Good day, and welcome to Magnite Third Quarter 2024 Earnings Call. All participants will be in a listen-only mode. [Operator Instructions]. After today's presentation, there will be an opportunity to ask questions. [Operator Instructions]. Please note this event is being recorded. I would now like to turn the conference over to Nick Kormeluk, Investor Relations. Please go ahead.

Thank you, operator, and good afternoon, everyone. Welcome to Magnite's third quarter 2024 earnings conference call. As a reminder, this conference call is being recorded. Joining me on the call today are Michael Barrett, CEO; and David Day, our CFO. I would like to point out that we have posted financial highlight slides on our Investor Relations website to accompany today's presentation. Before we get started, I'll remind you that our prepared remarks and answers to questions will include information that might be considered to be forward-looking statements, including, but not limited to statements concerning anticipated financial performance and strategic objectives, including the potential impacts of macroeconomic factors on our business. These statements are not guarantees of future performance that reflect our current views with respect to future events and are based on assumptions and estimates and subject to known and unknown risks, uncertainties and other factors that may cause actual results, performance or achievements to be materially different from expectations or results projected or implied by forward-looking statements. A discussion of these and other risks, uncertainties and assumptions is set forth in the company's periodic reports filed with the SEC, including our third quarter 2024 Quarterly Report on Form 10-Q and our 2023 Annual Report on Form 10-K. We undertake no obligation to update forward-looking statements or relative risks. Our commentary today will include non-GAAP financial measures, including contribution ex-TAC or less traffic acquisition costs, adjusted EBITDA and non-GAAP income per share. Reconciliations between GAAP and non-GAAP metrics for our reported results can be found in our earnings press release and in the financial highlights deck that is posted on our Investor Relations website. At times, in response to your questions, we may offer additional metrics to provide greater insight to the dynamics of our business. Please be advised that this additional detail may be one-time in nature, and we may or may not provide an update on the future of these metrics. I encourage you to visit our Investor Relations website to access our press release, financial highlights deck, periodic SEC reports, and the webcast replay of today's call to learn more about Magnite. I will now turn the call over to Michael. Please go ahead, Michael.

Thank you, Nick. We are pleased to deliver another strong quarter of CTV growth and a strong beat on adjusted EBITDA allowing us to boost full year numbers. For the quarter, our year-over-year growth rate and contribution ex-TAC from CTV accelerated to 23% from a 12% growth rate in Q2. Strong CTV performance was driven by overall ad spend growth, increasing programmatic adoption by the industry's largest players, ad serving strength, and solid contribution from political. Our growth rates in contribution ex-TAC from CTV and ad spend have significantly narrowed showing a stabilization in our mix and corresponding take rate. Robust programmatic CTV adoption is also continuing across a broad set of partners. As volumes continue to scale and CPM pressures persist, our clients are increasingly looking for us to help them add value to their inventory by layering on data to target and segment audiences. It is critically important for these partners to utilize programmatic selling to reach all demand sources including SMB advertisers that didn't historically play in linear. We believe these trends are very powerful in expanding industry usage of programmatic and growing our accessible TAM. Partners more fully embracing programmatic include Roku, Netflix, Paramount, Warner Discovery, and Disney. Next, I want to highlight our Netflix and Disney partnerships. Netflix's rollout of Magnite powered programmatic solutions continues to ramp and we anticipate the partnership to grow in revenue contribution through 2025. We recently announced a two-year extension and expansion of our relationship with Disney. The recently signed deal now expands our partnership to include live sports like college football, the Latin American region, and podcast for ESPN and ABC News. We will also make Disney inventory available through ClearLine. Magnite will continue to be a key partner for Disney in the years to come and we are…

Thanks, Michael. The third quarter was another strong quarter for Magnite, beating the high end of CTV top-line expectations and exceeding adjusted EBITDA expectations. Our results drove very strong free cash flow during the quarter. Our strong financial performance, debt refinancing, and early achievement of net leverage ratio goals, have allowed for a tighter focus on equity dilution management. Our business is in a very solid position and we are set up for a strong Q4, which is reflected in our updated guidance. Total revenue for Q3 was $162 million, up 8% from Q3 2023; contribution ex-TAC was $149 million, up 12%. CTV contribution ex-TAC was $64.4 million, up 23% year-over-year and again above the top end of our guidance range. Ad spend growth and strong momentum with our SpringServe ad serving and programmatic growth drove the outperformance in the third quarter. DV+ contribution ex-TAC was $85 million, an increase from $81 million or 5% compared to the third quarter last year. Our contribution ex-TAC mix for Q3 was 43% CTV, 40% mobile, and 17% desktop. From a vertical perspective, as expected, political was the strongest performing category at approximately 3.5% of contribution ex-TAC. News and retail were strong as well. Categories that did not perform as well were food and beverage and health and fitness. Total operating expenses, which includes cost of revenue for the third quarter, were $147 million, a decrease from $168 million for the same period last year. The primary driver of the decrease was the result of the SpotX acquired intangible assets that became fully amortized in the third quarter of last year. Adjusted EBITDA operating expense for the third quarter was $99 million, well below the low end of our guidance range. In Q3, we had a temporary benefit of approximately $1.5 million in…

Thank you. We will now begin the question-and-answer session. [Operator Instructions]. Our first question is from Laura Martin with Needham and Company. Please proceed.

Great, great numbers, you guys. My two questions are these often, net leverage 0.9x, are we on our way to zero? It feels like you guys are levered given your strong free cash flow. So that's my first question. My second question is on generative AI. One of the things we're seeing is that companies that are using generative AI are growing their costs much slower than they're growing revenue. So I'm wondering how you guys are using generative AI and what the roadmap for that is so we can get cost growth, like more operating leverage into the model. Thank you.

Great. I'll take the first one. This is David. Yes. No, we're really happy with where our balance sheet or capital structure is at this point in time. I think our goal was to get to 1x net leverage ratio of 1x or below. We're happy with where we're at. And in some ways our pivot is turning a little bit to equity dilution management. And so I suspect that our net leverage ratio will -- could come down over time, but there also would be a focus on share repurchases.

Yes, and hey, Laura, I'll take the generative AI question. Yes. So it's a journey for us as a company. We started a task force within the company comprised of engineering, go-to-market folks, product, et cetera, to look at a slew of tools that are out there as well as our own development. As you know, we have a very large data science team and a lot of the machine learning that we're doing isn't true generative AI to speak of, but that's really where it's driving our cost efficiencies as we kind of pointed out our volume on the DV+ platform continues to go through the roof, but our cost of processing has come down roughly 30%, largely because of machine learning, filtering, et cetera. And we have talked about AI that we've pushed to our customers in the form of demand manager that's going over very well in creating revenue lists up to 30% for some of our customers. So it's a like most companies, it's a work in progress and we see a lot of the efficiencies are going to be gained internally, probably through software development on our teams front. But we'll have more to say on that in the coming quarters.

Our next question is from Shyam Patil with Susquehanna. Please proceed.

Hey guys, nice job on the quarter. I had a couple of questions. Michael, you talked about, I think, both of these topics in your prepared remarks. I just wanted to see maybe if you could give more color on Netflix, obviously a great lend, great relationship. Just wondering if you could provide any more color on just how you expect that to ramp in 4Q next year and then also kind of beyond that. And then, on Disney, really good to see that partnership expansion, I guess, could you talk a little bit about just what led them to select Magnite for these additional areas and just what they really liked about the company, the technology that led to the renewal and just -- and I guess, maybe just kind of relate to that. Just do you expect further revenue growth from this relationship kind of over time? Thank you.

Yes, great questions. As it relates to Netflix, as we talked about, we're very careful on our messaging there. A lot of folks ask us a lot about Netflix and I'm not suggesting you're doing that, but to get a read-through on the Netflix ad business more so than read-through on Magnite's business. So I think what we stand by is that, we're with them, that we're so programmatic partner on the sell-side, as they expand into foreign markets, we are going with them. The extension, the implementation has been seamless to-date. They're in the very early stages of selling programmatic themselves. But we stand by the estimate that when 2025 dust settles, that Netflix could very well be one of our -- if not the biggest customer of Magnite. And as it relates to Disney, I think that you got to look at the length of our relationship with Disney and all the software that we've helped build for them that they use internally under DRAX and I just think it's more of an extension of the relationship that we've had as opposed to a full blown RFP. But I think Disney's gotten very comfortable with what they want to do proprietary and what they want to be able to offer their customers, but don't necessarily want to build it themselves. And so as you start to look at the expansion of the relationship, I think those are very good indicators that Magnite is a trusted partner of theirs and will be for years to come.

Our next question is from Jason Kreyer with Craig-Hallum Capital Group. Please proceed.

Great. Thank you. Michael, there's been a lot of focus on direct connections over the last couple of quarters. Just wondering if you can maybe revisit that and give us your thoughts on how direct connections evolve over the long-term.

Yes. I mean, as you well know, Jason, a little bit different for both platforms, right, so DV+, the dust is kind of settled there. It's a little bit easier to do a direct connection because of pre-bid software being predominantly used by all DV+ kind of publishers. And as we said, there hasn't been much change in the kind of two pipe strategy that Trade Desk has deployed. One pipe for the direct connection, the second pipe to compete in the unified auction that is pre-bid and continue to see large amounts of spend from Trade Desk in the pre-bid unified auction area. As it relates to CTV, there is no pre-bids, so to speak. And so it's a bespoke kind of connection to each player's ad server. And in many cases, as you know, we're either the primary ad server or we're the programmatic layer by which the DSP would connect through to the ad server. And so ergo, we don't see much loss in economics for Magnite in a direct connection scenario in CTV by the Trade Desk in non-proprietary ad serving CTV players. And even in those case, you're dealing with a two pipe scenario. So I think that one of the things that gets slightly conflated is the success that the Trade Desk is seeing with greater matching with UID. And I think a lot of people think; oh UID in CTV equals direct connection, which is very, very far from the truth, like look at Roku for instance. Roku talks about UID and how it's helping them generate more revenue through the Trade Desk pipe. But keep in mind, UID is in direct connect and UID is serviced by Magnite to the Trade Desk through Roku. And so I think people sometimes think if UID is mentioned, Magnite gets cut out of the picture and it's far from the truth.

Thank you. A lot of good color. You talked a little bit about curated audience too. Just curious I would think that that's relatively nascent in terms of adoption there, but can you talk about where there's some early success as far as DV+, CTV and kind of the scale of publishers that are looking at something like a curated audience solution?

Yes, I'll tell you, it's incredible. So Jason, we've been talking about this whole kind of moving audience targeting to the sell-side and I think that the -- what's been the hindrance to that is there's still third-party cookies, right? And most publishers are like, hey, until the third-party cookie goes away, I'm going to continue to transact the way I always did. But now that they feel that it is going away, even though the hiccup with Chrome deprecation of cookies, they are really leaning into it. So we've seen it is early days and its early revenue, but we've seen a big explosion in it from a theory to reality. And so you're seeing quite a bit of publishers adopting it. I mean, again, curation is you're looking for finite audiences, left handed jugglers, and there's only so many that any publisher would have. But if a Magnite can group together 1,000 publishers, now all of a sudden you have this curated left handed jugular segments that is scaled and buyers now can take advantage of that. And of course, in CTV, a lot of that world is first party data and almost everything's curated in that respect. So we think it's a logical outcome of what we've been talking about and that is when third-party cookies go away, the signal loss goes away. That's how the buy side has operated since the dawn of DSPs. We really think this is a very encouraging opportunity for publishers and their partners, like a Magnite to be able to participate in those economics. So early days, but very encouraging the direction it's heading in.

All right. Appreciate all your thoughts, Michael, thank you.

You're welcome.

Our next question is from Tim Nollen with Macquarie. Please proceed.

Hey, thanks very much. Michael, could you talk a little bit more about the transfer their shift into more bidded programmatic? It was only a year or so ago that there's this weird sort of fluctuation with a lot of direct sales, which I think coincided with linear TV advertising being soft and what was happening upfront at the time. But now linear TV advertising is still soft. There's a lot more demand or -- sorry, a lot more supply on the market from the likes of Amazon Prime and so on at CTV. And I think there's been a shift toward using more bidded tools, which should be good for you, for your business and for your take rate. And I wonder if you could just give us an update on transitions in the industry, please.

Yes, Tim, really good observation. And certainly publishers, I mean buyers have always wanted to be able to activate programmatic buys in kind of a biddable fashion because they're very used to that and they think it derives value for their advertisers. And the top premium publishers have been reluctant to play in that game for a number of reasons, largely driven by that's just not what they've done historically. So they much prefer to try to sell everything direct. And I think in an environment where we've had a surplus of CTV inventory as opposed to two years ago when you had, when it was supply constrained, you're not surprising, you're seeing behavior shift more towards what the buyers want. And you're still seeing, you're seeing it definitely occur more on the OEMs, the fasts, the streaming first players and then if you look at the plus streamers, the premium big programming, legacy broadcast folks bidding, biddable programmatic is becoming something that they're leaning into, but certainly still not the prevalent way they conduct business with the buyers. So a long way of saying there's still a lot of way room to grow in there, which I think is good from a Magnite economic standpoint.

And am I right in thinking that where CTV initially was a little bit, what's the word dilutive to your take rate versus online and mobile a few years ago, now with more bidded work, it should be actually positive for your take rate as this evolves.

Yes, I think you're right. As it evolves, I think we've always made a point of saying that it's not that we're under extreme take rate pressure from the marketplace, it's just that the type of product that's been popular, namely among the plus services has been publisher sold programmatic, which carries a very low take rate vis-à-vis Magnite sourcing demand and bringing it to those publishers. So I think you're exactly right. I think we're in the low end of our historic take rate in CTV and that will only improve as more biddable becomes activated.

Great, great. Thanks very much.

Thanks, Tim.

Our next question is from Omar Dessouky with Bank of America. Please proceed.

Hey, thanks a lot for taking my question. So happy to hear about the extension of the Disney. And I was just wondering if you could talk a little bit about the economics and if you can't do so in absolute terms, possibly in relative terms, versus the kind of the prior deal or relationship you had with them. In particular, is the take rate changing at all or were there any other kind of value exchanges in this extension?

Yes, Omar, obviously can't go into great detail on that. But suffice it to say, Disney's the products that we sell to Disney carry with it take rate bans that are similar to other premier clients. So in other words, Disney is predominantly sourced their own demand through their sales team, whether it's programmatic or if it's direct sold. And in those cases, that's again the very low range of our take rate, as you look at the expansion of our opportunities, some of those expansion opportunities carry with it the concept of Magnite bringing demand in. And when that happens, it comes at a more attractive take rate just like it would for any other publisher. So I think the expansion opportunities are great. Its validates our partnership and the strength of it. But more importantly, it probably does carry with it more attractive economics when they start to light up. So I think net-net, we view Disney as a big revenue grower for Magnite in the years to come.

Okay. That's really good to hear. And if I could just ask a quick follow-up, please feel free to skip this question if it's been asked before. Do you have any kind of update on your views given, I guess, the recent regulatory decisions and court decisions regarding Google and SSPs? And I apologize again if this question has been asked already.

No, Omar, it hasn't been asked and no real update. Obviously, the ad tech trial piece of it is the remedies are going to be argued coming soon. I think there's been a slight delay in it. But listen, I think that as we all know, this is going to be several years in the making before the dust settles. We can't imagine that when the dust settles, it's not more attractive for Magnite to be able to compete in an industry where Google has a stranglehold on share on the DV+ side. We've been very successful in competing against them on the CTV side. But on the DV+ side, we think there's significant opportunity for growth once some of the structural inefficiencies in the marketplace get worked out. Impossible to tell when that might be, but we think all in all we're quite positive about the direction it's heading in.

Okay. Good to hear and thanks for taking my questions.

Thanks, Omar.

Our next question is from Dan Kurnos with The Benchmark Company. Please proceed.

Yes, thanks. Michael, you’ve kind of touched on this and you raised the sort of confusion issue which I think is very prevalent out there. Maybe it would be helpful if you updated your thoughts on given the increasing prevalence of 1P ad tech stacks, how much you think the split of volume ends up being between the two, how much they lean on kind of 3P versus 1P. And obviously there's been noise on who has better signal strength between the two. And I think you guys have argued that you guys pretty much win every AV test. So I'd love to hear kind of your thoughts there. And then separately we continue to hear from other e-comm guys, Uber, Lyft for example, that they're willing to test out more programmatic. So it feels like United is just scratching the surface of what you can do in kind of broader e-comm. But love to hear, what you're thinking about incremental use cases as just the whole ecosystem continues to move more programmatic. Thanks.

Yes. Thanks, Dan. So on the first party data versus third-party data; I think you have to look at it in two buckets. I think that third-party data will be very prevalent in DV+. The -- if you look at the traditional DV+ publisher, whereas they do have a relationship with their end user, obviously the consumer that consumes the content. Oftentimes it's not a logged in registered relationship, and therefore, some of the fidelity the signals aren't nearly as strong as if you were to have a logged in relationship or even a credit card relationship. And that obviously is the CTV world. And so I think in CTV, it will be predominantly a first party world. And in the cases of third-party, I think you will see that's coming from the advertiser. So there's a lot of activity in clean rooms where you're matching Bank of America customers with a Paramount audience and trying to create audience segments specifically for BofA. And that's just an example, that's not a real life thing. And so I think when you start to look at it will play a very big role in creating signals in the DV+ world. But that's going to be a portfolio world. There's not going to be one killer data signal that's going to replace the deprecation of third-party cookies and will be part of that portfolio solution. But I think in CTV, you're going to see a very, very strong first party world and we'll be front and center in that. As it relates to commerce media, you give a good example of Uber and you can see it through across all kind of the travel industry per se. Anyone that owns their own screens, you're starting to see a real appetite to get into the commerce media business. And I think that we are expertly positioned as you look with the win a very contested RP with United. We have a video ad server. We have video monetization capabilities in the programmatic area. We are extremely experienced in this stuff and we can bring demand for them. And so I think that unlike some of the initial forays in kind of retail media networks, I think this commerce, travel, network business will be very much a sell-side focused business. And I think we're very well-positioned to capitalize on that.

Super helpful. Thanks, Michael.

Thanks.

Our next question is from Shweta Khajuria with Wolfe Research. Please proceed.

Hi guys, this is Brian Kraska for Shweta. Thank you for taking the questions and congrats on the quarter. Just wanted to see if you could provide an update on Trade Desk and OpenPath if there's any impact you're seeing there. I know you mentioned in the first quarter that you're gaining share in DV+ and accelerated growth. Just any update on that commentary and then we appreciate the color on the verticals in the quarter. Just wanted to see if you have any incremental color on those topics from the quarters today. Thank you.

Sure. Brian, I'll take the first and I'll let David talk about the verticals. Yes. No real update on OpenPath again, it's here to stay. We participate in the economics of OpenPath on the CTV side and we have a -- our share of spend with Trade Desk continues to grow in the DV+ side. So we feel as though you were well-positioned as a kind of leading omni-channel SSP. And David, I don't know if you have any more color on the verticals.

Yes. I mean, I think a couple things. I think first to, when you think about verticals for us, we're not disproportionately dependent on specific verticals. We have such a big footprint that overall spend, advertising spend in general, our spend tends to mimic that. And so when we talk about strength and weaker verticals, it's more on the margin. They're not always huge, huge drivers, so some context there. I think obviously political being 3.5% of our total revenue for the quarter was significant. And news and retail were the other strongest verticals. News, we were talking about this earlier. Hurricanes drove a little more news and probably politics as well. So that's, I think that's the insight that we have.

Our next question is from Zach Cummins with B. Riley. Please proceed.

Hi, good afternoon. Thanks for taking my questions. Is there any way you could provide any sort of incremental update around the Mediaocean partnership? I know it was something announced earlier this year, but any sort of incremental update on that would be appreciated.

Yes. Thanks, Zach. As we had said before, we think it's a great partnership filled with opportunity for both of us. But we caution that from a sales cycle standpoint and an activation standpoint, that it's going to take some time. And so I think we're pretty much right where we thought we would be. We have folks sampling the tool, we have folks actively coursing the spend through it, and it's more kind of a test mode and probably doesn't activate in terms of any substantial spend until you get into the kind of Q1, Q2 time frame of next year.

Understood. That's helpful. And realizing that DV+ is facing some near-term headwinds, what's the right way to think about a sustainable growth rate in that business as we go-forward from here? Just curious of how you're thinking about incrementally investing in opportunities on that side.

David, do you want to handle that?

Yes, sure. Excuse me. Yes, I think we feel good about DV+. There's no reason that we can't continue to be a share taker. And depending on the strength of the macro and overall spend, we certainly think that our DV+ growth rates could increase from the level that they're at now. We're investing in new formats and the curation that Michael mentioned earlier. We think we have the scale and the footprint and the technology that, that could be an additional driver for us both in the space and also for taking share.

Understood. Well, thanks for taking my questions and best of luck with the rest of the quarter.

Thanks, Zach.

Our next question is from Matt Swanson with RBC Capital Markets. Please proceed.

Hey guys, this is Simran on for Matt Swanson. Thanks for taking the question and congrats on the results. For my first one, it seems like you guys have partnered with almost everyone in CTV. So what would you say is the strategy to grow these partnerships from here?

Yes, great observation. There aren't too many folks that we aren't their primary programmatic partner. So I think what the strategy is that we ride the maturation of programmatic with our partners. So particularly for top premium partners, the plus services programmatic right now is almost exclusively sold by the publisher. And when we don't source the demand, our take rate is on the lower range of our take rates, but as it becomes more biddable, as more advertisers not just linear broadcast advertisers, but advertisers that have traditionally done social or OLV in open web as they start to test CTV and use it as a performance kind of middle funnel, bottom funnel advertising vehicle, that's where we start to pipe in the demand. We bring those advertisers in and that significantly improves the economics. So I think that the idea is to partner with everyone out there, which check that's occurred and now go with them on the journey and help bring them more monetization capabilities. And there you'll see the profile of our, the economics improve substantially.

Great. That's really helpful. And then one more for me. Do you see pressure on non-political spend? Do you see that towards the end of October and how are you thinking about it coming back in guidance? And then just speaking more broadly how has trends looked one month into Q4 and has that momentum from Q3 kind of continued?

Yes. So on the political side, there's always going to be dislocation, right? There's finite amount of ad times in CTV. And when political comes in, it does push out other folks. Also given the fact that it was like most of the elections in this country quite contentious, the mood of the programming, generally speaking with something that folks kind of shutdown on. So we did see a shutdown of like general ad brand spend, which we anticipate to return. Obviously, you're going into the holiday season. The strength of Q4 has always been there. So we anticipate that that will come back. As your -- the second part of the question, David, do you have any color as it relates to that?

Yes. As I mentioned, political is about 3.5% of our total revenue in Q3, and since the election is over, I guess, we can talk about the political in the past tense here. In absolute terms, we more than doubled Q3 spend concentrated in the pre-election time frame. And so if I pan back big picture, we had expectations of political coming in at around $20 million of contribution ex-TAC for the year and we came in right around those expectations.

Great. Thanks, guys. Congrats again.

Thanks.

Our next question is from Robert Coolbrith with Evercore ISI. Please proceed.

Hi, good afternoon. Thanks for taking our questions. I wanted to ask quickly if you have any update on the managed service business in the quarter and then also wanted to ask a follow-up on curation, maybe from more of an advertiser perspective. Our understanding is that with one of the major DSPs and their sort of preferred ID solution. There's probably a policy limitation on the use of third-party data segments. Are you seeing any advertisers trying to get ahead of that? And I'm wondering if there's anything you might be able to share in terms of share gain opportunity. If you're sort of lighting up these audience segments a different SSP solution isn't take rate opportunity or uplift opportunity, anything else you can tell us about that? Thank you very much.

Sure. I'll take the curation piece and then David; you can talk about managed service. Yes, so again, curation relative early stages really has gained popularity. There used to be a ton of services, a ton of, I'm not going to say networks, but some were network models, some were more of an exchange model. But as this SPO has continued with major buyers, basically, let's just say you were a company that had player technology, that a publisher would use your player -- your video player technology in return they would give you instead of paying for the technology, or they pay a little for the technology, but they give you some ad slacks in it. And so there's a case of technology and then there's a case of unique ad units. So some folks have unique spaces on the page. Increasingly, what buyers are saying and particularly the large agencies is like we would just assume conduct that through our exchange relationships that we have in place. So why don't you just go to Magnite and then Magnite will then activate through their marketplace, these special units, this special technology, et cetera. And therefore, we'll buy it that way. That's one quite common way of that occurring. The other is that the targeted solution of us finding audience segments and banding them together. I'm not aware of any restrictions vis-à-vis a DSP allowing that to occur. Generally speaking, it's not coming from the DSP per se, but it's coming from the advertiser behind the DSP and they're the ones that are driving that demand. So I feel hard pressed to believe that they would limit that, if they're finding success of that and trying to find the audience that they're looking for that's quite finite if we weren't able to stretch it across thousands of publishers.

Got it. Thank you.

And on the managed service front, in the third quarter, managed service was about 4% of our total contribution ex-TAC, and that's down about 20% year-over-year. As we look forward to the -- to the fourth quarter, we would also expect managed service to be down in those levels or perhaps even greater. So what's going on? As you recall, managed services, primarily our mid-market sales folks who are helping agencies and other buyers who don't have the in-house resources to run programmatically. But many of those agencies and buyers are growing in their ability to run programmatic, and so we're seeing some of those dollars shift into programmatic. The good news is most of the time that's continued, they continue to work with us. And so these relationships are very important from our managed service team. But you are seeing as we expected that, that business activity becoming programmatic, which is at a slower take rate. So in the short-term, it hurts us from a take rate perspective and a contribution ex-TAC perspective. But we believe these players over time are going to continue to grow and they're programmatic and that certainly offsets the negative component of that over time. And so we'll continue -- we believe that managed service will continue to become a smaller and smaller portion of our overall contribution ex-TAC as we continue to move forward.

Got it. Thank you very much.

Thanks.

Our next question is from Alec Brondolo with Wells Fargo Securities. Please proceed.

Thanks so much. Could you maybe double click on the 30% reduction in cost per ad request year-over-year? That seems like a pretty meaningful, it is a pretty meaningful improvement. I think you called out filtering and traffic shaping as the two drivers in the prepared remarks. Could you maybe just spend a little bit more time just elaborating on what specifically that means or what functionally is changing there? And then secondly, just any feedback on how growth spend trended in the quarter relative to contribution ex-TAC would be helpful. Thanks.

Yes. Great, Alec. I'll touch on the efficiency gains on the DV+ platform and then have David expand upon that and talk about gross spend. But essentially, our costs are when we receive an ad request that carries a little cost structure to it. But when you bring that ad request to the exchange and run an auction and you're sending that ad request out to multiple DSPs, obviously that's a different cost profile. So what you really want to do is figure out what are the most sought after ad requests, the ones that have been purchased in the past, whether it's a region, whether it's a device type, whether it's a inventory type, and for those that have never been bought, make sure that they don't make it to auction, where your costs escalate. And so the most of the filtering and the shaping occur when the trillions of ad requests hit the front door. And then you smartly allocate to DSP number two. Because DSP number two loves this type of inventory. We will send them that inventory first because we think we'll have a higher win rate. So all of this is going on in real time every day, recalibrated every 10 minutes based upon the trailing 10 minutes of activity. And that has really helped us become so much more efficient with our on-prem boxes. I don't know David, if you have more to add there or about the gross spend.

Yes. No, I think you covered it. From a gross and ad spend perspective; we'll look at DV+ and CTV separately. From a DV+ perspective, in general, our ad spend grows roughly at similar rates as our revenue growth, as our contribution ex-TAC. In CTV, as we've talked about, ad spend continues to grow at higher rates than our CTV contribution ex-TAC growth. But the difference between those growth rates has narrowed as our contribution ex-TAC has lifted up closer to the ad spend run rates. And so we're starting to see a little bit more stabilization. There's still a gap between those growth rates, but we're seeing more stabilization.

Thank you so much.

Thank you.

Our next question is from Max Michaelis with Lake Street Capital Markets. Please proceed.

Hey guys, just one from me. If we go back to that Disney renewal and you guys talked about sourcing your own demand and taking a higher take rate on that. Is that something you weren't previously doing or can you help me understand? Is that something new you're doing with Disney, I guess, anywhere, any comments you can provide? Thanks.

Yes, I think Max, good observation. I think the statement was more a statement about, it was twofold. Number one, it's just the growing evolution of premium publishers willingness to do programmatic and in biddable fashion, where -- when the first foray into programmatic, it was very much audiences defined by the publisher, pricing set by the publisher, and a one-to-one relationship in terms -- even though it was just executed programmatically, right? They knew the advertiser, they created the audience segment for them, it was just pushed through our pipes. So in that scenario, we're kind of providing value, but kind of the lower rung of value. And now that you're getting into more biddable opportunities, even if it's led by the publisher, when you're doing biddable, it carries it at a higher value perception, invite only auctions even could be 80 advertisers bidding in the same segment that's a different profile. So the maturation of the Disney relationship is not unlike any of the other relationships we have with the premium publishers, it's just a growing acceptance that biddable is a very viable way to increase yield and please buyers. The -- some of the properties that we expanded our relationship with in Disney particular are in markets that are either a) emerging, say for instance, audio or b) in international markets that may or may not have the same kind of sales footprint that you might have in North America. And they lend themselves more naturally to programmatic demand being sourced by Magnite.

Awesome. Thanks guys.

Thank you.

Our final question is from Tim Nollen with Macquarie. Please proceed. Tim, are you on the line?

We already had Tim and question on for Tim [indiscernible].

Operator, we can go ahead and move on to the next one if there's no others and we can wrap up.

Yes. This will conclude the question-and-answer session. I would like to turn the conference back over to Michael Barrett for any closing remarks.

Thank you, Sherry. Before wrapping up, I would like to thank the great Magnite team, around the world for their dedication and very hard work for delivering another strong quarter that exceeded expectations. Our team's success is publicly recognized by securing the highest score in the current offering category of the 10 vendors evaluated in the Forrester Wave Sell-Side Platforms Q4 2024 report. So congratulations to all the Magniters. We look forward to speaking with many of you at our upcoming investor events. We are participating in or at Wells Fargo Virtual Meetings tomorrow. Truist Virtual Conference on November 12, SIG meetings in Boston on November 13, Seaport Virtual Conference on November 18, Craig-Hallum Conference in New York on November 19, RBC Conference in New York on November 20, Macquarie Conference in Sydney on November 20, Wells Fargo Conference in Rancho Palos Verdes on December 3, Wolfe Conference in New York on December 4, Evercore meetings in San Francisco on December 9, and the Scotiabank Conference in San Francisco on December 10. Thank you all for joining and have a great evening.

Thank you. Thank you. This has now concluded. Thank you for attending today's presentation. You may now disconnect.