater focus, greater ownership and faster speeds. Q3 represented progress, but we are measuring ourselves against where we believe this company can go, not simply against the expectations we set 3 months ago. We have more work to do. At the same time, I believe the actions we've taken, the team we've assembled and the products we're bringing to market this quarter and in 2027 and beyond, position The Trade Desk to lead the open Internet and the advertising that funds it at higher levels than ever before. We've spent much of the last decade building the leading independent buying platform for the open Internet. Our focus is now turning that into the next phase of durable growth. With that, I'll turn the call over to Nate.

Thank you, Jeff, and good afternoon, everyone. Let me start with our third quarter results. In Q3, we delivered revenue of $736 million, down 0.5% year-over-year and meaningfully ahead of the outlook we provided in August. We saw particular demand strength in medical health and the political election business, which continued to grow cycle on cycle as expected. In addition, our Q3 results included a benefit to revenue totaling approximately 3 percentage points from business with a strategic partner that we do not expect to recur. Importantly, even excluding that onetime benefit, the underlying business still performed meaningfully better than we anticipated entering the quarter. We generated $266 million of adjusted EBITDA during the quarter, representing a 36% margin, also meaningfully ahead of our expectations. In terms of our channel mix, video, which includes CTV, represented a low 50s percent share of our business in Q3. Mobile represented a high 20s percent share of the business during the quarter, while display represented a low double-digit share. Audio represented around 7% of the business and posted much stronger year-over-year growth than any other channel. In our verticals, medical health, personal finance, travel and, of course, political related to the U.S. midterm elections posted the strongest year-over-year growth during the quarter. Now turning to expenses. Q3 operating expenses were $643 million. The quarter included $44 million of restructuring charges, primarily related to severance and other employee benefits associated with the organizational actions we announced in September. Excluding stock-based compensation and restructuring charges, Q3 operating expenses were $504 million. Income tax expense was $68 million in the third quarter, driven primarily by our pretax profitability and the impact of stock-based awards. Platform operations expense increased 23% year-over-year in Q3. There are 3 important factors driving the increase. First, we continue to invest in the infrastructure required to support the increasing scale and sophistication of our platform, including higher query volumes and greater use of AI and machine learning. Second, we are investing to optimize the mix between owned data center capacity and third-party hosting. Some of the costs associated with that transition are occurring ahead of the efficiencies we expect the infrastructure to generate over time. And third, as our platform and value-added services and data evolve, certain costs that historically were recognized as a reduction to revenue are now being recognized as a cost within platform operations expense. That affects the year-over-year OpEx comparison even though it does not represent a comparable increase in the underlying economic cost of operating the platform. We expect platform operations expense to remain an important area of investment, but we are intensely focused on improving the efficiency of that investment and generating improved operating leverage as the business scales. During the third quarter, we also took an important step toward creating a more focused and efficient operating model. As previously announced, we reduced our global workforce by approximately 15%. The actions resulted in approximately $44 million of restructuring expense in Q3. We currently expect these workforce actions to generate more than $100 million of gross savings in 2027. I do want to be clear about how we are thinking about those savings. Our objective is not simply to maximize near-term margins. We are building a more efficient operating model that allows us to continue investing aggressively in the areas where we see the greatest opportunity, which includes our platform, AI, measurement, data and our most important customer relationships while significantly improving the returns on those investments. Net income for the quarter was $38 million or $0.08 per diluted share. Adjusted net income for the quarter was $155 million or $0.33 per diluted share. Net cash provided by operating activities was $258 million, and free cash flow was $153 million in Q3. We ended the quarter with a strong cash and liquidity position. Our balance sheet had about $1.6 billion in cash, cash equivalents and short-term investments at the end of the quarter. We did not repurchase shares of our Class A common stock during Q3. We ended the quarter with $269 million remaining on our program authorization. As you saw in our press release, we announced an additional authorization under the program, bringing the total authorization available for future repurchases to $1 billion. We will continue to evaluate capital allocation opportunities with a focus on long-term shareholder returns, including offsetting dilution over time. Turning to our outlook for the fourth quarter. For Q4, we expect revenue to be at least $755 million. We estimate adjusted EBITDA for Q4 to be at least $280 million. Before I wrap up, I want to provide some additional context on how we are currently thinking about 2027. Given the actions we have taken and are currently evaluating, we estimate a meaningful moderation in our expense growth compared with 2026. We believe the changes we are making to our structure create a significantly better foundation for operating leverage as revenue growth improves. Our objective is to demonstrate that The Trade Desk can continue investing aggressively in innovation and growth while also translating incremental revenue into significantly greater incremental profitability over time. I look forward to sharing more details on our 2027 operating plan when we report in February. That concludes our prepared remarks. Operator, please open up the call for questions.

[Operator Instructions] Your first question is coming from Shyam Patil from Susquehanna.

Great to see the results. Jeff, I wanted to ask about your tactical plan to drive a turnaround. Q3 was clearly better than expected. The political performance, I think, highlighted the value of the platform to the marketplace. Bringing Pickles back is another very encouraging step for all of us that know what he meant. So I guess from here, what additional tactical changes do you think are needed to drive a broader turnaround? How confident are you that the company can execute? And how should investors think about the time frame for returning to stronger, more durable growth?

Yes. Thanks for the question, Shyam. Really appreciate it. I appreciate the kind words as well built into the question. So let me start big picture. First, of course, we're encouraged by the progress in Q3, but our growth is still below the level we expect from ourselves and what we think we are capable of and honestly, what the ecosystem needs from us. But of course, our focus is on what we can control, our execution, our product, our customer value, and of course, our resource allocation as well. And we understand how we can improve and grow, and we're making progress. So first of all, to do that, we need to make the power of The Trade Desk easier to use, easier to adopt and easier to prove. Zuma is improving our workflows and increasing AI adoption. And of course, open measurement is expected to better prove incremental outcomes. And with it, Audience Unlimited is simplifying the access to the high-quality data that's all over our data marketplace. While open measurement will be implemented this quarter, we expect the product to improve our growth next year and provide a tailwind for years to come. As you point out, Dave Pickles coming back is a really big deal, especially to me personally. He, of course, was the architect of the platform initially. He understands the technology at an extraordinary level, and he can help us move faster. But I'm especially excited about the 15-plus years of trust that he and I have working together full time, putting that to work once again as this company and our ecosystem is at a really important crossroads. But the objective of all of these changes is really clear and our investments have a really simple objective, and that is to help our clients grow. We're improving how we serve the world's largest advertisers and brands and agencies, as more decisions move higher in those organizations. Of course, our JBPs are an amazing and important proof point. The customers that use JBPs continue to grow meaningfully faster than the rest. And of course, we have more in our pipeline than ever because that tactic is working. We've, of course, reshaped our organization to improve our speed, our focus and our agility so that decisions get made faster and our teams are closer to the highest priority opportunities. At times, we've been too focused on doing 100 small things instead of 10 big things that can move the needle for our clients, and we're getting better focused on our resource allocation. I -- you talked about my confidence and timing. I am very confident in the opportunity ahead for us and our ability to execute against this. This is largely because of the team that we've assembled. We are well aware that one ingredient of the recipe of our recovery is clearly explaining to you, our investors, what you can expect from us. And our goal on that level is to demonstrate progress over and over again. Q3 was progress, but it was, of course, just 1 quarter. The important thing for us now is to build on that. And that's what we intend to do. If we keep executing against the changes that we've made and take advantage of the opportunities that are in front of us, we think that we can return The Trade Desk to the stronger, more durable story that we were in the past, and we think we can be bigger and better than we've ever been before. Really appreciate the question, Shyam.

Your next question is coming from Justin Patterson from KeyBanc.

It sounds like this is a much stronger political cycle for you. Jeff, could you talk about what drove that success and how you executed differently this cycle? And for Nate, can you size that political contribution for Q3 and Q4 and frame how we should think about that as we head into Q1 of next year?

You bet. Thanks for the question, Justin. So let me first just reiterate some of the things I said in the prepared remarks. One of the reasons that we have been so successful in political advertising is our objectivity and independence. I'm really proud of the fact that we've been able to support both sides of the aisle in the United States. And we've done this in democratic markets around the world. We view the democratic process of elections to be one that's really sacred to the functioning of our society, and we also view our role in that as sacred to create a fair marketplace of ideas and a place where both parties can come have a fair shake at sharing their ideas in the most premium parts of the media landscape. To that end, this was our biggest political election cycle yet. We had a bigger growth rate in political compared to the last presidential election. Of course, the presidential elections tend to be bigger than the midterms. But this is also a bigger growth rate and, of course, a total than our last midterm election as well. So in total dollars or growth rate, I think you can argue that this was our best political cycle yet. That's particularly significant because this, of course, was midterm, not a presidential. And it reinforces the increasing significant role that programmatic advertising plays in the political process. Of course, we want to make certain that political parties can continue to trust us and invest in us. And we think this particular cycle provides a more definitive proof point than we've ever provided to the market and to investors before. It's important to note, though, that political will always be cyclical. And of course, that's why Nate's answer is so important. So I'm going to give the hard part to Nate now.

Thanks, Jeff. Yes, I mean, political spend has been strong. And as Jeff said, it continues to grow cycle on cycle. It represented a mid-single-digit percentage of our business in Q3, and we expect it should be a little bit higher in Q4 than it was in Q3. I think that's really important to keep in mind, as you think about Q4, it's also very important to keep in mind, as you said, as you think about Q1. Typically, or I would say very consistently, Q1 is always seasonally lower than Q4. And in election years, the sequential decline has historically been about 25% on an absolute basis from Q4 to Q1. So if you think about Q1 and you start putting things together, you should definitely be mindful of that historical trend.

Your next question is coming from Vasily Karasyov from Cannonball Research.

Jeff, I wanted to ask you, could The Trade Desk expand more aggressively into the mid-market, maybe pursue new business models or use M&A to accelerate growth?

You bet. Thanks, Vasily. I appreciate the question. So let me go back to something that I said really since our IPO, advertising has a fat head and a long tail. Our business model has been very unique and our approach to the market has been very unique in that. We're one of the few that has ever started at the head and work our way down. Most tend to start with SMBs and then work their way up if they ever make their way up. But we started with the biggest knowing that we would work our way down. We've been working our way down that fat head towards the long tail over the last 1.5 decades. So we're in that process. We'll continue doing it. Some have argued we should do that more quickly, whether that's by developing new lines of business or through M&A. Of course, we're going to keep going as quickly as we can, but we don't want to lose sight of the fact that about half the TAM is contained inside of the top 1,000 advertisers, if you will. So there's still massive amounts of room to grow in the business that we've built over the last 17 years, and let's never lose sight of that because servicing those advertisers is way different than servicing SMBs. It's just a different business model or a different service model, even if the underlying technology is very similar. We think there's lots of opportunity there, and we'll keep going after it. As it relates to the M&A piece of it, our philosophy on M&A has not fundamentally changed. We are always looking. And I will just transparently say, we're looking more now than we historically have, partly because I just think there's a bunch of opportunity in the market that there hasn't been in the past. But that said, it's also critical to take away that our standards are also higher than ever. And the reason why our standards are higher than ever is we have an opportunity to execute and accelerate our growth with all the changes that we've made in our business. I would never want to derail that by making our 17th priority, our #1 priority simply because we bought a company. So by being super focused on the things that we look at and hold that acquisition standard to a very high level, we think we create the clearest opportunities ahead. We do think we have a very clear opportunity in front of us with one of the biggest TAMs in public markets today. And so we don't ever want to lose sight of the opportunity in front of us and don't want to get distracted through M&A, but at the same time, recognize that, that can help accelerate us. So we're always looking. I appreciate the question.

Our next question is coming from Shweta Khajuria from Wolfe Research.

Jeff, as we move into next year, why do you think The Trade Desk's pricing model, like the take rate model remains the right pricing model? And under what circumstances would you consider strategic adjustments to that?

You bet. Thanks, I was actually hoping somebody would ask about this in this Q&A, even though we've answered it many times over the years, partly because it's just been a topic of conversation seemingly in the peanut gallery, if you will, about our company more lately than ever. So just a couple of things, big picture. We've been public for about 10 years. And during that time, our take rate has remained pretty consistent. It's been up about half of those 10 years and it's been down about half of those 10 years, but all within 1 point, 1.5 points, if you will, of where the average is. So it's not moving around a lot. We, of course, have always been focused on giving our clients the very best outcomes. And whenever we launch a new product, especially if it comes with incremental cost, we want to make certain that it earns its keep, that it's adding more value to our advertisers than it costs. We want that economic -- we want that to be economically obvious for our clients to take advantage of. We think we've historically done that over and over again. But the reason why I think that there's a lot of discussion about should our take rate change is because we are the most transparent platform in our space. By giving transparency, we can incite debate where others will just bury their cost into media or hide it in some other way. That has never been our goal. That's not the way we operate. That's not what we do. But what we have to do going forward is do a better job of helping advertisers and agencies and the landscape have a conversation around true cost and not simply about pricing. Because when you start talking about true cost, it becomes really easy to see that The Trade Desk is not the most expensive platform in the space. Of course, the sticker price, in some cases, is. But when you look at the transparency and the fact that we break out prices instead of putting them into media, we create an advantage for our clients, and we preserve our objectivity where almost every one of our competitors have burned that to the ground. We think that as the ecosystem evolves, that the premium on objectivity, the premium on trust, especially in a world where their data is more valuable, exponentially more valuable in an AI world than it was before, that, that creates an opportunity for us to constantly be proving that our platform is worth its cost and that the true cost when evaluated apples-to-apples will always prove to be worth it. Now that said, let me just say this, if it's possible for us to lower our take rate and make it up -- make up more growth, of course, we're open to that. But in most cases, what we found is that people aren't really concerned about the cost. Instead, our transparency can create discussion or debate or in some cases, just get leverage for a negotiation. But when it really comes down to it, we're adding more value than we cost and that gets proven again and again. And as a result, we're confident that something close to where we're at today is the same way that we can grow in the future. But we're always open to changing if there's an easy way to prove that we can make up that growth by giving discounts or create more growth than we did in discounts. Thanks for the questions, Shweta.

Your next question is coming from Matt Swanson from RBC Capital Markets.

Great. Jeff, you mentioned in the prepared remarks the millions of opportunities in inventory. But thinking specifically about premium inventory as it opens up, it feels like some of the newer platforms we've seen emerge starting out with different DSP or ad tech partners first. Maybe historically, it seemed like Trade Desk was always the default. Could you just talk a little bit about what Trade Desk needs to do to kind of remain in those conversations as kind of a preferred premium partner?

Yes. Thank you, Matt. This was another question actually I was hoping we get. So I really appreciate the question. It's partly because there's been a little bit of a change in strategy or at least philosophy over what we've done in the past. So in 2017, '18, '19, up until the pandemic, really, we had a mantra in our inventory partnerships team that was go fast and be first. And our goal was to be the first programmatic partner of any sort of traditional media company or emerging media company that wanted to get into the world of programmatic. That's honestly changed in the last few years, in part because there is so much more supply than there is demand that it isn't -- there isn't a deep need for more supply. However, of course, we want quality supply, and we want to make certain that we have the right access. It's where our clients are really excited about putting our technology to work is where they can control decisioning and have access to media that they actually create price discovery. They know what they're buying. There are a whole bunch of platforms, partly because of that outsized supply to demand that are not trying to enable price discovery that don't want to provide any metadata so that an advertiser doesn't know what they're buying. We do not want to facilitate that. And whether or not you give access that way, we need to be careful of having a mantra like be first because if they're not getting the right access to the inventory and making it possible for us to have price discovery and compare one piece of inventory to another, then we don't want to do that. In general, the more premium side of the Internet, which is the premise of your question, they do want to provide metadata. They do want to be compared to everything else because they outperform. This is part of the reason why we're so excited by open measurement because a lot of the most premium sites -- or most premium outlets of the open Internet are especially excited to have another player help validate that their premium media is worth it, which is often what we're finding. It is very often worth what we pay for. And we just want to make certain that we're doing that with objectivity. We can only leverage our objectivity if we have enough metadata to enable price discovery. So we're doing that for the biggest advertisers on the most premium inventory. I think there's so much opportunity ahead that will come from the LLMs and others. I do expect that TAM to open up in part because I do expect their inventory to be really, really valuable, but it's also going to be very expensive in part because of all the processing power required for them to get great answers in an environment where that's really expensive, they want to get as much demand as possible. And of course, even though there's more supply than demand, there's always some amount of supply that has scarcity that there is outsized demand for that particular inventory than others. I think that will be true of the LLMs' inventory. So they are very early stage in monetizing through advertising, and there's tremendous opportunity for us in the future, I think for them as well. But we don't want to start off by doing it wrong, and we're okay to be patient. And so it's no longer about getting the first access. It is about getting the right access and enabling price discovery, making certain there's the right amount of metadata so we can do right by our customers. Many of our customers are not growing in -- over the last few years and in this environment. And we need to make certain that we're helping them to grow. And that's critical in the way we approach measurement, but that's also critical in the way that we approach inventory.

Your next question is coming from Youssef Squali from Truist Securities.

Maybe a couple for Nate. Nate, could you please talk about the operational efficiencies you're expecting in 2027 and how to think about OpEx growth versus revenue growth. I think in the prepared remarks, you talked about $100 million benefit from RIF, but what else is out there? And maybe also talk about the dilution from the recent options repricing. That's something -- that's a question we often get from investors.

Yes, sure. So I think looking out to the '27 on the expense side, I would say we expect expense growth to moderate meaningfully in '27 compared to '26. Part of that is the workforce actions that we took, which, as you mentioned, we expect to generate more than $100 million of gross savings. So some of that will be reinvested. Some of that will flow through. But at the same time, there's -- there are areas where we're going to keep investing aggressively because the best way to drive operating leverage is through revenue growth. So it's going to be a combination of some of that flowing through, but also for aggressive investments in areas where we see good line of sight to driving revenue growth and improving our competitive position. Really, I think what's important, though, is that we do that with discipline. It's about investing with discipline. It's about making each dollar of investment compete for that capital. So we'll do that throughout the winter here and then on into next year. I think the -- I think there's lots of areas other than headcount that we can look at. Obviously, platform operations is one area, software, SaaS, outside services, facilities, really across the board, we're looking at every area to find places where we can be more efficient. And then again, some of that efficiency or some of those savings will flow through, but I really see those as opportunities to fuel the investment that we need in product and in go-to-market to create more value for our customers. In terms of dilution from the option repricing, we actually did not issue any new shares through the option repricing. So there's no additional dilution from that. There's a little bit of additional stock-based compensation, but that was one of the factors we looked at as we considered different forms of compensation for our employees. And one of the attractive things about the repricing was the fact that it did not have a dilutive effect.

Your next question is coming from Mark Mahaney from Evercore.

A question for Nate and a question for Jeff. Nate, you mentioned a couple of verticals like medical health, personal finance, travel, political that were strong. Are there -- which ones would you call out as being weak? And were those due to execution issues that can be solved? Or were there some cyclical factors impacting those the softer verticals? And then, Jeff, we're further along into AI, AI overviews, AI mode, chatbots, and now in the last couple of months, the rise of these agents. And your thoughts on what the impact of all that has been on the advertising and marketing potential, the extent to which it challenges the TAM, expands the TAM or limits it? I know that's a broad general question, but the trends get stronger and your take on it would be really helpful.

Yes. So on the verticals, I would probably call out CPG and auto as a couple where we've seen some pressure. I think the higher inflation, higher interest rates impacts auto financing and things like that. And so you've seen some pressures in those markets, and that's been something that we've experienced from an end customer standpoint. I think there's always an element of execution as well. I mean, even when there's external factors like that, we're digging hard to find things that we can control and that we can improve. And Jeff called out some of the progress we've made here in Q3 across go-to-market and also on product. And I think we keep executing well on those things. I feel confident that we can return to stronger, more durable growth in the long term.

Awesome. Thanks, Nate. Mark, I really appreciate the question. Let me first just remind everybody, I personally think that the most interesting story of this earnings season until recently was Google search growth moving down so much and then, of course, pointing to the LLMs. And to me, if you zoom out on that, what that means is that there once was a monopoly that controlled all of search. And the LLMs have created competition. I think there are very few people that think that all of that will be owned by one company again. So as a result, when you create that competition, it makes it so that there is opportunity for us to unlock TAM that was previously locked up in a monopoly. So you asked the question, will the LLMs and chatbots and all of this unlock TAM? I think no way that it wouldn't. But even if you're just looking at it as probabilities, the probability of that being unlocked under monopoly are pretty low. The probability of those being unlocked in a competitive market, given that the open Internet is truly fueled by competition that's its earmark, I think it likely unlocks TAM. I also think that the programmatic advertising space is more conducive to agentic AI than almost any other field, maybe more than any other field. And if you just think of it as The Trade Desk has been operating an API for almost 15 years. That API has so many options, so many dials that you can turn that it is overwhelming. And we've, in a way, just given people quadrillions of permutations for every single campaign that is overwhelming for you to figure out what dials to turn and what to do. If you want to be really simple or reductive about it, agentic AI essentially makes it so you have an API that can reason. And by creating the 2 systems that can talk to each other and reason together, it makes it so that you can make better choices and take advantage of the quadrillions of permutations and thousands and millions of dials that you can turn in order to optimize your campaign. So I think an open platform like ours that enables the biggest brands in the world to keep and preserve their data while having more choices to actually participate in, we're already seeing so much being done, especially by the more agile companies, those that were not afraid of AI that have really embraced agentic. They're creating audiences with so much more speed and sophistication than before. They're optimizing campaigns faster than before and with more efficacy than before. So I think agentic really is a big unlock for us that would have taken much, much longer without agentic AI and the advent of agentic AI. So I think we're going to benefit tremendously from that. I think advertising in the open Internet will benefit tremendously from it. And I do think we're unlocking TAM because of all the competition that AI has brought, not just in LLMs and not just in agentic, but all forms of AI have created more competition. And that's good for any company that's built to compete in a competitive marketplace and win on objectivity and essentially data sovereignty, which is what we're trying to do for the biggest brands in the world. So I'm really excited about how AI has reshaped the landscape for us. Thanks, Mark.

Your next question is coming from Jason Helfstein from Oppenheimer.

To Jeff, I mean, obviously, you're always constantly getting questions about take rate. But is there a way to shift the conversation to outcome-based pricing? And just given everything you said about AI and kind of where we're going with data, I guess, how do you shift the conversation away from take rate to outcome-based pricing? And then Nate, really quick, just on the guide. I mean, I guess this quarter, the number is something like down 8% ex political and the onetime. And I think the guide is like minus 16% on that same basis, but it's against, let's say, a 3-point easier comp. So it's an 8-point slowdown against it. So I guess just what is the kind of puts and takes when you think about the fourth quarter outlook versus what you just did in the third quarter?

You bet. Jason, thanks for the question. So I do think that as a company, we can do a lot more in improving the efficacy of what we're doing. And that is a huge part of what we're trying to do with measurement. It is to reorient our biggest clients toward growing their brand faster, in some cases, growing their brand again. And it is on focusing on outcomes. Now that said, you said something in the question that I just want to point out. Most of the companies in advertising who have outcome-based pricing essentially go to clients and say, give me your data and give me your money, and I will just give you results. And then they're essentially giving results that they report on according to their own metrics and you're essentially renting your own data back and you're dependent forever on those platforms to produce those results, and you hope that they continue to do so. We've had marketing mantras over the years like, own your future, because we want to make certain that they own their models, that they own their data. And we've developed a system to preserve our objectivity and help them grow their businesses and know that they own the results and the insights that come from it. That's why we aim to be the most transparent in our space ever. And it's been a big part of our success. So of course, we want to align our interest with our clients. And in most cases, that means charging them in a way that enables us to together take on those problems and isn't just give me your money and I'll take it from here. Nate, anything you had to add?

Yes. I mean, I think Q4, as always, when we think about guidance, reflects the trends that we're seeing in the business today. So while Q3 performed a little bit better than we expected, we're not assuming that every positive indicator that we see immediately translates into revenue recognition. So as you said, there's puts and takes. There's some large global advertisers out there that continue to be under external pressure. We're seeing some momentum in some other areas, but still kind of a challenging market out there today. I think it's good to be prudent going into this -- what is typically our largest seasonal quarter. We're not seeing the normal -- we're not expecting the normal seasonal uplift Q3 to Q4 that we've seen in the past. We're expecting some growth. If you strip away the adjustments that you mentioned, you're still seeing some growth Q3 to Q4, but not at the level that we've seen historically. And that's what we're seeing right now in the business.

Thanks, Jason. And Matt, you can close out the call.

Thank you. That completes our Q&A session. Everyone, this concludes today's event. You may disconnect at this time, and have a wonderful day. Thank you for your participation.